d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
b7574f89a4
freebsd-nq/sys/isa/syscons.c

5191 lines
138 KiB
C
Raw Normal View History

Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
/*-
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
* Copyright (c) 1992-1997 S<EFBFBD>ren Schmidt
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
* notice, this list of conditions and the following disclaimer
Changed header slightly.
1994-09-29 08:29:21 +00:00
* in this position and unchanged.
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
* 3. The name of the author may not be used to endorse or promote products
* derived from this software withough specific prior written permission
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
*
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
*
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
* $Id: syscons.c,v 1.257 1998/04/04 13:24:51 phk Exp $
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
*/
Change all #includes to follow the current Berkeley style. Some of these ``changes'' are actually not changes at all, but CVS sometimes has trouble telling the difference. This also includes support for second-directory compiles. This is not quite complete yet, as `config' doesn't yet do the right thing. You can still make it work trivially, however, by doing the following: rm /sys/compile mkdir /usr/obj/sys/compile ln -s M-. /sys/compile cd /sys/i386/conf config MYKERNEL cd ../../compile/MYKERNEL ln -s /sys @ rm machine ln -s @/i386/include machine make depend make
1994-08-13 03:50:34 +00:00
#include "sc.h"
(a) remove the pointer to each driver's tty structure array from cdevsw (b) add a function callback vector to tty drivers that will return a pointer to a valid tty structure based upon a dev_t (c) make syscons structures the same size whether or not APM is enabled so utilities don't crash if NAPM changes (and make the damn kernel compile!) (d) rewrite /dev/snp ioctl interface so that it is device driver and i386 independant
1995-02-25 20:09:44 +00:00
#include "apm.h"
Convert DDB to new-style option.
1996-01-04 21:13:23 +00:00
#include "opt_ddb.h"
Make all file-system (MFS, FFS, NFS, LFS, DEVFS) related option new-style. This introduce an xxxFS_BOOT for each of the rootable filesystems. (Presently not required, but encouraged to allow a smooth move of option *FS to opt_dontuse.h later.) LFS is temporarily disabled, and will be re-enabled tomorrow.
1998-01-24 02:54:56 +00:00
#include "opt_devfs.h"
Add option SC_KBD_PROBE_WORKS to syscons driver. If you define this, it means your keyboard is actually probable using the brain-dammaged probe routine in syscons, and if the keyboard is NOT found, then you don't want syscons to activate itself further. This makes life sane for those of us who use serial consoles most of the time and want "the right thing" to happen when we plug a keyboard in.
1996-09-06 23:35:54 +00:00
#include "opt_syscons.h"
Convert DDB to new-style option.
1996-01-04 21:13:23 +00:00
Change all #includes to follow the current Berkeley style. Some of these ``changes'' are actually not changes at all, but CVS sometimes has trouble telling the difference. This also includes support for second-directory compiles. This is not quite complete yet, as `config' doesn't yet do the right thing. You can still make it work trivially, however, by doing the following: rm /sys/compile mkdir /usr/obj/sys/compile ln -s M-. /sys/compile cd /sys/i386/conf config MYKERNEL cd ../../compile/MYKERNEL ln -s /sys @ rm machine ln -s @/i386/include machine make depend make
1994-08-13 03:50:34 +00:00
#if NSC > 0
#include <sys/param.h>
The big 4.4BSD Lite to FreeBSD 2.0.0 (Development) patch. Reviewed by: Rodney W. Grimes Submitted by: John Dyson and David Greenman
1994-05-25 09:21:21 +00:00
#include <sys/systm.h>
Change all #includes to follow the current Berkeley style. Some of these ``changes'' are actually not changes at all, but CVS sometimes has trouble telling the difference. This also includes support for second-directory compiles. This is not quite complete yet, as `config' doesn't yet do the right thing. You can still make it work trivially, however, by doing the following: rm /sys/compile mkdir /usr/obj/sys/compile ln -s M-. /sys/compile cd /sys/i386/conf config MYKERNEL cd ../../compile/MYKERNEL ln -s /sys @ rm machine ln -s @/i386/include machine make depend make
1994-08-13 03:50:34 +00:00
#include <sys/conf.h>
#include <sys/proc.h>
Replaced #includes of <sys/user.h> by less gross headers, usually <sys/vm.h>. Many device drivers need only the definition of vtophys() from vm. Added nearby #includes of <sys/conf.h> where appropriate.
1995-12-06 23:52:35 +00:00
#include <sys/signalvar.h>
Change all #includes to follow the current Berkeley style. Some of these ``changes'' are actually not changes at all, but CVS sometimes has trouble telling the difference. This also includes support for second-directory compiles. This is not quite complete yet, as `config' doesn't yet do the right thing. You can still make it work trivially, however, by doing the following: rm /sys/compile mkdir /usr/obj/sys/compile ln -s M-. /sys/compile cd /sys/i386/conf config MYKERNEL cd ../../compile/MYKERNEL ln -s /sys @ rm machine ln -s @/i386/include machine make depend make
1994-08-13 03:50:34 +00:00
#include <sys/tty.h>
#include <sys/kernel.h>
#include <sys/malloc.h>
Pass 3 of the great devsw changes most devsw referenced functions are now static, as they are in the same file as their devsw structure. I've also added DEVFS support for nearly every device in the system, however many of the devices have 'incorrect' names under DEVFS because I couldn't quickly work out the correct naming conventions. (but devfs won't be coming on line for a month or so anyhow so that doesn't matter) If you "OWN" a device which would normally have an entry in /dev then search for the devfs_add_devsw() entries and munge to make them right.. check out similar devices to see what I might have done in them in you can't see what's going on.. for a laugh compare conf.c conf.h defore and after... :) I have not doen DEVFS entries for any DISKSLICE devices yet as that will be a much more complicated job.. (pass 5 :) pass 4 will be to make the devsw tables of type (cdevsw * ) rather than (cdevsw) seems to work here.. complaints to the usual places.. :)
1995-12-08 11:19:42 +00:00
#ifdef DEVFS
#include <sys/devfsext.h>
#endif
Finished device configuration database work for all ISA devices (except `ze') and all SCSI devices (except that it's not done quite the way I want). New information added includes: - A text description of the device - A ``state''---unknown, unconfigured, idle, or busy - A generic parent device (with support in the m.i. code) - An interrupt mask type field (which will hopefully go away) so that . ``doconfig'' can be written This requires a new version of the `lsdev' program as well (next commit).
1994-10-23 21:28:03 +00:00
Fix initialization of video_mode_ptr. It only worked if the segments in the far pointers are multiples of 4K (as is normal when the video BIOS is at seg 0xc000). Disallow mode switching if the pointer is bad. Use a new pa_to_va() macro for all BIOS and video addresses in syscons.
1994-10-26 21:51:22 +00:00
#include <machine/clock.h>
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
#include <machine/cons.h>
Oops, changed order of include lines... Reviewed by: Submitted by:
1994-08-17 19:32:23 +00:00
#include <machine/console.h>
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
#include <machine/mouse.h>
Added or restored #include of <machine/md_var.h>. Some declarations moved from <machine/cpufunc.h> to better places.
1996-07-01 20:29:10 +00:00
#include <machine/md_var.h>
Oops, changed order of include lines... Reviewed by: Submitted by:
1994-08-17 19:32:23 +00:00
#include <machine/psl.h>
#include <machine/frame.h>
#include <machine/pc/display.h>
(a) remove the pointer to each driver's tty structure array from cdevsw (b) add a function callback vector to tty drivers that will return a pointer to a valid tty structure based upon a dev_t (c) make syscons structures the same size whether or not APM is enabled so utilities don't crash if NAPM changes (and make the damn kernel compile!) (d) rewrite /dev/snp ioctl interface so that it is device driver and i386 independant
1995-02-25 20:09:44 +00:00
#include <machine/apm_bios.h>
Theodore Ts'po's random number gernerator for Linux, ported by me. This code will only be included in your kernel if you have 'options DEVRANDOM', but that will fall away in a couple of days. Obtained from: Theodore Ts'o, Linux
1995-10-28 16:58:05 +00:00
#include <machine/random.h>
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
#include <machine/bootinfo.h>
Finished device configuration database work for all ISA devices (except `ze') and all SCSI devices (except that it's not done quite the way I want). New information added includes: - A text description of the device - A ``state''---unknown, unconfigured, idle, or busy - A generic parent device (with support in the m.i. code) - An interrupt mask type field (which will hopefully go away) so that . ``doconfig'' can be written This requires a new version of the `lsdev' program as well (next commit).
1994-10-23 21:28:03 +00:00
Replaced #includes of <sys/user.h> by less gross headers, usually <sys/vm.h>. Many device drivers need only the definition of vtophys() from vm. Added nearby #includes of <sys/conf.h> where appropriate.
1995-12-06 23:52:35 +00:00
#include <vm/vm.h>
Untangled the vm.h include file spaghetti.
1995-12-07 12:48:31 +00:00
#include <vm/vm_param.h>
#include <vm/pmap.h>
Replaced #includes of <sys/user.h> by less gross headers, usually <sys/vm.h>. Many device drivers need only the definition of vtophys() from vm. Added nearby #includes of <sys/conf.h> where appropriate.
1995-12-06 23:52:35 +00:00
Change all #includes to follow the current Berkeley style. Some of these ``changes'' are actually not changes at all, but CVS sometimes has trouble telling the difference. This also includes support for second-directory compiles. This is not quite complete yet, as `config' doesn't yet do the right thing. You can still make it work trivially, however, by doing the following: rm /sys/compile mkdir /usr/obj/sys/compile ln -s M-. /sys/compile cd /sys/i386/conf config MYKERNEL cd ../../compile/MYKERNEL ln -s /sys @ rm machine ln -s @/i386/include machine make depend make
1994-08-13 03:50:34 +00:00
#include <i386/isa/isa.h>
#include <i386/isa/isa_device.h>
#include <i386/isa/timerreg.h>
Updated to latest 1.1.5.1+ version Reviewed by: Submitted by:
1994-08-17 08:51:59 +00:00
#include <i386/isa/kbdtables.h>
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
#include <i386/isa/kbdio.h>
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
#include <i386/isa/syscons.h>
This is syscons version 1.1
1993-10-28 05:27:36 +00:00
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
#if !defined(MAXCONS)
Third round in syscons update. Display update method changed, now allways write in memory buffer, then periodically update physical display. Speed improvements (now > 5 times faster than the old syscons). History now circular buffer, with changeable size. History scroll by up/down line, up/down page, home and end. Backtab proberly implemented. Now space for 96 function keys, 63 allocated standard, default now SCO/SYSV compat again as in the old days. New keyboard definition files ~share/syscons/keymaps/* Misc fixes for old "hacks" that broke SCO/SYSV compat. More that I forgot before writing this...
1995-01-28 22:18:05 +00:00
#define MAXCONS 16
Miscellaneous changes from Bill Paul: - /sys/i386/isa/if_ed.c doesn't quite know how to deal with SMC EtherEZ ethernet cards. The EtherEZ looks just like the Elite Ultra, except it has only 8K of shared memory. The only way to have it properly detected is to zero and test a few bytes of memory just about the first 8K region. If it clears properly, it's an Elite Ultra, otherwise it's an EtherEZ. I've also got an EtherEZ patch for netboot (Makefile, ether.c and ether.h). - /sys/i386/isa/syscons.c wraps at the next to the last column rather than the last column, like it should. You don't really notice this unless you use certain programs that write all the way out to, say, the 80th column, like VMSmail. Along with a one-line fix for this are some changes to implement a non-blinking cursor. Put 'options "NOBLINK_CURSOR"' in your config file and give it a try. :) Submitted by: wpaul
1994-12-31 17:09:58 +00:00
#endif
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
#if !defined(SC_MAX_HISTORY_SIZE)
#define SC_MAX_HISTORY_SIZE (1000 * MAXCONS)
#endif
#if !defined(SC_HISTORY_SIZE)
#define SC_HISTORY_SIZE (ROW * 4)
#endif
#if (SC_HISTORY_SIZE * MAXCONS) > SC_MAX_HISTORY_SIZE
#undef SC_MAX_HISTORY_SIZE
#define SC_MAX_HISTORY_SIZE (SC_HISTORY_SIZE * MAXCONS)
#endif
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
#if !defined(SC_MOUSE_CHAR)
#define SC_MOUSE_CHAR (0xd0)
#endif
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
#define COLD 0
#define WARM 1
Pass 3 of the great devsw changes most devsw referenced functions are now static, as they are in the same file as their devsw structure. I've also added DEVFS support for nearly every device in the system, however many of the devices have 'incorrect' names under DEVFS because I couldn't quickly work out the correct naming conventions. (but devfs won't be coming on line for a month or so anyhow so that doesn't matter) If you "OWN" a device which would normally have an entry in /dev then search for the devfs_add_devsw() entries and munge to make them right.. check out similar devices to see what I might have done in them in you can't see what's going on.. for a laugh compare conf.c conf.h defore and after... :) I have not doen DEVFS entries for any DISKSLICE devices yet as that will be a much more complicated job.. (pass 5 :) pass 4 will be to make the devsw tables of type (cdevsw * ) rather than (cdevsw) seems to work here.. complaints to the usual places.. :)
1995-12-08 11:19:42 +00:00
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
#define MODE_MAP_SIZE (M_VGA_CG320 + 1)
#define MODE_PARAM_SIZE 64
- Add binary compatibility support for obsolete console/mouse ioctls so that existing programs which were compiled before the introduction of the new mouse code and use these ioctls will run unmodified. Suggested by msmith.
1998-01-20 03:37:27 +00:00
/* for backward compatibility */
#define OLD_CONS_MOUSECTL _IOWR('c', 10, old_mouse_info_t)
typedef struct old_mouse_data {
int x;
int y;
int buttons;
} old_mouse_data_t;
typedef struct old_mouse_info {
int operation;
union {
struct old_mouse_data data;
struct mouse_mode mode;
} u;
} old_mouse_info_t;
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
/* XXX use sc_bcopy where video memory is concerned */
extern void generic_bcopy(const void *, void *, size_t);
Updated syscons to 0.2b - please test this! It won't be in the ALPHA release, but it's still worth testing.
1993-07-30 02:20:26 +00:00
static default_attr user_default = {
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(FG_LIGHTGREY | BG_BLACK) << 8,
(FG_BLACK | BG_LIGHTGREY) << 8
Updated syscons to 0.2b - please test this! It won't be in the ALPHA release, but it's still worth testing.
1993-07-30 02:20:26 +00:00
};
static default_attr kernel_default = {
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(FG_WHITE | BG_BLACK) << 8,
(FG_BLACK | BG_LIGHTGREY) << 8
Updated syscons to 0.2b - please test this! It won't be in the ALPHA release, but it's still worth testing.
1993-07-30 02:20:26 +00:00
};
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
static scr_stat main_console;
static scr_stat *console[MAXCONS];
Stomp another compiler warning: sc_devfs_token[] should only be declared if DEVFS is #defined.
1996-06-17 17:21:35 +00:00
#ifdef DEVFS
Pass 3 of the great devsw changes most devsw referenced functions are now static, as they are in the same file as their devsw structure. I've also added DEVFS support for nearly every device in the system, however many of the devices have 'incorrect' names under DEVFS because I couldn't quickly work out the correct naming conventions. (but devfs won't be coming on line for a month or so anyhow so that doesn't matter) If you "OWN" a device which would normally have an entry in /dev then search for the devfs_add_devsw() entries and munge to make them right.. check out similar devices to see what I might have done in them in you can't see what's going on.. for a laugh compare conf.c conf.h defore and after... :) I have not doen DEVFS entries for any DISKSLICE devices yet as that will be a much more complicated job.. (pass 5 :) pass 4 will be to make the devsw tables of type (cdevsw * ) rather than (cdevsw) seems to work here.. complaints to the usual places.. :)
1995-12-08 11:19:42 +00:00
static void *sc_devfs_token[MAXCONS];
add devfs entries for a few more syscons devices (e.g. sysmouse)
1998-01-12 03:28:36 +00:00
static void *sc_mouse_devfs_token;
static void *sc_console_devfs_token;
Stomp another compiler warning: sc_devfs_token[] should only be declared if DEVFS is #defined.
1996-06-17 17:21:35 +00:00
#endif
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scr_stat *cur_console;
static scr_stat *new_scp, *old_scp;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static term_stat kernel_console;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
static default_attr *current_default;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
static int flags = 0;
Fix UserConfig w/syscons. The 'getchar' function in syscons (sccngetc) is used by UserConfig to get keyboard input from the user. When it was modified to use the shared keyboard port routines it used the port passed in during the probe routine. Since the probe routine was not yet called, the port was set to 0, which is obviously not going to work. Pre-initialize sc_port to IO_KBD which is really a kludge, but it's how the previous driver did it's job. Found by: remote GDB
1996-11-19 17:08:10 +00:00
static int sc_port = IO_KBD;
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
static KBDC sc_kbdc = NULL;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
static char init_done = COLD;
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
static u_short sc_buffer[ROW*COL];
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
static char font_loading_in_progress = FALSE;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
static char switch_in_progress = FALSE;
static char write_in_progress = FALSE;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
static char blink_in_progress = FALSE;
static int blinkrate = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
u_int crtc_addr = MONO_BASE;
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
char crtc_type = KD_MONO;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
char crtc_vga = FALSE;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
static u_char shfts = 0, ctls = 0, alts = 0, agrs = 0, metas = 0;
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
static u_char accents = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
static u_char nlkcnt = 0, clkcnt = 0, slkcnt = 0, alkcnt = 0;
static const u_int n_fkey_tab = sizeof(fkey_tab) / sizeof(*fkey_tab);
static int delayed_next_scr = FALSE;
static long scrn_blank_time = 0; /* screen saver timeout value */
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
int scrn_blanked = 0; /* screen saver active flag */
Time changes mark 2: * Figure out UTC relative to boottime. Four new functions provide time relative to boottime. * move "runtime" into struct proc. This helps fix the calcru() problem in SMP. * kill mono_time. * add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!) * nanosleep, select & poll takes long sleeps one day at a time Reviewed by: bde Tested by: ache and others
1998-04-04 13:26:20 +00:00
static struct timeval scrn_time_stamp;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
u_char scr_map[256];
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
u_char scr_rmap[256];
char *video_mode_ptr = NULL;
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
static int vesa_mode;
Add option for compiling in a 8x16 font.
1997-07-25 11:53:30 +00:00
int fonts_loaded = 0
#ifdef STD8X16FONT
| FONT_16
#endif
;
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
char font_8[256*8];
char font_14[256*14];
Fix a brino in my last commit. Noticed by: smp
1997-07-26 07:58:29 +00:00
#ifdef STD8X16FONT
extern
#endif
unsigned char font_16[256*16];
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
char palette[256*3];
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
static char *mode_map[MODE_MAP_SIZE];
static char vgaregs[MODE_PARAM_SIZE];
static char vgaregs2[MODE_PARAM_SIZE];
static int rows_offset = 1;
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
static char *cut_buffer;
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
static int mouse_level = 0; /* sysmouse protocol level */
static mousestatus_t mouse_status = { 0, 0, 0, 0, 0, 0 };
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
static u_short mouse_and_mask[16] = {
0xc000, 0xe000, 0xf000, 0xf800,
0xfc00, 0xfe00, 0xff00, 0xff80,
0xfe00, 0x1e00, 0x1f00, 0x0f00,
0x0f00, 0x0000, 0x0000, 0x0000
};
static u_short mouse_or_mask[16] = {
0x0000, 0x4000, 0x6000, 0x7000,
0x7800, 0x7c00, 0x7e00, 0x6800,
0x0c00, 0x0c00, 0x0600, 0x0600,
0x0000, 0x0000, 0x0000, 0x0000
};
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
static int extra_history_size =
SC_MAX_HISTORY_SIZE - SC_HISTORY_SIZE * MAXCONS;
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
static void none_saver(int blank) { }
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
static void (*current_saver)(int blank) = none_saver;
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
int (*sc_user_ioctl)(dev_t dev, int cmd, caddr_t data,
int flag, struct proc *p) = NULL;
Make the screen savers runtime switchable. Everybody wants a different default saver, and the size increase in the kernel is minimal ( < 2.5K ).
1994-02-04 10:36:15 +00:00
From: sos@login.dkuug.dk (S|ren Schmidt) Subject: syscons-1.3 Date: Sat, 29 Jan 94 23:33:50 MET But here is the (hopefully) final syscons-1.3.... .... I've changed sgetc so it works as the pccons parallel (it now uses a scgetc internally). [ There were a couple changes that Bruce Evans sent me that were applied to this version along with some changes that S'ren didn't incorporate into the final version. There will be only minor changes if anything from this version to his final release. ]
1994-02-01 06:22:29 +00:00
/* OS specific stuff */
Third round in syscons update. Display update method changed, now allways write in memory buffer, then periodically update physical display. Speed improvements (now > 5 times faster than the old syscons). History now circular buffer, with changeable size. History scroll by up/down line, up/down page, home and end. Backtab proberly implemented. Now space for 96 function keys, 63 allocated standard, default now SCO/SYSV compat again as in the old days. New keyboard definition files ~share/syscons/keymaps/* Misc fixes for old "hacks" that broke SCO/SYSV compat. More that I forgot before writing this...
1995-01-28 22:18:05 +00:00
#ifdef not_yet_done
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
#define VIRTUAL_TTY(x) (sccons[x] = ttymalloc(sccons[x]))
struct CONSOLE_TTY (sccons[MAXCONS] = ttymalloc(sccons[MAXCONS]))
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
struct MOUSE_TTY (sccons[MAXCONS+1] = ttymalloc(sccons[MAXCONS+1]))
struct tty *sccons[MAXCONS+2];
The big 4.4BSD Lite to FreeBSD 2.0.0 (Development) patch. Reviewed by: Rodney W. Grimes Submitted by: John Dyson and David Greenman
1994-05-25 09:21:21 +00:00
#else
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
#define VIRTUAL_TTY(x) &sccons[x]
#define CONSOLE_TTY &sccons[MAXCONS]
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
#define MOUSE_TTY &sccons[MAXCONS+1]
static struct tty sccons[MAXCONS+2];
The big 4.4BSD Lite to FreeBSD 2.0.0 (Development) patch. Reviewed by: Rodney W. Grimes Submitted by: John Dyson and David Greenman
1994-05-25 09:21:21 +00:00
#endif
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
#define SC_MOUSE 128
#define SC_CONSOLE 255
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
#define MONO_BUF pa_to_va(0xB0000)
#define CGA_BUF pa_to_va(0xB8000)
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
u_short *Crtat;
The poor nsccons variable was gone agian this time hidden by an ifdef , wonder who broke it this time :) Submitted by: ache
1996-09-09 19:02:26 +00:00
static const int nsccons = MAXCONS+2;
From: sos@login.dkuug.dk (S|ren Schmidt) Subject: syscons-1.3 Date: Sat, 29 Jan 94 23:33:50 MET But here is the (hopefully) final syscons-1.3.... .... I've changed sgetc so it works as the pccons parallel (it now uses a scgetc internally). [ There were a couple changes that Bruce Evans sent me that were applied to this version along with some changes that S'ren didn't incorporate into the final version. There will be only minor changes if anything from this version to his final release. ]
1994-02-01 06:22:29 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
#define WRAPHIST(scp, pointer, offset)\
((scp->history) + ((((pointer) - (scp->history)) + (scp->history_size)\
+ (offset)) % (scp->history_size)))
make sure that the user supplied signals in struct vt_mode are actually valid signals, else return EINVAL for ioctl VT_SETMODE. this fixes a problem that anybody with vty access can panic the system. 2.2-Candidate (and 2.1.0 I believe) Reviewed-by: sos
1997-03-01 23:53:46 +00:00
#define ISSIGVALID(sig) ((sig) > 0 && (sig) < NSIG)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
/* this should really be in `rtc.h' */
#define RTC_EQUIPMENT 0x14
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
/* prototypes */
static int scattach(struct isa_device *dev);
static int scparam(struct tty *tp, struct termios *t);
static int scprobe(struct isa_device *dev);
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
static int scvidprobe(int unit, int flags);
static int sckbdprobe(int unit, int flags);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
static void scstart(struct tty *tp);
Hack workaround XFree86 switching failure when used with /dev/sysmouse and xdm, possibly in general. What was happening was that the server was doing a tcsetattr(.. TCSADRAIN) on the mouse fd after a write. Since /dev/sysmouse had a null t_oproc, the drain failed with EIO. Somehow this spammed XFree86 (!@&^#%*& binary release!!), and the driver was left in a bogus state (ie: switch_in_progress permanently TRUE). The simplest way out was to implement a dummy scmousestart() routine to accept any characters from the tty system and toss them into the void. It would probably be more correct to intercept scwrite()'s to the mouse device, but that's executed for every single write to the screen. Supplying a start routine to eat the characters is only executed for the mouse port during startup/shutdown, so it should be faster.
1996-09-10 19:14:49 +00:00
static void scmousestart(struct tty *tp);
Moved declarations of static functions to the correct file. This fixes hundreds of warnings from -Wunused in lkm/syscons/*.
1996-06-23 17:12:05 +00:00
static void scinit(void);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
static void map_mode_table(char *map[], char *table, int max);
static u_char map_mode_num(u_char mode);
static char *get_mode_param(scr_stat *scp, u_char mode);
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
static u_int scgetc(u_int flags);
#define SCGETC_CN 1
#define SCGETC_NONBLOCK 2
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
static void sccnupdate(scr_stat *scp);
Moved declarations of static functions to the correct file. This fixes hundreds of warnings from -Wunused in lkm/syscons/*.
1996-06-23 17:12:05 +00:00
static scr_stat *get_scr_stat(dev_t dev);
static scr_stat *alloc_scp(void);
static void init_scp(scr_stat *scp);
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
static void sc_bcopy(u_short *p, int from, int to, int mark);
Moved declarations of static functions to the correct file. This fixes hundreds of warnings from -Wunused in lkm/syscons/*.
1996-06-23 17:12:05 +00:00
static int get_scr_num(void);
Fixed the type of timeout functions and removed casts that hid the type mismatches. Not taking an arg in scrn_timer() broke `cc -mrtd'.
1997-04-20 16:05:33 +00:00
static timeout_t scrn_timer;
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
static void scrn_update(scr_stat *scp, int show_cursor);
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
static void stop_scrn_saver(void (*saver)(int));
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
static int wait_scrn_saver_stop(void);
Moved declarations of static functions to the correct file. This fixes hundreds of warnings from -Wunused in lkm/syscons/*.
1996-06-23 17:12:05 +00:00
static void clear_screen(scr_stat *scp);
static int switch_scr(scr_stat *scp, u_int next_scr);
static void exchange_scr(void);
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
static void move_crsr(scr_stat *scp, int x, int y);
Moved declarations of static functions to the correct file. This fixes hundreds of warnings from -Wunused in lkm/syscons/*.
1996-06-23 17:12:05 +00:00
static void scan_esc(scr_stat *scp, u_char c);
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
static void draw_cursor_image(scr_stat *scp);
static void remove_cursor_image(scr_stat *scp);
Moved declarations of static functions to the correct file. This fixes hundreds of warnings from -Wunused in lkm/syscons/*.
1996-06-23 17:12:05 +00:00
static void ansi_put(scr_stat *scp, u_char *buf, int len);
static u_char *get_fstr(u_int c, u_int *len);
static void history_to_screen(scr_stat *scp);
static int history_up_line(scr_stat *scp);
static int history_down_line(scr_stat *scp);
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
static int mask2attr(struct term_stat *term);
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
static void set_keyboard(int command, int data);
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
static void update_leds(int which);
Moved declarations of static functions to the correct file. This fixes hundreds of warnings from -Wunused in lkm/syscons/*.
1996-06-23 17:12:05 +00:00
static void set_vgaregs(char *modetable);
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
static void read_vgaregs(char *buf);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
#define COMP_IDENTICAL 0
#define COMP_SIMILAR 1
#define COMP_DIFFERENT 2
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
static int comp_vgaregs(u_char *buf1, u_char *buf2);
static void dump_vgaregs(u_char *buf);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
#define PARAM_BUFSIZE 6
static void set_font_mode(u_char *buf);
static void set_normal_mode(u_char *buf);
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
static void set_destructive_cursor(scr_stat *scp);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
static void set_mouse_pos(scr_stat *scp);
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
static int skip_spc_right(scr_stat *scp, u_short *p);
static int skip_spc_left(scr_stat *scp, u_short *p);
static void mouse_cut(scr_stat *scp);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
static void mouse_cut_start(scr_stat *scp);
static void mouse_cut_end(scr_stat *scp);
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
static void mouse_cut_word(scr_stat *scp);
static void mouse_cut_line(scr_stat *scp);
static void mouse_cut_extend(scr_stat *scp);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
static void mouse_paste(scr_stat *scp);
Moved declarations of static functions to the correct file. This fixes hundreds of warnings from -Wunused in lkm/syscons/*.
1996-06-23 17:12:05 +00:00
static void draw_mouse_image(scr_stat *scp);
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
static void remove_mouse_image(scr_stat *scp);
static void draw_cutmarking(scr_stat *scp);
static void remove_cutmarking(scr_stat *scp);
Moved declarations of static functions to the correct file. This fixes hundreds of warnings from -Wunused in lkm/syscons/*.
1996-06-23 17:12:05 +00:00
static void save_palette(void);
static void do_bell(scr_stat *scp, int pitch, int duration);
Fixed the type of timeout functions and removed casts that hid the type mismatches. Not taking an arg in scrn_timer() broke `cc -mrtd'.
1997-04-20 16:05:33 +00:00
static timeout_t blink_screen;
#ifdef another instance of toggle_splash_screen() which Soren missed.
1996-10-01 07:38:14 +00:00
#ifdef SC_SPLASH_SCREEN
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
static void toggle_splash_screen(scr_stat *scp);
#ifdef another instance of toggle_splash_screen() which Soren missed.
1996-10-01 07:38:14 +00:00
#endif
Moved declarations of static functions to the correct file. This fixes hundreds of warnings from -Wunused in lkm/syscons/*.
1996-06-23 17:12:05 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
struct isa_driver scdriver = {
scprobe, scattach, "sc", 1
Updated syscons to 0.2b - please test this! It won't be in the ALPHA release, but it's still worth testing.
1993-07-30 02:20:26 +00:00
};
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
static d_open_t scopen;
static d_close_t scclose;
Removed new alias d_size_t for d_psize_t. Removed old aliases d_rdwr_t and d_ttycv_t for d_read_t/d_write_t and d_devtotty_t. Sorted declarations of switch functions into switch order. Removed duplicated comments and declarations of nonexistent switch functions.
1995-12-10 15:55:34 +00:00
static d_read_t scread;
static d_write_t scwrite;
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
static d_ioctl_t scioctl;
Removed new alias d_size_t for d_psize_t. Removed old aliases d_rdwr_t and d_ttycv_t for d_read_t/d_write_t and d_devtotty_t. Sorted declarations of switch functions into switch order. Removed duplicated comments and declarations of nonexistent switch functions.
1995-12-10 15:55:34 +00:00
static d_devtotty_t scdevtotty;
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
static d_mmap_t scmmap;
Removed my devsw access functions [un]register_cdev() and getmajorbyname() which were a better (sigh) temporary interface to the going-away devswitches. Note that SYSINIT()s to initialize the devswitches would be fatal in syscons.c and pcvt_drv.c (and are bogus elsewhere) because they get called independently of whether the device is attached; thus devices that share a major clobber each other's devswitch entries until the last one wins. conf.c: Removed stale #includes and comments.
1995-12-14 22:03:12 +00:00
#define CDEV_MAJOR 12
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
static struct cdevsw scdevsw = {
scopen, scclose, scread, scwrite,
scioctl, nullstop, noreset, scdevtotty,
Update select -> poll in drivers.
1997-09-14 03:19:42 +00:00
ttpoll, scmmap, nostrategy, "sc", NULL, -1 };
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
/*
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
* These functions need to be before calls to them so they can be inlined.
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
*/
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
static void
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
draw_cursor_image(scr_stat *scp)
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
{
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
u_short cursor_image, *ptr = Crtat + (scp->cursor_pos - scp->scr_buf);
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
u_short prev_image;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
if (vesa_mode) {
sc_bcopy(scp->scr_buf, scp->cursor_pos - scp->scr_buf,
scp->cursor_pos - scp->scr_buf, 1);
return;
}
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
/* do we have a destructive cursor ? */
if (flags & CHAR_CURSOR) {
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
prev_image = scp->cursor_saveunder;
cursor_image = *ptr & 0x00ff;
if (cursor_image == DEAD_CHAR)
cursor_image = prev_image & 0x00ff;
cursor_image |= *(scp->cursor_pos) & 0xff00;
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
scp->cursor_saveunder = cursor_image;
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
/* update the cursor bitmap if the char under the cursor has changed */
if (prev_image != cursor_image)
set_destructive_cursor(scp);
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
/* modify cursor_image */
if (!(flags & BLINK_CURSOR)||((flags & BLINK_CURSOR)&&(blinkrate & 4))){
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
/*
* When the mouse pointer is at the same position as the cursor,
* the cursor bitmap needs to be updated even if the char under
* the cursor hasn't changed, because the mouse pionter may
* have moved by a few dots within the cursor cel.
*/
if ((prev_image == cursor_image)
&& (cursor_image != *(scp->cursor_pos)))
set_destructive_cursor(scp);
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
cursor_image &= 0xff00;
cursor_image |= DEAD_CHAR;
}
Fix a white-space nit which I happened to notice.
1998-02-12 20:52:24 +00:00
} else {
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
cursor_image = (*(ptr) & 0x00ff) | *(scp->cursor_pos) & 0xff00;
scp->cursor_saveunder = cursor_image;
if (!(flags & BLINK_CURSOR)||((flags & BLINK_CURSOR)&&(blinkrate & 4))){
if ((cursor_image & 0x7000) == 0x7000) {
cursor_image &= 0x8fff;
if(!(cursor_image & 0x0700))
cursor_image |= 0x0700;
} else {
cursor_image |= 0x7000;
if ((cursor_image & 0x0700) == 0x0700)
cursor_image &= 0xf0ff;
}
}
}
*ptr = cursor_image;
}
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
static void
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
remove_cursor_image(scr_stat *scp)
{
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
if (vesa_mode)
Hmm, it is generally an advantage to commit the most recent version of ones changes: A faster character painter procedure and fix on cursor bug.
1998-02-13 11:31:34 +00:00
sc_bcopy(scp->scr_buf, scp->cursor_oldpos - scp->scr_buf,
scp->cursor_oldpos - scp->scr_buf, 0);
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
else
*(Crtat + (scp->cursor_oldpos - scp->scr_buf)) = scp->cursor_saveunder;
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
}
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
static void
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
move_crsr(scr_stat *scp, int x, int y)
{
if (x < 0)
x = 0;
if (y < 0)
y = 0;
if (x >= scp->xsize)
x = scp->xsize-1;
if (y >= scp->ysize)
y = scp->ysize-1;
scp->xpos = x;
scp->ypos = y;
scp->cursor_pos = scp->scr_buf + scp->ypos * scp->xsize + scp->xpos;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
}
Staticize and cleanup.
1995-12-10 13:40:44 +00:00
static int
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
scprobe(struct isa_device *dev)
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
{
if (!scvidprobe(dev->id_unit, dev->id_flags)) {
if (bootverbose)
printf("sc%d: no video adapter is found.\n", dev->id_unit);
return (0);
}
sc_port = dev->id_iobase;
if (sckbdprobe(dev->id_unit, dev->id_flags))
return (IO_KBDSIZE);
else
return ((dev->id_flags & DETECT_KBD) ? 0 : IO_KBDSIZE);
}
/* probe video adapters, return TRUE if found */
static int
scvidprobe(int unit, int flags)
{
/*
* XXX don't try to `printf' anything here, the console may not have
* been configured yet.
*/
u_short volatile *cp;
u_short was;
u_long pa;
u_long segoff;
/* do this test only once */
if (init_done != COLD)
return (Crtat != 0);
/*
* Finish defaulting crtc variables for a mono screen. Crtat is a
* bogus common variable so that it can be shared with pcvt, so it
* can't be statically initialized. XXX.
*/
Crtat = (u_short *)MONO_BUF;
crtc_type = KD_MONO;
/* If CGA memory seems to work, switch to color. */
cp = (u_short *)CGA_BUF;
was = *cp;
*cp = (u_short) 0xA55A;
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
if (bootinfo.bi_vesa == 0x102) {
vesa_mode = bootinfo.bi_vesa;
Crtat = (u_short *)pa_to_va(0xA0000);
crtc_type = KD_PIXEL;
bzero(Crtat, 800*600/8);
} else if (*cp == 0xA55A) {
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
Crtat = (u_short *)CGA_BUF;
crtc_addr = COLOR_BASE;
crtc_type = KD_CGA;
} else {
cp = Crtat;
was = *cp;
*cp = (u_short) 0xA55A;
if (*cp != 0xA55A) {
/* no screen at all, bail out */
Crtat = 0;
return FALSE;
}
}
*cp = was;
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
if (crtc_type != KD_PIXEL) {
/*
* Check rtc and BIOS date area.
* XXX: don't use BIOSDATA_EQUIPMENT, it is not a dead copy
* of RTC_EQUIPMENT. The bit 4 and 5 of the ETC_EQUIPMENT are
* zeros for EGA and VGA. However, the EGA/VGA BIOS will set
* these bits in BIOSDATA_EQUIPMENT according to the monitor
* type detected.
*/
switch ((rtcin(RTC_EQUIPMENT) >> 4) & 3) { /* bit 4 and 5 */
case 0: /* EGA/VGA, or nothing */
crtc_type = KD_EGA;
/* the color adapter may be in the 40x25 mode... XXX */
break;
case 1: /* CGA 40x25 */
/* switch to the 80x25 mode? XXX */
/* FALL THROUGH */
case 2: /* CGA 80x25 */
/* `crtc_type' has already been set... */
/* crtc_type = KD_CGA; */
break;
case 3: /* MDA */
/* `crtc_type' has already been set... */
/* crtc_type = KD_MONO; */
break;
}
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
/* is this a VGA or higher ? */
outb(crtc_addr, 7);
if (inb(crtc_addr) == 7) {
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
crtc_type = KD_VGA;
crtc_vga = TRUE;
read_vgaregs(vgaregs);
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
/* Get the BIOS video mode pointer */
segoff = *(u_long *)pa_to_va(0x4a8);
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
pa = (((segoff & 0xffff0000) >> 12) + (segoff & 0xffff));
Fix a couple of lines that dropped out in my merge yesterday. Complained about by: ache
1998-02-13 17:54:53 +00:00
if (ISMAPPED(pa, sizeof(u_long))) {
segoff = *(u_long *)pa_to_va(pa);
pa = (((segoff & 0xffff0000) >> 12) + (segoff & 0xffff));
if (ISMAPPED(pa, MODE_PARAM_SIZE))
video_mode_ptr = (char *)pa_to_va(pa);
}
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
}
}
return TRUE;
}
/* probe the keyboard, return TRUE if found */
static int
sckbdprobe(int unit, int flags)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
int codeset;
int c = -1;
int m;
Updated to latest 1.1.5.1+ version Reviewed by: Submitted by:
1994-08-17 08:51:59 +00:00
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
sc_kbdc = kbdc_open(sc_port);
if (!kbdc_lock(sc_kbdc, TRUE)) {
/* driver error? */
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
printf("sc%d: unable to lock the controller.\n", unit);
return ((flags & DETECT_KBD) ? FALSE : TRUE);
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
}
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
/* discard anything left after UserConfig */
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
empty_both_buffers(sc_kbdc, 10);
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
/* save the current keyboard controller command byte */
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
m = kbdc_get_device_mask(sc_kbdc) & ~KBD_KBD_CONTROL_BITS;
c = get_controller_command_byte(sc_kbdc);
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
if (c == -1) {
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
/* CONTROLLER ERROR */
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
printf("sc%d: unable to get the current command byte value.\n", unit);
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
goto fail;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
if (bootverbose)
printf("sc%d: the current keyboard controller command byte %04x\n",
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
unit, c);
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
#if 0
/* override the keyboard lock switch */
c |= KBD_OVERRIDE_KBD_LOCK;
#endif
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
/*
* The keyboard may have been screwed up by the boot block.
* We may just be able to recover from error by testing the controller
* and the keyboard port. The controller command byte needs to be saved
* before this recovery operation, as some controllers seem to set
* the command byte to particular values.
*/
test_controller(sc_kbdc);
test_kbd_port(sc_kbdc);
Preserve some bits in the keyboard controller command byte when resetting the keyboard. Well, sorry, this bug is totally my fault. I DID intend to preserve them, but somehow I failed. The bug puts some old keyboard controllers in a strange state, resulting in keyboard freeze or random key input. The fix closes PR kern/3067.
1997-04-10 12:26:50 +00:00
/* enable the keyboard port, but disable the keyboard intr. */
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
if (!set_controller_command_byte(sc_kbdc,
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
KBD_KBD_CONTROL_BITS,
Preserve some bits in the keyboard controller command byte when resetting the keyboard. Well, sorry, this bug is totally my fault. I DID intend to preserve them, but somehow I failed. The bug puts some old keyboard controllers in a strange state, resulting in keyboard freeze or random key input. The fix closes PR kern/3067.
1997-04-10 12:26:50 +00:00
KBD_ENABLE_KBD_PORT | KBD_DISABLE_KBD_INT)) {
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
/* CONTROLLER ERROR
* there is very little we can do...
*/
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
printf("sc%d: unable to set the command byte.\n", unit);
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
goto fail;
}
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
/*
* Check if we have an XT keyboard before we attempt to reset it.
* The procedure assumes that the keyboard and the controller have
* been set up properly by BIOS and have not been messed up
* during the boot process.
*/
codeset = -1;
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
if (flags & XT_KEYBD)
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
/* the user says there is a XT keyboard */
codeset = 1;
#ifdef DETECT_XT_KEYBOARD
else if ((c & KBD_TRANSLATION) == 0) {
/* SET_SCANCODE_SET is not always supported; ignore error */
if (send_kbd_command_and_data(sc_kbdc, KBDC_SET_SCANCODE_SET, 0)
== KBD_ACK)
codeset = read_kbd_data(sc_kbdc);
}
if (bootverbose)
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
printf("sc%d: keyboard scancode set %d\n", unit, codeset);
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
#endif /* DETECT_XT_KEYBOARD */
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
dev->id_flags --> flags (not compiles otherwise)
1997-06-29 22:23:32 +00:00
if (flags & KBD_NORESET) {
A fix/work-around for ThinkPad 535. Add a new configuration flag, KBD_NORESET (0x20) to tell scprobe() not to reset the keyboard. IBM ThinkPad 535 has the `Fn' key with which the user can perform certain functions in conjunction with other keys. For example, `Fn' + PageUP/PageDOWN adjust speaker volume, `Fn' + Home/End change brightness of LCD screen. It can also be used to suspend the system. It appears that these functions are implemented at the keyboard level or the keyboard controller level and totally independent from BIOS or OS. But, if the keyboard is reset (as is done in scprobe()), they become unavailable. (There are other laptops which have similar functions associated with the `Fn' key. But, they aren't affected by keyboard reset.) ThinkPad 535 doesn't have switches or buttons to adjust brightness and volume, or to put the system into the suspend mode. Therefore, it is essential to preserve these `Fn' key functions in FreeBSD. The new flag make scprobe() skip keyboard reset. If this flag is not set, scprobe() behaves in the same say as before. (If we only knew a way to detect ThinkPad 535, we could skip keyboard reset automatically, but...)
1997-06-29 15:11:40 +00:00
write_kbd_command(sc_kbdc, KBDC_ECHO);
if (read_kbd_data(sc_kbdc) != KBD_ECHO) {
empty_both_buffers(sc_kbdc, 10);
test_controller(sc_kbdc);
test_kbd_port(sc_kbdc);
if (bootverbose)
printf("sc%d: failed to get response from the keyboard.\n",
unit);
goto fail;
}
} else {
/* reset keyboard hardware */
if (!reset_kbd(sc_kbdc)) {
/* KEYBOARD ERROR
* Keyboard reset may fail either because the keyboard doen't
* exist, or because the keyboard doesn't pass the self-test,
* or the keyboard controller on the motherboard and the keyboard
* somehow fail to shake hands. It is just possible, particularly
* in the last case, that the keyoard controller may be left
* in a hung state. test_controller() and test_kbd_port() appear
* to bring the keyboard controller back (I don't know why and
* how, though.)
*/
empty_both_buffers(sc_kbdc, 10);
test_controller(sc_kbdc);
test_kbd_port(sc_kbdc);
/* We could disable the keyboard port and interrupt... but,
* the keyboard may still exist (see above).
*/
if (bootverbose)
printf("sc%d: failed to reset the keyboard.\n", unit);
goto fail;
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
Allow us to enable the 'XT_KEYBOARD' code using a configuration flag. This allows the user to add modify syscons's configuration flags using UserConfig that will allow older/quirky hardware (most notably older IBM ThinkPad laptops) to work with the standard boot kernel. Inspired by: The Nomads
1996-11-10 16:44:13 +00:00
/*
* Allow us to set the XT_KEYBD flag in UserConfig so that keyboards
* such as those on the IBM ThinkPad laptop computers can be used
* with the standard console driver.
*/
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
if (codeset == 1) {
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
if (send_kbd_command_and_data(
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
sc_kbdc, KBDC_SET_SCANCODE_SET, codeset) == KBD_ACK) {
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
/* XT kbd doesn't need scan code translation */
c &= ~KBD_TRANSLATION;
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
} else {
/* KEYBOARD ERROR
* The XT kbd isn't usable unless the proper scan code set
* is selected.
*/
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
printf("sc%d: unable to set the XT keyboard mode.\n", unit);
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
goto fail;
}
Allow us to enable the 'XT_KEYBOARD' code using a configuration flag. This allows the user to add modify syscons's configuration flags using UserConfig that will allow older/quirky hardware (most notably older IBM ThinkPad laptops) to work with the standard boot kernel. Inspired by: The Nomads
1996-11-10 16:44:13 +00:00
}
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
/* enable the keyboard port and intr. */
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
if (!set_controller_command_byte(sc_kbdc,
Preserve some bits in the keyboard controller command byte when resetting the keyboard. Well, sorry, this bug is totally my fault. I DID intend to preserve them, but somehow I failed. The bug puts some old keyboard controllers in a strange state, resulting in keyboard freeze or random key input. The fix closes PR kern/3067.
1997-04-10 12:26:50 +00:00
KBD_KBD_CONTROL_BITS | KBD_TRANSLATION | KBD_OVERRIDE_KBD_LOCK,
(c & (KBD_TRANSLATION | KBD_OVERRIDE_KBD_LOCK))
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
| KBD_ENABLE_KBD_PORT | KBD_ENABLE_KBD_INT)) {
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
/* CONTROLLER ERROR
* This is serious; we are left with the disabled keyboard intr.
*/
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
printf("sc%d: unable to enable the keyboard port and intr.\n", unit);
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
goto fail;
}
Remove SC_KBD_PROBE_WORKS option and replace it with a simple run-time flag bit (0x0008) in the sc driver configuration line. This way it's easy to boink a generic kernel. Also, document and place in an opt_ file the #define's for overriding which serial port is the system console. Approved by: sos
1996-10-23 07:29:44 +00:00
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
kbdc_set_device_mask(sc_kbdc, m | KBD_KBD_CONTROL_BITS),
kbdc_lock(sc_kbdc, FALSE);
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
return TRUE;
Remove SC_KBD_PROBE_WORKS option and replace it with a simple run-time flag bit (0x0008) in the sc driver configuration line. This way it's easy to boink a generic kernel. Also, document and place in an opt_ file the #define's for overriding which serial port is the system console. Approved by: sos
1996-10-23 07:29:44 +00:00
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
fail:
Alot of fixes from kazu: 1. All the suggestions earlier made by Bruce: renaming some symbols, stricter error checking, removing redundant code, etc. 2. The `psm' driver preserves the default counter resolution and report rate, whatever they are after reset. (Based on reports and suggestion from Nate and Rob Bolin). 3. The `psm' driver now does not check the so-called sync. bit in the first byte of the data packet by default, so that the tapping feature of ALPUS GlidePoint works (based on reports from Louis Mamakos). I tested the code with ALPUS Desktop GlidePoint (M/N GP101) and found no problem; tapping worked. It appears ALPUS produces several models of GlidePoint. I hope the other models are OK too. The check code can still be activated by defining the PSM_CHECKSYNC option in the config file. (The bit checking slightly reduces, if not completely eliminates, weird mouse behavior cased by unsynchronized mouse data packets. It also helps us to detect if the mouse interrupt can ever be lost. But, well, if there are devices which cannot be supported this way...) 4. The `psm' driver does not include the protocol emulation code by default. The code can still be compiled in if the PSM_EMULATION option is specified in the config file. Louis Mamakos suggests the emulation code is putting too much in the kernel, and `moused' works well. I will think about this later and decide if the entire emulation code should be removed. 5. And, of course, the fix in `scprobe()' from Bruce to cure the UserConfig problem. My code in `kbdio.c' is slightly different from his patch, but has the same effect. There still is a possibility that `scprobe()' gets confused, if, for whatever reasons, the user holds down a key for very long time during the boot process. But we cannot cope with everything, can we? Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-12-01 19:05:50 +00:00
if (c != -1)
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
/* try to restore the command byte as before, if possible */
set_controller_command_byte(sc_kbdc, 0xff, c);
kbdc_set_device_mask(sc_kbdc,
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
(flags & DETECT_KBD) ? m : m | KBD_KBD_CONTROL_BITS);
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
kbdc_lock(sc_kbdc, FALSE);
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
return FALSE;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
scresume erroneously used before declared. Move the function rather than add another APM ifdef just for the forward decl. [Boy, sure a lot of warnings in here!]
1996-10-03 00:42:27 +00:00
#if NAPM > 0
static int
scresume(void *dummy)
{
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
shfts = ctls = alts = agrs = metas = accents = 0;
scresume erroneously used before declared. Move the function rather than add another APM ifdef just for the forward decl. [Boy, sure a lot of warnings in here!]
1996-10-03 00:42:27 +00:00
return 0;
}
#endif
Staticize and cleanup.
1995-12-10 13:40:44 +00:00
static int
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
scattach(struct isa_device *dev)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
scr_stat *scp;
dev_t cdev = makedev(CDEV_MAJOR, 0);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
char *p;
Moved the initialization of the devfs devices to after the initialization of the devswitch so that it actually works.
1996-03-27 19:11:41 +00:00
#ifdef DEVFS
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
int vc;
Moved the initialization of the devfs devices to after the initialization of the devswitch so that it actually works.
1996-03-27 19:11:41 +00:00
#endif
Third round in syscons update. Display update method changed, now allways write in memory buffer, then periodically update physical display. Speed improvements (now > 5 times faster than the old syscons). History now circular buffer, with changeable size. History scroll by up/down line, up/down page, home and end. Backtab proberly implemented. Now space for 96 function keys, 63 allocated standard, default now SCO/SYSV compat again as in the old days. New keyboard definition files ~share/syscons/keymaps/* Misc fixes for old "hacks" that broke SCO/SYSV compat. More that I forgot before writing this...
1995-01-28 22:18:05 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scinit();
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
flags = dev->id_flags;
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
if (!crtc_vga)
flags &= ~CHAR_CURSOR;
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp = console[0];
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
if (crtc_vga) {
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
cut_buffer = (char *)malloc(scp->xsize*scp->ysize, M_DEVBUF, M_NOWAIT);
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->scr_buf = (u_short *)malloc(scp->xsize*scp->ysize*sizeof(u_short),
M_DEVBUF, M_NOWAIT);
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
/* copy temporary buffer to final buffer */
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(sc_buffer, scp->scr_buf, scp->xsize * scp->ysize * sizeof(u_short));
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
scp->cursor_pos = scp->cursor_oldpos =
scp->scr_buf + scp->xpos + scp->ypos * scp->xsize;
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
scp->mouse_pos = scp->mouse_oldpos =
scp->scr_buf + ((scp->mouse_ypos/scp->font_size)*scp->xsize +
scp->mouse_xpos/8);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* initialize history buffer & pointers */
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
scp->history_head = scp->history_pos =
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(u_short *)malloc(scp->history_size*sizeof(u_short),
M_DEVBUF, M_NOWAIT);
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
if (scp->history_head != NULL)
bzero(scp->history_head, scp->history_size*sizeof(u_short));
scp->history = scp->history_head;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
/* initialize cursor stuff */
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
if (!(scp->status & UNKNOWN_MODE))
Fixed the "missing updates" reported by ache. Moved a little closer to having a splash page capability.
1996-10-01 23:24:09 +00:00
draw_cursor_image(scp);
Allow all characters except "real" control chars. Fix NULL reference when serial console (thanks bruce).
1995-02-03 21:09:44 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* get screen update going */
Fixed the type of timeout functions and removed casts that hid the type mismatches. Not taking an arg in scrn_timer() broke `cc -mrtd'.
1997-04-20 16:05:33 +00:00
scrn_timer(NULL);
Allow all characters except "real" control chars. Fix NULL reference when serial console (thanks bruce).
1995-02-03 21:09:44 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
update_leds(scp->status);
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
if ((crtc_type == KD_VGA) && bootverbose) {
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
printf("sc%d: BIOS video mode:%d\n",
dev->id_unit, *(u_char *)pa_to_va(0x449));
printf("sc%d: VGA registers upon power-up\n", dev->id_unit);
dump_vgaregs(vgaregs);
printf("sc%d: video mode:%d\n", dev->id_unit, scp->mode);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
printf("sc%d: VGA registers in BIOS for mode:%d\n",
dev->id_unit, scp->mode);
dump_vgaregs(vgaregs2);
p = get_mode_param(scp, scp->mode);
if (p != NULL) {
printf("sc%d: VGA registers to be used for mode:%d\n",
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
dev->id_unit, scp->mode);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
dump_vgaregs(p);
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
}
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
printf("sc%d: rows_offset:%d\n", dev->id_unit, rows_offset);
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
}
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
if ((crtc_type == KD_VGA) && (video_mode_ptr == NULL))
printf("sc%d: WARNING: video mode switching is only partially supported\n",
dev->id_unit);
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
printf("sc%d: ", dev->id_unit);
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
switch(crtc_type) {
case KD_VGA:
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
if (crtc_addr == MONO_BASE)
printf("VGA mono");
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
else
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
printf("VGA color");
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
break;
case KD_EGA:
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
if (crtc_addr == MONO_BASE)
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
printf("EGA mono");
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
else
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
printf("EGA color");
break;
case KD_CGA:
printf("CGA");
break;
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
case KD_PIXEL:
printf("Graphics display (VESA mode = 0x%x)", vesa_mode);
break;
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
case KD_MONO:
case KD_HERCULES:
default:
printf("MDA/hercules");
break;
}
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
printf(" <%d virtual consoles, flags=0x%x>\n", MAXCONS, flags);
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
This is the latest version of the APM stuff from HOSOKAWA, I have looked briefly over it, and see some serious architectural issues in this stuff. On the other hand, I doubt that we will have any solution to these issues before 2.1, so we might as well leave this in. Most of the stuff is bracketed by #ifdef's so it shouldn't matter too much in the normal case. Reviewed by: phk Submitted by: HOSOKAWA, Tatsumi <hosokawa@mt.cs.keio.ac.jp>
1995-02-17 02:22:57 +00:00
#if NAPM > 0
(a) remove the pointer to each driver's tty structure array from cdevsw (b) add a function callback vector to tty drivers that will return a pointer to a valid tty structure based upon a dev_t (c) make syscons structures the same size whether or not APM is enabled so utilities don't crash if NAPM changes (and make the damn kernel compile!) (d) rewrite /dev/snp ioctl interface so that it is device driver and i386 independant
1995-02-25 20:09:44 +00:00
scp->r_hook.ah_fun = scresume;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->r_hook.ah_arg = NULL;
scp->r_hook.ah_name = "system keyboard";
scp->r_hook.ah_order = APM_MID_ORDER;
apm_hook_establish(APM_HOOK_RESUME , &scp->r_hook);
#endif
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
cdevsw_add(&cdev, &scdevsw, NULL);
Removed my devsw access functions [un]register_cdev() and getmajorbyname() which were a better (sigh) temporary interface to the going-away devswitches. Note that SYSINIT()s to initialize the devswitches would be fatal in syscons.c and pcvt_drv.c (and are bogus elsewhere) because they get called independently of whether the device is attached; thus devices that share a major clobber each other's devswitch entries until the last one wins. conf.c: Removed stale #includes and comments.
1995-12-14 22:03:12 +00:00
Moved the initialization of the devfs devices to after the initialization of the devswitch so that it actually works.
1996-03-27 19:11:41 +00:00
#ifdef DEVFS
for (vc = 0; vc < MAXCONS; vc++)
add devfs entries for a few more syscons devices (e.g. sysmouse)
1998-01-12 03:28:36 +00:00
sc_devfs_token[vc] = devfs_add_devswf(&scdevsw, vc, DV_CHR,
UID_ROOT, GID_WHEEL, 0600, "ttyv%n", vc);
sc_mouse_devfs_token = devfs_add_devswf(&scdevsw, SC_MOUSE, DV_CHR,
UID_ROOT, GID_WHEEL, 0600, "sysmouse");
sc_console_devfs_token = devfs_add_devswf(&scdevsw, SC_CONSOLE, DV_CHR,
UID_ROOT, GID_WHEEL, 0600, "consolectl");
Moved the initialization of the devfs devices to after the initialization of the devswitch so that it actually works.
1996-03-27 19:11:41 +00:00
#endif
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
struct tty
(a) remove the pointer to each driver's tty structure array from cdevsw (b) add a function callback vector to tty drivers that will return a pointer to a valid tty structure based upon a dev_t (c) make syscons structures the same size whether or not APM is enabled so utilities don't crash if NAPM changes (and make the damn kernel compile!) (d) rewrite /dev/snp ioctl interface so that it is device driver and i386 independant
1995-02-25 20:09:44 +00:00
*scdevtotty(dev_t dev)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
int unit = minor(dev);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
if (init_done == COLD)
Close kern/627 - panic if syscons not attached
1996-02-08 06:30:31 +00:00
return(NULL);
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
if (unit == SC_CONSOLE)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return CONSOLE_TTY;
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
if (unit == SC_MOUSE)
return MOUSE_TTY;
if (unit >= MAXCONS || unit < 0)
return(NULL);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return VIRTUAL_TTY(unit);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
int
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
scopen(dev_t dev, int flag, int mode, struct proc *p)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
(a) remove the pointer to each driver's tty structure array from cdevsw (b) add a function callback vector to tty drivers that will return a pointer to a valid tty structure based upon a dev_t (c) make syscons structures the same size whether or not APM is enabled so utilities don't crash if NAPM changes (and make the damn kernel compile!) (d) rewrite /dev/snp ioctl interface so that it is device driver and i386 independant
1995-02-25 20:09:44 +00:00
struct tty *tp = scdevtotty(dev);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (!tp)
return(ENXIO);
Hack workaround XFree86 switching failure when used with /dev/sysmouse and xdm, possibly in general. What was happening was that the server was doing a tcsetattr(.. TCSADRAIN) on the mouse fd after a write. Since /dev/sysmouse had a null t_oproc, the drain failed with EIO. Somehow this spammed XFree86 (!@&^#%*& binary release!!), and the driver was left in a bogus state (ie: switch_in_progress permanently TRUE). The simplest way out was to implement a dummy scmousestart() routine to accept any characters from the tty system and toss them into the void. It would probably be more correct to intercept scwrite()'s to the mouse device, but that's executed for every single write to the screen. Supplying a start routine to eat the characters is only executed for the mouse port during startup/shutdown, so it should be faster.
1996-09-10 19:14:49 +00:00
tp->t_oproc = (minor(dev) == SC_MOUSE) ? scmousestart : scstart;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
tp->t_param = scparam;
tp->t_dev = dev;
if (!(tp->t_state & TS_ISOPEN)) {
ttychars(tp);
Make the default VERASE key the <-- key and remove root's `stty's. 'nuff said. Inventor: joerg@FreeBSD.org Reviewer: sos@FreeBSD.org
1997-04-03 21:42:42 +00:00
/* Use the current setting of the <-- key as default VERASE. */
/* If the Delete key is preferable, an stty is necessary */
tp->t_cc[VERASE] = key_map.key[0x0e].map[0];
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
tp->t_iflag = TTYDEF_IFLAG;
tp->t_oflag = TTYDEF_OFLAG;
tp->t_cflag = TTYDEF_CFLAG;
tp->t_lflag = TTYDEF_LFLAG;
tp->t_ispeed = tp->t_ospeed = TTYDEF_SPEED;
scparam(tp, &tp->t_termios);
ttsetwater(tp);
Obtained from: partly from ancient patches of mine via 1.1.5 Introduce TS_CONNECTED and TS_ZOMBIE states. TS_CONNECTED is set while a connection is established. It is set while (TS_CARR_ON or CLOCAL is set) and TS_ZOMBIE is clear. TS_ZOMBIE is set for on to off transitions of TS_CARR_ON that occur when CLOCAL is clear and is cleared for off to on transitions of CLOCAL. I/o can only occur while TS_CONNECTED is set. TS_ZOMBIE prevents further i/o. Split the input-event sleep address TSA_CARR_ON(tp) into TSA_CARR_ON(tp) and TSA_HUP_OR_INPUT(tp). The former address is now used only for off to on carrier transitions and equivalent CLOCAL transitions. The latter is used for all input events, all carrier transitions and certain CLOCAL transitions. There are some harmless extra wakeups for rare connection- related events. Previously there were too many extra wakeups for non-rare input events. Drivers now call l_modem() instead of setting TS_CARR_ON directly to handle even the initial off to on transition of carrier. They should always have done this. l_modem() now handles TS_CONNECTED and TS_ZOMBIE as well as TS_CARR_ON. gnu/isdn/iitty.c: Set TS_CONNECTED for first open ourself to go with bogusly setting CLOCAL. i386/isa/syscons.c, i386/isa/pcvt/pcvt_drv.c: We fake carrier, so don't also fake CLOCAL. kern/tty.c: Testing TS_CONNECTED instead of TS_CARR_ON fixes TIOCCONS forgetting to test CLOCAL. TS_ISOPEN was tested instead, but that broke when we disabled the clearing of TS_ISOPEN for certain transitions of CLOCAL. Testing TS_CONNECTED fixes ttyselect() returning false success for output to devices in state !TS_CARR_ON && !CLOCAL. Optimize the other selwakeup() call (this is not related to the other changes). kern/tty_pty.c: ptcopen() can be declared in traditional C now that dev_t isn't short.
1995-07-31 21:02:00 +00:00
(*linesw[tp->t_line].l_modem)(tp, 1);
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
if (minor(dev) == SC_MOUSE)
mouse_level = 0; /* XXX */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
}
else
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (tp->t_state & TS_XCLUDE && p->p_ucred->cr_uid != 0)
return(EBUSY);
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
if (minor(dev) < MAXCONS && !console[minor(dev)]) {
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
console[minor(dev)] = alloc_scp();
Fixed the "missing updates" reported by ache. Moved a little closer to having a splash page capability.
1996-10-01 23:24:09 +00:00
}
Corrected the setting of winsize on open(). Problem pointed out by David Nugent - davidn@blaze.net.au.
1996-10-15 20:27:07 +00:00
if (minor(dev)<MAXCONS && !tp->t_winsize.ws_col && !tp->t_winsize.ws_row) {
tp->t_winsize.ws_col = console[minor(dev)]->xsize;
tp->t_winsize.ws_row = console[minor(dev)]->ysize;
}
Fixed the "missing updates" reported by ache. Moved a little closer to having a splash page capability.
1996-10-01 23:24:09 +00:00
return ((*linesw[tp->t_line].l_open)(dev, tp));
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
int
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
scclose(dev_t dev, int flag, int mode, struct proc *p)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
(a) remove the pointer to each driver's tty structure array from cdevsw (b) add a function callback vector to tty drivers that will return a pointer to a valid tty structure based upon a dev_t (c) make syscons structures the same size whether or not APM is enabled so utilities don't crash if NAPM changes (and make the damn kernel compile!) (d) rewrite /dev/snp ioctl interface so that it is device driver and i386 independant
1995-02-25 20:09:44 +00:00
struct tty *tp = scdevtotty(dev);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
struct scr_stat *scp;
if (!tp)
return(ENXIO);
if (minor(dev) < MAXCONS) {
scp = get_scr_stat(tp->t_dev);
if (scp->status & SWITCH_WAIT_ACQ)
wakeup((caddr_t)&scp->smode);
Third round in syscons update. Display update method changed, now allways write in memory buffer, then periodically update physical display. Speed improvements (now > 5 times faster than the old syscons). History now circular buffer, with changeable size. History scroll by up/down line, up/down page, home and end. Backtab proberly implemented. Now space for 96 function keys, 63 allocated standard, default now SCO/SYSV compat again as in the old days. New keyboard definition files ~share/syscons/keymaps/* Misc fixes for old "hacks" that broke SCO/SYSV compat. More that I forgot before writing this...
1995-01-28 22:18:05 +00:00
#if not_yet_done
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (scp == &main_console) {
scp->pid = 0;
scp->proc = NULL;
scp->smode.mode = VT_AUTO;
}
else {
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
free(scp->scr_buf, M_DEVBUF);
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
if (scp->history != NULL) {
YAMF2.2: Handle indirect results of "kbdcontrol -h X" where X < 25.
1997-07-07 13:42:42 +00:00
free(scp->history, M_DEVBUF);
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
if (scp->history_size / scp->xsize
> imax(SC_HISTORY_SIZE, scp->ysize))
extra_history_size += scp->history_size / scp->xsize
- imax(SC_HISTORY_SIZE, scp->ysize);
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
free(scp, M_DEVBUF);
console[minor(dev)] = NULL;
}
Third round in syscons update. Display update method changed, now allways write in memory buffer, then periodically update physical display. Speed improvements (now > 5 times faster than the old syscons). History now circular buffer, with changeable size. History scroll by up/down line, up/down page, home and end. Backtab proberly implemented. Now space for 96 function keys, 63 allocated standard, default now SCO/SYSV compat again as in the old days. New keyboard definition files ~share/syscons/keymaps/* Misc fixes for old "hacks" that broke SCO/SYSV compat. More that I forgot before writing this...
1995-01-28 22:18:05 +00:00
#else
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->pid = 0;
scp->proc = NULL;
scp->smode.mode = VT_AUTO;
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
#endif
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
1. Avoid a race in scclose(). tty.c has kludges so that the race is actually harmless. 2. Fixed code to match comment in scintr(). 3. Don't allow even root to take control of the machine when securelevel > 0. I've secured the accesses to PSL_IOPL in all drivers and asked pst to review it, but he seems to be busy. Write access to /dev/kmem and other critival devices currently leaks across raisings of securelevel via open fd's, so there may as well be a similar leak for PSL_IOPL. 4. (Most important.) Don't corrupt memory beyond the screen buffers if the cursor happens to be off the 80x25 screen when syscons starts. 5. Fix console cursor update (not perfect yet). Submitted by: bruce ~
1996-11-11 22:21:03 +00:00
spltty();
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(*linesw[tp->t_line].l_close)(tp, flag);
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
ttyclose(tp);
1. Avoid a race in scclose(). tty.c has kludges so that the race is actually harmless. 2. Fixed code to match comment in scintr(). 3. Don't allow even root to take control of the machine when securelevel > 0. I've secured the accesses to PSL_IOPL in all drivers and asked pst to review it, but he seems to be busy. Write access to /dev/kmem and other critival devices currently leaks across raisings of securelevel via open fd's, so there may as well be a similar leak for PSL_IOPL. 4. (Most important.) Don't corrupt memory beyond the screen buffers if the cursor happens to be off the 80x25 screen when syscons starts. 5. Fix console cursor update (not perfect yet). Submitted by: bruce ~
1996-11-11 22:21:03 +00:00
spl0();
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return(0);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
int
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
scread(dev_t dev, struct uio *uio, int flag)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
(a) remove the pointer to each driver's tty structure array from cdevsw (b) add a function callback vector to tty drivers that will return a pointer to a valid tty structure based upon a dev_t (c) make syscons structures the same size whether or not APM is enabled so utilities don't crash if NAPM changes (and make the damn kernel compile!) (d) rewrite /dev/snp ioctl interface so that it is device driver and i386 independant
1995-02-25 20:09:44 +00:00
struct tty *tp = scdevtotty(dev);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (!tp)
return(ENXIO);
return((*linesw[tp->t_line].l_read)(tp, uio, flag));
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
int
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
scwrite(dev_t dev, struct uio *uio, int flag)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
(a) remove the pointer to each driver's tty structure array from cdevsw (b) add a function callback vector to tty drivers that will return a pointer to a valid tty structure based upon a dev_t (c) make syscons structures the same size whether or not APM is enabled so utilities don't crash if NAPM changes (and make the damn kernel compile!) (d) rewrite /dev/snp ioctl interface so that it is device driver and i386 independant
1995-02-25 20:09:44 +00:00
struct tty *tp = scdevtotty(dev);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (!tp)
return(ENXIO);
return((*linesw[tp->t_line].l_write)(tp, uio, flag));
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
void
Updated to latest 1.1.5.1+ version Reviewed by: Submitted by:
1994-08-17 08:51:59 +00:00
scintr(int unit)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
static struct tty *cur_tty;
int c, len;
u_char *cp;
Fix the hanging keyboard problem under Xaccel. Apprently we are loosing an interrupt somewhere. The solution here is to check for keyboard input each time the screen update timer ticks. Not pretty, but works.
1996-11-04 21:01:08 +00:00
/*
* Loop while there is still input to get from the keyboard.
* I don't think this is nessesary, and it doesn't fix
* the Xaccel-2.1 keyboard hang, but it can't hurt. XXX
*/
while ((c = scgetc(SCGETC_NONBLOCK)) != NOKEY) {
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
Fix the hanging keyboard problem under Xaccel. Apprently we are loosing an interrupt somewhere. The solution here is to check for keyboard input each time the screen update timer ticks. Not pretty, but works.
1996-11-04 21:01:08 +00:00
cur_tty = VIRTUAL_TTY(get_scr_num());
if (!(cur_tty->t_state & TS_ISOPEN))
if (!((cur_tty = CONSOLE_TTY)->t_state & TS_ISOPEN))
1. Avoid a race in scclose(). tty.c has kludges so that the race is actually harmless. 2. Fixed code to match comment in scintr(). 3. Don't allow even root to take control of the machine when securelevel > 0. I've secured the accesses to PSL_IOPL in all drivers and asked pst to review it, but he seems to be busy. Write access to /dev/kmem and other critival devices currently leaks across raisings of securelevel via open fd's, so there may as well be a similar leak for PSL_IOPL. 4. (Most important.) Don't corrupt memory beyond the screen buffers if the cursor happens to be off the 80x25 screen when syscons starts. 5. Fix console cursor update (not perfect yet). Submitted by: bruce ~
1996-11-11 22:21:03 +00:00
continue;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
Fix the hanging keyboard problem under Xaccel. Apprently we are loosing an interrupt somewhere. The solution here is to check for keyboard input each time the screen update timer ticks. Not pretty, but works.
1996-11-04 21:01:08 +00:00
switch (c & 0xff00) {
case 0x0000: /* normal key */
(*linesw[cur_tty->t_line].l_rint)(c & 0xFF, cur_tty);
break;
case FKEY: /* function key, return string */
if (cp = get_fstr((u_int)c, (u_int *)&len)) {
while (len-- > 0)
(*linesw[cur_tty->t_line].l_rint)(*cp++ & 0xFF, cur_tty);
}
1. Avoid a race in scclose(). tty.c has kludges so that the race is actually harmless. 2. Fixed code to match comment in scintr(). 3. Don't allow even root to take control of the machine when securelevel > 0. I've secured the accesses to PSL_IOPL in all drivers and asked pst to review it, but he seems to be busy. Write access to /dev/kmem and other critival devices currently leaks across raisings of securelevel via open fd's, so there may as well be a similar leak for PSL_IOPL. 4. (Most important.) Don't corrupt memory beyond the screen buffers if the cursor happens to be off the 80x25 screen when syscons starts. 5. Fix console cursor update (not perfect yet). Submitted by: bruce ~
1996-11-11 22:21:03 +00:00
break;
Fix the hanging keyboard problem under Xaccel. Apprently we are loosing an interrupt somewhere. The solution here is to check for keyboard input each time the screen update timer ticks. Not pretty, but works.
1996-11-04 21:01:08 +00:00
case MKEY: /* meta is active, prepend ESC */
(*linesw[cur_tty->t_line].l_rint)(0x1b, cur_tty);
(*linesw[cur_tty->t_line].l_rint)(c & 0xFF, cur_tty);
break;
case BKEY: /* backtab fixed sequence (esc [ Z) */
(*linesw[cur_tty->t_line].l_rint)(0x1b, cur_tty);
(*linesw[cur_tty->t_line].l_rint)('[', cur_tty);
(*linesw[cur_tty->t_line].l_rint)('Z', cur_tty);
break;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
}
Fix the hanging keyboard problem under Xaccel. Apprently we are loosing an interrupt somewhere. The solution here is to check for keyboard input each time the screen update timer ticks. Not pretty, but works.
1996-11-04 21:01:08 +00:00
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
if (cur_console->status & MOUSE_ENABLED) {
cur_console->status &= ~MOUSE_VISIBLE;
remove_mouse_image(cur_console);
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Staticize and cleanup.
1995-12-10 13:40:44 +00:00
static int
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
scparam(struct tty *tp, struct termios *t)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
tp->t_ispeed = t->c_ispeed;
tp->t_ospeed = t->c_ospeed;
tp->t_cflag = t->c_cflag;
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
int
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
scioctl(dev_t dev, int cmd, caddr_t data, int flag, struct proc *p)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Corrected the setting of winsize on open(). Problem pointed out by David Nugent - davidn@blaze.net.au.
1996-10-15 20:27:07 +00:00
int error;
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
u_int i;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
struct tty *tp;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
scr_stat *scp;
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
u_short *usp;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
char *mp;
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
int s;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
(a) remove the pointer to each driver's tty structure array from cdevsw (b) add a function callback vector to tty drivers that will return a pointer to a valid tty structure based upon a dev_t (c) make syscons structures the same size whether or not APM is enabled so utilities don't crash if NAPM changes (and make the damn kernel compile!) (d) rewrite /dev/snp ioctl interface so that it is device driver and i386 independant
1995-02-25 20:09:44 +00:00
tp = scdevtotty(dev);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (!tp)
return ENXIO;
scp = get_scr_stat(tp->t_dev);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
/* If there is a user_ioctl function call that first */
if (sc_user_ioctl) {
if (error = (*sc_user_ioctl)(dev, cmd, data, flag, p))
return error;
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
switch (cmd) { /* process console hardware related ioctl's */
Added more SCO compat ioctl's.
1994-10-17 22:12:06 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case GIO_ATTR: /* get current attributes */
Fix GIO_ATTR ioctl return to match SYSV
1996-07-30 15:20:08 +00:00
*(int*)data = (scp->term.cur_attr >> 8) & 0xFF;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
Added more SCO compat ioctl's.
1994-10-17 22:12:06 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case GIO_COLOR: /* is this a color console ? */
if (crtc_addr == COLOR_BASE)
*(int*)data = 1;
else
*(int*)data = 0;
return 0;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case CONS_CURRENT: /* get current adapter type */
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
*(int *)data = crtc_type;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
Added more SCO compat ioctl's.
1994-10-17 22:12:06 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case CONS_GET: /* get current video mode */
*(int*)data = scp->mode;
return 0;
Added support for many more videomodes, including graphic modes up til 320x200 256col VGA. This is nessesary for the iBCS stuff to work right. (And we get the benefit of more video modes). Uses the videocard BIOS to optain mode tables. Added a "green" saver, switches off the syncs for "green" monitors. Reviewed by: Submitted by: Obtained from:
1994-09-15 07:26:40 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case CONS_BLANKTIME: /* set screen saver timeout (0 = no saver) */
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
if (*(int *)data < 0)
return EINVAL;
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
scrn_blank_time = *(int *)data;
if (scrn_blank_time == 0)
Time changes mark 2: * Figure out UTC relative to boottime. Four new functions provide time relative to boottime. * move "runtime" into struct proc. This helps fix the calcru() problem in SMP. * kill mono_time. * add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!) * nanosleep, select & poll takes long sleeps one day at a time Reviewed by: bde Tested by: ache and others
1998-04-04 13:26:20 +00:00
getmicroruntime(&scrn_time_stamp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
Added support for many more videomodes, including graphic modes up til 320x200 256col VGA. This is nessesary for the iBCS stuff to work right. (And we get the benefit of more video modes). Uses the videocard BIOS to optain mode tables. Added a "green" saver, switches off the syncs for "green" monitors. Reviewed by: Submitted by: Obtained from:
1994-09-15 07:26:40 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case CONS_CURSORTYPE: /* set cursor type blink/noblink */
if ((*(int*)data) & 0x01)
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
flags |= BLINK_CURSOR;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
else
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
flags &= ~BLINK_CURSOR;
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
if ((*(int*)data) & 0x02) {
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
if (!crtc_vga)
return ENXIO;
flags |= CHAR_CURSOR;
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
} else
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
flags &= ~CHAR_CURSOR;
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
/*
* The cursor shape is global property; all virtual consoles
* are affected. Update the cursor in the current console...
*/
if (!(cur_console->status & UNKNOWN_MODE)) {
remove_cursor_image(cur_console);
if (flags & CHAR_CURSOR)
set_destructive_cursor(cur_console);
draw_cursor_image(cur_console);
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
Added support for many more videomodes, including graphic modes up til 320x200 256col VGA. This is nessesary for the iBCS stuff to work right. (And we get the benefit of more video modes). Uses the videocard BIOS to optain mode tables. Added a "green" saver, switches off the syncs for "green" monitors. Reviewed by: Submitted by: Obtained from:
1994-09-15 07:26:40 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case CONS_BELLTYPE: /* set bell type sound/visual */
if (*data)
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
flags |= VISUAL_BELL;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
else
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
flags &= ~VISUAL_BELL;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case CONS_HISTORY: /* set history size */
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
if (*(int *)data > 0) {
int lines; /* buffer size to allocate */
int lines0; /* current buffer size */
lines = imax(*(int *)data, scp->ysize);
lines0 = (scp->history != NULL) ?
scp->history_size / scp->xsize : scp->ysize;
/*
* syscons unconditionally allocates buffers upto SC_HISTORY_SIZE
* lines or scp->ysize lines, whichever is larger. A value
* greater than that is allowed, subject to extra_history_size.
*/
if (lines > imax(lines0, SC_HISTORY_SIZE) + extra_history_size)
return EINVAL;
Don't allow CONS_HISTORY ioctl to alter history pointers while in "scroll mode" (return EBUSY). Suggested by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-07-09 14:10:19 +00:00
if (cur_console->status & BUFFER_SAVED)
return EBUSY;
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
usp = scp->history;
scp->history = NULL;
if (usp != NULL)
free(usp, M_DEVBUF);
scp->history_size = lines * scp->xsize;
/*
* extra_history_size +=
* (lines0 > imax(SC_HISTORY_SIZE, scp->ysize)) ?
* lines0 - imax(SC_HISTORY_SIZE, scp->ysize)) : 0;
* extra_history_size -=
* (lines > imax(SC_HISTORY_SIZE, scp->ysize)) ?
* lines - imax(SC_HISTORY_SIZE, scp->ysize)) : 0;
* lines0 >= ysize && lines >= ysize... Hey, the above can be
* reduced to the following...
*/
extra_history_size +=
imax(lines0, SC_HISTORY_SIZE) - imax(lines, SC_HISTORY_SIZE);
usp = (u_short *)malloc(scp->history_size * sizeof(u_short),
M_DEVBUF, M_WAITOK);
bzero(usp, scp->history_size * sizeof(u_short));
scp->history_head = scp->history_pos = usp;
scp->history = usp;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
}
else
return EINVAL;
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case CONS_MOUSECTL: /* control mouse arrow */
- Add binary compatibility support for obsolete console/mouse ioctls so that existing programs which were compiled before the introduction of the new mouse code and use these ioctls will run unmodified. Suggested by msmith.
1998-01-20 03:37:27 +00:00
case OLD_CONS_MOUSECTL:
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
{
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
/* MOUSE_BUTTON?DOWN -> MOUSE_MSC_BUTTON?UP */
static butmap[8] = {
MOUSE_MSC_BUTTON1UP | MOUSE_MSC_BUTTON2UP
| MOUSE_MSC_BUTTON3UP,
MOUSE_MSC_BUTTON2UP | MOUSE_MSC_BUTTON3UP,
MOUSE_MSC_BUTTON1UP | MOUSE_MSC_BUTTON3UP,
MOUSE_MSC_BUTTON3UP,
MOUSE_MSC_BUTTON1UP | MOUSE_MSC_BUTTON2UP,
MOUSE_MSC_BUTTON2UP,
MOUSE_MSC_BUTTON1UP,
0,
};
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
mouse_info_t *mouse = (mouse_info_t*)data;
- Add binary compatibility support for obsolete console/mouse ioctls so that existing programs which were compiled before the introduction of the new mouse code and use these ioctls will run unmodified. Suggested by msmith.
1998-01-20 03:37:27 +00:00
mouse_info_t buf;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
if (!crtc_vga)
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
return ENODEV;
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
- Add binary compatibility support for obsolete console/mouse ioctls so that existing programs which were compiled before the introduction of the new mouse code and use these ioctls will run unmodified. Suggested by msmith.
1998-01-20 03:37:27 +00:00
if (cmd == OLD_CONS_MOUSECTL) {
static unsigned char swapb[] = { 0, 4, 2, 6, 1, 5, 3, 7 };
old_mouse_info_t *old_mouse = (old_mouse_info_t *)data;
mouse = &buf;
mouse->operation = old_mouse->operation;
switch (mouse->operation) {
case MOUSE_MODE:
mouse->u.mode = old_mouse->u.mode;
break;
case MOUSE_SHOW:
case MOUSE_HIDE:
break;
case MOUSE_MOVEABS:
case MOUSE_MOVEREL:
case MOUSE_ACTION:
mouse->u.data.x = old_mouse->u.data.x;
mouse->u.data.y = old_mouse->u.data.y;
mouse->u.data.z = 0;
mouse->u.data.buttons = swapb[old_mouse->u.data.buttons & 0x7];
break;
case MOUSE_GETINFO:
old_mouse->u.data.x = scp->mouse_xpos;
old_mouse->u.data.y = scp->mouse_ypos;
old_mouse->u.data.buttons = swapb[scp->mouse_buttons & 0x7];
break;
default:
return EINVAL;
}
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
switch (mouse->operation) {
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
case MOUSE_MODE:
make sure that the user supplied signals in struct vt_mode are actually valid signals, else return EINVAL for ioctl VT_SETMODE. this fixes a problem that anybody with vty access can panic the system. 2.2-Candidate (and 2.1.0 I believe) Reviewed-by: sos
1997-03-01 23:53:46 +00:00
if (ISSIGVALID(mouse->u.mode.signal)) {
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
scp->mouse_signal = mouse->u.mode.signal;
scp->mouse_proc = p;
scp->mouse_pid = p->p_pid;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
else {
scp->mouse_signal = 0;
scp->mouse_proc = NULL;
scp->mouse_pid = 0;
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
}
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
break;
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
case MOUSE_SHOW:
if (!(scp->status & MOUSE_ENABLED)) {
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
scp->status |= (MOUSE_ENABLED | MOUSE_VISIBLE);
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
scp->mouse_oldpos = scp->mouse_pos;
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
mark_all(scp);
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
}
else
return EINVAL;
break;
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
case MOUSE_HIDE:
if (scp->status & MOUSE_ENABLED) {
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
scp->status &= ~(MOUSE_ENABLED | MOUSE_VISIBLE);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
mark_all(scp);
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
}
else
return EINVAL;
break;
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
case MOUSE_MOVEABS:
scp->mouse_xpos = mouse->u.data.x;
scp->mouse_ypos = mouse->u.data.y;
set_mouse_pos(scp);
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
break;
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
case MOUSE_MOVEREL:
scp->mouse_xpos += mouse->u.data.x;
scp->mouse_ypos += mouse->u.data.y;
set_mouse_pos(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
Added more SCO compat ioctl's.
1994-10-17 22:12:06 +00:00
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
case MOUSE_GETINFO:
mouse->u.data.x = scp->mouse_xpos;
mouse->u.data.y = scp->mouse_ypos;
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
mouse->u.data.z = 0;
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
mouse->u.data.buttons = scp->mouse_buttons;
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
break;
Added more SCO compat ioctl's.
1994-10-17 22:12:06 +00:00
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
case MOUSE_ACTION:
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
case MOUSE_MOTION_EVENT:
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
/* this should maybe only be settable from /dev/consolectl SOS */
/* send out mouse event on /dev/sysmouse */
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
mouse_status.dx += mouse->u.data.x;
mouse_status.dy += mouse->u.data.y;
mouse_status.dz += mouse->u.data.z;
if (mouse->operation == MOUSE_ACTION)
mouse_status.button = mouse->u.data.buttons;
mouse_status.flags |=
((mouse->u.data.x || mouse->u.data.y || mouse->u.data.z) ?
MOUSE_POSCHANGED : 0)
| (mouse_status.obutton ^ mouse_status.button);
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
if (cur_console->status & MOUSE_ENABLED)
cur_console->status |= MOUSE_VISIBLE;
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
if ((MOUSE_TTY)->t_state & TS_ISOPEN) {
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
u_char buf[MOUSE_SYS_PACKETSIZE];
Rename some local variables to avoid shadowing other local variables. Found by: -Wshadow
1997-11-07 09:21:01 +00:00
int j;
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
/* the first five bytes are compatible with MouseSystems' */
buf[0] = MOUSE_MSC_SYNC
| butmap[mouse_status.button & MOUSE_STDBUTTONS];
j = imax(imin(mouse->u.data.x, 255), -256);
buf[1] = j >> 1;
buf[3] = j - buf[1];
j = -imax(imin(mouse->u.data.y, 255), -256);
buf[2] = j >> 1;
buf[4] = j - buf[2];
for (j = 0; j < MOUSE_MSC_PACKETSIZE; j++)
(*linesw[(MOUSE_TTY)->t_line].l_rint)(buf[j],MOUSE_TTY);
if (mouse_level >= 1) { /* extended part */
j = imax(imin(mouse->u.data.z, 127), -128);
buf[5] = (j >> 1) & 0x7f;
buf[6] = (j - (j >> 1)) & 0x7f;
/* buttons 4-10 */
buf[7] = (~mouse_status.button >> 3) & 0x7f;
for (j = MOUSE_MSC_PACKETSIZE;
j < MOUSE_SYS_PACKETSIZE; j++)
Rename some local variables to avoid shadowing other local variables. Found by: -Wshadow
1997-11-07 09:21:01 +00:00
(*linesw[(MOUSE_TTY)->t_line].l_rint)(buf[j],MOUSE_TTY);
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
}
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
}
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
if (cur_console->mouse_signal) {
cur_console->mouse_buttons = mouse->u.data.buttons;
/* has controlling process died? */
if (cur_console->mouse_proc &&
(cur_console->mouse_proc != pfind(cur_console->mouse_pid))){
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
cur_console->mouse_signal = 0;
cur_console->mouse_proc = NULL;
cur_console->mouse_pid = 0;
}
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
else
psignal(cur_console->mouse_proc, cur_console->mouse_signal);
}
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
else if (mouse->operation == MOUSE_ACTION) {
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
/* process button presses */
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
if ((cur_console->mouse_buttons ^ mouse->u.data.buttons) &&
!(cur_console->status & UNKNOWN_MODE)) {
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
cur_console->mouse_buttons = mouse->u.data.buttons;
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
if (cur_console->mouse_buttons & MOUSE_BUTTON1DOWN)
mouse_cut_start(cur_console);
else
mouse_cut_end(cur_console);
if (cur_console->mouse_buttons & MOUSE_BUTTON2DOWN ||
cur_console->mouse_buttons & MOUSE_BUTTON3DOWN)
mouse_paste(cur_console);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
}
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
}
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
Fix the checks for screenborder for the mousepointer.
1997-08-09 19:24:03 +00:00
if (mouse->u.data.x != 0 || mouse->u.data.y != 0) {
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
cur_console->mouse_xpos += mouse->u.data.x;
cur_console->mouse_ypos += mouse->u.data.y;
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
set_mouse_pos(cur_console);
Fix the checks for screenborder for the mousepointer.
1997-08-09 19:24:03 +00:00
}
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
break;
case MOUSE_BUTTON_EVENT:
if ((mouse->u.event.id & MOUSE_BUTTONS) == 0)
return EINVAL;
if (mouse->u.event.value < 0)
return EINVAL;
if (mouse->u.event.value > 0) {
cur_console->mouse_buttons |= mouse->u.event.id;
mouse_status.button |= mouse->u.event.id;
} else {
cur_console->mouse_buttons &= ~mouse->u.event.id;
mouse_status.button &= ~mouse->u.event.id;
}
mouse_status.flags |=
((mouse->u.data.x || mouse->u.data.y || mouse->u.data.z) ?
MOUSE_POSCHANGED : 0)
| (mouse_status.obutton ^ mouse_status.button);
if (cur_console->status & MOUSE_ENABLED)
cur_console->status |= MOUSE_VISIBLE;
if ((MOUSE_TTY)->t_state & TS_ISOPEN) {
u_char buf[8];
int i;
buf[0] = MOUSE_MSC_SYNC
| butmap[mouse_status.button & MOUSE_STDBUTTONS];
buf[7] = (~mouse_status.button >> 3) & 0x7f;
buf[1] = buf[2] = buf[3] = buf[4] = buf[5] = buf[6] = 0;
for (i = 0;
i < ((mouse_level >= 1) ? MOUSE_SYS_PACKETSIZE
: MOUSE_MSC_PACKETSIZE); i++)
(*linesw[(MOUSE_TTY)->t_line].l_rint)(buf[i],MOUSE_TTY);
}
if (cur_console->mouse_signal) {
if (cur_console->mouse_proc &&
(cur_console->mouse_proc != pfind(cur_console->mouse_pid))){
cur_console->mouse_signal = 0;
cur_console->mouse_proc = NULL;
cur_console->mouse_pid = 0;
}
else
psignal(cur_console->mouse_proc, cur_console->mouse_signal);
break;
}
if (cur_console->status & UNKNOWN_MODE)
break;
switch (mouse->u.event.id) {
case MOUSE_BUTTON1DOWN:
switch (mouse->u.event.value % 4) {
case 0: /* up */
mouse_cut_end(cur_console);
break;
case 1:
mouse_cut_start(cur_console);
break;
case 2:
mouse_cut_word(cur_console);
break;
case 3:
mouse_cut_line(cur_console);
break;
}
break;
case MOUSE_BUTTON2DOWN:
switch (mouse->u.event.value) {
case 0: /* up */
break;
default:
mouse_paste(cur_console);
break;
}
break;
case MOUSE_BUTTON3DOWN:
switch (mouse->u.event.value) {
case 0: /* up */
if (!(cur_console->mouse_buttons & MOUSE_BUTTON1DOWN))
mouse_cut_end(cur_console);
break;
default:
mouse_cut_extend(cur_console);
break;
}
break;
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
default:
return EINVAL;
}
/* make screensaver happy */
Time changes mark 2: * Figure out UTC relative to boottime. Four new functions provide time relative to boottime. * move "runtime" into struct proc. This helps fix the calcru() problem in SMP. * kill mono_time. * add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!) * nanosleep, select & poll takes long sleeps one day at a time Reviewed by: bde Tested by: ache and others
1998-04-04 13:26:20 +00:00
getmicroruntime(&scrn_time_stamp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
/* MOUSE_XXX: /dev/sysmouse ioctls */
case MOUSE_GETHWINFO: /* get device information */
{
mousehw_t *hw = (mousehw_t *)data;
if (tp != MOUSE_TTY)
return ENOTTY;
hw->buttons = 10; /* XXX unknown */
hw->iftype = MOUSE_IF_SYSMOUSE;
hw->type = MOUSE_MOUSE;
hw->model = MOUSE_MODEL_GENERIC;
hw->hwid = 0;
return 0;
}
case MOUSE_GETMODE: /* get protocol/mode */
{
mousemode_t *mode = (mousemode_t *)data;
if (tp != MOUSE_TTY)
return ENOTTY;
mode->level = mouse_level;
switch (mode->level) {
case 0:
/* at this level, sysmouse emulates MouseSystems protocol */
mode->protocol = MOUSE_PROTO_MSC;
mode->rate = -1; /* unknown */
mode->resolution = -1; /* unknown */
mode->accelfactor = 0; /* disabled */
mode->packetsize = MOUSE_MSC_PACKETSIZE;
mode->syncmask[0] = MOUSE_MSC_SYNCMASK;
mode->syncmask[1] = MOUSE_MSC_SYNC;
break;
case 1:
/* at this level, sysmouse uses its own protocol */
mode->protocol = MOUSE_PROTO_SYSMOUSE;
mode->rate = -1;
mode->resolution = -1;
mode->accelfactor = 0;
mode->packetsize = MOUSE_SYS_PACKETSIZE;
mode->syncmask[0] = MOUSE_SYS_SYNCMASK;
mode->syncmask[1] = MOUSE_SYS_SYNC;
break;
}
return 0;
}
case MOUSE_SETMODE: /* set protocol/mode */
{
mousemode_t *mode = (mousemode_t *)data;
if (tp != MOUSE_TTY)
return ENOTTY;
if ((mode->level < 0) || (mode->level > 1))
return EINVAL;
mouse_level = mode->level;
return 0;
}
case MOUSE_GETLEVEL: /* get operation level */
if (tp != MOUSE_TTY)
return ENOTTY;
*(int *)data = mouse_level;
return 0;
case MOUSE_SETLEVEL: /* set operation level */
if (tp != MOUSE_TTY)
return ENOTTY;
if ((*(int *)data < 0) || (*(int *)data > 1))
return EINVAL;
mouse_level = *(int *)data;
return 0;
case MOUSE_GETSTATUS: /* get accumulated mouse events */
if (tp != MOUSE_TTY)
return ENOTTY;
s = spltty();
*(mousestatus_t *)data = mouse_status;
mouse_status.flags = 0;
mouse_status.obutton = mouse_status.button;
mouse_status.dx = 0;
mouse_status.dy = 0;
mouse_status.dz = 0;
splx(s);
return 0;
#if notyet
case MOUSE_GETVARS: /* get internal mouse variables */
case MOUSE_SETVARS: /* set internal mouse variables */
if (tp != MOUSE_TTY)
return ENOTTY;
return ENODEV;
#endif
case MOUSE_READSTATE: /* read status from the device */
case MOUSE_READDATA: /* read data from the device */
if (tp != MOUSE_TTY)
return ENOTTY;
return ENODEV;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case CONS_GETINFO: /* get current (virtual) console info */
{
vid_info_t *ptr = (vid_info_t*)data;
if (ptr->size == sizeof(struct vid_info)) {
ptr->m_num = get_scr_num();
ptr->mv_col = scp->xpos;
ptr->mv_row = scp->ypos;
ptr->mv_csz = scp->xsize;
ptr->mv_rsz = scp->ysize;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
ptr->mv_norm.fore = (scp->term.std_color & 0x0f00)>>8;
ptr->mv_norm.back = (scp->term.std_color & 0xf000)>>12;
ptr->mv_rev.fore = (scp->term.rev_color & 0x0f00)>>8;
ptr->mv_rev.back = (scp->term.rev_color & 0xf000)>>12;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
ptr->mv_grfc.fore = 0; /* not supported */
ptr->mv_grfc.back = 0; /* not supported */
ptr->mv_ovscan = scp->border;
ptr->mk_keylock = scp->status & LOCK_KEY_MASK;
return 0;
}
return EINVAL;
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case CONS_GETVERS: /* get version number */
*(int*)data = 0x200; /* version 2.0 */
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* VGA TEXT MODES */
case SW_VGA_C40x25:
case SW_VGA_C80x25: case SW_VGA_M80x25:
case SW_VGA_C80x30: case SW_VGA_M80x30:
case SW_VGA_C80x50: case SW_VGA_M80x50:
case SW_VGA_C80x60: case SW_VGA_M80x60:
case SW_B40x25: case SW_C40x25:
case SW_B80x25: case SW_C80x25:
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
case SW_ENH_B40x25: case SW_ENH_C40x25:
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case SW_ENH_B80x25: case SW_ENH_C80x25:
case SW_ENH_B80x43: case SW_ENH_C80x43:
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
case SW_EGAMONO80x25:
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
if (!crtc_vga)
return ENODEV;
mp = get_mode_param(scp, cmd & 0xff);
if (mp == NULL)
return ENODEV;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
s = spltty();
if ((error = wait_scrn_saver_stop())) {
splx(s);
return error;
}
scp->status &= ~MOUSE_VISIBLE;
remove_cutmarking(scp);
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
if (scp->history != NULL)
i = imax(scp->history_size / scp->xsize
- imax(SC_HISTORY_SIZE, scp->ysize), 0);
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
else
i = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
switch (cmd & 0xff) {
case M_VGA_C80x60: case M_VGA_M80x60:
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
if (!(fonts_loaded & FONT_8)) {
splx(s);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return EINVAL;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
}
/*
* This is a kludge to fend off scrn_update() while we
* muck around with scp. XXX
*/
scp->status |= UNKNOWN_MODE;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize = 80;
scp->ysize = 60;
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
scp->font_size = 8;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case M_VGA_C80x50: case M_VGA_M80x50:
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
if (!(fonts_loaded & FONT_8)) {
splx(s);
Syscons bug fix: tab now right (thanks bruce!) Mouse arrow support almost finished.
1995-02-01 21:56:32 +00:00
return EINVAL;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
}
scp->status |= UNKNOWN_MODE;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize = 80;
scp->ysize = 50;
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
scp->font_size = 8;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case M_ENH_B80x43: case M_ENH_C80x43:
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
if (!(fonts_loaded & FONT_8)) {
splx(s);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return EINVAL;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
}
scp->status |= UNKNOWN_MODE;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize = 80;
scp->ysize = 43;
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
scp->font_size = 8;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case M_VGA_C80x30: case M_VGA_M80x30:
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
scp->status |= UNKNOWN_MODE;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize = 80;
scp->ysize = 30;
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
scp->font_size = mp[2];
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
case M_ENH_C40x25: case M_ENH_B40x25:
case M_ENH_C80x25: case M_ENH_B80x25:
case M_EGAMONO80x25:
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
if (!(fonts_loaded & FONT_14)) {
splx(s);
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
return EINVAL;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
}
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
/* FALL THROUGH */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
default:
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
if ((cmd & 0xff) > M_VGA_CG320) {
splx(s);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return EINVAL;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
}
scp->status |= UNKNOWN_MODE;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
scp->xsize = mp[0];
scp->ysize = mp[1] + rows_offset;
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
scp->font_size = mp[2];
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
}
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->mode = cmd & 0xff;
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
scp->xpixel = scp->xsize * 8;
scp->ypixel = scp->ysize * scp->font_size;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
free(scp->scr_buf, M_DEVBUF);
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
scp->scr_buf = (u_short *)
malloc(scp->xsize*scp->ysize*sizeof(u_short), M_DEVBUF, M_WAITOK);
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
/* move the text cursor to the home position */
move_crsr(scp, 0, 0);
/* move the mouse cursor at the center of the screen */
scp->mouse_xpos = scp->xpixel / 2;
scp->mouse_ypos = scp->ypixel / 2;
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
scp->mouse_pos = scp->mouse_oldpos =
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
scp->scr_buf + (scp->mouse_ypos / scp->font_size) * scp->xsize
+ scp->mouse_xpos / 8;
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
free(cut_buffer, M_DEVBUF);
cut_buffer = (char *)malloc(scp->xsize*scp->ysize, M_DEVBUF, M_NOWAIT);
cut_buffer[0] = 0x00;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
splx(s);
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
usp = scp->history;
scp->history = NULL;
if (usp != NULL) {
free(usp, M_DEVBUF);
extra_history_size += i;
}
scp->history_size = imax(SC_HISTORY_SIZE, scp->ysize) * scp->xsize;
usp = (u_short *)malloc(scp->history_size * sizeof(u_short),
M_DEVBUF, M_NOWAIT);
if (usp != NULL)
bzero(usp, scp->history_size * sizeof(u_short));
scp->history_head = scp->history_pos = usp;
scp->history = usp;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (scp == cur_console)
set_mode(scp);
clear_screen(scp);
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
scp->status &= ~UNKNOWN_MODE;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
if (tp->t_winsize.ws_col != scp->xsize
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
|| tp->t_winsize.ws_row != scp->ysize) {
tp->t_winsize.ws_col = scp->xsize;
tp->t_winsize.ws_row = scp->ysize;
pgsignal(tp->t_pgrp, SIGWINCH, 1);
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
return 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* GRAPHICS MODES */
case SW_BG320: case SW_BG640:
case SW_CG320: case SW_CG320_D: case SW_CG640_E:
case SW_CG640x350: case SW_ENH_CG640:
case SW_BG640x480: case SW_CG640x480: case SW_VGA_CG320:
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
if (!crtc_vga)
return ENODEV;
mp = get_mode_param(scp, cmd & 0xff);
if (mp == NULL)
return ENODEV;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
s = spltty();
if ((error = wait_scrn_saver_stop())) {
splx(s);
return error;
}
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
scp->status &= ~MOUSE_VISIBLE;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
remove_cutmarking(scp);
scp->status |= UNKNOWN_MODE; /* graphics mode */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->mode = cmd & 0xFF;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
scp->xpixel = mp[0] * 8;
scp->ypixel = (mp[1] + rows_offset) * mp[2];
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
scp->font_size = FONT_NONE;
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
/* move the mouse cursor at the center of the screen */
scp->mouse_xpos = scp->xpixel / 2;
scp->mouse_ypos = scp->ypixel / 2;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
splx(s);
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
if (scp == cur_console)
set_mode(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* clear_graphics();*/
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
if (tp->t_winsize.ws_xpixel != scp->xpixel
|| tp->t_winsize.ws_ypixel != scp->ypixel) {
tp->t_winsize.ws_xpixel = scp->xpixel;
tp->t_winsize.ws_ypixel = scp->ypixel;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
pgsignal(tp->t_pgrp, SIGWINCH, 1);
}
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Yeah I'm back hacking syscons !! Add support for MODEX 320x240x256color with "unchained" adressing, giving access to all 256K on all VGA's, those with that much memory that is :) Also make sysmouse use the right resolution in graphics modes.
1997-08-08 22:52:30 +00:00
case SW_VGA_MODEX:
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
if (!crtc_vga)
return ENODEV;
mp = get_mode_param(scp, cmd & 0xff);
if (mp == NULL)
return ENODEV;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
s = spltty();
if ((error = wait_scrn_saver_stop())) {
splx(s);
return error;
}
scp->status &= ~MOUSE_VISIBLE;
remove_cutmarking(scp);
scp->status |= UNKNOWN_MODE; /* graphics mode */
Yeah I'm back hacking syscons !! Add support for MODEX 320x240x256color with "unchained" adressing, giving access to all 256K on all VGA's, those with that much memory that is :) Also make sysmouse use the right resolution in graphics modes.
1997-08-08 22:52:30 +00:00
scp->mode = cmd & 0xFF;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
scp->xpixel = 320;
scp->ypixel = 240;
scp->font_size = FONT_NONE;
splx(s);
Yeah I'm back hacking syscons !! Add support for MODEX 320x240x256color with "unchained" adressing, giving access to all 256K on all VGA's, those with that much memory that is :) Also make sysmouse use the right resolution in graphics modes.
1997-08-08 22:52:30 +00:00
if (scp == cur_console)
set_mode(scp);
/* clear_graphics();*/
if (tp->t_winsize.ws_xpixel != scp->xpixel
|| tp->t_winsize.ws_ypixel != scp->ypixel) {
tp->t_winsize.ws_xpixel = scp->xpixel;
tp->t_winsize.ws_ypixel = scp->ypixel;
pgsignal(tp->t_pgrp, SIGWINCH, 1);
}
return 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case VT_SETMODE: /* set screen switcher mode */
make sure that the user supplied signals in struct vt_mode are actually valid signals, else return EINVAL for ioctl VT_SETMODE. this fixes a problem that anybody with vty access can panic the system. 2.2-Candidate (and 2.1.0 I believe) Reviewed-by: sos
1997-03-01 23:53:46 +00:00
{
struct vt_mode *mode;
mode = (struct vt_mode *)data;
if (ISSIGVALID(mode->relsig) && ISSIGVALID(mode->acqsig) &&
ISSIGVALID(mode->frsig)) {
bcopy(data, &scp->smode, sizeof(struct vt_mode));
if (scp->smode.mode == VT_PROCESS) {
scp->proc = p;
scp->pid = scp->proc->p_pid;
}
return 0;
} else
return EINVAL;
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case VT_GETMODE: /* get screen switcher mode */
bcopy(&scp->smode, data, sizeof(struct vt_mode));
return 0;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case VT_RELDISP: /* screen switcher ioctl */
switch(*data) {
case VT_FALSE: /* user refuses to release screen, abort */
if (scp == old_scp && (scp->status & SWITCH_WAIT_REL)) {
old_scp->status &= ~SWITCH_WAIT_REL;
switch_in_progress = FALSE;
return 0;
}
return EINVAL;
case VT_TRUE: /* user has released screen, go on */
if (scp == old_scp && (scp->status & SWITCH_WAIT_REL)) {
scp->status &= ~SWITCH_WAIT_REL;
exchange_scr();
if (new_scp->smode.mode == VT_PROCESS) {
new_scp->status |= SWITCH_WAIT_ACQ;
psignal(new_scp->proc, new_scp->smode.acqsig);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
else
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
switch_in_progress = FALSE;
This is syscons version 1.1
1993-10-28 05:27:36 +00:00
return 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
return EINVAL;
This is syscons version 1.1
1993-10-28 05:27:36 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case VT_ACKACQ: /* acquire acknowledged, switch completed */
if (scp == new_scp && (scp->status & SWITCH_WAIT_ACQ)) {
scp->status &= ~SWITCH_WAIT_ACQ;
switch_in_progress = FALSE;
Syscons bug fix: tab now right (thanks bruce!) Mouse arrow support almost finished.
1995-02-01 21:56:32 +00:00
return 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
return EINVAL;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
default:
return EINVAL;
}
/* NOT REACHED */
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case VT_OPENQRY: /* return free virtual console */
for (i = 0; i < MAXCONS; i++) {
tp = VIRTUAL_TTY(i);
if (!(tp->t_state & TS_ISOPEN)) {
*data = i + 1;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
return 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return EINVAL;
case VT_ACTIVATE: /* switch to screen *data */
return switch_scr(scp, (*data) - 1);
case VT_WAITACTIVE: /* wait for switch to occur */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
if (*data > MAXCONS || *data < 0)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return EINVAL;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
if (minor(dev) == (*data) - 1)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
if (*data == 0) {
if (scp == cur_console)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
return 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
else
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp = console[(*data) - 1];
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
while ((error=tsleep((caddr_t)&scp->smode, PZERO|PCATCH,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
"waitvt", 0)) == ERESTART) ;
return error;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case VT_GETACTIVE:
*data = get_scr_num()+1;
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KDENABIO: /* allow io operations */
Don't allow i/o operations for non-root users. Add prototypes.
1995-10-14 07:08:09 +00:00
error = suser(p->p_ucred, &p->p_acflag);
if (error != 0)
return error;
1. Avoid a race in scclose(). tty.c has kludges so that the race is actually harmless. 2. Fixed code to match comment in scintr(). 3. Don't allow even root to take control of the machine when securelevel > 0. I've secured the accesses to PSL_IOPL in all drivers and asked pst to review it, but he seems to be busy. Write access to /dev/kmem and other critival devices currently leaks across raisings of securelevel via open fd's, so there may as well be a similar leak for PSL_IOPL. 4. (Most important.) Don't corrupt memory beyond the screen buffers if the cursor happens to be off the 80x25 screen when syscons starts. 5. Fix console cursor update (not perfect yet). Submitted by: bruce ~
1996-11-11 22:21:03 +00:00
if (securelevel > 0)
return EPERM;
simplify IOPL gain/remove privs code. It's easier with md_regs being a trapframe.
1997-05-07 20:02:38 +00:00
p->p_md.md_regs->tf_eflags |= PSL_IOPL;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KDDISABIO: /* disallow io operations (default) */
simplify IOPL gain/remove privs code. It's easier with md_regs being a trapframe.
1997-05-07 20:02:38 +00:00
p->p_md.md_regs->tf_eflags &= ~PSL_IOPL;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KDSETMODE: /* set current mode of this (virtual) console */
switch (*data) {
case KD_TEXT: /* switch to TEXT (known) mode */
/* restore fonts & palette ! */
if (crtc_vga) {
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
#if 0
/*
* FONT KLUDGE
* Don't load fonts for now... XXX
*/
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (fonts_loaded & FONT_8)
copy_font(LOAD, FONT_8, font_8);
if (fonts_loaded & FONT_14)
copy_font(LOAD, FONT_14, font_14);
if (fonts_loaded & FONT_16)
copy_font(LOAD, FONT_16, font_16);
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
#endif
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
load_palette(palette);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
Move the hardware currsor off screen when getting back from an unknown mode. Fix warning on uninitialized var.
1997-11-25 12:44:44 +00:00
/* move hardware cursor out of the way */
outb(crtc_addr, 14);
outb(crtc_addr + 1, 0xff);
outb(crtc_addr, 15);
outb(crtc_addr + 1, 0xff);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* FALL THROUGH */
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KD_TEXT1: /* switch to TEXT (known) mode */
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
s = spltty();
if ((error = wait_scrn_saver_stop())) {
splx(s);
return error;
}
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
scp->status &= ~MOUSE_VISIBLE;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
remove_cutmarking(scp);
scp->status |= UNKNOWN_MODE;
splx(s);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* no restore fonts & palette */
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
if (crtc_vga)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
set_mode(scp);
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
scp->status &= ~UNKNOWN_MODE;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
clear_screen(scp);
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KD_GRAPHICS: /* switch to GRAPHICS (unknown) mode */
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
s = spltty();
if ((error = wait_scrn_saver_stop())) {
splx(s);
return error;
}
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
scp->status &= ~MOUSE_VISIBLE;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
remove_cutmarking(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->status |= UNKNOWN_MODE;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
splx(s);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
default:
return EINVAL;
}
/* NOT REACHED */
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KDGETMODE: /* get current mode of this (virtual) console */
*data = (scp->status & UNKNOWN_MODE) ? KD_GRAPHICS : KD_TEXT;
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KDSBORDER: /* set border color of this (virtual) console */
scp->border = *data;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
if (scp == cur_console)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
set_border(scp->border);
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KDSKBSTATE: /* set keyboard state (locks) */
if (*data >= 0 && *data <= LOCK_KEY_MASK) {
scp->status &= ~LOCK_KEY_MASK;
scp->status |= *data;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
if (scp == cur_console)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
update_leds(scp->status);
return 0;
}
return EINVAL;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KDGKBSTATE: /* get keyboard state (locks) */
*data = scp->status & LOCK_KEY_MASK;
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KDSETRAD: /* set keyboard repeat & delay rates */
if (*data & 0x80)
return EINVAL;
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
if (sc_kbdc != NULL)
set_keyboard(KBDC_SET_TYPEMATIC, *data);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KDSKBMODE: /* set keyboard mode */
switch (*data) {
case K_RAW: /* switch to RAW scancode mode */
Add a new keyboard mode K_CODE. Returns a single byte for each key much like the scancode mode. However the keys that (for no good reason) returns extension codes etc, are translated into singlebyte codes. Needed by libvgl. This makes life ALOT easier, also the XFree86 folks could use this.
1997-10-01 20:46:29 +00:00
scp->status &= ~KBD_CODE_MODE;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->status |= KBD_RAW_MODE;
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Add a new keyboard mode K_CODE. Returns a single byte for each key much like the scancode mode. However the keys that (for no good reason) returns extension codes etc, are translated into singlebyte codes. Needed by libvgl. This makes life ALOT easier, also the XFree86 folks could use this.
1997-10-01 20:46:29 +00:00
case K_CODE: /* switch to CODE mode */
scp->status &= ~KBD_RAW_MODE;
scp->status |= KBD_CODE_MODE;
return 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case K_XLATE: /* switch to XLT ascii mode */
Fix the bug that caused CTRL & ALT keys to be hanging sometimes after an X seesion. Really stupid error of me, and I've been looking at this code SO many times. Thanks to Kazutaka YOKOTA for seeing this.. Submitted by: Kazutaka YOKOTA
1997-01-20 08:05:15 +00:00
if (scp == cur_console && scp->status & KBD_RAW_MODE)
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
shfts = ctls = alts = agrs = metas = accents = 0;
Add a new keyboard mode K_CODE. Returns a single byte for each key much like the scancode mode. However the keys that (for no good reason) returns extension codes etc, are translated into singlebyte codes. Needed by libvgl. This makes life ALOT easier, also the XFree86 folks could use this.
1997-10-01 20:46:29 +00:00
scp->status &= ~(KBD_RAW_MODE | KBD_CODE_MODE);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
default:
return EINVAL;
}
/* NOT REACHED */
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KDGKBMODE: /* get keyboard mode */
Add a new keyboard mode K_CODE. Returns a single byte for each key much like the scancode mode. However the keys that (for no good reason) returns extension codes etc, are translated into singlebyte codes. Needed by libvgl. This makes life ALOT easier, also the XFree86 folks could use this.
1997-10-01 20:46:29 +00:00
*data = (scp->status & KBD_RAW_MODE) ? K_RAW :
((scp->status & KBD_CODE_MODE) ? K_CODE : K_XLATE);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KDMKTONE: /* sound the bell */
if (*(int*)data)
do_bell(scp, (*(int*)data)&0xffff,
(((*(int*)data)>>16)&0xffff)*hz/1000);
else
do_bell(scp, scp->bell_pitch, scp->bell_duration);
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case KIOCSOUND: /* make tone (*data) hz */
if (scp == cur_console) {
if (*(int*)data) {
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
int pitch = timer_freq / *(int*)data;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* set command for counter 2, 2 byte write */
if (acquire_timer2(TIMER_16BIT|TIMER_SQWAVE))
return EBUSY;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* set pitch */
outb(TIMER_CNTR2, pitch);
outb(TIMER_CNTR2, (pitch>>8));
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* enable counter 2 output to speaker */
outb(IO_PPI, inb(IO_PPI) | 3);
}
else {
/* disable counter 2 output to speaker */
outb(IO_PPI, inb(IO_PPI) & 0xFC);
release_timer2();
}
}
return 0;
case KDGKBTYPE: /* get keyboard type */
*data = 0; /* type not known (yet) */
return 0;
case KDSETLED: /* set keyboard LED status */
if (*data >= 0 && *data <= LED_MASK) {
scp->status &= ~LED_MASK;
scp->status |= *data;
if (scp == cur_console)
update_leds(scp->status);
return 0;
}
return EINVAL;
case KDGETLED: /* get keyboard LED status */
*data = scp->status & LED_MASK;
return 0;
case GETFKEY: /* get functionkey string */
if (*(u_short*)data < n_fkey_tab) {
fkeyarg_t *ptr = (fkeyarg_t*)data;
bcopy(&fkey_tab[ptr->keynum].str, ptr->keydef,
fkey_tab[ptr->keynum].len);
ptr->flen = fkey_tab[ptr->keynum].len;
return 0;
}
else
return EINVAL;
case SETFKEY: /* set functionkey string */
if (*(u_short*)data < n_fkey_tab) {
fkeyarg_t *ptr = (fkeyarg_t*)data;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
bcopy(ptr->keydef, &fkey_tab[ptr->keynum].str,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
min(ptr->flen, MAXFK));
fkey_tab[ptr->keynum].len = min(ptr->flen, MAXFK);
return 0;
}
else
return EINVAL;
case GIO_SCRNMAP: /* get output translation table */
bcopy(&scr_map, data, sizeof(scr_map));
return 0;
case PIO_SCRNMAP: /* set output translation table */
bcopy(data, &scr_map, sizeof(scr_map));
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
for (i=0; i<sizeof(scr_map); i++)
scr_rmap[scr_map[i]] = i;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
case GIO_KEYMAP: /* get keyboard translation table */
bcopy(&key_map, data, sizeof(key_map));
return 0;
case PIO_KEYMAP: /* set keyboard translation table */
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
accents = 0;
bzero(&accent_map, sizeof(accent_map));
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
bcopy(data, &key_map, sizeof(key_map));
return 0;
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
case GIO_DEADKEYMAP: /* get accent key translation table */
bcopy(&accent_map, data, sizeof(accent_map));
return 0;
case PIO_DEADKEYMAP: /* set accent key translation table */
accents = 0;
bcopy(data, &accent_map, sizeof(accent_map));
return 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case PIO_FONT8x8: /* set 8x8 dot font */
if (!crtc_vga)
return ENXIO;
bcopy(data, font_8, 8*256);
fonts_loaded |= FONT_8;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
/*
* FONT KLUDGE
* Always use the font page #0. XXX
* Don't load if the current font size is not 8x8.
*/
if (!(cur_console->status & UNKNOWN_MODE)
&& (cur_console->font_size < 14)) {
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
copy_font(LOAD, FONT_8, font_8);
if (flags & CHAR_CURSOR)
set_destructive_cursor(cur_console);
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
case GIO_FONT8x8: /* get 8x8 dot font */
if (!crtc_vga)
return ENXIO;
if (fonts_loaded & FONT_8) {
bcopy(font_8, data, 8*256);
return 0;
}
else
return ENXIO;
case PIO_FONT8x14: /* set 8x14 dot font */
if (!crtc_vga)
return ENXIO;
bcopy(data, font_14, 14*256);
fonts_loaded |= FONT_14;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
/*
* FONT KLUDGE
* Always use the font page #0. XXX
* Don't load if the current font size is not 8x14.
*/
if (!(cur_console->status & UNKNOWN_MODE)
&& (cur_console->font_size >= 14) && (cur_console->font_size < 16)) {
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
copy_font(LOAD, FONT_14, font_14);
if (flags & CHAR_CURSOR)
set_destructive_cursor(cur_console);
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
case GIO_FONT8x14: /* get 8x14 dot font */
if (!crtc_vga)
return ENXIO;
if (fonts_loaded & FONT_14) {
bcopy(font_14, data, 14*256);
return 0;
}
else
return ENXIO;
case PIO_FONT8x16: /* set 8x16 dot font */
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
if (!crtc_vga && crtc_type != KD_PIXEL)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return ENXIO;
bcopy(data, font_16, 16*256);
fonts_loaded |= FONT_16;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
/*
* FONT KLUDGE
* Always use the font page #0. XXX
* Don't load if the current font size is not 8x16.
*/
if (crtc_vga && !(cur_console->status & UNKNOWN_MODE)
&& (cur_console->font_size >= 16)) {
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
copy_font(LOAD, FONT_16, font_16);
if (flags & CHAR_CURSOR)
set_destructive_cursor(cur_console);
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return 0;
case GIO_FONT8x16: /* get 8x16 dot font */
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
if (!crtc_vga && crtc_type != KD_PIXEL)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return ENXIO;
if (fonts_loaded & FONT_16) {
bcopy(font_16, data, 16*256);
return 0;
}
else
return ENXIO;
default:
break;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
error = (*linesw[tp->t_line].l_ioctl)(tp, cmd, data, flag, p);
Use ENOIOCTL instead of -1 (= ERESTART) for tty ioctls that are not handled at a particular level. This fixes mainly restarting of interrupted TIOCDRAINs and TIOCSETA{W,F}s.
1997-12-06 13:25:01 +00:00
if (error != ENOIOCTL)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return(error);
error = ttioctl(tp, cmd, data, flag);
Use ENOIOCTL instead of -1 (= ERESTART) for tty ioctls that are not handled at a particular level. This fixes mainly restarting of interrupted TIOCDRAINs and TIOCSETA{W,F}s.
1997-12-06 13:25:01 +00:00
if (error != ENOIOCTL)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return(error);
return(ENOTTY);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Staticize and cleanup.
1995-12-10 13:40:44 +00:00
static void
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
scstart(struct tty *tp)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
struct clist *rbp;
Fix compiler warnings.
1995-11-20 12:13:32 +00:00
int s, len;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
u_char buf[PCBURST];
scr_stat *scp = get_scr_stat(tp->t_dev);
This is syscons version 1.1
1993-10-28 05:27:36 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (scp->status & SLKED || blink_in_progress)
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
return; /* XXX who repeats the call when the above flags are cleared? */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
s = spltty();
Fix races in scstart(). q_to_b() wasn't called at spltty(), so there were two races: - q_to_b() might unexpectedly return 0 (e.g, after a keyboard signal flushes the output queue and isn't echoed). ansi_put() interprets 0 bytes as 4GB... - more output (e.g. for echoes) might arrive afer q_to_b() returns 0. Then scstart() returns presumably and the new output might not be handled for a long time. Remove unused function scxint(). Fix prototypes (foo() isn't a prototype).
1995-07-11 18:34:30 +00:00
if (!(tp->t_state & (TS_TIMEOUT | TS_BUSY | TS_TTSTOP))) {
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
tp->t_state |= TS_BUSY;
rbp = &tp->t_outq;
while (rbp->c_cc) {
len = q_to_b(rbp, buf, PCBURST);
Fix races in scstart(). q_to_b() wasn't called at spltty(), so there were two races: - q_to_b() might unexpectedly return 0 (e.g, after a keyboard signal flushes the output queue and isn't echoed). ansi_put() interprets 0 bytes as 4GB... - more output (e.g. for echoes) might arrive afer q_to_b() returns 0. Then scstart() returns presumably and the new output might not be handled for a long time. Remove unused function scxint(). Fix prototypes (foo() isn't a prototype).
1995-07-11 18:34:30 +00:00
splx(s);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
ansi_put(scp, buf, len);
Fix races in scstart(). q_to_b() wasn't called at spltty(), so there were two races: - q_to_b() might unexpectedly return 0 (e.g, after a keyboard signal flushes the output queue and isn't echoed). ansi_put() interprets 0 bytes as 4GB... - more output (e.g. for echoes) might arrive afer q_to_b() returns 0. Then scstart() returns presumably and the new output might not be handled for a long time. Remove unused function scxint(). Fix prototypes (foo() isn't a prototype).
1995-07-11 18:34:30 +00:00
s = spltty();
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
tp->t_state &= ~TS_BUSY;
Move the inline code for waking up writers to a new function ttwwakeup(). The conditions for doing the wakeup will soon become more complicated and I don't want them duplicated in all drivers. It's probably not worth making ttwwakeup() a macro or an inline function. The cost of the function call is relatively small when there is a process to wake up. There is usually a process to wake up for large writes and the system call overhead dwarfs the function call overhead for small writes.
1995-07-22 01:30:45 +00:00
ttwwakeup(tp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
splx(s);
}
Hack workaround XFree86 switching failure when used with /dev/sysmouse and xdm, possibly in general. What was happening was that the server was doing a tcsetattr(.. TCSADRAIN) on the mouse fd after a write. Since /dev/sysmouse had a null t_oproc, the drain failed with EIO. Somehow this spammed XFree86 (!@&^#%*& binary release!!), and the driver was left in a bogus state (ie: switch_in_progress permanently TRUE). The simplest way out was to implement a dummy scmousestart() routine to accept any characters from the tty system and toss them into the void. It would probably be more correct to intercept scwrite()'s to the mouse device, but that's executed for every single write to the screen. Supplying a start routine to eat the characters is only executed for the mouse port during startup/shutdown, so it should be faster.
1996-09-10 19:14:49 +00:00
static void
scmousestart(struct tty *tp)
{
struct clist *rbp;
int s;
u_char buf[PCBURST];
s = spltty();
if (!(tp->t_state & (TS_TIMEOUT | TS_BUSY | TS_TTSTOP))) {
tp->t_state |= TS_BUSY;
rbp = &tp->t_outq;
while (rbp->c_cc) {
q_to_b(rbp, buf, PCBURST);
}
tp->t_state &= ~TS_BUSY;
ttwwakeup(tp);
}
splx(s);
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
void
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
sccnprobe(struct consdev *cp)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
struct isa_device *dvp;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
/*
* Take control if we are the highest priority enabled display device.
*/
dvp = find_display();
Don't claim the console when the driver is disabled. The getc/putc part of the console driver usually works when the driver is disabled, but the normal read/write part doesn't (it caused a panic).
1996-10-09 15:24:21 +00:00
if (dvp == NULL || dvp->id_driver != &scdriver) {
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
cp->cn_pri = CN_DEAD;
return;
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
if (!scvidprobe(dvp->id_unit, dvp->id_flags)) {
cp->cn_pri = CN_DEAD;
return;
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* initialize required fields */
Make syscons replicate a mousesystems mouse on minor 128.. This enables other consumers of the mouse, to get it info via moused/syscons. In order to use it run moused (from sysconfig), and then tell your Xserver that it should use /dev/sysmouse (mknod sysmouse c 12 128) and it a mousesystems mouse. Everybody will be happy then :) Remember that moused still needs to know what kind of mouse you have.. Comments welcome, as is test results...
1996-09-08 21:31:56 +00:00
cp->cn_dev = makedev(CDEV_MAJOR, SC_CONSOLE);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
cp->cn_pri = CN_INTERNAL;
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
sc_kbdc = kbdc_open(sc_port);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
void
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
sccninit(struct consdev *cp)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scinit();
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
void
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
sccnputc(dev_t dev, int c)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Correct the type of the `c' arg to pccnputc(). Move declarations of console functions to cons.h so that they can't be defined inconsistently in several places. They should be config(8)ed.
1995-04-23 10:15:42 +00:00
u_char buf[1];
Fixed a race that caused panics in kernel printfs when the screen timeout routine changed the screen pointers while output was in progress.
1996-05-01 03:58:21 +00:00
int s;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scr_stat *scp = console[0];
term_stat save = scp->term;
scp->term = kernel_console;
current_default = &kernel_default;
1. Avoid a race in scclose(). tty.c has kludges so that the race is actually harmless. 2. Fixed code to match comment in scintr(). 3. Don't allow even root to take control of the machine when securelevel > 0. I've secured the accesses to PSL_IOPL in all drivers and asked pst to review it, but he seems to be busy. Write access to /dev/kmem and other critival devices currently leaks across raisings of securelevel via open fd's, so there may as well be a similar leak for PSL_IOPL. 4. (Most important.) Don't corrupt memory beyond the screen buffers if the cursor happens to be off the 80x25 screen when syscons starts. 5. Fix console cursor update (not perfect yet). Submitted by: bruce ~
1996-11-11 22:21:03 +00:00
if (scp == cur_console && !(scp->status & UNKNOWN_MODE))
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
remove_cursor_image(scp);
Don't convert \n to \r\n in pccnputc(). This is the responsibility of cnputc().
1995-06-14 05:16:12 +00:00
buf[0] = c;
ansi_put(scp, buf, 1);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
kernel_console = scp->term;
current_default = &user_default;
scp->term = save;
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
s = spltty(); /* block scintr and scrn_timer */
sccnupdate(scp);
Fixed a race that caused panics in kernel printfs when the screen timeout routine changed the screen pointers while output was in progress.
1996-05-01 03:58:21 +00:00
splx(s);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
int
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
sccngetc(dev_t dev)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
int s = spltty(); /* block scintr and scrn_timer while we poll */
int c;
/*
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
* Stop the screen saver and update the screen if necessary.
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
* What if we have been running in the screen saver code... XXX
*/
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
sccnupdate(cur_console);
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
c = scgetc(SCGETC_CN);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
splx(s);
return(c);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
int
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
sccncheckc(dev_t dev)
Peter Dufaults comconsole changes. Submitted by: Peter Dufault
1994-10-20 00:08:31 +00:00
{
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
int s = spltty(); /* block scintr and scrn_timer while we poll */
int c;
Changed cncheckc() interface so that it is 8-bit clean - return -1 instead of 0 if there is no input. syscons.c: Added missing spl locking in sccncheckc(). Return the same value as sccngetc() would. It is wrong for sccngetc() to return non-ASCII, but stripping the non-ASCII bits doesn't help.
1996-09-14 04:27:46 +00:00
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
sccnupdate(cur_console);
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
c = scgetc(SCGETC_CN | SCGETC_NONBLOCK);
Changed cncheckc() interface so that it is 8-bit clean - return -1 instead of 0 if there is no input. syscons.c: Added missing spl locking in sccncheckc(). Return the same value as sccngetc() would. It is wrong for sccngetc() to return non-ASCII, but stripping the non-ASCII bits doesn't help.
1996-09-14 04:27:46 +00:00
splx(s);
return(c == NOKEY ? -1 : c); /* c == -1 can't happen */
1) Restore removed SNAKE_SAVER (Nate, WHY you do it?) 2) Make SNAKE_SAVER like default, if no saver specified in "options" 3) Remove #ifdef STAR_SAVER before line /* make screensaver happy */ this code needed in any case.
1994-02-01 10:43:02 +00:00
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
static void
sccnupdate(scr_stat *scp)
{
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
if (scp == cur_console && !font_loading_in_progress) {
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
if (scrn_blanked > 0)
stop_scrn_saver(current_saver);
if (!(scp->status & UNKNOWN_MODE) && scrn_blanked <= 0
&& !blink_in_progress && !switch_in_progress)
scrn_update(scp, TRUE);
}
}
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
static scr_stat
*get_scr_stat(dev_t dev)
{
int unit = minor(dev);
if (unit == SC_CONSOLE)
return console[0];
if (unit >= MAXCONS || unit < 0)
return(NULL);
return console[unit];
}
static int
get_scr_num()
{
int i = 0;
while ((i < MAXCONS) && (cur_console != console[i]))
i++;
return i < MAXCONS ? i : 0;
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static void
Fixed the type of timeout functions and removed casts that hid the type mismatches. Not taking an arg in scrn_timer() broke `cc -mrtd'.
1997-04-20 16:05:33 +00:00
scrn_timer(void *arg)
o Restore removed hardware cursor support and make it _default_ case, software cursor now optional case. Driver must provide raw things (what hardware do for us, exactly) as default case, all driver features must be _optional_. Modern VGAs have internal configuration utilities to set cursor shape/blinking which stored into cards ROM, and syscons nuke out such features completely by forcing software cursor. Moreover, software cursor is hard to distinguish on standouted (or near standouted) fields and tends to disappearse from the screen. Set "flags 0x4" to enable software cursor now. o Cleanup screen savers. o Don't draw cursor if saver or blinker is active. o Duplicated code moved to functons. o Add more checks for blinker in progress, character lost otherwise when blinker restore old contents. o Reduce blinking counter to 3, too slow in old variant. o Fix timeout code in scrn_timer(), old variant can reenter iself, if action takes too long time. o Disable visual bell for scroll lock mode, saved screen becomes overwritted otherwise.
1995-01-26 04:56:25 +00:00
{
Time changes mark 2: * Figure out UTC relative to boottime. Four new functions provide time relative to boottime. * move "runtime" into struct proc. This helps fix the calcru() problem in SMP. * kill mono_time. * add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!) * nanosleep, select & poll takes long sleeps one day at a time Reviewed by: bde Tested by: ache and others
1998-04-04 13:26:20 +00:00
struct timeval tv;
Syscons bug fix: tab now right (thanks bruce!) Mouse arrow support almost finished.
1995-02-01 21:56:32 +00:00
scr_stat *scp = cur_console;
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
int s;
/* don't do anything when we are touching font */
if (font_loading_in_progress) {
timeout(scrn_timer, NULL, hz / 10);
return;
}
s = spltty();
o Restore removed hardware cursor support and make it _default_ case, software cursor now optional case. Driver must provide raw things (what hardware do for us, exactly) as default case, all driver features must be _optional_. Modern VGAs have internal configuration utilities to set cursor shape/blinking which stored into cards ROM, and syscons nuke out such features completely by forcing software cursor. Moreover, software cursor is hard to distinguish on standouted (or near standouted) fields and tends to disappearse from the screen. Set "flags 0x4" to enable software cursor now. o Cleanup screen savers. o Don't draw cursor if saver or blinker is active. o Duplicated code moved to functons. o Add more checks for blinker in progress, character lost otherwise when blinker restore old contents. o Reduce blinking counter to 3, too slow in old variant. o Fix timeout code in scrn_timer(), old variant can reenter iself, if action takes too long time. o Disable visual bell for scroll lock mode, saved screen becomes overwritted otherwise.
1995-01-26 04:56:25 +00:00
Fix the hanging keyboard problem under Xaccel. Apprently we are loosing an interrupt somewhere. The solution here is to check for keyboard input each time the screen update timer ticks. Not pretty, but works.
1996-11-04 21:01:08 +00:00
/*
* With release 2.1 of the Xaccel server, the keyboard is left
Only poll the keyboard if the data left in the buffer is from the kbd, not if its from the psm device.
1996-11-15 08:45:24 +00:00
* hanging pretty often. Apparently an interrupt from the
Fix the hanging keyboard problem under Xaccel. Apprently we are loosing an interrupt somewhere. The solution here is to check for keyboard input each time the screen update timer ticks. Not pretty, but works.
1996-11-04 21:01:08 +00:00
* keyboard is lost, and I don't know why (yet).
Only poll the keyboard if the data left in the buffer is from the kbd, not if its from the psm device.
1996-11-15 08:45:24 +00:00
* This ugly hack calls scintr if input is ready for the keyboard
* and conveniently hides the problem. XXX
Fix the hanging keyboard problem under Xaccel. Apprently we are loosing an interrupt somewhere. The solution here is to check for keyboard input each time the screen update timer ticks. Not pretty, but works.
1996-11-04 21:01:08 +00:00
*/
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
/* Try removing anything stuck in the keyboard controller; whether
* it's a keyboard scan code or mouse data. `scintr()' doesn't
* read the mouse data directly, but `kbdio' routines will, as a
* side effect.
*/
if (kbdc_lock(sc_kbdc, TRUE)) {
/*
* We have seen the lock flag is not set. Let's reset the flag early;
* otherwise `update_led()' failes which may want the lock
* during `scintr()'.
*/
kbdc_lock(sc_kbdc, FALSE);
if (kbdc_data_ready(sc_kbdc))
scintr(0);
}
Fix the hanging keyboard problem under Xaccel. Apprently we are loosing an interrupt somewhere. The solution here is to check for keyboard input each time the screen update timer ticks. Not pretty, but works.
1996-11-04 21:01:08 +00:00
Syscons bug fix: tab now right (thanks bruce!) Mouse arrow support almost finished.
1995-02-01 21:56:32 +00:00
/* should we just return ? */
Fixed the annoing bug that caused writes to /dev/console to go to the active virtual terminal (Boy, do I have little hair left). Misc other fixes, try to enable cursor earlier in the game.
1995-02-05 23:44:25 +00:00
if ((scp->status&UNKNOWN_MODE) || blink_in_progress || switch_in_progress) {
Fixed the type of timeout functions and removed casts that hid the type mismatches. Not taking an arg in scrn_timer() broke `cc -mrtd'.
1997-04-20 16:05:33 +00:00
timeout(scrn_timer, NULL, hz / 10);
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
splx(s);
Syscons bug fix: tab now right (thanks bruce!) Mouse arrow support almost finished.
1995-02-01 21:56:32 +00:00
return;
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
/* should we stop the screen saver? */
Time changes mark 2: * Figure out UTC relative to boottime. Four new functions provide time relative to boottime. * move "runtime" into struct proc. This helps fix the calcru() problem in SMP. * kill mono_time. * add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!) * nanosleep, select & poll takes long sleeps one day at a time Reviewed by: bde Tested by: ache and others
1998-04-04 13:26:20 +00:00
getmicroruntime(&tv);
- Ignore KBD_RAW_MODE and KBD_CODE_MODE in scgetc() when it's called from the low-level console routines sccngetc() and sccncheckc(). Submitted by: bde (a long time ago) - Don't try to ring bell and immediately return from do_bell() while device probe is in progress at boot time; the timeout queue is not functional yet. PR: kern/2424 - Stop running the screen saver after panic() is called: check if `panicstr' is non-NULL during scrn_timer(). PR: kern/5314 - A new option: SC_DISABLE_REBOOT The reboot key (usually Ctl-Alt-Del) will be ignored if this option is defined. You may still have the reboot key defined in the keymap and it won't cause error when the keymap is loaded, but it will be quietly treated as nop. OKed by: sos
1998-02-11 14:56:02 +00:00
if (panicstr)
Time changes mark 2: * Figure out UTC relative to boottime. Four new functions provide time relative to boottime. * move "runtime" into struct proc. This helps fix the calcru() problem in SMP. * kill mono_time. * add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!) * nanosleep, select & poll takes long sleeps one day at a time Reviewed by: bde Tested by: ache and others
1998-04-04 13:26:20 +00:00
scrn_time_stamp = tv;
if (tv.tv_sec <= scrn_time_stamp.tv_sec + scrn_blank_time)
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
if (scrn_blanked > 0)
stop_scrn_saver(current_saver);
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
scp = cur_console;
if (scrn_blanked <= 0)
scrn_update(scp, TRUE);
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
/* should we activate the screen saver? */
if ((scrn_blank_time != 0)
Time changes mark 2: * Figure out UTC relative to boottime. Four new functions provide time relative to boottime. * move "runtime" into struct proc. This helps fix the calcru() problem in SMP. * kill mono_time. * add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!) * nanosleep, select & poll takes long sleeps one day at a time Reviewed by: bde Tested by: ache and others
1998-04-04 13:26:20 +00:00
&& (tv.tv_sec > scrn_time_stamp.tv_sec + scrn_blank_time))
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(*current_saver)(TRUE);
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
Fixed the type of timeout functions and removed casts that hid the type mismatches. Not taking an arg in scrn_timer() broke `cc -mrtd'.
1997-04-20 16:05:33 +00:00
timeout(scrn_timer, NULL, hz / 25);
Fixed the "missing updates" reported by ache. Moved a little closer to having a splash page capability.
1996-10-01 23:24:09 +00:00
splx(s);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
static void
scrn_update(scr_stat *scp, int show_cursor)
{
/* update screen image */
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
if (scp->start <= scp->end)
sc_bcopy(scp->scr_buf, scp->start, scp->end, 0);
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
/* we are not to show the cursor and the mouse pointer... */
if (!show_cursor) {
scp->end = 0;
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
scp->start = scp->xsize*scp->ysize - 1;
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
return;
}
/* update "pseudo" mouse pointer image */
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
if (scp->status & MOUSE_VISIBLE) {
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
/* did mouse move since last time ? */
if (scp->status & MOUSE_MOVED) {
/* do we need to remove old mouse pointer image ? */
if (scp->mouse_cut_start != NULL ||
(scp->mouse_pos-scp->scr_buf) <= scp->start ||
(scp->mouse_pos+scp->xsize + 1 - scp->scr_buf) >= scp->end) {
remove_mouse_image(scp);
}
scp->status &= ~MOUSE_MOVED;
draw_mouse_image(scp);
}
else {
/* mouse didn't move, has it been overwritten ? */
if ((scp->mouse_pos+scp->xsize + 1 - scp->scr_buf) >= scp->start &&
(scp->mouse_pos - scp->scr_buf) <= scp->end) {
draw_mouse_image(scp);
}
}
}
/* update cursor image */
if (scp->status & CURSOR_ENABLED) {
/* did cursor move since last time ? */
if (scp->cursor_pos != scp->cursor_oldpos) {
/* do we need to remove old cursor image ? */
if ((scp->cursor_oldpos - scp->scr_buf) < scp->start ||
((scp->cursor_oldpos - scp->scr_buf) > scp->end)) {
remove_cursor_image(scp);
}
scp->cursor_oldpos = scp->cursor_pos;
draw_cursor_image(scp);
}
else {
/* cursor didn't move, has it been overwritten ? */
if (scp->cursor_pos - scp->scr_buf >= scp->start &&
scp->cursor_pos - scp->scr_buf <= scp->end) {
draw_cursor_image(scp);
} else {
/* if its a blinking cursor, we may have to update it */
if (flags & BLINK_CURSOR)
draw_cursor_image(scp);
}
}
blinkrate++;
}
if (scp->mouse_cut_start != NULL)
draw_cutmarking(scp);
scp->end = 0;
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
scp->start = scp->xsize*scp->ysize - 1;
- Break scrn_timer() into two pieces. Now screen update is done in a separate routine: scupdate() called from scrn_timer(). - Make sure that the screen is updated for the low-level console routines sccngetc() and sccncheckc(). A new routine, sccnupdate(), is introduced and will call scupdate() above. Requested by: bde and msmith OKed by: sos
1998-02-11 14:58:15 +00:00
}
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
int
add_scrn_saver(void (*this_saver)(int))
{
if (current_saver != none_saver)
return EBUSY;
current_saver = this_saver;
return 0;
}
int
remove_scrn_saver(void (*this_saver)(int))
{
if (current_saver != this_saver)
return EINVAL;
/*
* In order to prevent `current_saver' from being called by
* the timeout routine `scrn_timer()' while we manipulate
* the saver list, we shall set `current_saver' to `none_saver'
* before stopping the current saver, rather than blocking by `splXX()'.
*/
current_saver = none_saver;
if (scrn_blanked > 0)
stop_scrn_saver(this_saver);
return 0;
}
static void
stop_scrn_saver(void (*saver)(int))
{
(*saver)(FALSE);
Time changes mark 2: * Figure out UTC relative to boottime. Four new functions provide time relative to boottime. * move "runtime" into struct proc. This helps fix the calcru() problem in SMP. * kill mono_time. * add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!) * nanosleep, select & poll takes long sleeps one day at a time Reviewed by: bde Tested by: ache and others
1998-04-04 13:26:20 +00:00
getmicroruntime(&scrn_time_stamp);
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
mark_all(cur_console);
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
wakeup((caddr_t)&scrn_blanked);
}
static int
wait_scrn_saver_stop(void)
{
int error = 0;
getmicroruntime(&scrn_time_stamp);
while (scrn_blanked > 0) {
error = tsleep((caddr_t)&scrn_blanked, PZERO | PCATCH, "scrsav", 0);
getmicroruntime(&scrn_time_stamp);
if (error != ERESTART)
break;
}
return error;
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static void
Updated to latest 1.1.5.1+ version Reviewed by: Submitted by:
1994-08-17 08:51:59 +00:00
clear_screen(scr_stat *scp)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
move_crsr(scp, 0, 0);
Fixed a panic when switching to 40x25 mode, and cursor was beyond the new buffer.
1996-09-04 22:24:19 +00:00
scp->cursor_oldpos = scp->cursor_pos;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20], scp->scr_buf,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize * scp->ysize);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_all(scp);
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
remove_cutmarking(scp);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static int
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
switch_scr(scr_stat *scp, u_int next_scr)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (switch_in_progress && (cur_console->proc != pfind(cur_console->pid)))
switch_in_progress = FALSE;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
if (next_scr >= MAXCONS || switch_in_progress ||
(cur_console->smode.mode == VT_AUTO
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
&& cur_console->status & UNKNOWN_MODE)) {
do_bell(scp, BELL_PITCH, BELL_DURATION);
return EINVAL;
}
From: sos@login.dkuug.dk (S|ren Schmidt) Subject: syscons-1.3 Date: Sat, 29 Jan 94 23:33:50 MET But here is the (hopefully) final syscons-1.3.... .... I've changed sgetc so it works as the pccons parallel (it now uses a scgetc internally). [ There were a couple changes that Bruce Evans sent me that were applied to this version along with some changes that S'ren didn't incorporate into the final version. There will be only minor changes if anything from this version to his final release. ]
1994-02-01 06:22:29 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* is the wanted virtual console open ? */
if (next_scr) {
struct tty *tp = VIRTUAL_TTY(next_scr);
if (!(tp->t_state & TS_ISOPEN)) {
do_bell(scp, BELL_PITCH, BELL_DURATION);
return EINVAL;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
/* delay switch if actively updating screen */
if (write_in_progress || blink_in_progress) {
delayed_next_scr = next_scr+1;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
return 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
switch_in_progress = TRUE;
old_scp = cur_console;
new_scp = console[next_scr];
wakeup((caddr_t)&new_scp->smode);
if (new_scp == old_scp) {
switch_in_progress = FALSE;
delayed_next_scr = FALSE;
return 0;
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* has controlling process died? */
if (old_scp->proc && (old_scp->proc != pfind(old_scp->pid)))
old_scp->smode.mode = VT_AUTO;
if (new_scp->proc && (new_scp->proc != pfind(new_scp->pid)))
new_scp->smode.mode = VT_AUTO;
Hawaii-Five-Typo
1996-12-19 00:26:22 +00:00
/* check the modes and switch appropriately */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (old_scp->smode.mode == VT_PROCESS) {
old_scp->status |= SWITCH_WAIT_REL;
psignal(old_scp->proc, old_scp->smode.relsig);
}
else {
exchange_scr();
if (new_scp->smode.mode == VT_PROCESS) {
new_scp->status |= SWITCH_WAIT_ACQ;
psignal(new_scp->proc, new_scp->smode.acqsig);
}
else
switch_in_progress = FALSE;
}
return 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static void
exchange_scr(void)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
move_crsr(old_scp, old_scp->xpos, old_scp->ypos);
cur_console = new_scp;
if (old_scp->mode != new_scp->mode || (old_scp->status & UNKNOWN_MODE)){
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
if (crtc_vga)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
set_mode(new_scp);
}
move_crsr(new_scp, new_scp->xpos, new_scp->ypos);
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
if (!(new_scp->status & UNKNOWN_MODE) && (flags & CHAR_CURSOR))
set_destructive_cursor(new_scp);
if ((old_scp->status & UNKNOWN_MODE) && crtc_vga)
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
load_palette(palette);
Add a new keyboard mode K_CODE. Returns a single byte for each key much like the scancode mode. However the keys that (for no good reason) returns extension codes etc, are translated into singlebyte codes. Needed by libvgl. This makes life ALOT easier, also the XFree86 folks could use this.
1997-10-01 20:46:29 +00:00
if (old_scp->status & KBD_RAW_MODE || new_scp->status & KBD_RAW_MODE ||
old_scp->status & KBD_CODE_MODE || new_scp->status & KBD_CODE_MODE)
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
shfts = ctls = alts = agrs = metas = accents = 0;
- Slightly change the way the border color register is updated so that flicker won't occur when set_border() is called. - Properly restore the border color when switching virtual consoles. Pointed out by: tony@dell.com OKed by: sos
1997-10-26 07:36:13 +00:00
set_border(new_scp->border);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
update_leds(new_scp->status);
delayed_next_scr = FALSE;
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
mark_all(new_scp);
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
if (vesa_mode == 0x102) {
bzero(Crtat, 800*600/8);
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static void
Updated to latest 1.1.5.1+ version Reviewed by: Submitted by:
1994-08-17 08:51:59 +00:00
scan_esc(scr_stat *scp, u_char c)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static u_char ansi_col[16] =
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
{0, 4, 2, 6, 1, 5, 3, 7, 8, 12, 10, 14, 9, 13, 11, 15};
int i, n;
u_short *src, *dst, count;
tiny update of ESC sequence parser. - some addition of comments (for readability) - iso-2022 G0 designation support. This does almost nothing. Just for avoiding garbled screen when got "ESC ( B". (how about G1/2/3 designation? I'm not sure)
1997-09-26 15:27:55 +00:00
if (scp->term.esc == 1) { /* seen ESC */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
switch (c) {
Add save/restore cursor as pr SCO screen(HW) manpage. Fix ESC[2J to not move cursor home Clear mouse cutmarking on more cases. Minor changes by me. Submitted by: ache
1997-01-23 20:00:45 +00:00
case '7': /* Save cursor position */
scp->saved_xpos = scp->xpos;
scp->saved_ypos = scp->ypos;
break;
case '8': /* Restore saved cursor position */
if (scp->saved_xpos >= 0 && scp->saved_ypos >= 0)
move_crsr(scp, scp->saved_xpos, scp->saved_ypos);
break;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case '[': /* Start ESC [ sequence */
scp->term.esc = 2;
scp->term.last_param = -1;
for (i = scp->term.num_param; i < MAX_ESC_PAR; i++)
scp->term.param[i] = 1;
scp->term.num_param = 0;
return;
case 'M': /* Move cursor up 1 line, scroll if at top */
if (scp->ypos > 0)
move_crsr(scp, scp->xpos, scp->ypos - 1);
else {
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(scp->scr_buf, scp->scr_buf + scp->xsize,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(scp->ysize - 1) * scp->xsize * sizeof(u_short));
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20],
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->scr_buf, scp->xsize);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_all(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
break;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
#if notyet
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case 'Q':
scp->term.esc = 4;
tiny update of ESC sequence parser. - some addition of comments (for readability) - iso-2022 G0 designation support. This does almost nothing. Just for avoiding garbled screen when got "ESC ( B". (how about G1/2/3 designation? I'm not sure)
1997-09-26 15:27:55 +00:00
return;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
#endif
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case 'c': /* Clear screen & home */
clear_screen(scp);
break;
tiny update of ESC sequence parser. - some addition of comments (for readability) - iso-2022 G0 designation support. This does almost nothing. Just for avoiding garbled screen when got "ESC ( B". (how about G1/2/3 designation? I'm not sure)
1997-09-26 15:27:55 +00:00
case '(': /* iso-2022: designate 94 character set to G0 */
scp->term.esc = 5;
return;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
tiny update of ESC sequence parser. - some addition of comments (for readability) - iso-2022 G0 designation support. This does almost nothing. Just for avoiding garbled screen when got "ESC ( B". (how about G1/2/3 designation? I'm not sure)
1997-09-26 15:27:55 +00:00
else if (scp->term.esc == 2) { /* seen ESC [ */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (c >= '0' && c <= '9') {
if (scp->term.num_param < MAX_ESC_PAR) {
if (scp->term.last_param != scp->term.num_param) {
scp->term.last_param = scp->term.num_param;
scp->term.param[scp->term.num_param] = 0;
}
else
scp->term.param[scp->term.num_param] *= 10;
scp->term.param[scp->term.num_param] += c - '0';
return;
}
}
scp->term.num_param = scp->term.last_param + 1;
switch (c) {
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case ';':
if (scp->term.num_param < MAX_ESC_PAR)
return;
break;
case '=':
scp->term.esc = 3;
scp->term.last_param = -1;
for (i = scp->term.num_param; i < MAX_ESC_PAR; i++)
scp->term.param[i] = 1;
scp->term.num_param = 0;
return;
case 'A': /* up n rows */
n = scp->term.param[0]; if (n < 1) n = 1;
move_crsr(scp, scp->xpos, scp->ypos - n);
break;
case 'B': /* down n rows */
n = scp->term.param[0]; if (n < 1) n = 1;
move_crsr(scp, scp->xpos, scp->ypos + n);
break;
case 'C': /* right n columns */
n = scp->term.param[0]; if (n < 1) n = 1;
move_crsr(scp, scp->xpos + n, scp->ypos);
break;
case 'D': /* left n columns */
n = scp->term.param[0]; if (n < 1) n = 1;
move_crsr(scp, scp->xpos - n, scp->ypos);
break;
case 'E': /* cursor to start of line n lines down */
n = scp->term.param[0]; if (n < 1) n = 1;
move_crsr(scp, 0, scp->ypos + n);
break;
case 'F': /* cursor to start of line n lines up */
n = scp->term.param[0]; if (n < 1) n = 1;
move_crsr(scp, 0, scp->ypos - n);
break;
case 'f': /* Cursor move */
case 'H':
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
if (scp->term.num_param == 0)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
move_crsr(scp, 0, 0);
else if (scp->term.num_param == 2)
move_crsr(scp, scp->term.param[1] - 1, scp->term.param[0] - 1);
break;
case 'J': /* Clear all or part of display */
if (scp->term.num_param == 0)
n = 0;
else
n = scp->term.param[0];
switch (n) {
case 0: /* clear form cursor to end of display */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20],
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
scp->cursor_pos,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->scr_buf + scp->xsize * scp->ysize - scp->cursor_pos);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
mark_for_update(scp, scp->xsize * scp->ysize);
Add save/restore cursor as pr SCO screen(HW) manpage. Fix ESC[2J to not move cursor home Clear mouse cutmarking on more cases. Minor changes by me. Submitted by: ache
1997-01-23 20:00:45 +00:00
remove_cutmarking(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 1: /* clear from beginning of display to cursor */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20],
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
scp->scr_buf,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->cursor_pos - scp->scr_buf);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, 0);
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
Add save/restore cursor as pr SCO screen(HW) manpage. Fix ESC[2J to not move cursor home Clear mouse cutmarking on more cases. Minor changes by me. Submitted by: ache
1997-01-23 20:00:45 +00:00
remove_cutmarking(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 2: /* clear entire display */
Add save/restore cursor as pr SCO screen(HW) manpage. Fix ESC[2J to not move cursor home Clear mouse cutmarking on more cases. Minor changes by me. Submitted by: ache
1997-01-23 20:00:45 +00:00
fillw(scp->term.cur_color | scr_map[0x20], scp->scr_buf,
scp->xsize * scp->ysize);
mark_all(scp);
remove_cutmarking(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
}
break;
case 'K': /* Clear all or part of line */
if (scp->term.num_param == 0)
n = 0;
else
n = scp->term.param[0];
switch (n) {
case 0: /* clear form cursor to end of line */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20],
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
scp->cursor_pos,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize - scp->xpos);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
mark_for_update(scp, scp->cursor_pos - scp->scr_buf +
scp->xsize - scp->xpos);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 1: /* clear from beginning of line to cursor */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20],
Fix ^[[1K (clear from beginning of line to cursor) and ^[[2K (clear whole line), they not work _initially_
1996-05-02 21:47:50 +00:00
scp->cursor_pos - scp->xpos,
scp->xpos + 1);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->ypos * scp->xsize);
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 2: /* clear entire line */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20],
Fix ^[[1K (clear from beginning of line to cursor) and ^[[2K (clear whole line), they not work _initially_
1996-05-02 21:47:50 +00:00
scp->cursor_pos - scp->xpos,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->ypos * scp->xsize);
mark_for_update(scp, (scp->ypos + 1) * scp->xsize);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
}
break;
case 'L': /* Insert n lines */
n = scp->term.param[0]; if (n < 1) n = 1;
if (n > scp->ysize - scp->ypos)
n = scp->ysize - scp->ypos;
src = scp->scr_buf + scp->ypos * scp->xsize;
dst = src + n * scp->xsize;
count = scp->ysize - (scp->ypos + n);
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(src, dst, count * scp->xsize * sizeof(u_short));
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20], src,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
n * scp->xsize);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->ypos * scp->xsize);
mark_for_update(scp, scp->xsize * scp->ysize);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 'M': /* Delete n lines */
n = scp->term.param[0]; if (n < 1) n = 1;
if (n > scp->ysize - scp->ypos)
n = scp->ysize - scp->ypos;
dst = scp->scr_buf + scp->ypos * scp->xsize;
src = dst + n * scp->xsize;
count = scp->ysize - (scp->ypos + n);
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(src, dst, count * scp->xsize * sizeof(u_short));
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
src = dst + count * scp->xsize;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20], src,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
n * scp->xsize);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->ypos * scp->xsize);
mark_for_update(scp, scp->xsize * scp->ysize);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 'P': /* Delete n chars */
n = scp->term.param[0]; if (n < 1) n = 1;
if (n > scp->xsize - scp->xpos)
n = scp->xsize - scp->xpos;
dst = scp->cursor_pos;
src = dst + n;
count = scp->xsize - (scp->xpos + n);
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(src, dst, count * sizeof(u_short));
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
src = dst + count;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20], src, n);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf + n + count);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case '@': /* Insert n chars */
n = scp->term.param[0]; if (n < 1) n = 1;
if (n > scp->xsize - scp->xpos)
n = scp->xsize - scp->xpos;
src = scp->cursor_pos;
dst = src + n;
count = scp->xsize - (scp->xpos + n);
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(src, dst, count * sizeof(u_short));
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20], src, n);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
Fixes to the hardware cursor emulation. Submitted by: ache
1995-04-04 20:06:26 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf + n + count);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 'S': /* scroll up n lines */
n = scp->term.param[0]; if (n < 1) n = 1;
if (n > scp->ysize)
n = scp->ysize;
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(scp->scr_buf + (scp->xsize * n),
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
scp->scr_buf,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize * (scp->ysize - n) * sizeof(u_short));
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20],
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
scp->scr_buf + scp->xsize * (scp->ysize - n),
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize * n);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_all(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 'T': /* scroll down n lines */
n = scp->term.param[0]; if (n < 1) n = 1;
if (n > scp->ysize)
n = scp->ysize;
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(scp->scr_buf,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->scr_buf + (scp->xsize * n),
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
scp->xsize * (scp->ysize - n) *
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
sizeof(u_short));
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20],
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->scr_buf, scp->xsize * n);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_all(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
Fix count in mark_for_update() for insert-char(s) and delete-char(s). Everything from the cursor to the end of the line must be updated. Fix comment about erase-char(s).
1995-04-01 19:57:36 +00:00
case 'X': /* erase n characters in line */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
n = scp->term.param[0]; if (n < 1) n = 1;
if (n > scp->xsize - scp->xpos)
n = scp->xsize - scp->xpos;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20],
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
scp->cursor_pos, n);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
mark_for_update(scp, scp->cursor_pos - scp->scr_buf + n);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 'Z': /* move n tabs backwards */
n = scp->term.param[0]; if (n < 1) n = 1;
if ((i = scp->xpos & 0xf8) == scp->xpos)
i -= 8*n;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
else
i -= 8*(n-1);
if (i < 0)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
i = 0;
move_crsr(scp, i, scp->ypos);
break;
case '`': /* move cursor to column n */
n = scp->term.param[0]; if (n < 1) n = 1;
move_crsr(scp, n - 1, scp->ypos);
break;
case 'a': /* move cursor n columns to the right */
n = scp->term.param[0]; if (n < 1) n = 1;
move_crsr(scp, scp->xpos + n, scp->ypos);
break;
case 'd': /* move cursor to row n */
n = scp->term.param[0]; if (n < 1) n = 1;
move_crsr(scp, scp->xpos, n - 1);
break;
case 'e': /* move cursor n rows down */
n = scp->term.param[0]; if (n < 1) n = 1;
move_crsr(scp, scp->xpos, scp->ypos + n);
break;
case 'm': /* change attribute */
if (scp->term.num_param == 0) {
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.attr_mask = NORMAL_ATTR;
scp->term.cur_attr =
scp->term.cur_color = scp->term.std_color;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
}
for (i = 0; i < scp->term.num_param; i++) {
switch (n = scp->term.param[i]) {
case 0: /* back to normal */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.attr_mask = NORMAL_ATTR;
scp->term.cur_attr =
scp->term.cur_color = scp->term.std_color;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
case 1: /* bold */
scp->term.attr_mask |= BOLD_ATTR;
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
case 4: /* underline */
scp->term.attr_mask |= UNDERLINE_ATTR;
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 5: /* blink */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.attr_mask |= BLINK_ATTR;
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 7: /* reverse video */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.attr_mask |= REVERSE_ATTR;
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 30: case 31: /* set fg color */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
case 32: case 33: case 34:
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case 35: case 36: case 37:
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.attr_mask |= FOREGROUND_CHANGED;
scp->term.cur_color =
(scp->term.cur_color&0xF000) | (ansi_col[(n-30)&7]<<8);
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 40: case 41: /* set bg color */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
case 42: case 43: case 44:
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case 45: case 46: case 47:
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.attr_mask |= BACKGROUND_CHANGED;
scp->term.cur_color =
(scp->term.cur_color&0x0F00) | (ansi_col[(n-40)&7]<<12);
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
break;
Add save/restore cursor as pr SCO screen(HW) manpage. Fix ESC[2J to not move cursor home Clear mouse cutmarking on more cases. Minor changes by me. Submitted by: ache
1997-01-23 20:00:45 +00:00
case 's': /* Save cursor position */
scp->saved_xpos = scp->xpos;
scp->saved_ypos = scp->ypos;
break;
case 'u': /* Restore saved cursor position */
if (scp->saved_xpos >= 0 && scp->saved_ypos >= 0)
move_crsr(scp, scp->saved_xpos, scp->saved_ypos);
break;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case 'x':
if (scp->term.num_param == 0)
n = 0;
else
n = scp->term.param[0];
switch (n) {
case 0: /* reset attributes */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.attr_mask = NORMAL_ATTR;
scp->term.cur_attr =
scp->term.cur_color = scp->term.std_color =
current_default->std_color;
scp->term.rev_color = current_default->rev_color;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 1: /* set ansi background */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.attr_mask &= ~BACKGROUND_CHANGED;
scp->term.cur_color = scp->term.std_color =
(scp->term.std_color & 0x0F00) |
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(ansi_col[(scp->term.param[1])&0x0F]<<12);
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 2: /* set ansi foreground */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.attr_mask &= ~FOREGROUND_CHANGED;
scp->term.cur_color = scp->term.std_color =
(scp->term.std_color & 0xF000) |
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(ansi_col[(scp->term.param[1])&0x0F]<<8);
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 3: /* set ansi attribute directly */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.attr_mask &= ~(FOREGROUND_CHANGED|BACKGROUND_CHANGED);
scp->term.cur_color = scp->term.std_color =
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(scp->term.param[1]&0xFF)<<8;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 5: /* set ansi reverse video background */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.rev_color =
(scp->term.rev_color & 0x0F00) |
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(ansi_col[(scp->term.param[1])&0x0F]<<12);
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 6: /* set ansi reverse video foreground */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.rev_color =
(scp->term.rev_color & 0xF000) |
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(ansi_col[(scp->term.param[1])&0x0F]<<8);
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 7: /* set ansi reverse video directly */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.rev_color =
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(scp->term.param[1]&0xFF)<<8;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.cur_attr = mask2attr(&scp->term);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
}
break;
case 'z': /* switch to (virtual) console n */
if (scp->term.num_param == 1)
switch_scr(scp, scp->term.param[0]);
break;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
tiny update of ESC sequence parser. - some addition of comments (for readability) - iso-2022 G0 designation support. This does almost nothing. Just for avoiding garbled screen when got "ESC ( B". (how about G1/2/3 designation? I'm not sure)
1997-09-26 15:27:55 +00:00
else if (scp->term.esc == 3) { /* seen ESC [0-9]+ = */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (c >= '0' && c <= '9') {
if (scp->term.num_param < MAX_ESC_PAR) {
if (scp->term.last_param != scp->term.num_param) {
scp->term.last_param = scp->term.num_param;
scp->term.param[scp->term.num_param] = 0;
}
else
scp->term.param[scp->term.num_param] *= 10;
scp->term.param[scp->term.num_param] += c - '0';
return;
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->term.num_param = scp->term.last_param + 1;
switch (c) {
case ';':
if (scp->term.num_param < MAX_ESC_PAR)
return;
break;
case 'A': /* set display border color */
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
if (scp->term.num_param == 1) {
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->border=scp->term.param[0] & 0xff;
if (scp == cur_console)
set_border(scp->border);
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 'B': /* set bell pitch and duration */
if (scp->term.num_param == 2) {
scp->bell_pitch = scp->term.param[0];
scp->bell_duration = scp->term.param[1]*10;
}
break;
case 'C': /* set cursor type & shape */
if (scp->term.num_param == 1) {
if (scp->term.param[0] & 0x01)
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
flags |= BLINK_CURSOR;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
else
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
flags &= ~BLINK_CURSOR;
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
if ((scp->term.param[0] & 0x02) && crtc_vga)
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
flags |= CHAR_CURSOR;
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
else
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
flags &= ~CHAR_CURSOR;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
else if (scp->term.num_param == 2) {
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
scp->cursor_start = scp->term.param[0] & 0x1F;
scp->cursor_end = scp->term.param[1] & 0x1F;
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
}
/*
* The cursor shape is global property; all virtual consoles
* are affected. Update the cursor in the current console...
*/
if (!(cur_console->status & UNKNOWN_MODE)) {
remove_cursor_image(cur_console);
if (crtc_vga && (flags & CHAR_CURSOR))
set_destructive_cursor(cur_console);
draw_cursor_image(cur_console);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
break;
case 'F': /* set ansi foreground */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
if (scp->term.num_param == 1) {
scp->term.attr_mask &= ~FOREGROUND_CHANGED;
scp->term.cur_color = scp->term.std_color =
(scp->term.std_color & 0xF000)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
| ((scp->term.param[0] & 0x0F) << 8);
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.cur_attr = mask2attr(&scp->term);
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 'G': /* set ansi background */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
if (scp->term.num_param == 1) {
scp->term.attr_mask &= ~BACKGROUND_CHANGED;
scp->term.cur_color = scp->term.std_color =
(scp->term.std_color & 0x0F00)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
| ((scp->term.param[0] & 0x0F) << 12);
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.cur_attr = mask2attr(&scp->term);
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 'H': /* set ansi reverse video foreground */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
if (scp->term.num_param == 1) {
scp->term.rev_color =
(scp->term.rev_color & 0xF000)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
| ((scp->term.param[0] & 0x0F) << 8);
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.cur_attr = mask2attr(&scp->term);
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 'I': /* set ansi reverse video background */
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
if (scp->term.num_param == 1) {
scp->term.rev_color =
(scp->term.rev_color & 0x0F00)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
| ((scp->term.param[0] & 0x0F) << 12);
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.cur_attr = mask2attr(&scp->term);
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
}
}
tiny update of ESC sequence parser. - some addition of comments (for readability) - iso-2022 G0 designation support. This does almost nothing. Just for avoiding garbled screen when got "ESC ( B". (how about G1/2/3 designation? I'm not sure)
1997-09-26 15:27:55 +00:00
#if notyet
else if (scp->term.esc == 4) { /* seen ESC Q */
/* to be filled */
}
#endif
else if (scp->term.esc == 5) { /* seen ESC ( */
switch (c) {
case 'B': /* iso-2022: desginate ASCII into G0 */
break;
/* other items to be filled */
default:
break;
}
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->term.esc = 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static void
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
ansi_put(scr_stat *scp, u_char *buf, int len)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
u_char *ptr = buf;
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* make screensaver happy */
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
if (scp == cur_console)
Time changes mark 2: * Figure out UTC relative to boottime. Four new functions provide time relative to boottime. * move "runtime" into struct proc. This helps fix the calcru() problem in SMP. * kill mono_time. * add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!) * nanosleep, select & poll takes long sleeps one day at a time Reviewed by: bde Tested by: ache and others
1998-04-04 13:26:20 +00:00
getmicroruntime(&scrn_time_stamp);
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
write_in_progress++;
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
outloop:
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (scp->term.esc) {
scan_esc(scp, *ptr++);
len--;
}
else if (PRINTABLE(*ptr)) { /* Print only printables */
Minor update to syscons. Let "grey delete" be a function key (default is 0x7f) Fix the xor cursor again.. Made the backspace key generate del as default Made CTRL-space generate nul as default.
1995-03-03 08:37:52 +00:00
int cnt = len <= (scp->xsize-scp->xpos) ? len : (scp->xsize-scp->xpos);
u_short cur_attr = scp->term.cur_attr;
u_short *cursor_pos = scp->cursor_pos;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
do {
Speed up the inner loop of ansi_put() by a few percent. syscons' output is now only about 4-5 times slower than I want. It loses a factor of 2 for scrolling output by unnecessarily copying the screen buffer, a factor of 4/3 for dumb OPOST processing, and a factor of 3/2 for clist processing.
1995-07-11 17:59:22 +00:00
/*
* gcc-2.6.3 generates poor (un)sign extension code. Casting the
* pointers in the following to volatile should have no effect,
* but in fact speeds up this inner loop from 26 to 18 cycles
* (+ cache misses) on i486's.
*/
#define UCVP(ucp) ((u_char volatile *)(ucp))
*cursor_pos++ = UCVP(scr_map)[*UCVP(ptr)] | cur_attr;
ptr++;
Minor update to syscons. Let "grey delete" be a function key (default is 0x7f) Fix the xor cursor again.. Made the backspace key generate del as default Made CTRL-space generate nul as default.
1995-03-03 08:37:52 +00:00
cnt--;
} while (cnt && PRINTABLE(*ptr));
len -= (cursor_pos - scp->cursor_pos);
scp->xpos += (cursor_pos - scp->cursor_pos);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
mark_for_update(scp, cursor_pos - scp->scr_buf);
Minor update to syscons. Let "grey delete" be a function key (default is 0x7f) Fix the xor cursor again.. Made the backspace key generate del as default Made CTRL-space generate nul as default.
1995-03-03 08:37:52 +00:00
scp->cursor_pos = cursor_pos;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (scp->xpos >= scp->xsize) {
scp->xpos = 0;
scp->ypos++;
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
else {
switch(*ptr) {
case 0x07:
do_bell(scp, scp->bell_pitch, scp->bell_duration);
break;
case 0x08: /* non-destructive backspace */
if (scp->cursor_pos > scp->scr_buf) {
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->cursor_pos--;
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (scp->xpos > 0)
scp->xpos--;
else {
scp->xpos += scp->xsize - 1;
scp->ypos--;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
break;
case 0x09: /* non-destructive tab */
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
scp->cursor_pos += (8 - scp->xpos % 8u);
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
if ((scp->xpos += (8 - scp->xpos % 8u)) >= scp->xsize) {
scp->xpos = 0;
scp->ypos++;
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case 0x0a: /* newline, same pos */
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->cursor_pos += scp->xsize;
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->ypos++;
break;
case 0x0c: /* form feed, clears screen */
clear_screen(scp);
break;
case 0x0d: /* return, return to pos 0 */
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->cursor_pos -= scp->xpos;
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_for_update(scp, scp->cursor_pos - scp->scr_buf);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xpos = 0;
break;
case 0x1b: /* start escape sequence */
scp->term.esc = 1;
scp->term.num_param = 0;
break;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
ptr++; len--;
}
/* do we have to scroll ?? */
if (scp->cursor_pos >= scp->scr_buf + scp->ysize * scp->xsize) {
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
remove_cutmarking(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (scp->history) {
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(scp->scr_buf, scp->history_head,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize * sizeof(u_short));
scp->history_head += scp->xsize;
if (scp->history_head + scp->xsize >
scp->history + scp->history_size)
scp->history_head = scp->history;
}
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(scp->scr_buf + scp->xsize, scp->scr_buf,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize * (scp->ysize - 1) * sizeof(u_short));
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(scp->term.cur_color | scr_map[0x20],
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
scp->scr_buf + scp->xsize * (scp->ysize - 1),
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize);
scp->cursor_pos -= scp->xsize;
scp->ypos--;
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_all(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
if (len)
goto outloop;
write_in_progress--;
if (delayed_next_scr)
switch_scr(scp, delayed_next_scr - 1);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static void
Updated to latest 1.1.5.1+ version Reviewed by: Submitted by:
1994-08-17 08:51:59 +00:00
scinit(void)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Corrected the setting of winsize on open(). Problem pointed out by David Nugent - davidn@blaze.net.au.
1996-10-15 20:27:07 +00:00
u_int hw_cursor;
u_int i;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
if (init_done != COLD)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
init_done = WARM;
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
/*
* Ensure a zero start address. This is mainly to recover after
Fixed a comment.
1995-11-24 14:56:00 +00:00
* switching from pcvt using userconfig(). The registers are w/o
Make pcvt and syscons live in the same kernel. If both are enabled, then the first one in the config has priority. They can be switched using userconfig(). i386/i386/conf.c: Initialize the shared syscons/pcvt cdevsw entry to `nx'. Add cdevsw registration functions. Use devsw functions of the correct type if they exist. i386/i386/cons.c: Add renamed syscons entry points to constab. i386/i386/cons.h: Declare the renamed syscons entry points. i386/i386/machdep.c: Repeat console initialization after userconfig() in case the current console has become wrong. This depends on cn functions not wiring down anything important. sys/conf.h: Declare new functions. i386/isa/isa.[ch]: Add a function to decide which display driver has priority. Should be done better. i386/isa/syscons.c: Rename pccn* -> sccn*. Initialize CRTC start address in case the previous driver has moved it. i386/isa/syscons.c, i386/isa/pcvt/* Initialize the bogusly shared variable Crtat dynamically in case the stored value was changed by the previous driver. Initialize cdevsw table from a template. Don't grab the console if another display driver has priority. i386/isa/syscons.h, i386/isa/pcvt/pcvt_hdr.h: Don't externally declare now-static cdevsw functions. i386/isa/pcvt/pcvt_hdr.h: Set the sensitive hardware flag so that pcvt doesn't always have lower priority than syscons. This also fixes the "stupid" detection of the display after filling the display with text. i386/isa/pcvt/pcvt_out.c: Don't be confused the off-screen cursor offset 0xffff set by syscons. kern/subr_xxx.c: Add enough nxio/nodev/null devsw functions of the correct type for syscons and pcvt.
1995-09-10 21:36:12 +00:00
* for old hardware so it's too hard to relocate the active screen
* memory.
*/
outb(crtc_addr, 12);
outb(crtc_addr + 1, 0);
outb(crtc_addr, 13);
outb(crtc_addr + 1, 0);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* extract cursor location */
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
outb(crtc_addr, 14);
hw_cursor = inb(crtc_addr + 1) << 8;
outb(crtc_addr, 15);
hw_cursor |= inb(crtc_addr + 1);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
1. Avoid a race in scclose(). tty.c has kludges so that the race is actually harmless. 2. Fixed code to match comment in scintr(). 3. Don't allow even root to take control of the machine when securelevel > 0. I've secured the accesses to PSL_IOPL in all drivers and asked pst to review it, but he seems to be busy. Write access to /dev/kmem and other critival devices currently leaks across raisings of securelevel via open fd's, so there may as well be a similar leak for PSL_IOPL. 4. (Most important.) Don't corrupt memory beyond the screen buffers if the cursor happens to be off the 80x25 screen when syscons starts. 5. Fix console cursor update (not perfect yet). Submitted by: bruce ~
1996-11-11 22:21:03 +00:00
/*
* Validate cursor location. It may be off the screen. Then we must
* not use it for the initial buffer offset.
*/
if (hw_cursor >= ROW * COL)
hw_cursor = (ROW - 1) * COL;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* move hardware cursor out of the way */
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
outb(crtc_addr, 14);
outb(crtc_addr + 1, 0xff);
outb(crtc_addr, 15);
outb(crtc_addr + 1, 0xff);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
The syscons driver doesn't really check the presence of the display adapter during the system boot. It always assumes there is at least a monochrome adapter. This is rather strange assumption. If there is no dispaly adapter, the console driver cannot be any good... In this patch, scinit() is split into two parts; the first part is now called scvidprobe() which will detect the presence of video card at the CGA or MONO buffer address and returns TRUE if found. It is called during sccnprobe() and scprobe(). Both will fail if no video card is found. The second part, whose name stays the same as before, scinit(), is called from sccninit() and scattach() to complete initialization of the found video card. The keyboard probe code is moved from scprobe() to sckbdprobe(); scprobe() now calls scvidprobe() and sckbdprobe() to carry out device probe. (This is rather a cosmetic change, but it sure makes the code look better organized.) The problem pointed out by Joerg.
1997-06-22 12:04:36 +00:00
/* set up the first console */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
current_default = &user_default;
console[0] = &main_console;
init_scp(console[0]);
Bruce pointed out a bogon: cur_console is NULL when copy_font() is first called from scinit(). This is apparently harmless when scinit() is called early from sccninit() - page 0 is apparently mapped r/w then, and 0->status contains suitable garbage. However, when there is a serial console, scinit() is first called from scattach() when the page tables are completely initialized, so the NULL pointer causes a panic. Submitted by: bruce
1997-01-17 15:49:13 +00:00
cur_console = console[0];
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
/* discard the video mode table if we are not familiar with it... */
if (video_mode_ptr) {
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
bzero(mode_map, sizeof(mode_map));
bcopy(video_mode_ptr + MODE_PARAM_SIZE*console[0]->mode,
vgaregs2, sizeof(vgaregs2));
switch (comp_vgaregs(vgaregs, video_mode_ptr
+ MODE_PARAM_SIZE*console[0]->mode)) {
case COMP_IDENTICAL:
map_mode_table(mode_map, video_mode_ptr, M_VGA_CG320 + 1);
/*
* This is a kludge for Toshiba DynaBook SS433 whose BIOS video
* mode table entry has the actual # of rows at the offset 1;
* BIOSes from other manufacturers store the # of rows - 1 there.
* XXX
*/
rows_offset = vgaregs[1] + 1
- video_mode_ptr[MODE_PARAM_SIZE*console[0]->mode + 1];
break;
case COMP_SIMILAR:
map_mode_table(mode_map, video_mode_ptr, M_VGA_CG320 + 1);
mode_map[console[0]->mode] = vgaregs;
rows_offset = vgaregs[1] + 1
- video_mode_ptr[MODE_PARAM_SIZE*console[0]->mode + 1];
vgaregs[1] -= rows_offset - 1;
break;
case COMP_DIFFERENT:
default:
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
video_mode_ptr = NULL;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
mode_map[console[0]->mode] = vgaregs;
rows_offset = 1;
break;
}
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
}
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
/* copy screen to temporary buffer */
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
if (crtc_type != KD_PIXEL)
generic_bcopy(Crtat, sc_buffer,
console[0]->xsize * console[0]->ysize * sizeof(u_short));
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
console[0]->scr_buf = console[0]->mouse_pos = console[0]->mouse_oldpos
= sc_buffer;
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
console[0]->cursor_pos = console[0]->cursor_oldpos = sc_buffer + hw_cursor;
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
console[0]->cursor_saveunder = *console[0]->cursor_pos;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
console[0]->xpos = hw_cursor % COL;
console[0]->ypos = hw_cursor / COL;
for (i=1; i<MAXCONS; i++)
console[i] = NULL;
kernel_console.esc = 0;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
kernel_console.attr_mask = NORMAL_ATTR;
kernel_console.cur_attr =
kernel_console.cur_color = kernel_console.std_color =
kernel_default.std_color;
kernel_console.rev_color = kernel_default.rev_color;
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
/* initialize mapscrn arrays to a one to one map */
for (i=0; i<sizeof(scr_map); i++) {
scr_map[i] = scr_rmap[i] = i;
}
Fixed the "missing updates" reported by ache. Moved a little closer to having a splash page capability.
1996-10-01 23:24:09 +00:00
Oops, the position of the savefont code was wrong, cur_console wasn't set yet.
1997-01-19 17:34:14 +00:00
/* Save font and palette if VGA */
if (crtc_vga) {
Add option for compiling in a 8x16 font.
1997-07-25 11:53:30 +00:00
if (fonts_loaded & FONT_16) {
copy_font(LOAD, FONT_16, font_16);
} else {
copy_font(SAVE, FONT_16, font_16);
fonts_loaded = FONT_16;
}
Oops, the position of the savefont code was wrong, cur_console wasn't set yet.
1997-01-19 17:34:14 +00:00
save_palette();
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
set_destructive_cursor(console[0]);
Oops, the position of the savefont code was wrong, cur_console wasn't set yet.
1997-01-19 17:34:14 +00:00
}
Fixed the "missing updates" reported by ache. Moved a little closer to having a splash page capability.
1996-10-01 23:24:09 +00:00
#ifdef SC_SPLASH_SCREEN
/*
* Now put up a graphics image, and maybe cycle a
* couble of palette entries for simple animation.
*/
toggle_splash_screen(cur_console);
#endif
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
static void
map_mode_table(char *map[], char *table, int max)
{
int i;
for(i = 0; i < max; ++i)
map[i] = table + i*MODE_PARAM_SIZE;
for(; i < MODE_MAP_SIZE; ++i)
map[i] = NULL;
}
static u_char
map_mode_num(u_char mode)
{
static struct {
u_char from;
u_char to;
} mode_map[] = {
{ M_ENH_B80x43, M_ENH_B80x25 },
{ M_ENH_C80x43, M_ENH_C80x25 },
{ M_VGA_M80x30, M_VGA_M80x25 },
{ M_VGA_C80x30, M_VGA_C80x25 },
{ M_VGA_M80x50, M_VGA_M80x25 },
{ M_VGA_C80x50, M_VGA_C80x25 },
{ M_VGA_M80x60, M_VGA_M80x25 },
{ M_VGA_C80x60, M_VGA_C80x25 },
{ M_VGA_MODEX, M_VGA_CG320 },
};
int i;
for (i = 0; i < sizeof(mode_map)/sizeof(mode_map[0]); ++i) {
if (mode_map[i].from == mode)
return mode_map[i].to;
}
return mode;
}
static char
*get_mode_param(scr_stat *scp, u_char mode)
{
if (mode >= MODE_MAP_SIZE)
mode = map_mode_num(mode);
if (mode < MODE_MAP_SIZE)
return mode_map[mode];
else
return NULL;
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static scr_stat
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
*alloc_scp()
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scr_stat *scp;
Fixed a problem that malloc(..,..,M_NOWAIT) was being called without checking for return values. It just so happens that in the cases where it is likely to fail, it is okay to change the M_NOWAIT to M_WAITOK -- and all will be well. This problem was manfest as a panic very regularly on a 4MB system right after bootup.
1995-08-08 05:14:40 +00:00
scp = (scr_stat *)malloc(sizeof(scr_stat), M_DEVBUF, M_WAITOK);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
init_scp(scp);
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
scp->scr_buf = scp->cursor_pos = scp->cursor_oldpos =
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(u_short *)malloc(scp->xsize*scp->ysize*sizeof(u_short),
Fixed a problem that malloc(..,..,M_NOWAIT) was being called without checking for return values. It just so happens that in the cases where it is likely to fail, it is okay to change the M_NOWAIT to M_WAITOK -- and all will be well. This problem was manfest as a panic very regularly on a 4MB system right after bootup.
1995-08-08 05:14:40 +00:00
M_DEVBUF, M_WAITOK);
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
scp->mouse_pos = scp->mouse_oldpos =
scp->scr_buf + ((scp->mouse_ypos/scp->font_size)*scp->xsize +
scp->mouse_xpos/8);
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
scp->history_head = scp->history_pos =
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
(u_short *)malloc(scp->history_size*sizeof(u_short),
Fixed a problem that malloc(..,..,M_NOWAIT) was being called without checking for return values. It just so happens that in the cases where it is likely to fail, it is okay to change the M_NOWAIT to M_WAITOK -- and all will be well. This problem was manfest as a panic very regularly on a 4MB system right after bootup.
1995-08-08 05:14:40 +00:00
M_DEVBUF, M_WAITOK);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
bzero(scp->history_head, scp->history_size*sizeof(u_short));
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
scp->history = scp->history_head;
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
/* SOS
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (crtc_vga && video_mode_ptr)
set_mode(scp);
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
*/
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
clear_screen(scp);
Fixes annoying behavior and a bug regarding the destructive cursor (CHAR_CURSOR) 1. Reduced the number of calls to set_destructive_cursor(). The destructive cursor produced noticeable overhead on the system. It was caused by draw_cursor_image() calling set_destructive_cursor() every so often. set_destructive_cursor() absolutely needs to be called when a) the character code under the cursor has changed either because the cursor moved or because the screen was updated or the mouse pointer overlapped the cursor. b) Or a new font has been loaded, c) or the video mode has been changed, d) or the cursor shape has been changed, e) or the user switched virtual consoles. 2. Turn off the configuration flag CHAR_CURSOR (destructive cursor) in scattach() if we have a non-VGA card. The destructive cursor works only for VGA. 3. Removed redundant calls to set_destructive_cursor() in some places. 4. Fixed the "disappearing mouse pointer" problem. The mouse pointer looked hidden under the destructive cursor when it overlaped the cursor. A slightly different version of the patch was reviewd and OKed by sos and ache.
1997-06-30 13:31:49 +00:00
scp->cursor_saveunder = *scp->cursor_pos;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return scp;
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
}
static void
init_scp(scr_stat *scp)
{
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
switch(crtc_type) {
case KD_VGA:
Fix the broken support for monochrome VGA's. It was MY fault after all, damn..
1996-12-07 22:33:05 +00:00
if (crtc_addr == MONO_BASE)
scp->mode = M_VGA_M80x25;
else
Bruce pointed out a bogon: cur_console is NULL when copy_font() is first called from scinit(). This is apparently harmless when scinit() is called early from sccninit() - page 0 is apparently mapped r/w then, and 0->status contains suitable garbage. However, when there is a serial console, scinit() is first called from scattach() when the page tables are completely initialized, so the NULL pointer causes a panic. Submitted by: bruce
1997-01-17 15:49:13 +00:00
scp->mode = M_VGA_C80x25;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
scp->font_size = 16;
break;
case KD_CGA:
if (crtc_addr == MONO_BASE)
scp->mode = M_B80x25;
else
scp->mode = M_C80x25;
scp->font_size = 8;
break;
case KD_EGA:
Fix the broken support for monochrome VGA's. It was MY fault after all, damn..
1996-12-07 22:33:05 +00:00
if (crtc_addr == MONO_BASE)
scp->mode = M_B80x25;
else
scp->mode = M_C80x25;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
scp->font_size = 14;
break;
case KD_MONO:
case KD_HERCULES:
default:
scp->mode = M_EGAMONO80x25;
scp->font_size = 14;
break;
}
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
scp->initial_mode = scp->mode;
Fix the broken support for monochrome VGA's. It was MY fault after all, damn..
1996-12-07 22:33:05 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->xsize = COL;
scp->ysize = ROW;
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
scp->xpixel = scp->xsize * 8;
scp->ypixel = scp->ysize * scp->font_size;
Add save/restore cursor as pr SCO screen(HW) manpage. Fix ESC[2J to not move cursor home Clear mouse cutmarking on more cases. Minor changes by me. Submitted by: ache
1997-01-23 20:00:45 +00:00
scp->xpos = scp->ypos = 0;
scp->saved_xpos = scp->saved_ypos = -1;
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
scp->start = scp->xsize * scp->ysize;
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
scp->end = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->term.esc = 0;
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
scp->term.attr_mask = NORMAL_ATTR;
scp->term.cur_attr =
scp->term.cur_color = scp->term.std_color =
current_default->std_color;
scp->term.rev_color = current_default->rev_color;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->border = BG_BLACK;
Emulate hw cursor closely, and get start&end scanlines from BIOS.
1995-03-30 14:32:31 +00:00
scp->cursor_start = *(char *)pa_to_va(0x461);
scp->cursor_end = *(char *)pa_to_va(0x460);
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
scp->mouse_xpos = scp->xsize*8/2;
scp->mouse_ypos = scp->ysize*scp->font_size/2;
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
scp->mouse_cut_start = scp->mouse_cut_end = NULL;
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
scp->mouse_signal = 0;
scp->mouse_pid = 0;
scp->mouse_proc = NULL;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->bell_pitch = BELL_PITCH;
scp->bell_duration = BELL_DURATION;
scp->status = (*(char *)pa_to_va(0x417) & 0x20) ? NLKED : 0;
scp->status |= CURSOR_ENABLED;
scp->pid = 0;
scp->proc = NULL;
scp->smode.mode = VT_AUTO;
scp->history_head = scp->history_pos = scp->history = NULL;
Reject unreasonable values passed to CONS_HISTORY ioctl. It did not check the value and caused kernel panic when a large value was given. - Move the configuration option SC_HISTORY_SIZE from syscons.h to syscons.c. - Define the maximum total number of history lines of all consoles. It is SC_HISTORY_SIZE*MAXCONS or 1000*MAXCONS; whichever is larger. CONS_HISTORY will allow the user to set the history size up to SC_HISTORY_SIZE unconditionally (or the current height of the console if it is larger than SC_HISTORY_SIZE). If the user requests a larger buffer, it will be granted only if the total number of all allocated history lines and the requested number of lines won't exceed the maximum. - Don't free the previous history buffer and leave the history buffer pointer holding a invalid pointer. Set the pointer to NULL first, then free the buffer. PR: bin/4592
1997-10-23 03:23:50 +00:00
scp->history_size = imax(SC_HISTORY_SIZE, scp->ysize) * scp->xsize;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static u_char
Updated to latest 1.1.5.1+ version Reviewed by: Submitted by:
1994-08-17 08:51:59 +00:00
*get_fstr(u_int c, u_int *len)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
u_int i;
if (!(c & FKEY))
return(NULL);
i = (c & 0xFF) - F_FN;
if (i > n_fkey_tab)
return(NULL);
*len = fkey_tab[i].len;
return(fkey_tab[i].str);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Third round in syscons update. Display update method changed, now allways write in memory buffer, then periodically update physical display. Speed improvements (now > 5 times faster than the old syscons). History now circular buffer, with changeable size. History scroll by up/down line, up/down page, home and end. Backtab proberly implemented. Now space for 96 function keys, 63 allocated standard, default now SCO/SYSV compat again as in the old days. New keyboard definition files ~share/syscons/keymaps/* Misc fixes for old "hacks" that broke SCO/SYSV compat. More that I forgot before writing this...
1995-01-28 22:18:05 +00:00
static void
history_to_screen(scr_stat *scp)
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
int i;
for (i=0; i<scp->ysize; i++)
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(scp->history + (((scp->history_pos - scp->history) +
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scp->history_size-((i+1)*scp->xsize))%scp->history_size),
scp->scr_buf + (scp->xsize * (scp->ysize-1 - i)),
scp->xsize * sizeof(u_short));
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
mark_all(scp);
Third round in syscons update. Display update method changed, now allways write in memory buffer, then periodically update physical display. Speed improvements (now > 5 times faster than the old syscons). History now circular buffer, with changeable size. History scroll by up/down line, up/down page, home and end. Backtab proberly implemented. Now space for 96 function keys, 63 allocated standard, default now SCO/SYSV compat again as in the old days. New keyboard definition files ~share/syscons/keymaps/* Misc fixes for old "hacks" that broke SCO/SYSV compat. More that I forgot before writing this...
1995-01-28 22:18:05 +00:00
}
static int
history_up_line(scr_stat *scp)
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (WRAPHIST(scp, scp->history_pos, -(scp->xsize*scp->ysize)) !=
scp->history_head) {
scp->history_pos = WRAPHIST(scp, scp->history_pos, -scp->xsize);
history_to_screen(scp);
return 0;
}
else
return -1;
Third round in syscons update. Display update method changed, now allways write in memory buffer, then periodically update physical display. Speed improvements (now > 5 times faster than the old syscons). History now circular buffer, with changeable size. History scroll by up/down line, up/down page, home and end. Backtab proberly implemented. Now space for 96 function keys, 63 allocated standard, default now SCO/SYSV compat again as in the old days. New keyboard definition files ~share/syscons/keymaps/* Misc fixes for old "hacks" that broke SCO/SYSV compat. More that I forgot before writing this...
1995-01-28 22:18:05 +00:00
}
static int
history_down_line(scr_stat *scp)
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (scp->history_pos != scp->history_head) {
scp->history_pos = WRAPHIST(scp, scp->history_pos, scp->xsize);
history_to_screen(scp);
return 0;
}
else
return -1;
Third round in syscons update. Display update method changed, now allways write in memory buffer, then periodically update physical display. Speed improvements (now > 5 times faster than the old syscons). History now circular buffer, with changeable size. History scroll by up/down line, up/down page, home and end. Backtab proberly implemented. Now space for 96 function keys, 63 allocated standard, default now SCO/SYSV compat again as in the old days. New keyboard definition files ~share/syscons/keymaps/* Misc fixes for old "hacks" that broke SCO/SYSV compat. More that I forgot before writing this...
1995-01-28 22:18:05 +00:00
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
/*
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
* scgetc(flags) - get character from keyboard.
* If flags & SCGETC_CN, then avoid harmful side effects.
* If flags & SCGETC_NONBLOCK, then wait until a key is pressed, else
* return NOKEY if there is nothing there.
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
*/
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
static u_int
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
scgetc(u_int flags)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
struct key_t *key;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
u_char scancode, keycode;
u_int state, action;
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
int c;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
static u_char esc_flag = 0, compose = 0;
static u_int chr = 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
next_code:
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
/* first see if there is something in the keyboard port */
if (flags & SCGETC_NONBLOCK) {
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
c = read_kbd_data_no_wait(sc_kbdc);
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
if (c == -1)
return(NOKEY);
} else {
do {
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
c = read_kbd_data(sc_kbdc);
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
} while(c == -1);
Fix the hanging keyboard problem under Xaccel. Apprently we are loosing an interrupt somewhere. The solution here is to check for keyboard input each time the screen update timer ticks. Not pretty, but works.
1996-11-04 21:01:08 +00:00
}
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
scancode = (u_char)c;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
- Ignore KBD_RAW_MODE and KBD_CODE_MODE in scgetc() when it's called from the low-level console routines sccngetc() and sccncheckc(). Submitted by: bde (a long time ago) - Don't try to ring bell and immediately return from do_bell() while device probe is in progress at boot time; the timeout queue is not functional yet. PR: kern/2424 - Stop running the screen saver after panic() is called: check if `panicstr' is non-NULL during scrn_timer(). PR: kern/5314 - A new option: SC_DISABLE_REBOOT The reboot key (usually Ctl-Alt-Del) will be ignored if this option is defined. You may still have the reboot key defined in the keymap and it won't cause error when the keymap is loaded, but it will be quietly treated as nop. OKed by: sos
1998-02-11 14:56:02 +00:00
/* make screensaver happy */
if (!(scancode & 0x80))
Time changes mark 2: * Figure out UTC relative to boottime. Four new functions provide time relative to boottime. * move "runtime" into struct proc. This helps fix the calcru() problem in SMP. * kill mono_time. * add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!) * nanosleep, select & poll takes long sleeps one day at a time Reviewed by: bde Tested by: ache and others
1998-04-04 13:26:20 +00:00
getmicroruntime(&scrn_time_stamp);
- Ignore KBD_RAW_MODE and KBD_CODE_MODE in scgetc() when it's called from the low-level console routines sccngetc() and sccncheckc(). Submitted by: bde (a long time ago) - Don't try to ring bell and immediately return from do_bell() while device probe is in progress at boot time; the timeout queue is not functional yet. PR: kern/2424 - Stop running the screen saver after panic() is called: check if `panicstr' is non-NULL during scrn_timer(). PR: kern/5314 - A new option: SC_DISABLE_REBOOT The reboot key (usually Ctl-Alt-Del) will be ignored if this option is defined. You may still have the reboot key defined in the keymap and it won't cause error when the keymap is loaded, but it will be quietly treated as nop. OKed by: sos
1998-02-11 14:56:02 +00:00
if (!(flags & SCGETC_CN)) {
/* do the /dev/random device a favour */
Use the calibrated/adjustable i8254 frequency `timer_freq' instead of TIMER_FREQ. Fixed missing splx() in scrn_timer(). The bug was harmless because of the undocumented behaviour that the ipl is automatically restored for timeout functions (see softclock()). Perhaps we should depend on this behaviour. Fixed the ddb fix in rev.1.176. The in_debugger flag was no use because it only works when the debugger is entered via the keyboard hotkey. The debugger may be entered for breakpoints and traps, and the console putc routine has no way of knowing when it was, so the console putc routine must (almost?) always remove the cursor image. Not fixed: console switching in ddb doesn't work (ISTR it working), and console 0 shouldn't be switched to for the debugger hotkey unless console 0 is /dev/console. Fixed side effects from calling add_keyboard_randomness() in the console getc routine by not calling it. add_keyboard_randomness() currently always reenables interrupts on 386's and 486's. This is very bad if the console getc routine is called from the debugger and the debugger was entered with interrupts disabled. Fixed preservation of initial screen and now-bogus comment about it. It was broken by setting the initial scr_buf to `buffer' instead of Crtat. `buffer' was full of nulls and the first scroll cleared everything above the things written through syscons. Submitted by: bruce (bde@freebsd.org)
1996-10-26 20:16:58 +00:00
add_keyboard_randomness(scancode);
Theodore Ts'po's random number gernerator for Linux, ported by me. This code will only be included in your kernel if you have 'options DEVRANDOM', but that will fall away in a couple of days. Obtained from: Theodore Ts'o, Linux
1995-10-28 16:58:05 +00:00
- Ignore KBD_RAW_MODE and KBD_CODE_MODE in scgetc() when it's called from the low-level console routines sccngetc() and sccncheckc(). Submitted by: bde (a long time ago) - Don't try to ring bell and immediately return from do_bell() while device probe is in progress at boot time; the timeout queue is not functional yet. PR: kern/2424 - Stop running the screen saver after panic() is called: check if `panicstr' is non-NULL during scrn_timer(). PR: kern/5314 - A new option: SC_DISABLE_REBOOT The reboot key (usually Ctl-Alt-Del) will be ignored if this option is defined. You may still have the reboot key defined in the keymap and it won't cause error when the keymap is loaded, but it will be quietly treated as nop. OKed by: sos
1998-02-11 14:56:02 +00:00
if (cur_console->status & KBD_RAW_MODE)
return scancode;
}
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
keycode = scancode & 0x7F;
switch (esc_flag) {
case 0x00: /* normal scancode */
switch(scancode) {
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
case 0xB8: /* left alt (compose key) */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (compose) {
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
compose = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (chr > 255) {
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
do_bell(cur_console,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
BELL_PITCH, BELL_DURATION);
chr = 0;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
break;
case 0x38:
if (!compose) {
compose = 1;
chr = 0;
}
break;
case 0xE0:
case 0xE1:
esc_flag = scancode;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
goto next_code;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
break;
case 0xE0: /* 0xE0 prefix */
esc_flag = 0;
switch (keycode) {
case 0x1C: /* right enter key */
keycode = 0x59;
break;
case 0x1D: /* right ctrl key */
keycode = 0x5A;
break;
case 0x35: /* keypad divide key */
keycode = 0x5B;
break;
case 0x37: /* print scrn key */
keycode = 0x5C;
break;
case 0x38: /* right alt key (alt gr) */
keycode = 0x5D;
break;
case 0x47: /* grey home key */
keycode = 0x5E;
break;
case 0x48: /* grey up arrow key */
keycode = 0x5F;
break;
case 0x49: /* grey page up key */
keycode = 0x60;
break;
case 0x4B: /* grey left arrow key */
keycode = 0x61;
break;
case 0x4D: /* grey right arrow key */
keycode = 0x62;
break;
case 0x4F: /* grey end key */
keycode = 0x63;
break;
case 0x50: /* grey down arrow key */
keycode = 0x64;
break;
case 0x51: /* grey page down key */
keycode = 0x65;
break;
case 0x52: /* grey insert key */
keycode = 0x66;
break;
case 0x53: /* grey delete key */
keycode = 0x67;
break;
/* the following 3 are only used on the MS "Natural" keyboard */
case 0x5b: /* left Window key */
keycode = 0x69;
break;
case 0x5c: /* right Window key */
keycode = 0x6a;
break;
case 0x5d: /* menu key */
keycode = 0x6b;
break;
default: /* ignore everything else */
goto next_code;
}
break;
case 0xE1: /* 0xE1 prefix */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
esc_flag = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (keycode == 0x1D)
esc_flag = 0x1D;
goto next_code;
/* NOT REACHED */
case 0x1D: /* pause / break */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
esc_flag = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (keycode != 0x45)
goto next_code;
keycode = 0x68;
break;
}
- Ignore KBD_RAW_MODE and KBD_CODE_MODE in scgetc() when it's called from the low-level console routines sccngetc() and sccncheckc(). Submitted by: bde (a long time ago) - Don't try to ring bell and immediately return from do_bell() while device probe is in progress at boot time; the timeout queue is not functional yet. PR: kern/2424 - Stop running the screen saver after panic() is called: check if `panicstr' is non-NULL during scrn_timer(). PR: kern/5314 - A new option: SC_DISABLE_REBOOT The reboot key (usually Ctl-Alt-Del) will be ignored if this option is defined. You may still have the reboot key defined in the keymap and it won't cause error when the keymap is loaded, but it will be quietly treated as nop. OKed by: sos
1998-02-11 14:56:02 +00:00
if (!(flags & SCGETC_CN) && (cur_console->status & KBD_CODE_MODE))
Add a new keyboard mode K_CODE. Returns a single byte for each key much like the scancode mode. However the keys that (for no good reason) returns extension codes etc, are translated into singlebyte codes. Needed by libvgl. This makes life ALOT easier, also the XFree86 folks could use this.
1997-10-01 20:46:29 +00:00
return (keycode | (scancode & 0x80));
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* if scroll-lock pressed allow history browsing */
if (cur_console->history && cur_console->status & SLKED) {
int i;
cur_console->status &= ~CURSOR_ENABLED;
if (!(cur_console->status & BUFFER_SAVED)) {
cur_console->status |= BUFFER_SAVED;
cur_console->history_save = cur_console->history_head;
/* copy screen into top of history buffer */
for (i=0; i<cur_console->ysize; i++) {
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(cur_console->scr_buf + (cur_console->xsize * i),
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
cur_console->history_head,
cur_console->xsize * sizeof(u_short));
cur_console->history_head += cur_console->xsize;
if (cur_console->history_head + cur_console->xsize >
cur_console->history + cur_console->history_size)
cur_console->history_head=cur_console->history;
}
cur_console->history_pos = cur_console->history_head;
history_to_screen(cur_console);
}
switch (scancode) {
case 0x47: /* home key */
cur_console->history_pos = cur_console->history_head;
history_to_screen(cur_console);
goto next_code;
case 0x4F: /* end key */
cur_console->history_pos =
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
WRAPHIST(cur_console, cur_console->history_head,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
cur_console->xsize*cur_console->ysize);
history_to_screen(cur_console);
goto next_code;
case 0x48: /* up arrow key */
if (history_up_line(cur_console))
do_bell(cur_console, BELL_PITCH, BELL_DURATION);
goto next_code;
case 0x50: /* down arrow key */
if (history_down_line(cur_console))
do_bell(cur_console, BELL_PITCH, BELL_DURATION);
goto next_code;
case 0x49: /* page up key */
for (i=0; i<cur_console->ysize; i++)
if (history_up_line(cur_console)) {
do_bell(cur_console, BELL_PITCH, BELL_DURATION);
Updated syscons to 0.2b - please test this! It won't be in the ALPHA release, but it's still worth testing.
1993-07-30 02:20:26 +00:00
break;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
goto next_code;
case 0x51: /* page down key */
for (i=0; i<cur_console->ysize; i++)
if (history_down_line(cur_console)) {
do_bell(cur_console, BELL_PITCH, BELL_DURATION);
Updated syscons to 0.2b - please test this! It won't be in the ALPHA release, but it's still worth testing.
1993-07-30 02:20:26 +00:00
break;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
goto next_code;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (compose) {
switch (scancode) {
/* key pressed process it */
case 0x47: case 0x48: case 0x49: /* keypad 7,8,9 */
chr = (scancode - 0x40) + chr*10;
goto next_code;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
case 0x4B: case 0x4C: case 0x4D: /* keypad 4,5,6 */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
chr = (scancode - 0x47) + chr*10;
goto next_code;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
case 0x4F: case 0x50: case 0x51: /* keypad 1,2,3 */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
chr = (scancode - 0x4E) + chr*10;
goto next_code;
case 0x52: /* keypad 0 */
chr *= 10;
goto next_code;
/* key release, no interest here */
case 0xC7: case 0xC8: case 0xC9: /* keypad 7,8,9 */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
case 0xCB: case 0xCC: case 0xCD: /* keypad 4,5,6 */
case 0xCF: case 0xD0: case 0xD1: /* keypad 1,2,3 */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case 0xD2: /* keypad 0 */
goto next_code;
case 0x38: /* left alt key */
break;
default:
if (chr) {
compose = chr = 0;
do_bell(cur_console, BELL_PITCH, BELL_DURATION);
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
goto next_code;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
break;
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
state = (shfts ? 1 : 0 ) | (2 * (ctls ? 1 : 0)) | (4 * (alts ? 1 : 0));
if ((!agrs && (cur_console->status & ALKED))
|| (agrs && !(cur_console->status & ALKED)))
keycode += ALTGR_OFFSET;
key = &key_map.key[keycode];
if ( ((key->flgs & FLAG_LOCK_C) && (cur_console->status & CLKED))
|| ((key->flgs & FLAG_LOCK_N) && (cur_console->status & NLKED)) )
state ^= 1;
/* Check for make/break */
action = key->map[state];
if (scancode & 0x80) { /* key released */
scgetc() referred to `spcl' in different ways when a key is pressed and released. It should use `spcl' consistently in both cases, otherwise shift/control/alt state may not be correctly set/reset. (Even with this fix, you can still make syscons confused and fail to change internal state if you really want to, by installing a really arcane and artificial keymap.) PR: i386/4030 Reviewed by: sos
1997-07-14 03:36:50 +00:00
if (key->spcl & (0x80>>state)) {
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
switch (action) {
case LSH:
shfts &= ~1;
break;
case RSH:
shfts &= ~2;
break;
case LCTR:
ctls &= ~1;
break;
case RCTR:
ctls &= ~2;
break;
case LALT:
alts &= ~1;
break;
case RALT:
alts &= ~2;
break;
case NLK:
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
nlkcnt = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case CLK:
clkcnt = 0;
break;
case SLK:
slkcnt = 0;
break;
case ASH:
agrs = 0;
break;
case ALK:
alkcnt = 0;
break;
case META:
metas = 0;
break;
}
Updated syscons to 0.2b - please test this! It won't be in the ALPHA release, but it's still worth testing.
1993-07-30 02:20:26 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (chr && !compose) {
action = chr;
chr = 0;
return(action);
}
} else {
/* key pressed */
if (key->spcl & (0x80>>state)) {
switch (action) {
/* LOCKING KEYS */
case NLK:
#ifdef another instance of toggle_splash_screen() which Soren missed.
1996-10-01 07:38:14 +00:00
#ifdef SC_SPLASH_SCREEN
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
toggle_splash_screen(cur_console); /* SOS XXX */
#ifdef another instance of toggle_splash_screen() which Soren missed.
1996-10-01 07:38:14 +00:00
#endif
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (!nlkcnt) {
nlkcnt++;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
if (cur_console->status & NLKED)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
cur_console->status &= ~NLKED;
else
cur_console->status |= NLKED;
update_leds(cur_console->status);
}
break;
case CLK:
if (!clkcnt) {
clkcnt++;
if (cur_console->status & CLKED)
cur_console->status &= ~CLKED;
else
cur_console->status |= CLKED;
update_leds(cur_console->status);
}
break;
case SLK:
if (!slkcnt) {
slkcnt++;
if (cur_console->status & SLKED) {
cur_console->status &= ~SLKED;
if (cur_console->status & BUFFER_SAVED){
int i;
u_short *ptr = cur_console->history_save;
for (i=0; i<cur_console->ysize; i++) {
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(ptr,
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
cur_console->scr_buf +
(cur_console->xsize*i),
cur_console->xsize * sizeof(u_short));
ptr += cur_console->xsize;
Remove useless comparison. Suggested by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-07-08 13:38:58 +00:00
if (ptr + cur_console->xsize >
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
cur_console->history +
cur_console->history_size)
ptr = cur_console->history;
}
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
cur_console->status &= ~BUFFER_SAVED;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
cur_console->history_head=cur_console->history_save;
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
cur_console->status |= CURSOR_ENABLED;
Added mark_all() call so that screen is proberly updated when scroll-lock history is disengaged.
1995-04-25 10:22:28 +00:00
mark_all(cur_console);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
scstart(VIRTUAL_TTY(get_scr_num()));
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
}
else
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
cur_console->status |= SLKED;
update_leds(cur_console->status);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case ALK:
if (!alkcnt) {
alkcnt++;
if (cur_console->status & ALKED)
cur_console->status &= ~ALKED;
else
cur_console->status |= ALKED;
update_leds(cur_console->status);
Updated syscons to 0.2b - please test this! It won't be in the ALPHA release, but it's still worth testing.
1993-07-30 02:20:26 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
/* NON-LOCKING KEYS */
case NOP:
break;
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
case SPSC:
#ifdef SC_SPLASH_SCREEN
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
accents = 0;
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
toggle_splash_screen(cur_console);
#endif
break;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case RBT:
- Ignore KBD_RAW_MODE and KBD_CODE_MODE in scgetc() when it's called from the low-level console routines sccngetc() and sccncheckc(). Submitted by: bde (a long time ago) - Don't try to ring bell and immediately return from do_bell() while device probe is in progress at boot time; the timeout queue is not functional yet. PR: kern/2424 - Stop running the screen saver after panic() is called: check if `panicstr' is non-NULL during scrn_timer(). PR: kern/5314 - A new option: SC_DISABLE_REBOOT The reboot key (usually Ctl-Alt-Del) will be ignored if this option is defined. You may still have the reboot key defined in the keymap and it won't cause error when the keymap is loaded, but it will be quietly treated as nop. OKed by: sos
1998-02-11 14:56:02 +00:00
#ifndef SC_DISABLE_REBOOT
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
accents = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
shutdown_nice();
- Ignore KBD_RAW_MODE and KBD_CODE_MODE in scgetc() when it's called from the low-level console routines sccngetc() and sccncheckc(). Submitted by: bde (a long time ago) - Don't try to ring bell and immediately return from do_bell() while device probe is in progress at boot time; the timeout queue is not functional yet. PR: kern/2424 - Stop running the screen saver after panic() is called: check if `panicstr' is non-NULL during scrn_timer(). PR: kern/5314 - A new option: SC_DISABLE_REBOOT The reboot key (usually Ctl-Alt-Del) will be ignored if this option is defined. You may still have the reboot key defined in the keymap and it won't cause error when the keymap is loaded, but it will be quietly treated as nop. OKed by: sos
1998-02-11 14:56:02 +00:00
#endif
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
break;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case SUSP:
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
#if NAPM > 0
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
accents = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
apm_suspend();
This is the latest version of the APM stuff from HOSOKAWA, I have looked briefly over it, and see some serious architectural issues in this stuff. On the other hand, I doubt that we will have any solution to these issues before 2.1, so we might as well leave this in. Most of the stuff is bracketed by #ifdef's so it shouldn't matter too much in the normal case. Reviewed by: phk Submitted by: HOSOKAWA, Tatsumi <hosokawa@mt.cs.keio.ac.jp>
1995-02-17 02:22:57 +00:00
#endif
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case DBG:
#ifdef DDB /* try to switch to console 0 */
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
accents = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (cur_console->smode.mode == VT_AUTO &&
console[0]->smode.mode == VT_AUTO)
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
switch_scr(cur_console, 0);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
Debugger("manual escape to debugger");
This is syscons version 1.1
1993-10-28 05:27:36 +00:00
#else
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
printf("No debugger in kernel\n");
This is syscons version 1.1
1993-10-28 05:27:36 +00:00
#endif
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case LSH:
shfts |= 1;
break;
case RSH:
shfts |= 2;
break;
case LCTR:
ctls |= 1;
break;
case RCTR:
ctls |= 2;
break;
case LALT:
alts |= 1;
break;
case RALT:
alts |= 2;
break;
case ASH:
agrs = 1;
break;
case META:
metas = 1;
break;
case NEXT:
Fixed the "switch to next screen" command (normally bound to the key "print scrn". It used to stop at the first non-open vty, now it skips the non-open ones and thereby enable one to cycle around all open vty by pressing "print scrn".
1997-01-30 15:12:17 +00:00
{
int next, this = get_scr_num();
- Produce the accent letter if the user hits the accent key twice. (accent_key + space does still print the accent letter too, as in the previous commit.) Requested by a couple of users. - Clear the accent flag when the next_screen key is pressed. - Added some comment lines regarding accent key processing.
1998-01-09 09:06:55 +00:00
accents = 0;
Fixed the "switch to next screen" command (normally bound to the key "print scrn". It used to stop at the first non-open vty, now it skips the non-open ones and thereby enable one to cycle around all open vty by pressing "print scrn".
1997-01-30 15:12:17 +00:00
for (next = this+1; next != this; next = (next+1)%MAXCONS) {
struct tty *tp = VIRTUAL_TTY(next);
if (tp->t_state & TS_ISOPEN) {
switch_scr(cur_console, next);
break;
}
}
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
case BTAB:
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
accents = 0;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
return(BKEY);
default:
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
if (action >= F_ACC && action <= L_ACC) {
- Produce the accent letter if the user hits the accent key twice. (accent_key + space does still print the accent letter too, as in the previous commit.) Requested by a couple of users. - Clear the accent flag when the next_screen key is pressed. - Added some comment lines regarding accent key processing.
1998-01-09 09:06:55 +00:00
/* turn it into an index */
action -= F_ACC - 1;
if ((action > accent_map.n_accs)
|| (accent_map.acc[action - 1].accchar == 0)) {
/*
* The index is out of range or pointing to an
* empty entry.
*/
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
accents = 0;
do_bell(cur_console, BELL_PITCH, BELL_DURATION);
}
- Produce the accent letter if the user hits the accent key twice. (accent_key + space does still print the accent letter too, as in the previous commit.) Requested by a couple of users. - Clear the accent flag when the next_screen key is pressed. - Added some comment lines regarding accent key processing.
1998-01-09 09:06:55 +00:00
/*
* If the same accent key has been hit twice,
* produce the accent char itself.
*/
if (action == accents) {
action = accent_map.acc[accents - 1].accchar;
accents = 0;
if (metas)
action |= MKEY;
return (action);
}
/* remember the index and wait for the next key stroke */
accents = action;
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
break;
}
if (accents > 0) {
accents = 0;
do_bell(cur_console, BELL_PITCH, BELL_DURATION);
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (action >= F_SCR && action <= L_SCR) {
switch_scr(cur_console, action - F_SCR);
break;
This is syscons version 1.1
1993-10-28 05:27:36 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
if (action >= F_FN && action <= L_FN)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
action |= FKEY;
return(action);
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
else {
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
if (accents) {
struct acc_t *acc;
int i;
acc = &accent_map.acc[accents - 1];
accents = 0;
- Produce the accent letter if the user hits the accent key twice. (accent_key + space does still print the accent letter too, as in the previous commit.) Requested by a couple of users. - Clear the accent flag when the next_screen key is pressed. - Added some comment lines regarding accent key processing.
1998-01-09 09:06:55 +00:00
/*
* If the accent key is followed by the space key,
* produce the accent char itself.
*/
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
if (action == ' ') {
action = acc->accchar;
if (metas)
action |= MKEY;
return (action);
}
for (i = 0; i < NUM_ACCENTCHARS; ++i) {
- Produce the accent letter if the user hits the accent key twice. (accent_key + space does still print the accent letter too, as in the previous commit.) Requested by a couple of users. - Clear the accent flag when the next_screen key is pressed. - Added some comment lines regarding accent key processing.
1998-01-09 09:06:55 +00:00
if (acc->map[i][0] == 0) /* end of the map entry */
Added accent (dead) key support to syscons and kbdcontrol. With a keymap with accent key definitions loaded to syscons, you press an accent key followed by a regular letter key to produce an accented letter. Press an accent key followed by the space bar to get the accent letter itself. Code is based on the ideas and work by jmrueda@diatel.upm.es and totii@est.is. PR: i386/4016 console.h - Defined structures and constants for accent (dead) keys. syscons.c, kbdtables.h - When an accent key is pressed, set the corresponding index to `accents'. If the next key is the space key, produce the accent char itself. Otherwise search the accent key map entry, indexed by `accents', for a matching pair of a regular char and an accented char. - Added ioctl functions to set and get the accent key map (PIO_DEADKEYMAP and GIO_DEADKEYMAP).
1998-01-07 08:40:34 +00:00
break;
if (acc->map[i][0] == action) {
action = acc->map[i][1];
if (metas)
action |= MKEY;
return (action);
}
}
do_bell(cur_console, BELL_PITCH, BELL_DURATION);
goto next_code;
}
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (metas)
action |= MKEY;
return(action);
}
}
goto next_code;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
int
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
scmmap(dev_t dev, int offset, int nprot)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (offset > 0x20000 - PAGE_SIZE)
return -1;
return i386_btop((VIDEOMEM + offset));
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
/*
* Calculate hardware attributes word using logical attributes mask and
* hardware colors
*/
static int
mask2attr(struct term_stat *term)
{
int attr, mask = term->attr_mask;
if (mask & REVERSE_ATTR) {
attr = ((mask & FOREGROUND_CHANGED) ?
((term->cur_color & 0xF000) >> 4) :
(term->rev_color & 0x0F00)) |
((mask & BACKGROUND_CHANGED) ?
((term->cur_color & 0x0F00) << 4) :
(term->rev_color & 0xF000));
} else
attr = term->cur_color;
/* XXX: underline mapping for Hercules adapter can be better */
if (mask & (BOLD_ATTR | UNDERLINE_ATTR))
attr ^= 0x0800;
if (mask & BLINK_ATTR)
attr ^= 0x8000;
return attr;
}
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
static void
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
set_keyboard(int command, int data)
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
{
int s;
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
if (sc_kbdc == NULL)
return;
/* prevent the timeout routine from polling the keyboard */
if (!kbdc_lock(sc_kbdc, TRUE))
return;
/* disable the keyboard and mouse interrupt */
s = spltty();
Interim patch/kludge for keyboard lock-up when `kbdcontrol' tries to change typematic rate, or the X server (XFree86 or Accelerated X) starts up. So far, there have been two independent reports from Dell Latitude XPi notebook/laptop owners. The Latitude seems to be the only system which suffers from this problem. (I don't know the problem is with the entire Latitude line or with only some Latitude models) No problem report has been heard about other systems (I certainly cannot reproduce the problem in my -current and 2.2 systems). In 3.0-CURRENT, 2.2-RELEASE and 2.2-GAMMA-970310, when programming the keyboard LED/repeat-rate, `set_keyboard()' in `syscons' tells the keyboard controller not to generate keyboard interrupt (IRQ1) and then enable tty interrupts, expecting the keyboard interrupt doesn't occur. It appears that somehow Latitude's keyboard controller still generates the keyboard interrupt thereafter, and `set_keyboard()' doesn't see the return code from the keyboard because it is consumed by the keyboard interrupt handler. The patch entirely disables tty interrupts while setting LED and typematic rate in `set_keyboard()', making the routine behave more like the previous versions of `syscons' (versions in 2.1.X and 2.2-ALPHA, -BETA, and some -GAMMAs). The reporter said this patch eliminated the problem. (I also found another typo/bug, but the reporter and I found that it wasn't the cause of the problem...) This should go into RELENG_2_2.
1997-03-28 10:11:24 +00:00
#if 0
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
c = get_controller_command_byte(sc_kbdc);
if ((c == -1)
|| !set_controller_command_byte(sc_kbdc,
kbdc_get_device_mask(sc_kbdc),
Interim patch/kludge for keyboard lock-up when `kbdcontrol' tries to change typematic rate, or the X server (XFree86 or Accelerated X) starts up. So far, there have been two independent reports from Dell Latitude XPi notebook/laptop owners. The Latitude seems to be the only system which suffers from this problem. (I don't know the problem is with the entire Latitude line or with only some Latitude models) No problem report has been heard about other systems (I certainly cannot reproduce the problem in my -current and 2.2 systems). In 3.0-CURRENT, 2.2-RELEASE and 2.2-GAMMA-970310, when programming the keyboard LED/repeat-rate, `set_keyboard()' in `syscons' tells the keyboard controller not to generate keyboard interrupt (IRQ1) and then enable tty interrupts, expecting the keyboard interrupt doesn't occur. It appears that somehow Latitude's keyboard controller still generates the keyboard interrupt thereafter, and `set_keyboard()' doesn't see the return code from the keyboard because it is consumed by the keyboard interrupt handler. The patch entirely disables tty interrupts while setting LED and typematic rate in `set_keyboard()', making the routine behave more like the previous versions of `syscons' (versions in 2.1.X and 2.2-ALPHA, -BETA, and some -GAMMAs). The reporter said this patch eliminated the problem. (I also found another typo/bug, but the reporter and I found that it wasn't the cause of the problem...) This should go into RELENG_2_2.
1997-03-28 10:11:24 +00:00
KBD_DISABLE_KBD_PORT | KBD_DISABLE_KBD_INT
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
| KBD_DISABLE_AUX_PORT | KBD_DISABLE_AUX_INT)) {
/* CONTROLLER ERROR */
kbdc_lock(sc_kbdc, FALSE);
splx(s);
return;
}
/*
* Now that the keyboard controller is told not to generate
* the keyboard and mouse interrupts, call `splx()' to allow
* the other tty interrupts. The clock interrupt may also occur,
* but the timeout routine (`scrn_timer()') will be blocked
* by the lock flag set via `kbdc_lock()'
*/
splx(s);
Interim patch/kludge for keyboard lock-up when `kbdcontrol' tries to change typematic rate, or the X server (XFree86 or Accelerated X) starts up. So far, there have been two independent reports from Dell Latitude XPi notebook/laptop owners. The Latitude seems to be the only system which suffers from this problem. (I don't know the problem is with the entire Latitude line or with only some Latitude models) No problem report has been heard about other systems (I certainly cannot reproduce the problem in my -current and 2.2 systems). In 3.0-CURRENT, 2.2-RELEASE and 2.2-GAMMA-970310, when programming the keyboard LED/repeat-rate, `set_keyboard()' in `syscons' tells the keyboard controller not to generate keyboard interrupt (IRQ1) and then enable tty interrupts, expecting the keyboard interrupt doesn't occur. It appears that somehow Latitude's keyboard controller still generates the keyboard interrupt thereafter, and `set_keyboard()' doesn't see the return code from the keyboard because it is consumed by the keyboard interrupt handler. The patch entirely disables tty interrupts while setting LED and typematic rate in `set_keyboard()', making the routine behave more like the previous versions of `syscons' (versions in 2.1.X and 2.2-ALPHA, -BETA, and some -GAMMAs). The reporter said this patch eliminated the problem. (I also found another typo/bug, but the reporter and I found that it wasn't the cause of the problem...) This should go into RELENG_2_2.
1997-03-28 10:11:24 +00:00
#endif
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
Interim patch/kludge for keyboard lock-up when `kbdcontrol' tries to change typematic rate, or the X server (XFree86 or Accelerated X) starts up. So far, there have been two independent reports from Dell Latitude XPi notebook/laptop owners. The Latitude seems to be the only system which suffers from this problem. (I don't know the problem is with the entire Latitude line or with only some Latitude models) No problem report has been heard about other systems (I certainly cannot reproduce the problem in my -current and 2.2 systems). In 3.0-CURRENT, 2.2-RELEASE and 2.2-GAMMA-970310, when programming the keyboard LED/repeat-rate, `set_keyboard()' in `syscons' tells the keyboard controller not to generate keyboard interrupt (IRQ1) and then enable tty interrupts, expecting the keyboard interrupt doesn't occur. It appears that somehow Latitude's keyboard controller still generates the keyboard interrupt thereafter, and `set_keyboard()' doesn't see the return code from the keyboard because it is consumed by the keyboard interrupt handler. The patch entirely disables tty interrupts while setting LED and typematic rate in `set_keyboard()', making the routine behave more like the previous versions of `syscons' (versions in 2.1.X and 2.2-ALPHA, -BETA, and some -GAMMAs). The reporter said this patch eliminated the problem. (I also found another typo/bug, but the reporter and I found that it wasn't the cause of the problem...) This should go into RELENG_2_2.
1997-03-28 10:11:24 +00:00
if (send_kbd_command_and_data(sc_kbdc, command, data) != KBD_ACK)
send_kbd_command(sc_kbdc, KBDC_ENABLE_KBD);
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
Interim patch/kludge for keyboard lock-up when `kbdcontrol' tries to change typematic rate, or the X server (XFree86 or Accelerated X) starts up. So far, there have been two independent reports from Dell Latitude XPi notebook/laptop owners. The Latitude seems to be the only system which suffers from this problem. (I don't know the problem is with the entire Latitude line or with only some Latitude models) No problem report has been heard about other systems (I certainly cannot reproduce the problem in my -current and 2.2 systems). In 3.0-CURRENT, 2.2-RELEASE and 2.2-GAMMA-970310, when programming the keyboard LED/repeat-rate, `set_keyboard()' in `syscons' tells the keyboard controller not to generate keyboard interrupt (IRQ1) and then enable tty interrupts, expecting the keyboard interrupt doesn't occur. It appears that somehow Latitude's keyboard controller still generates the keyboard interrupt thereafter, and `set_keyboard()' doesn't see the return code from the keyboard because it is consumed by the keyboard interrupt handler. The patch entirely disables tty interrupts while setting LED and typematic rate in `set_keyboard()', making the routine behave more like the previous versions of `syscons' (versions in 2.1.X and 2.2-ALPHA, -BETA, and some -GAMMAs). The reporter said this patch eliminated the problem. (I also found another typo/bug, but the reporter and I found that it wasn't the cause of the problem...) This should go into RELENG_2_2.
1997-03-28 10:11:24 +00:00
#if 0
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
/* restore the interrupts */
if (!set_controller_command_byte(sc_kbdc,
kbdc_get_device_mask(sc_kbdc),
c & (KBD_KBD_CONTROL_BITS | KBD_AUX_CONTROL_BITS))) {
/* CONTROLLER ERROR */
}
Interim patch/kludge for keyboard lock-up when `kbdcontrol' tries to change typematic rate, or the X server (XFree86 or Accelerated X) starts up. So far, there have been two independent reports from Dell Latitude XPi notebook/laptop owners. The Latitude seems to be the only system which suffers from this problem. (I don't know the problem is with the entire Latitude line or with only some Latitude models) No problem report has been heard about other systems (I certainly cannot reproduce the problem in my -current and 2.2 systems). In 3.0-CURRENT, 2.2-RELEASE and 2.2-GAMMA-970310, when programming the keyboard LED/repeat-rate, `set_keyboard()' in `syscons' tells the keyboard controller not to generate keyboard interrupt (IRQ1) and then enable tty interrupts, expecting the keyboard interrupt doesn't occur. It appears that somehow Latitude's keyboard controller still generates the keyboard interrupt thereafter, and `set_keyboard()' doesn't see the return code from the keyboard because it is consumed by the keyboard interrupt handler. The patch entirely disables tty interrupts while setting LED and typematic rate in `set_keyboard()', making the routine behave more like the previous versions of `syscons' (versions in 2.1.X and 2.2-ALPHA, -BETA, and some -GAMMAs). The reporter said this patch eliminated the problem. (I also found another typo/bug, but the reporter and I found that it wasn't the cause of the problem...) This should go into RELENG_2_2.
1997-03-28 10:11:24 +00:00
#else
splx(s);
#endif
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
kbdc_lock(sc_kbdc, FALSE);
}
static void
update_leds(int which)
{
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
static u_char xlate_leds[8] = { 0, 4, 2, 6, 1, 5, 3, 7 };
/* replace CAPS led with ALTGR led for ALTGR keyboards */
if (key_map.n_keys > ALTGR_OFFSET) {
if (which & ALKED)
which |= CLKED;
else
which &= ~CLKED;
}
Finally a start at sharing the kdb controller routines between syscons and psm, curtesy Kazutaka Yokota with minor changes by me. This contains an update of the psm driver as well. This also fixes the breakage that I introduced to the psm driver by making syscons poll for keyboard events in the atempt to fix the hanging keyboard problem. It works perfectly for me, and I'd like to hear from all that have had keyboard/ps/2 mouse problems if this is the cure... Submitted by: Kazutaka YOKOTA (yokota@zodiac.mech.utsunomiya-u.ac.jp)
1996-11-14 22:19:17 +00:00
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
set_keyboard(KBDC_SET_LEDS, xlate_leds[which & LED_MASK]);
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
}
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
void
Updated to latest 1.1.5.1+ version Reviewed by: Submitted by:
1994-08-17 08:51:59 +00:00
set_mode(scr_stat *scp)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
char special_modetable[MODE_PARAM_SIZE];
char *mp;
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
int s;
int i;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (scp != cur_console)
return;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
/*
* even if mode switching is disabled, we can change back
* to the initial mode or the custom mode based on the initial
* mode if we have saved register values upon start-up.
*/
mp = get_mode_param(scp, scp->mode);
if (mp == NULL)
return;
bcopy(mp, &special_modetable, sizeof(special_modetable));
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* setup video hardware for the given mode */
switch (scp->mode) {
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
case M_VGA_C80x60: case M_VGA_M80x60:
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
special_modetable[2] = 0x08;
special_modetable[19] = 0x47;
goto special_480l;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
case M_VGA_C80x30: case M_VGA_M80x30:
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
special_modetable[19] = 0x4f;
special_480l:
special_modetable[9] |= 0xc0;
special_modetable[16] = 0x08;
special_modetable[17] = 0x3e;
special_modetable[26] = 0xea;
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
special_modetable[28] = 0xdf;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
special_modetable[31] = 0xe7;
special_modetable[32] = 0x04;
goto setup_mode;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
case M_ENH_C80x43: case M_ENH_B80x43:
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
special_modetable[28] = 87;
goto special_80x50;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
case M_VGA_C80x50: case M_VGA_M80x50:
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
special_80x50:
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
special_modetable[2] = 8;
special_modetable[19] = 7;
goto setup_mode;
case M_VGA_C40x25: case M_VGA_C80x25:
case M_VGA_M80x25:
case M_B40x25: case M_C40x25:
case M_B80x25: case M_C80x25:
case M_ENH_B40x25: case M_ENH_C40x25:
case M_ENH_B80x25: case M_ENH_C80x25:
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
case M_EGAMONO80x25:
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
Added support for many more videomodes, including graphic modes up til 320x200 256col VGA. This is nessesary for the iBCS stuff to work right. (And we get the benefit of more video modes). Uses the videocard BIOS to optain mode tables. Added a "green" saver, switches off the syncs for "green" monitors. Reviewed by: Submitted by: Obtained from:
1994-09-15 07:26:40 +00:00
setup_mode:
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
set_vgaregs(special_modetable);
scp->font_size = special_modetable[2];
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* set font type (size) */
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
if (scp->font_size < 14) {
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
if (fonts_loaded & FONT_8)
copy_font(LOAD, FONT_8, font_8);
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
i = 0x0a; /* font 2 */
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
} else if (scp->font_size >= 16) {
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
if (fonts_loaded & FONT_16)
copy_font(LOAD, FONT_16, font_16);
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
i = 0x00; /* font 0 */
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
} else {
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
if (fonts_loaded & FONT_14)
copy_font(LOAD, FONT_14, font_14);
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
i = 0x05; /* font 1 */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
/*
* FONT KLUDGE:
* This is an interim kludge to display correct font.
* Always use the font page #0 on the video plane 2.
* Somehow we cannot show the font in other font pages on
* some video cards... XXX
*/
i = 0x00;
s = splhigh();
outb(TSIDX, 0x00); outb(TSREG, 0x01);
outb(TSIDX, 0x03); outb(TSREG, i);
outb(TSIDX, 0x00); outb(TSREG, 0x03);
splx(s);
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
if (flags & CHAR_CURSOR)
set_destructive_cursor(scp);
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
mark_all(scp);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
Added support for many more videomodes, including graphic modes up til 320x200 256col VGA. This is nessesary for the iBCS stuff to work right. (And we get the benefit of more video modes). Uses the videocard BIOS to optain mode tables. Added a "green" saver, switches off the syncs for "green" monitors. Reviewed by: Submitted by: Obtained from:
1994-09-15 07:26:40 +00:00
Yeah I'm back hacking syscons !! Add support for MODEX 320x240x256color with "unchained" adressing, giving access to all 256K on all VGA's, those with that much memory that is :) Also make sysmouse use the right resolution in graphics modes.
1997-08-08 22:52:30 +00:00
case M_VGA_MODEX:
/* "unchain" the VGA mode */
special_modetable[5-1+0x04] &= 0xf7;
special_modetable[5-1+0x04] |= 0x04;
/* turn off doubleword mode */
special_modetable[10+0x14] &= 0xbf;
/* turn off word adressing */
special_modetable[10+0x17] |= 0x40;
/* set logical screen width */
special_modetable[10+0x13] = 80;
/* set 240 lines */
special_modetable[10+0x11] = 0x2c;
special_modetable[10+0x06] = 0x0d;
special_modetable[10+0x07] = 0x3e;
special_modetable[10+0x10] = 0xea;
special_modetable[10+0x11] = 0xac;
special_modetable[10+0x12] = 0xdf;
special_modetable[10+0x15] = 0xe7;
special_modetable[10+0x16] = 0x06;
/* set vertical sync polarity to reflect aspect ratio */
special_modetable[9] = 0xe3;
goto setup_grmode;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
case M_BG320: case M_CG320: case M_BG640:
case M_CG320_D: case M_CG640_E:
case M_CG640x350: case M_ENH_CG640:
case M_BG640x480: case M_CG640x480: case M_VGA_CG320:
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
Yeah I'm back hacking syscons !! Add support for MODEX 320x240x256color with "unchained" adressing, giving access to all 256K on all VGA's, those with that much memory that is :) Also make sysmouse use the right resolution in graphics modes.
1997-08-08 22:52:30 +00:00
setup_grmode:
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
set_vgaregs(special_modetable);
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
scp->font_size = FONT_NONE;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
break;
Added support for many more videomodes, including graphic modes up til 320x200 256col VGA. This is nessesary for the iBCS stuff to work right. (And we get the benefit of more video modes). Uses the videocard BIOS to optain mode tables. Added a "green" saver, switches off the syncs for "green" monitors. Reviewed by: Submitted by: Obtained from:
1994-09-15 07:26:40 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
default:
/* call user defined function XXX */
break;
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* set border color for this (virtual) console */
set_border(scp->border);
return;
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
void
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
set_border(u_char color)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
switch (crtc_type) {
case KD_EGA:
case KD_VGA:
inb(crtc_addr + 6); /* reset flip-flop */
- Slightly change the way the border color register is updated so that flicker won't occur when set_border() is called. - Properly restore the border color when switching virtual consoles. Pointed out by: tony@dell.com OKed by: sos
1997-10-26 07:36:13 +00:00
outb(ATC, 0x31); outb(ATC, color);
Screen saver related fixes. 1. Add new interface, add_scrn_saver()/remove_scrn_saver(), to declare loading/unloading of a screen saver. The screen saver calls these functions to notify syscons of loading/unloading events. It was possible to load multiple savers each of which will try to remember the previous saver in a local variable (`old_saver'). The scheme breaks easily if the user load two savers and unload them in a wrong order; if the first saver is unloaded first, `old_saver' in the second saver points to nowhere. Now only one screen saver is allowed in memory at a time. Soeren will be looking into this issue again later. syscons is becoming too heavy. It's time to cut things down, rather than adding more... 2. Make scrn_timer() to be the primary caller of the screen saver (*current_saver)(). scintr(), scioctl() and ansi_put() update `scrn_time_stamp' to indicate that they want to stop the screen saver. There are three exceptions, however. One is remove_scrn_saver() which need to stop the current screen saver if it is running. To guard against scrn_timer() calling the saver during this operation, `current_saver' is set to `none_saver' early. The others are sccngetc() and sccncheckc(); they will unblank the screen too. When the kernel enters DDB (via the hot key or a break point), the screen saver will be stopped by sccngetc(). However, we have a reentrancy problem here. If the system has been in the middle of the screen saver... (The screen saver reentrancy problem has always been with sccnputc() and sccngetc() in the -current source. So, the new code is doing no worse, I reckon.) 3. Use `mono_time' rather than `time'. 4. Make set_border() work for EGA and CGA in addition to VGA. Do nothing for MDA. Changes to the LKM screen saver modules will follow shortly. YOU NEED TO RECOMPILE BOTH SCREEN SAVERS AND KERNEL AS OF THESE CHANGES. Reviewed by: sos and bde
1997-07-15 14:43:27 +00:00
break;
case KD_CGA:
outb(crtc_addr + 5, color & 0x0f); /* color select register */
break;
case KD_MONO:
case KD_HERCULES:
default:
break;
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Added support for many more videomodes, including graphic modes up til 320x200 256col VGA. This is nessesary for the iBCS stuff to work right. (And we get the benefit of more video modes). Uses the videocard BIOS to optain mode tables. Added a "green" saver, switches off the syncs for "green" monitors. Reviewed by: Submitted by: Obtained from:
1994-09-15 07:26:40 +00:00
static void
set_vgaregs(char *modetable)
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
int i, s = splhigh();
Added support for many more videomodes, including graphic modes up til 320x200 256col VGA. This is nessesary for the iBCS stuff to work right. (And we get the benefit of more video modes). Uses the videocard BIOS to optain mode tables. Added a "green" saver, switches off the syncs for "green" monitors. Reviewed by: Submitted by: Obtained from:
1994-09-15 07:26:40 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
outb(TSIDX, 0x00); outb(TSREG, 0x01); /* stop sequencer */
for (i=0; i<4; i++) { /* program sequencer */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
outb(TSIDX, i+1);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
outb(TSREG, modetable[i+5]);
}
outb(MISC, modetable[9]); /* set dot-clock */
outb(TSIDX, 0x00); outb(TSREG, 0x03); /* start sequencer */
outb(crtc_addr, 0x11);
outb(crtc_addr+1, inb(crtc_addr+1) & 0x7F);
for (i=0; i<25; i++) { /* program crtc */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
outb(crtc_addr, i);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (i == 14 || i == 15) /* no hardware cursor */
outb(crtc_addr+1, 0xff);
else
outb(crtc_addr+1, modetable[i+10]);
}
inb(crtc_addr+6); /* reset flip-flop */
for (i=0; i<20; i++) { /* program attribute ctrl */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
outb(ATC, i);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
outb(ATC, modetable[i+35]);
}
for (i=0; i<9; i++) { /* program graph data ctrl */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
outb(GDCIDX, i);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
outb(GDCREG, modetable[i+55]);
}
inb(crtc_addr+6); /* reset flip-flop */
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
outb(ATC, 0x20); /* enable palette */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
splx(s);
Added support for many more videomodes, including graphic modes up til 320x200 256col VGA. This is nessesary for the iBCS stuff to work right. (And we get the benefit of more video modes). Uses the videocard BIOS to optain mode tables. Added a "green" saver, switches off the syncs for "green" monitors. Reviewed by: Submitted by: Obtained from:
1994-09-15 07:26:40 +00:00
}
Updated to latest 1.1.5.1+ version Reviewed by: Submitted by:
1994-08-17 08:51:59 +00:00
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
static void
read_vgaregs(char *buf)
{
int i, j;
int s;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
bzero(buf, MODE_PARAM_SIZE);
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
s = splhigh();
outb(TSIDX, 0x00); outb(TSREG, 0x01); /* stop sequencer */
for (i=0, j=5; i<4; i++) {
outb(TSIDX, i+1);
buf[j++] = inb(TSREG);
}
buf[9] = inb(MISC + 10); /* dot-clock */
outb(TSIDX, 0x00); outb(TSREG, 0x03); /* start sequencer */
for (i=0, j=10; i<25; i++) { /* crtc */
outb(crtc_addr, i);
buf[j++] = inb(crtc_addr+1);
}
for (i=0, j=35; i<20; i++) { /* attribute ctrl */
inb(crtc_addr+6); /* reset flip-flop */
outb(ATC, i);
buf[j++] = inb(ATC + 1);
}
for (i=0, j=55; i<9; i++) { /* graph data ctrl */
outb(GDCIDX, i);
buf[j++] = inb(GDCREG);
}
inb(crtc_addr+6); /* reset flip-flop */
outb(ATC, 0x20); /* enable palette */
buf[0] = *(char *)pa_to_va(0x44a); /* COLS */
buf[1] = *(char *)pa_to_va(0x484); /* ROWS */
buf[2] = *(char *)pa_to_va(0x485); /* POINTS */
buf[3] = *(char *)pa_to_va(0x44c);
buf[4] = *(char *)pa_to_va(0x44d);
splx(s);
}
static int
comp_vgaregs(u_char *buf1, u_char *buf2)
{
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
static struct {
u_char mask;
} params[MODE_PARAM_SIZE] = {
0xff, 0x00, 0xff, /* COLS, ROWS, POINTS */
0xff, 0xff, /* page length */
0xfe, 0xff, 0xff, 0xff, /* sequencer registers */
0xf3, /* misc register */
0xff, 0xff, 0xff, 0x7f, 0xff, /* CRTC */
0xff, 0xff, 0xff, 0x7f, 0xff,
0x00, 0x00, 0x00, 0x00, 0x00,
0x00, 0xff, 0x7f, 0xff, 0xff,
0x7f, 0xff, 0xff, 0xef, 0xff,
0xff, 0xff, 0xff, 0xff, 0xff, /* attribute controller registers */
0xff, 0xff, 0xff, 0xff, 0xff,
0xff, 0xff, 0xff, 0xff, 0xff,
0xff, 0xff, 0xff, 0xff, 0xf0,
0xff, 0xff, 0xff, 0xff, 0xff, /* GDC register */
0xff, 0xff, 0xff, 0xff,
};
int identical = TRUE;
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
int i;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
for (i = 0; i < sizeof(params)/sizeof(params[0]); ++i) {
if (params[i].mask == 0) /* don't care */
continue;
if ((buf1[i] & params[i].mask) != (buf2[i] & params[i].mask))
return COMP_DIFFERENT;
if (buf1[i] != buf2[i])
identical = FALSE;
}
return (identical) ? COMP_IDENTICAL : COMP_SIMILAR;
#if 0
In comp_vgaregs skip cursor shape and 4 unused bytes (6 total) instead of 2 unused. Previus code broke syscons on Diamond Stealth 3D 3000 (video mode ptr was NULL with no real reason)
1997-04-30 01:19:21 +00:00
for(i = 0; i < 20; ++i) {
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
if (*buf1++ != *buf2++)
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
return COMP_DIFFERENT;
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
}
Restore CRTC start address check back after looking into more sources
1997-04-30 15:31:22 +00:00
buf1 += 2; /* skip the cursor shape */
buf2 += 2;
for(i = 22; i < 24; ++i) {
if (*buf1++ != *buf2++)
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
return COMP_DIFFERENT;
Restore CRTC start address check back after looking into more sources
1997-04-30 15:31:22 +00:00
}
buf1 += 2; /* skip the cursor position */
buf2 += 2;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
for(i = 26; i < MODE_PARAM_SIZE; ++i) {
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
if (*buf1++ != *buf2++)
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
return COMP_DIFFERENT;
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
}
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
return COMP_IDENTICAL;
#endif
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
}
static void
dump_vgaregs(u_char *buf)
{
int i;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
for(i = 0; i < MODE_PARAM_SIZE;) {
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
printf("%02x ", buf[i]);
if ((++i % 16) == 0)
printf("\n");
}
}
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
static void
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
set_font_mode(u_char *buf)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Bruce pointed out a bogon: cur_console is NULL when copy_font() is first called from scinit(). This is apparently harmless when scinit() is called early from sccninit() - page 0 is apparently mapped r/w then, and 0->status contains suitable garbage. However, when there is a serial console, scinit() is first called from scattach() when the page tables are completely initialized, so the NULL pointer causes a panic. Submitted by: bruce
1997-01-17 15:49:13 +00:00
int s = splhigh();
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
font_loading_in_progress = TRUE;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
/* save register values */
outb(TSIDX, 0x02); buf[0] = inb(TSREG);
outb(TSIDX, 0x04); buf[1] = inb(TSREG);
outb(GDCIDX, 0x04); buf[2] = inb(GDCREG);
outb(GDCIDX, 0x05); buf[3] = inb(GDCREG);
outb(GDCIDX, 0x06); buf[4] = inb(GDCREG);
inb(crtc_addr + 6);
outb(ATC, 0x10); buf[5] = inb(ATC + 1);
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
/* setup vga for loading fonts */
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
inb(crtc_addr+6); /* reset flip-flop */
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
outb(ATC, 0x10); outb(ATC, buf[5] & ~0x01);
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
inb(crtc_addr+6); /* reset flip-flop */
outb(ATC, 0x20); /* enable palette */
Bruce pointed out a bogon: cur_console is NULL when copy_font() is first called from scinit(). This is apparently harmless when scinit() is called early from sccninit() - page 0 is apparently mapped r/w then, and 0->status contains suitable garbage. However, when there is a serial console, scinit() is first called from scattach() when the page tables are completely initialized, so the NULL pointer causes a panic. Submitted by: bruce
1997-01-17 15:49:13 +00:00
Third round in syscons update. Display update method changed, now allways write in memory buffer, then periodically update physical display. Speed improvements (now > 5 times faster than the old syscons). History now circular buffer, with changeable size. History scroll by up/down line, up/down page, home and end. Backtab proberly implemented. Now space for 96 function keys, 63 allocated standard, default now SCO/SYSV compat again as in the old days. New keyboard definition files ~share/syscons/keymaps/* Misc fixes for old "hacks" that broke SCO/SYSV compat. More that I forgot before writing this...
1995-01-28 22:18:05 +00:00
#if SLOW_VGA
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
outb(TSIDX, 0x00); outb(TSREG, 0x01);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
outb(TSIDX, 0x02); outb(TSREG, 0x04);
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
outb(TSIDX, 0x04); outb(TSREG, 0x07);
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
outb(TSIDX, 0x00); outb(TSREG, 0x03);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
outb(GDCIDX, 0x04); outb(GDCREG, 0x02);
outb(GDCIDX, 0x05); outb(GDCREG, 0x00);
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
outb(GDCIDX, 0x06); outb(GDCREG, 0x04);
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
#else
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
outw(TSIDX, 0x0100);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
outw(TSIDX, 0x0402);
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
outw(TSIDX, 0x0704);
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
outw(TSIDX, 0x0300);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
outw(GDCIDX, 0x0204);
outw(GDCIDX, 0x0005);
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
outw(GDCIDX, 0x0406); /* addr = a0000, 64kb */
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
#endif
Bruce pointed out a bogon: cur_console is NULL when copy_font() is first called from scinit(). This is apparently harmless when scinit() is called early from sccninit() - page 0 is apparently mapped r/w then, and 0->status contains suitable garbage. However, when there is a serial console, scinit() is first called from scattach() when the page tables are completely initialized, so the NULL pointer causes a panic. Submitted by: bruce
1997-01-17 15:49:13 +00:00
splx(s);
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
}
static void
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
set_normal_mode(u_char *buf)
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
{
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
char *modetable;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
int s = splhigh();
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
/* setup vga for normal operation mode again */
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
inb(crtc_addr+6); /* reset flip-flop */
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
outb(ATC, 0x10); outb(ATC, buf[5]);
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
inb(crtc_addr+6); /* reset flip-flop */
outb(ATC, 0x20); /* enable palette */
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
Third round in syscons update. Display update method changed, now allways write in memory buffer, then periodically update physical display. Speed improvements (now > 5 times faster than the old syscons). History now circular buffer, with changeable size. History scroll by up/down line, up/down page, home and end. Backtab proberly implemented. Now space for 96 function keys, 63 allocated standard, default now SCO/SYSV compat again as in the old days. New keyboard definition files ~share/syscons/keymaps/* Misc fixes for old "hacks" that broke SCO/SYSV compat. More that I forgot before writing this...
1995-01-28 22:18:05 +00:00
#if SLOW_VGA
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
outb(TSIDX, 0x00); outb(TSREG, 0x01);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
outb(TSIDX, 0x02); outb(TSREG, buf[0]);
outb(TSIDX, 0x04); outb(TSREG, buf[1]);
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
outb(TSIDX, 0x00); outb(TSREG, 0x03);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
outb(GDCIDX, 0x04); outb(GDCREG, buf[2]);
outb(GDCIDX, 0x05); outb(GDCREG, buf[3]);
Fix the broken support for monochrome VGA's. It was MY fault after all, damn..
1996-12-07 22:33:05 +00:00
if (crtc_addr == MONO_BASE) {
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
outb(GDCIDX, 0x06); outb(GDCREG,(buf[4] & 0x03) | 0x08);
} else {
outb(GDCIDX, 0x06); outb(GDCREG,(buf[4] & 0x03) | 0x0c);
Fix the broken support for monochrome VGA's. It was MY fault after all, damn..
1996-12-07 22:33:05 +00:00
}
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
#else
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
outw(TSIDX, 0x0100);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
outw(TSIDX, 0x0002 | (buf[0] << 8));
outw(TSIDX, 0x0004 | (buf[1] << 8));
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
outw(TSIDX, 0x0300);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
outw(GDCIDX, 0x0004 | (buf[2] << 8));
outw(GDCIDX, 0x0005 | (buf[3] << 8));
Fix the broken support for monochrome VGA's. It was MY fault after all, damn..
1996-12-07 22:33:05 +00:00
if (crtc_addr == MONO_BASE)
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
outw(GDCIDX, 0x0006 | (((buf[4] & 0x03) | 0x08)<<8));
Fix the broken support for monochrome VGA's. It was MY fault after all, damn..
1996-12-07 22:33:05 +00:00
else
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
outw(GDCIDX, 0x0006 | (((buf[4] & 0x03) | 0x0c)<<8));
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
#endif
- A new semaphore: font_loading_in_progress. Don't touch/update the screen while manipulating font data. Possibly related to PR: kern/4271 - Set up VGA in alphanumeric mode rather than graphics mode when loading font into video memory. This will drastically reduce flicker. PR: bin/2977 - Set up scp->font_size properly during video mode switch caused by ioctl.
1998-02-11 15:00:24 +00:00
font_loading_in_progress = FALSE;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
splx(s);
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
}
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
void
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
copy_font(int operation, int font_type, char* font_image)
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
int ch, line, segment, fontsize;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
u_char buf[PARAM_BUFSIZE];
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
u_char val;
switch (font_type) {
default:
case FONT_8:
segment = 0x8000;
fontsize = 8;
break;
case FONT_14:
segment = 0x4000;
fontsize = 14;
break;
case FONT_16:
segment = 0x0000;
fontsize = 16;
break;
}
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
/*
* FONT KLUDGE
* Always use the font page #0. XXX
*/
segment = 0x0000;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
outb(TSIDX, 0x01); val = inb(TSREG); /* disable screen */
outb(TSIDX, 0x01); outb(TSREG, val | 0x20);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
set_font_mode(buf);
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
for (ch=0; ch < 256; ch++)
for (line=0; line < fontsize; line++)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (operation)
*(char *)pa_to_va(VIDEOMEM+(segment)+(ch*32)+line) =
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
font_image[(ch*fontsize)+line];
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
else
font_image[(ch*fontsize)+line] =
*(char *)pa_to_va(VIDEOMEM+(segment)+(ch*32)+line);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
set_normal_mode(buf);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
outb(TSIDX, 0x01); outb(TSREG, val & 0xDF); /* enable screen */
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static void
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
set_destructive_cursor(scr_stat *scp)
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
{
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
u_char buf[PARAM_BUFSIZE];
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
u_char cursor[32];
caddr_t address;
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
int i;
Emulate hw cursor closely, and get start&end scanlines from BIOS.
1995-03-30 14:32:31 +00:00
char *font_buffer;
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
if (scp->font_size < 14) {
Emulate hw cursor closely, and get start&end scanlines from BIOS.
1995-03-30 14:32:31 +00:00
font_buffer = font_8;
address = (caddr_t)VIDEOMEM + 0x8000;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
}
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
else if (scp->font_size >= 16) {
Emulate hw cursor closely, and get start&end scanlines from BIOS.
1995-03-30 14:32:31 +00:00
font_buffer = font_16;
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
address = (caddr_t)VIDEOMEM;
}
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
else {
font_buffer = font_14;
address = (caddr_t)VIDEOMEM + 0x4000;
}
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
/*
* FONT KLUDGE
* Always use the font page #0. XXX
*/
address = (caddr_t)VIDEOMEM;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
if (scp->status & MOUSE_VISIBLE) {
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
if ((scp->cursor_saveunder & 0xff) == SC_MOUSE_CHAR)
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(&scp->mouse_cursor[0], cursor, scp->font_size);
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
else if ((scp->cursor_saveunder & 0xff) == SC_MOUSE_CHAR + 1)
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(&scp->mouse_cursor[32], cursor, scp->font_size);
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
else if ((scp->cursor_saveunder & 0xff) == SC_MOUSE_CHAR + 2)
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(&scp->mouse_cursor[64], cursor, scp->font_size);
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
else if ((scp->cursor_saveunder & 0xff) == SC_MOUSE_CHAR + 3)
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(&scp->mouse_cursor[96], cursor, scp->font_size);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
else
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(font_buffer+((scp->cursor_saveunder & 0xff)*scp->font_size),
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
cursor, scp->font_size);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
}
else
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(font_buffer + ((scp->cursor_saveunder & 0xff) * scp->font_size),
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
cursor, scp->font_size);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
for (i=0; i<32; i++)
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
if ((i >= scp->cursor_start && i <= scp->cursor_end) ||
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
(scp->cursor_start >= scp->font_size && i == scp->font_size - 1))
Emulate hw cursor closely, and get start&end scanlines from BIOS.
1995-03-30 14:32:31 +00:00
cursor[i] |= 0xff;
Oops, the position of the savefont code was wrong, cur_console wasn't set yet.
1997-01-19 17:34:14 +00:00
#if 1
while (!(inb(crtc_addr+6) & 0x08)) /* wait for vertical retrace */ ;
#endif
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
set_font_mode(buf);
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
generic_bcopy(cursor, (char *)pa_to_va(address) + DEAD_CHAR * 32, 32);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
set_normal_mode(buf);
Optimized the way physical screen updates are done. Now only update what has actually been touched. This should speed up screen access on slow hardware. Introduced setting of "destructive" cursor size, much like the old hardware cursor.
1995-03-29 20:55:03 +00:00
}
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
static void
set_mouse_pos(scr_stat *scp)
{
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
static int last_xpos = -1, last_ypos = -1;
Fix the checks for screenborder for the mousepointer.
1997-08-09 19:24:03 +00:00
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
if (scp->mouse_xpos < 0)
scp->mouse_xpos = 0;
if (scp->mouse_ypos < 0)
scp->mouse_ypos = 0;
Yeah I'm back hacking syscons !! Add support for MODEX 320x240x256color with "unchained" adressing, giving access to all 256K on all VGA's, those with that much memory that is :) Also make sysmouse use the right resolution in graphics modes.
1997-08-08 22:52:30 +00:00
if (scp->status & UNKNOWN_MODE) {
Fix the checks for screenborder for the mousepointer.
1997-08-09 19:24:03 +00:00
if (scp->mouse_xpos > scp->xpixel-1)
Yeah I'm back hacking syscons !! Add support for MODEX 320x240x256color with "unchained" adressing, giving access to all 256K on all VGA's, those with that much memory that is :) Also make sysmouse use the right resolution in graphics modes.
1997-08-08 22:52:30 +00:00
scp->mouse_xpos = scp->xpixel-1;
Fix the checks for screenborder for the mousepointer.
1997-08-09 19:24:03 +00:00
if (scp->mouse_ypos > scp->ypixel-1)
Yeah I'm back hacking syscons !! Add support for MODEX 320x240x256color with "unchained" adressing, giving access to all 256K on all VGA's, those with that much memory that is :) Also make sysmouse use the right resolution in graphics modes.
1997-08-08 22:52:30 +00:00
scp->mouse_ypos = scp->ypixel-1;
return;
}
Fix the checks for screenborder for the mousepointer.
1997-08-09 19:24:03 +00:00
if (scp->mouse_xpos > (scp->xsize*8)-1)
scp->mouse_xpos = (scp->xsize*8)-1;
if (scp->mouse_ypos > (scp->ysize*scp->font_size)-1)
scp->mouse_ypos = (scp->ysize*scp->font_size)-1;
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
if (scp->mouse_xpos != last_xpos || scp->mouse_ypos != last_ypos) {
scp->status |= MOUSE_MOVED;
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
scp->mouse_pos = scp->scr_buf +
((scp->mouse_ypos/scp->font_size)*scp->xsize + scp->mouse_xpos/8);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
if ((scp->status & MOUSE_VISIBLE) && (scp->status & MOUSE_CUTTING))
mouse_cut(scp);
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
}
}
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
#define isspace(c) (((c) & 0xff) == ' ')
static int
skip_spc_right(scr_stat *scp, u_short *p)
{
int i;
for (i = (p - scp->scr_buf) % scp->xsize; i < scp->xsize; ++i) {
if (!isspace(*p))
break;
++p;
}
return i;
}
static int
skip_spc_left(scr_stat *scp, u_short *p)
{
int i;
for (i = (p-- - scp->scr_buf) % scp->xsize - 1; i >= 0; --i) {
if (!isspace(*p))
break;
--p;
}
return i;
}
static void
mouse_cut(scr_stat *scp)
{
u_short *end;
u_short *p;
int i = 0;
int j = 0;
scp->mouse_cut_end = (scp->mouse_pos >= scp->mouse_cut_start) ?
scp->mouse_pos + 1 : scp->mouse_pos;
end = (scp->mouse_cut_start > scp->mouse_cut_end) ?
scp->mouse_cut_start : scp->mouse_cut_end;
for (p = (scp->mouse_cut_start > scp->mouse_cut_end) ?
scp->mouse_cut_end : scp->mouse_cut_start; p < end; ++p) {
cut_buffer[i] = *p & 0xff;
/* remember the position of the last non-space char */
if (!isspace(cut_buffer[i++]))
j = i;
/* trim trailing blank when crossing lines */
if (((p - scp->scr_buf) % scp->xsize) == (scp->xsize - 1)) {
cut_buffer[j++] = '\n';
i = j;
}
}
cut_buffer[i] = '\0';
/* scan towards the end of the last line */
--p;
for (i = (p - scp->scr_buf) % scp->xsize; i < scp->xsize; ++i) {
if (!isspace(*p))
break;
++p;
}
/* if there is nothing but blank chars, trim them, but mark towards eol */
if (i >= scp->xsize) {
if (scp->mouse_cut_start > scp->mouse_cut_end)
scp->mouse_cut_start = p;
else
scp->mouse_cut_end = p;
cut_buffer[j++] = '\n';
cut_buffer[j] = '\0';
}
mark_for_update(scp, scp->mouse_cut_start - scp->scr_buf);
mark_for_update(scp, scp->mouse_cut_end - scp->scr_buf);
}
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
static void
mouse_cut_start(scr_stat *scp)
{
int i;
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
if (scp->status & MOUSE_VISIBLE) {
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
if (scp->mouse_pos == scp->mouse_cut_start &&
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
scp->mouse_cut_start == scp->mouse_cut_end - 1) {
cut_buffer[0] = '\0';
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
remove_cutmarking(scp);
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
} else if (skip_spc_right(scp, scp->mouse_pos) >= scp->xsize) {
/* if the pointer is on trailing blank chars, mark towards eol */
i = skip_spc_left(scp, scp->mouse_pos) + 1;
scp->mouse_cut_start = scp->scr_buf +
((scp->mouse_pos - scp->scr_buf) / scp->xsize) * scp->xsize + i;
scp->mouse_cut_end = scp->scr_buf +
((scp->mouse_pos - scp->scr_buf) / scp->xsize + 1) * scp->xsize;
cut_buffer[0] = '\n';
cut_buffer[1] = '\0';
scp->status |= MOUSE_CUTTING;
} else {
scp->mouse_cut_start = scp->mouse_pos;
scp->mouse_cut_end = scp->mouse_cut_start + 1;
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
cut_buffer[0] = *scp->mouse_cut_start & 0xff;
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
cut_buffer[1] = '\0';
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
scp->status |= MOUSE_CUTTING;
}
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
mark_all(scp);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
/* delete all other screens cut markings */
for (i=0; i<MAXCONS; i++) {
if (console[i] == NULL || console[i] == scp)
continue;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
remove_cutmarking(console[i]);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
}
}
}
static void
mouse_cut_end(scr_stat *scp)
{
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
if (scp->status & MOUSE_VISIBLE) {
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
scp->status &= ~MOUSE_CUTTING;
}
}
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
static void
mouse_cut_word(scr_stat *scp)
{
u_short *p;
u_short *sol;
u_short *eol;
int i;
/*
* Because we don't have locale information in the kernel,
* we only distinguish space char and non-space chars. Punctuation
* chars, symbols and other regular chars are all treated alike.
*/
if (scp->status & MOUSE_VISIBLE) {
sol = scp->scr_buf
+ ((scp->mouse_pos - scp->scr_buf) / scp->xsize) * scp->xsize;
eol = sol + scp->xsize;
if (isspace(*scp->mouse_pos)) {
for (p = scp->mouse_pos; p >= sol; --p)
if (!isspace(*p))
break;
scp->mouse_cut_start = ++p;
for (p = scp->mouse_pos; p < eol; ++p)
if (!isspace(*p))
break;
scp->mouse_cut_end = p;
} else {
for (p = scp->mouse_pos; p >= sol; --p)
if (isspace(*p))
break;
scp->mouse_cut_start = ++p;
for (p = scp->mouse_pos; p < eol; ++p)
if (isspace(*p))
break;
scp->mouse_cut_end = p;
}
for (i = 0, p = scp->mouse_cut_start; p < scp->mouse_cut_end; ++p)
cut_buffer[i++] = *p & 0xff;
cut_buffer[i] = '\0';
scp->status |= MOUSE_CUTTING;
}
}
static void
mouse_cut_line(scr_stat *scp)
{
u_short *p;
int i;
if (scp->status & MOUSE_VISIBLE) {
scp->mouse_cut_start = scp->scr_buf
+ ((scp->mouse_pos - scp->scr_buf) / scp->xsize) * scp->xsize;
scp->mouse_cut_end = scp->mouse_cut_start + scp->xsize;
for (i = 0, p = scp->mouse_cut_start; p < scp->mouse_cut_end; ++p)
cut_buffer[i++] = *p & 0xff;
cut_buffer[i++] = '\n';
cut_buffer[i] = '\0';
scp->status |= MOUSE_CUTTING;
}
}
static void
mouse_cut_extend(scr_stat *scp)
{
if ((scp->status & MOUSE_VISIBLE) && !(scp->status & MOUSE_CUTTING)
&& (scp->mouse_cut_start != NULL)) {
mouse_cut(scp);
scp->status |= MOUSE_CUTTING;
}
}
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
static void
mouse_paste(scr_stat *scp)
{
Changed mouse functionality a bit, now the pointer disappears if there is keyboard input. The mousepointer is shown again immediately if moved. Also a function pointer used to install a userwritten extra ioctl handler (sc_user_ioctl). This way its is possible to install user defined videomodes etc etc. No further changes should be in the kernel.
1996-10-18 18:51:37 +00:00
if (scp->status & MOUSE_VISIBLE) {
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
struct tty *tp;
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
u_char *ptr = cut_buffer;
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
tp = VIRTUAL_TTY(get_scr_num());
while (*ptr)
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
(*linesw[tp->t_line].l_rint)(scr_rmap[*ptr++], tp);
Change the way moused talk to syscons, now its only delivering mouseevents via an ioctl (MOUSE_ACTION). Fixed a couple of bugs (destructive cursor, uncut, jitter). Now applications can use the mouse via the MOUSE_MODE ioctl, its possible to have a signal sent on mouseevents, makeing an event loop in the application take over mouseevents.
1996-06-25 08:54:57 +00:00
}
}
Second round in syscons update: Support for pseudo graphic mouse cursor (not complete yet) Some cheap speed fixes. More cleanups. Call ourselves scxxxx finally.
1995-01-20 08:35:32 +00:00
static void
draw_mouse_image(scr_stat *scp)
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
caddr_t address;
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
int i;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
char *font_buffer;
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
u_char buf[PARAM_BUFSIZE];
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
u_short buffer[32];
u_short xoffset, yoffset;
u_short *crt_pos = Crtat + (scp->mouse_pos - scp->scr_buf);
Fixed bug in pasting 8bit char (ache). Added linefeeds in cuts that extend beyond one line. Prepared for the mousefunctions to be used in nontext modes.
1996-06-26 13:04:53 +00:00
int font_size = scp->font_size;
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
if (font_size < 14) {
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
font_buffer = font_8;
Emulate hw cursor closely, and get start&end scanlines from BIOS.
1995-03-30 14:32:31 +00:00
address = (caddr_t)VIDEOMEM + 0x8000;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
}
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
else if (font_size >= 16) {
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
font_buffer = font_16;
address = (caddr_t)VIDEOMEM;
}
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
else {
font_buffer = font_14;
address = (caddr_t)VIDEOMEM + 0x4000;
}
More fixes to deal with fonts: - Set the correct value scp->font_size in init_scp(). - Set scp->font_size to FONT_NONE for VGA_MODEX. Interim fix for a font problem: - A kludge to display the correct font on some video cards. We should be able to load multiple fonts to the VGA plane #2 and switch between fonts by setting the font select register in the VGA sequencer. It appears that the current code isn't functioning as expected on some VGA cards (I have reports on Millenium and Mach64 cards). This is either a bug in syscons or a hardware compatibility problem ;-< This kludge will always load only one font set at a time and always use the font page #0 on the plane #2. It is an interim kludge until we find the exact cause and solution. Small adjustment for mouse cursor handling: - Turn off the mouse cursor early when changing video modes. Video mode switch fixes: - Stop the screen saver when changing video modes. - Enclose the critical section with a pair of spltty()/splx(). - A kludge to prevent scrn_update() from accessing video memory in less- critical sections in video mode change; artificially turn on the UNKNOWN_MODE flag. PR: bin/5899, bin/5907 Tested by: ache and a couple of users OKed by: sos
1998-04-04 16:26:53 +00:00
/*
* FONT KLUDGE
* Always use the font page #0. XXX
*/
address = (caddr_t)VIDEOMEM;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
xoffset = scp->mouse_xpos % 8;
yoffset = scp->mouse_ypos % font_size;
/* prepare mousepointer char's bitmaps */
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(font_buffer + ((*(scp->mouse_pos) & 0xff) * font_size),
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
&scp->mouse_cursor[0], font_size);
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(font_buffer + ((*(scp->mouse_pos+1) & 0xff) * font_size),
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
&scp->mouse_cursor[32], font_size);
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(font_buffer + ((*(scp->mouse_pos+scp->xsize) & 0xff) * font_size),
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
&scp->mouse_cursor[64], font_size);
Now that we have replaced video-related block copy calls with sc_bcopy(), which is generic_bcopy() in reality, there is no point in retaining bcopyw() calls in the source code. OKed by : sos
1997-05-07 10:42:06 +00:00
bcopy(font_buffer + ((*(scp->mouse_pos+scp->xsize+1) & 0xff) * font_size),
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
&scp->mouse_cursor[96], font_size);
for (i=0; i<font_size; i++) {
buffer[i] = scp->mouse_cursor[i]<<8 | scp->mouse_cursor[i+32];
buffer[i+font_size]=scp->mouse_cursor[i+64]<<8|scp->mouse_cursor[i+96];
}
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
/* now and-or in the mousepointer image */
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
for (i=0; i<16; i++) {
buffer[i+yoffset] =
( buffer[i+yoffset] & ~(mouse_and_mask[i] >> xoffset))
| (mouse_or_mask[i] >> xoffset);
}
for (i=0; i<font_size; i++) {
scp->mouse_cursor[i] = (buffer[i] & 0xff00) >> 8;
scp->mouse_cursor[i+32] = buffer[i] & 0xff;
scp->mouse_cursor[i+64] = (buffer[i+font_size] & 0xff00) >> 8;
scp->mouse_cursor[i+96] = buffer[i+font_size] & 0xff;
}
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
scp->mouse_oldpos = scp->mouse_pos;
Oops, the position of the savefont code was wrong, cur_console wasn't set yet.
1997-01-19 17:34:14 +00:00
#if 1
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
/* wait for vertical retrace to avoid jitter on some videocards */
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
while (!(inb(crtc_addr+6) & 0x08)) /* idle */ ;
Oops, the position of the savefont code was wrong, cur_console wasn't set yet.
1997-01-19 17:34:14 +00:00
#endif
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
set_font_mode(buf);
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
generic_bcopy(scp->mouse_cursor, (char *)pa_to_va(address) + SC_MOUSE_CHAR * 32, 128);
Make comp_vgaregs() less strict about VGA register values when checking the BIOS video mode paramter table. Now syscons uses the parameter table even if some bits in the table are different from the current VGA register settings. Even if comp_vgaregs() finds that the BIOS video parameter table looks totally unfamiliar to it, syscons allows the user to change the current video mode to some modes which are based on the VGA 80x25 mode. They are VGA 80x30, VGA 80x50, VGA 80x60. In this case the user will be warned, during boot, that video mode switching is only paritally supported on his machine. PR: bin/4477
1997-11-21 11:37:07 +00:00
set_normal_mode(buf);
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
*(crt_pos) = (*(scp->mouse_pos) & 0xff00) | SC_MOUSE_CHAR;
*(crt_pos+scp->xsize) =
(*(scp->mouse_pos + scp->xsize) & 0xff00) | (SC_MOUSE_CHAR + 2);
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
if (scp->mouse_xpos < (scp->xsize-1)*8) {
- Move the mouse cursor to the center of the screen after video mode switch in ioctl. Possibly related to PR: kern/4271 - A kludge: initialize scp->xpixel and ypixel even in the text mode. If the console enters the `unknown' graphics mode via the ioctl KDSETMODE (KD_GRAPHICS), these fields are not set (because syscons cannot know the correct values), but set_mouse_pos() need to refer to these field to adjust the mouse position. - Turn off MOUSE_VISIBLE when switching video mode by ioctl. - another new option: SC_MOUSE_CHAR Define the first character code of four consecutive codes to be used for the mouse cursor. Default codes are 0xd0 through 0xd3. Beware that if you decide to use any codes outside the range of 0xc0-0xdf, the mouse cursor may not look good, because of the way VGA displays characters in 9-dot-wide character cells. Requested by several people. (This patch was tested by a person who recently reported, in the -current ML, a page fault problem in the kernel (draw_mouse_iamge()) after X server shutdown. The patch cured his problem.)
1998-02-11 15:02:40 +00:00
*(crt_pos + 1) = (*(scp->mouse_pos + 1) & 0xff00) | (SC_MOUSE_CHAR + 1);
*(crt_pos+scp->xsize + 1) =
(*(scp->mouse_pos + scp->xsize + 1) & 0xff00) | (SC_MOUSE_CHAR + 3);
Some news for syscons (long overdue): Real support for a Textmode mousecursor, works by reprogramming the charset. Together with this support for cut&paste in text mode. To use it a userland daemon is needed (moused), which provides the interface to the various mice protokols. Bug fixes here and there, all known PR's closed by this update.
1996-06-21 07:19:18 +00:00
}
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
mark_for_update(scp, scp->mouse_pos - scp->scr_buf);
mark_for_update(scp, scp->mouse_pos + scp->xsize + 1 - scp->scr_buf);
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
}
static void
remove_mouse_image(scr_stat *scp)
{
u_short *crt_pos = Crtat + (scp->mouse_oldpos - scp->scr_buf);
*(crt_pos) = *(scp->mouse_oldpos);
*(crt_pos+1) = *(scp->mouse_oldpos+1);
*(crt_pos+scp->xsize) = *(scp->mouse_oldpos+scp->xsize);
*(crt_pos+scp->xsize+1) = *(scp->mouse_oldpos+scp->xsize+1);
Fixed the userconfig problem (and one with ddb as well). The rudimentary support for a splash page is there, and works, it just needs a splash page...
1996-10-02 22:00:38 +00:00
mark_for_update(scp, scp->mouse_oldpos - scp->scr_buf);
mark_for_update(scp, scp->mouse_oldpos + scp->xsize + 1 - scp->scr_buf);
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
}
static void
draw_cutmarking(scr_stat *scp)
{
u_short *ptr;
u_short och, nch;
for (ptr=scp->scr_buf; ptr<=(scp->scr_buf+(scp->xsize*scp->ysize)); ptr++) {
nch = och = *(Crtat + (ptr - scp->scr_buf));
/* are we outside the selected area ? */
if ( ptr < (scp->mouse_cut_start > scp->mouse_cut_end ?
scp->mouse_cut_end : scp->mouse_cut_start) ||
- Add support for the following mice to psm/moused/sysmouse: MS IntelliMouse, Kensington Thinking Mouse, Genius NetScroll, Genius NetMouse, Genius NetMouse Pro, ALPS GlidePoint, ASCII MieMouse, Logitech MouseMan+, FirstMouse+ - The `psm' driver is made to recognize various models of PS/2 mice and enable their extra features so that their additional buttons and wheel/roller are recognized. The name of the detected model will be printed at boot time. - A set of new ioctl functions are added to the `psm', `mse' and `sysmouse' drivers so that the userland program (such as the X server) can query device information and change driver settings. - The wheel/roller movement is handled as the `Z' axis movement by the mouse drivers and the moused daemon. The Z axis movement may be mapped to another axis movement or buttons. - The mouse drivers support a new, standard mouse data format, MOUSE_PROTO_SYSMOUSE format which can encode x, y, and x axis movement and up to 10 buttons. /sys/i386/include/mouse.h - Added some fields to `mousestatus_t' to store Z axis movement and flag bits. - Added the field `model' to `mousehw_t' to store mouse model code. Defined model codes. - Extended `mousemode_t'. - Added new protocols and some constants for them. - Added new ioctl functions and structures. - Removed obsolete ioctl definitions. /sys/i386/include/console.h - Added `dz' field to the structure `mouse_data' to pass Z axis movement to `syscons/sysmouse'. - Removed LEFT_BUTTON, MIDDLE_BUTTON and RIGHT_BUTTON. Use button bits defined in `mouse.h' instead. /sys/i386/isa/psm.c - Added a set of functions to detect various mice which have additional features (wheel and buttons) unavailable in the standard PS/2 mouse. - Refined existing ioctl functions and added new ones. Most important of all is MOUSE_SETLEVEL which manipulates the output level of the driver. While the output level remains zero, the output from the `psm' driver is in the standard PS/2 mouse format (three bytes long). When the level is set to one, the `psm' driver will send data in the extended format. At the level two the driver uses the format which is native to the connected mouse is used. (Meaning that the output from the device is passed to the caller as is, unmodified.) The `psm' driver will pass such extended data format as is to the caller if the output level is two, but emulates the standard format if the output level is zero. - Added kernel configuration flags to set initial resolution (PSM_CONFIG_RESOLUTION) and acceleration (PSM_CONFIG_ACCEL). - Removed the compile options PSM_ACCEL, PSM_CHECKSYNC and PSM_EMULATION. Acceleration ratio is now specified by the kernel configuration flags stated above. Sync check logic is refined and now standard. The sync check can be turned off by the new kernel configuration flags PSM_CONFIG_NOCHECKSYNC (0x100). PSM_EMULATION has been of little use. - Summer clean up :-) Removed unused code and obsolete comments. /sys/i386/isa/mse.c - Created mseioctl() to deal with ioctl functions MOUSE_XXXX. Most importantly, the MOUSE_SETLEVEL ioctl will change the output format from the 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Use constants defined in `mouse.h' rather than magic numbers. /sys/i386/isa/syscons.c - Changed scioctl() to reflect the new `console.h' and some of the new ioctls defined in `mouse.h'. Most importantly, the MOUSE_SETLEVEL ioctl will change the `sysmouse' output format from the MouseSystems 5 byte format to the new, extended format so that the caller can take advantage of Z axis movement and additional buttons. - Added support for double/triple click actions of the left button and single click action of the right button in the virtual console. The left button double click will select a word under the mouse pointer. The triple click will select a line and the single click of the right button will extend the selected region to the current position of the mouse pointer. This will make the cut/paste support more compatible with xterm. /sys/i386/isa/kbdio.h - Added PSM_INTELLI_ID.
1997-12-07 08:09:19 +00:00
ptr >= (scp->mouse_cut_start > scp->mouse_cut_end ?
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
scp->mouse_cut_start : scp->mouse_cut_end)) {
if (ptr != scp->cursor_pos)
nch = (och & 0xff) | (*ptr & 0xff00);
}
else {
/* are we clear of the cursor image ? */
if (ptr != scp->cursor_pos)
nch = (och & 0x88ff) | (*ptr & 0x7000)>>4 | (*ptr & 0x0700)<<4;
else {
if (flags & CHAR_CURSOR)
nch = (och & 0x88ff)|(*ptr & 0x7000)>>4|(*ptr & 0x0700)<<4;
else
if (!(flags & BLINK_CURSOR))
nch = (och & 0xff) | (*ptr & 0xff00);
}
}
if (nch != och)
*(Crtat + (ptr - scp->scr_buf)) = nch;
}
}
static void
remove_cutmarking(scr_stat *scp)
{
scp->mouse_cut_start = scp->mouse_cut_end = NULL;
scp->status &= ~MOUSE_CUTTING;
mark_all(scp);
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
static void
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
save_palette(void)
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
int i;
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
outb(PALRADR, 0x00);
for (i=0x00; i<0x300; i++)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
palette[i] = inb(PALDATA);
inb(crtc_addr+6); /* reset flip/flop */
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
}
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
void
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
load_palette(char *palette)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
int i;
outb(PIXMASK, 0xFF); /* no pixelmask */
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
outb(PALWADR, 0x00);
for (i=0x00; i<0x300; i++)
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
outb(PALDATA, palette[i]);
inb(crtc_addr+6); /* reset flip/flop */
outb(ATC, 0x20); /* enable palette */
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
static void
do_bell(scr_stat *scp, int pitch, int duration)
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
{
- Ignore KBD_RAW_MODE and KBD_CODE_MODE in scgetc() when it's called from the low-level console routines sccngetc() and sccncheckc(). Submitted by: bde (a long time ago) - Don't try to ring bell and immediately return from do_bell() while device probe is in progress at boot time; the timeout queue is not functional yet. PR: kern/2424 - Stop running the screen saver after panic() is called: check if `panicstr' is non-NULL during scrn_timer(). PR: kern/5314 - A new option: SC_DISABLE_REBOOT The reboot key (usually Ctl-Alt-Del) will be ignored if this option is defined. You may still have the reboot key defined in the keymap and it won't cause error when the keymap is loaded, but it will be quietly treated as nop. OKed by: sos
1998-02-11 14:56:02 +00:00
if (cold)
return;
Fixed a couple of bugs in the mousepointer code. Changed update strategy slightly. Make set_mode & copy_font externally visible.
1996-09-01 18:16:06 +00:00
if (flags & VISUAL_BELL) {
Deliver all bells to current console, use different pitch for non-originating bells.
1996-02-05 14:08:39 +00:00
if (blink_in_progress)
return;
blink_in_progress = 4;
if (scp != cur_console)
blink_in_progress += 2;
blink_screen(cur_console);
} else {
if (scp != cur_console)
pitch *= 2;
sysbeep(pitch, duration);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
static void
Fixed the type of timeout functions and removed casts that hid the type mismatches. Not taking an arg in scrn_timer() broke `cc -mrtd'.
1997-04-20 16:05:33 +00:00
blink_screen(void *arg)
First round in syscons update. Several new features has been added: No kernel config options anymore besides keyboard language layout. Virtual consoles are now dynamically allocated, no NCONS anymore. Software cursor blinking/nonblinking. Visual bell for laptops (don't beep at meetings :-). Cursor/bell default type setable via config "flags" instead of as defines. Cursor/bell type setable via ioctl's. New video modes 80x30 80x60 for some laptops, and those with multisync monitors. Scroll-lock history (length currently fixed at 100 lines). Lots of cleanups, some only commented out for now (will goaway soon). Support for new features in vidcontrol/kbdcontrol. Updated manpages.
1995-01-12 11:47:05 +00:00
{
Fixed the type of timeout functions and removed casts that hid the type mismatches. Not taking an arg in scrn_timer() broke `cc -mrtd'.
1997-04-20 16:05:33 +00:00
scr_stat *scp = arg;
1) font loading (two fixes) When an ioctl command SW_XXXX is issued, scioctl() checks if the font appropriate for the specified mode is already loaded. The check was correctly done for 8 line and 16 line fonts, but not for 14 line font. The symbols FONT_8, FONT_14 and FONT_16 were defined as numbers but were sometimes treated as bit flags. They are now defined as bit flags. 2) screen blinking (two fixes) Removed a redundant call to timeout() in do_bell(). Don't let blink_screen() write to the video buffer if the screen is in the graphics (UNKNOWN) mode. 3) screen saver timeout The ioctl command CONS_BLANKTIME sets the screen saver's timeout. The value of zero will disable the screen saver. If the screen saver is currently running it should be stopped. 4) border color and destructive cursor (two fixes) The border color and the cursor type can be changed via escape sequences. But only VGA can change the border color and set the cursor type to destructive (CHAR_CURSOR) in the current syscons. scan_esc() failed to check this. Reviewed by: sos
1997-05-15 05:43:59 +00:00
if ((scp->status & UNKNOWN_MODE) || (blink_in_progress <= 1)) {
blink_in_progress = FALSE;
mark_all(scp);
if (delayed_next_scr)
switch_scr(scp, delayed_next_scr - 1);
}
else {
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
if (blink_in_progress & 1)
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(kernel_default.std_color | scr_map[0x20],
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
Crtat, scp->xsize * scp->ysize);
else
Separate colors & attributes as Terry points Reviewed by: soren
1995-11-28 00:17:32 +00:00
fillw(kernel_default.rev_color | scr_map[0x20],
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
Crtat, scp->xsize * scp->ysize);
blink_in_progress--;
Fixed the type of timeout functions and removed casts that hid the type mismatches. Not taking an arg in scrn_timer() broke `cc -mrtd'.
1997-04-20 16:05:33 +00:00
timeout(blink_screen, scp, hz / 10);
Next syscons update (given up on numbering :) Removed screensavers from syscons, they are now LKM's. This makes it possible to do some really "interesting" screensavers... Fixed bug that sometimes caused garbage to appear when leaving "scroll-lock" history. Reformattet indentation, it got too deep for a normal 80 pos screen. Split up in syscons.c & syscons.h for use with the saver-lkm's. Temporarily removed -s option from vidcontrol, savers should now be loaded with modload.
1995-02-22 13:48:07 +00:00
}
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
}
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
void
sc_bcopy(u_short *p, int from, int to, int mark)
{
if (!vesa_mode) {
generic_bcopy(p+from, Crtat+from, (to-from+1)*sizeof (u_short));
} else if (vesa_mode == 0x102) {
Hmm, it is generally an advantage to commit the most recent version of ones changes: A faster character painter procedure and fix on cursor bug.
1998-02-13 11:31:34 +00:00
u_char *d, *e;
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
int i,j;
Hmm, it is generally an advantage to commit the most recent version of ones changes: A faster character painter procedure and fix on cursor bug.
1998-02-13 11:31:34 +00:00
if (mark)
mark = 255;
d = (u_char *)Crtat;
d += 10 + 6*16*100 + (from%80) + 16*100*(from/80);
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
for (i = from ; i <= to ; i++) {
Hmm, it is generally an advantage to commit the most recent version of ones changes: A faster character painter procedure and fix on cursor bug.
1998-02-13 11:31:34 +00:00
e = d;
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
for (j = 0 ; j < 16; j++) {
Hmm, it is generally an advantage to commit the most recent version of ones changes: A faster character painter procedure and fix on cursor bug.
1998-02-13 11:31:34 +00:00
*e = mark^font_16[(p[i]&0xff)*16+j];
e+=100;
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
}
Hmm, it is generally an advantage to commit the most recent version of ones changes: A faster character painter procedure and fix on cursor bug.
1998-02-13 11:31:34 +00:00
d++;
if ((i % 80) == 79)
d += 20 + 15*100;
Add support for VESA mode 0x102 (800x600x4) in syscons. You can activate this using option "-b" to the boot blocks. It is smartest to compile a font into your kernel (See LINT), but not mandatory, but apart from the cursor you will see nothing on the screen until you load a font. This mode allows XF86_VGA16 to run in 800x600 mode on otherwise unsupported graphics hardware. A number of buglets in the cursor handling in syscons may become visible this way.
1998-02-12 22:05:08 +00:00
}
}
}
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
#ifdef SC_SPLASH_SCREEN
static void
toggle_splash_screen(scr_stat *scp)
{
static int toggle = 0;
static u_char save_mode;
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
int s;
if (video_mode_ptr == NULL)
return;
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
Addresses the following two problems. 1) i586_bcopy() problem There have been a number of reports that the syscons doesn't work properly if i586_bcopy() is enabled. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is that the system looks frozen during device probe or just before the main installation menu. The workaround was to specify the flag 0x01 to the npx device so that i586_bcopy() is disabled. The patch forces the syscons to call generic_bcopy() when copying to/from the video memory, even if CPU is Pentium and i586_bcopy() is enabled. i586_bcopy() is still called for copy operations between non-video memory regions. PR: kern/2277, kern/3066, kern/3107, kern/3134 2) video mode parameter table problem The syscons reads and uses the video mode parameter table provided by the VGA BIOS to set VGA registers when changing video mode and modifying font data. It appears that in some VGA BIOSes the table is not ordered as the syscons expects, and this leads to screen corruption. The problem prevented users from installing 2.2(.1)-RELEASE. The symptom is the corrupt screen or strange vertical lines soon after the kernel is loaded into memory (just after the kernel decompression). The patch performs simplistic test and if it fails, set video_mode_ptr to NULL so that the video mode switching won't happen. This is an interim kludge. There should be a better way to deal with the problem. PR: kern/2498, conf/2775, conf/3354 Reviewed by: sos Tested by: PR originators (not all of them, though)
1997-04-27 09:03:10 +00:00
s = splhigh();
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
if (toggle) {
scp->mode = save_mode;
scp->status &= ~UNKNOWN_MODE;
set_mode(scp);
Upgrade the kbdio rutines to provide queued kbd & mouse events. Minor other updates to syscons by me. Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
1997-01-15 18:16:32 +00:00
load_palette(palette);
Fix a couble of nasties regarding mouse pointer and different resolutions. Allow middle mouse button to be used for pasting. Also added the beginnings of support for a splash page.
1996-09-30 23:00:58 +00:00
toggle = 0;
}
else {
save_mode = scp->mode;
scp->mode = M_VGA_CG320;
scp->status |= UNKNOWN_MODE;
set_mode(scp);
/* load image */
toggle = 1;
}
splx(s);
}
#endif
Added necessary files for syscons (may move font include later, but for now it's easier to leave there since I don't feel like breaking anything).
1993-07-26 21:21:11 +00:00
#endif /* NSC */
Reference in New Issue Copy Permalink