1993-06-12 14:58:17 +00:00
|
|
|
/*-
|
|
|
|
|
* Copyright (c) 1991 The Regents of the University of California.
|
|
|
|
|
* All rights reserved.
|
|
|
|
|
*
|
|
|
|
|
* This code is derived from software contributed to Berkeley by
|
|
|
|
|
* William Jolitz.
|
|
|
|
|
*
|
|
|
|
|
* 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
|
|
|
|
|
* notice, this list of conditions and the following disclaimer.
|
|
|
|
|
* 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.
|
|
|
|
|
* 3. All advertising materials mentioning features or use of this software
|
|
|
|
|
* must display the following acknowledgement:
|
|
|
|
|
* This product includes software developed by the University of
|
|
|
|
|
* California, Berkeley and its contributors.
|
|
|
|
|
* 4. Neither the name of the University nor the names of its contributors
|
|
|
|
|
* may be used to endorse or promote products derived from this software
|
|
|
|
|
* without specific prior written permission.
|
|
|
|
|
*
|
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``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 REGENTS OR CONTRIBUTORS 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.
|
|
|
|
|
*
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
* from: @(#)isa.c 7.2 (Berkeley) 5/13/91
|
1998-11-29 15:42:40 +00:00
|
|
|
* $Id: isa.c,v 1.116 1998/10/22 05:58:39 bde Exp $
|
1993-06-12 14:58:17 +00:00
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* code to manage AT bus
|
|
|
|
|
*
|
|
|
|
|
* 92/08/18 Frank P. MacLachlan (fpm@crash.cts.com):
|
|
|
|
|
* Fixed uninitialized variable problem and added code to deal
|
|
|
|
|
* with DMA page boundaries in isa_dmarangecheck(). Fixed word
|
|
|
|
|
* mode DMA count compution and reorganized DMA setup code in
|
|
|
|
|
* isa_dmastart()
|
|
|
|
|
*/
|
|
|
|
|
|
1994-08-13 03:50:34 +00:00
|
|
|
#include <sys/param.h>
|
1996-01-27 01:56:30 +00:00
|
|
|
#include <sys/systm.h>
|
1994-08-13 03:50:34 +00:00
|
|
|
#include <sys/buf.h>
|
|
|
|
|
#include <sys/malloc.h>
|
1997-05-31 09:27:31 +00:00
|
|
|
#include <machine/ipl.h>
|
1996-06-25 20:31:01 +00:00
|
|
|
#include <machine/md_var.h>
|
1997-06-25 21:00:00 +00:00
|
|
|
#ifdef APIC_IO
|
1997-04-26 11:46:25 +00:00
|
|
|
#include <machine/smp.h>
|
|
|
|
|
#endif /* APIC_IO */
|
1994-08-13 03:50:34 +00:00
|
|
|
#include <vm/vm.h>
|
1995-12-07 12:48:31 +00:00
|
|
|
#include <vm/vm_param.h>
|
|
|
|
|
#include <vm/pmap.h>
|
1994-08-13 03:50:34 +00:00
|
|
|
#include <i386/isa/isa_device.h>
|
1997-06-02 08:19:06 +00:00
|
|
|
#include <i386/isa/intr_machdep.h>
|
1994-08-13 03:50:34 +00:00
|
|
|
#include <i386/isa/isa.h>
|
|
|
|
|
#include <i386/isa/ic/i8237.h>
|
1993-06-12 14:58:17 +00:00
|
|
|
|
1997-05-26 14:42:24 +00:00
|
|
|
#include <sys/interrupt.h>
|
|
|
|
|
|
1997-09-19 15:20:25 +00:00
|
|
|
#include "pnp.h"
|
|
|
|
|
#if NPNP > 0
|
|
|
|
|
#include <i386/isa/pnp.h>
|
|
|
|
|
#endif
|
|
|
|
|
|
1993-06-12 14:58:17 +00:00
|
|
|
/*
|
|
|
|
|
** Register definitions for DMA controller 1 (channels 0..3):
|
|
|
|
|
*/
|
|
|
|
|
#define DMA1_CHN(c) (IO_DMA1 + 1*(2*(c))) /* addr reg for channel c */
|
|
|
|
|
#define DMA1_SMSK (IO_DMA1 + 1*10) /* single mask register */
|
|
|
|
|
#define DMA1_MODE (IO_DMA1 + 1*11) /* mode register */
|
|
|
|
|
#define DMA1_FFC (IO_DMA1 + 1*12) /* clear first/last FF */
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
** Register definitions for DMA controller 2 (channels 4..7):
|
|
|
|
|
*/
|
1993-10-18 18:45:48 +00:00
|
|
|
#define DMA2_CHN(c) (IO_DMA2 + 2*(2*(c))) /* addr reg for channel c */
|
1993-06-12 14:58:17 +00:00
|
|
|
#define DMA2_SMSK (IO_DMA2 + 2*10) /* single mask register */
|
|
|
|
|
#define DMA2_MODE (IO_DMA2 + 2*11) /* mode register */
|
|
|
|
|
#define DMA2_FFC (IO_DMA2 + 2*12) /* clear first/last FF */
|
|
|
|
|
|
1994-08-18 05:09:36 +00:00
|
|
|
static void config_isadev __P((struct isa_device *isdp, u_int *mp));
|
1995-03-28 07:58:53 +00:00
|
|
|
static void config_isadev_c __P((struct isa_device *isdp, u_int *mp,
|
|
|
|
|
int reconfig));
|
1994-08-18 05:09:36 +00:00
|
|
|
static void conflict __P((struct isa_device *dvp, struct isa_device *tmpdvp,
|
|
|
|
|
int item, char const *whatnot, char const *reason,
|
|
|
|
|
char const *format));
|
|
|
|
|
static int haveseen __P((struct isa_device *dvp, struct isa_device *tmpdvp,
|
|
|
|
|
u_int checkbits));
|
1996-04-08 19:38:57 +00:00
|
|
|
static int isa_dmarangecheck __P((caddr_t va, u_int length, int chan));
|
1993-06-12 14:58:17 +00:00
|
|
|
|
|
|
|
|
/*
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
* print a conflict message
|
1993-06-12 14:58:17 +00:00
|
|
|
*/
|
1994-08-18 05:09:36 +00:00
|
|
|
static void
|
|
|
|
|
conflict(dvp, tmpdvp, item, whatnot, reason, format)
|
|
|
|
|
struct isa_device *dvp;
|
|
|
|
|
struct isa_device *tmpdvp;
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
int item;
|
1994-08-18 05:09:36 +00:00
|
|
|
char const *whatnot;
|
|
|
|
|
char const *reason;
|
|
|
|
|
char const *format;
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
{
|
1994-08-18 05:09:36 +00:00
|
|
|
printf("%s%d not %sed due to %s conflict with %s%d at ",
|
|
|
|
|
dvp->id_driver->name, dvp->id_unit, whatnot, reason,
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
tmpdvp->id_driver->name, tmpdvp->id_unit);
|
|
|
|
|
printf(format, item);
|
|
|
|
|
printf("\n");
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
1995-05-11 02:15:55 +00:00
|
|
|
* Check to see if things are already in use, like IRQ's, I/O addresses
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
* and Memory addresses.
|
1993-06-12 14:58:17 +00:00
|
|
|
*/
|
1994-08-18 05:09:36 +00:00
|
|
|
static int
|
|
|
|
|
haveseen(dvp, tmpdvp, checkbits)
|
|
|
|
|
struct isa_device *dvp;
|
|
|
|
|
struct isa_device *tmpdvp;
|
|
|
|
|
u_int checkbits;
|
1993-06-12 14:58:17 +00:00
|
|
|
{
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
/*
|
1997-08-25 22:52:59 +00:00
|
|
|
* Ignore all conflicts except IRQ ones if conflicts are allowed.
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
*/
|
1997-08-25 22:52:59 +00:00
|
|
|
if (dvp->id_conflicts)
|
|
|
|
|
checkbits &= ~(CC_DRQ | CC_IOADDR | CC_MEMADDR);
|
|
|
|
|
/*
|
|
|
|
|
* Only check against devices that have already been found.
|
|
|
|
|
*/
|
|
|
|
|
if (tmpdvp->id_alive) {
|
1994-08-18 05:09:36 +00:00
|
|
|
char const *whatnot;
|
|
|
|
|
|
1998-01-31 03:29:00 +00:00
|
|
|
/*
|
|
|
|
|
* Check for device driver & unit conflict; just drop probing
|
|
|
|
|
* a device which has already probed true. This is usually
|
|
|
|
|
* not strictly a conflict, but rather the case of somebody
|
|
|
|
|
* having specified several mutually exclusive configurations
|
|
|
|
|
* for a single device.
|
|
|
|
|
*/
|
|
|
|
|
if (tmpdvp->id_driver == dvp->id_driver &&
|
|
|
|
|
tmpdvp->id_unit == dvp->id_unit) {
|
|
|
|
|
return 1;
|
|
|
|
|
}
|
|
|
|
|
|
1994-09-02 22:13:34 +00:00
|
|
|
whatnot = checkbits & CC_ATTACH ? "attach" : "prob";
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
/*
|
|
|
|
|
* Check for I/O address conflict. We can only check the
|
|
|
|
|
* starting address of the device against the range of the
|
|
|
|
|
* device that has already been probed since we do not
|
|
|
|
|
* know how many I/O addresses this device uses.
|
|
|
|
|
*/
|
1994-08-18 05:09:36 +00:00
|
|
|
if (checkbits & CC_IOADDR && tmpdvp->id_alive != -1) {
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
if ((dvp->id_iobase >= tmpdvp->id_iobase) &&
|
|
|
|
|
(dvp->id_iobase <=
|
|
|
|
|
(tmpdvp->id_iobase + tmpdvp->id_alive - 1))) {
|
1998-10-12 13:12:45 +00:00
|
|
|
if (!(checkbits & CC_QUIET))
|
|
|
|
|
conflict(dvp, tmpdvp, dvp->id_iobase,
|
|
|
|
|
whatnot, "I/O address", "0x%x");
|
1994-10-17 21:16:41 +00:00
|
|
|
return 1;
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
}
|
|
|
|
|
/*
|
|
|
|
|
* Check for Memory address conflict. We can check for
|
|
|
|
|
* range overlap, but it will not catch all cases since the
|
|
|
|
|
* driver may adjust the msize paramater during probe, for
|
|
|
|
|
* now we just check that the starting address does not
|
|
|
|
|
* fall within any allocated region.
|
|
|
|
|
* XXX could add a second check after the probe for overlap,
|
|
|
|
|
* since at that time we would know the full range.
|
|
|
|
|
* XXX KERNBASE is a hack, we should have vaddr in the table!
|
|
|
|
|
*/
|
1994-08-18 05:09:36 +00:00
|
|
|
if (checkbits & CC_MEMADDR && tmpdvp->id_maddr) {
|
|
|
|
|
if ((KERNBASE + dvp->id_maddr >= tmpdvp->id_maddr) &&
|
|
|
|
|
(KERNBASE + dvp->id_maddr <=
|
|
|
|
|
(tmpdvp->id_maddr + tmpdvp->id_msize - 1))) {
|
1998-10-12 13:12:45 +00:00
|
|
|
if (!(checkbits & CC_QUIET))
|
|
|
|
|
conflict(dvp, tmpdvp,
|
|
|
|
|
(int)dvp->id_maddr, whatnot,
|
|
|
|
|
"maddr", "0x%x");
|
1994-10-17 21:16:41 +00:00
|
|
|
return 1;
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
}
|
|
|
|
|
/*
|
|
|
|
|
* Check for IRQ conflicts.
|
|
|
|
|
*/
|
1994-08-18 05:09:36 +00:00
|
|
|
if (checkbits & CC_IRQ && tmpdvp->id_irq) {
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
if (tmpdvp->id_irq == dvp->id_irq) {
|
1998-10-12 13:12:45 +00:00
|
|
|
if (!(checkbits & CC_QUIET))
|
|
|
|
|
conflict(dvp, tmpdvp,
|
|
|
|
|
ffs(dvp->id_irq) - 1, whatnot,
|
|
|
|
|
"irq", "%d");
|
1994-10-17 21:16:41 +00:00
|
|
|
return 1;
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
}
|
|
|
|
|
/*
|
|
|
|
|
* Check for DRQ conflicts.
|
|
|
|
|
*/
|
1994-08-18 05:09:36 +00:00
|
|
|
if (checkbits & CC_DRQ && tmpdvp->id_drq != -1) {
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
if (tmpdvp->id_drq == dvp->id_drq) {
|
1998-10-12 13:12:45 +00:00
|
|
|
if (!(checkbits & CC_QUIET))
|
|
|
|
|
conflict(dvp, tmpdvp, dvp->id_drq,
|
|
|
|
|
whatnot, "drq", "%d");
|
1994-10-17 21:16:41 +00:00
|
|
|
return 1;
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
}
|
|
|
|
|
}
|
1994-10-17 21:16:41 +00:00
|
|
|
return 0;
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
}
|
1993-06-12 14:58:17 +00:00
|
|
|
|
1997-05-26 14:42:24 +00:00
|
|
|
#ifdef RESOURCE_CHECK
|
|
|
|
|
#include <sys/drvresource.h>
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
checkone (struct isa_device *dvp, int type, addr_t low, addr_t high,
|
|
|
|
|
char *resname, char *resfmt, int attaching)
|
|
|
|
|
{
|
|
|
|
|
int result = 0;
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
if (low == high)
|
|
|
|
|
printf("\tcheck %s: 0x%x\n", resname, low);
|
|
|
|
|
else
|
|
|
|
|
printf("\tcheck %s: 0x%x to 0x%x\n",
|
|
|
|
|
resname, low, high);
|
|
|
|
|
}
|
|
|
|
|
if (resource_check(type, RESF_NONE, low, high) != NULL) {
|
|
|
|
|
char *whatnot = attaching ? "attach" : "prob";
|
|
|
|
|
static struct isa_device dummydev;
|
|
|
|
|
static struct isa_driver dummydrv;
|
|
|
|
|
struct isa_device *tmpdvp = &dummydev;
|
|
|
|
|
|
|
|
|
|
dummydev.id_driver = &dummydrv;
|
|
|
|
|
dummydev.id_unit = 0;
|
|
|
|
|
dummydrv.name = "pci";
|
|
|
|
|
conflict(dvp, tmpdvp, low, whatnot, resname, resfmt);
|
|
|
|
|
result = 1;
|
|
|
|
|
} else if (attaching) {
|
|
|
|
|
if (low == high)
|
|
|
|
|
printf("\tregister %s: 0x%x\n", resname, low);
|
|
|
|
|
else
|
|
|
|
|
printf("\tregister %s: 0x%x to 0x%x\n",
|
|
|
|
|
resname, low, high);
|
|
|
|
|
resource_claim(dvp, type, RESF_NONE, low, high);
|
|
|
|
|
}
|
|
|
|
|
return (result);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
check_pciconflict(struct isa_device *dvp, int checkbits)
|
|
|
|
|
{
|
|
|
|
|
int result = 0;
|
|
|
|
|
int attaching = (checkbits & CC_ATTACH) != 0;
|
|
|
|
|
|
|
|
|
|
if (checkbits & CC_MEMADDR) {
|
|
|
|
|
long maddr = dvp->id_maddr;
|
|
|
|
|
long msize = dvp->id_msize;
|
|
|
|
|
if (msize > 0) {
|
|
|
|
|
if (checkone(dvp, REST_MEM, maddr, maddr + msize - 1,
|
|
|
|
|
"maddr", "0x%x", attaching) != 0) {
|
|
|
|
|
result = 1;
|
|
|
|
|
attaching = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (checkbits & CC_IOADDR) {
|
|
|
|
|
unsigned iobase = dvp->id_iobase;
|
|
|
|
|
unsigned iosize = dvp->id_alive;
|
|
|
|
|
if (iosize == -1)
|
|
|
|
|
iosize = 1; /* XXX can't do much about this ... */
|
|
|
|
|
if (iosize > 0) {
|
|
|
|
|
if (checkone(dvp, REST_PORT, iobase, iobase + iosize -1,
|
|
|
|
|
"I/O address", "0x%x", attaching) != 0) {
|
|
|
|
|
result = 1;
|
|
|
|
|
attaching = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (checkbits & CC_IRQ) {
|
|
|
|
|
int irq = ffs(dvp->id_irq) - 1;
|
|
|
|
|
if (irq >= 0) {
|
|
|
|
|
if (checkone(dvp, REST_INT, irq, irq,
|
|
|
|
|
"irq", "%d", attaching) != 0) {
|
|
|
|
|
result = 1;
|
|
|
|
|
attaching = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (checkbits & CC_DRQ) {
|
|
|
|
|
int drq = dvp->id_drq;
|
|
|
|
|
if (drq >= 0) {
|
|
|
|
|
if (checkone(dvp, REST_DMA, drq, drq,
|
|
|
|
|
"drq", "%d", attaching) != 0) {
|
|
|
|
|
result = 1;
|
|
|
|
|
attaching = 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (result != 0)
|
|
|
|
|
resource_free (dvp);
|
|
|
|
|
return (result);
|
|
|
|
|
}
|
|
|
|
|
#endif /* RESOURCE_CHECK */
|
|
|
|
|
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
/*
|
|
|
|
|
* Search through all the isa_devtab_* tables looking for anything that
|
|
|
|
|
* conflicts with the current device.
|
|
|
|
|
*/
|
1995-04-23 09:13:08 +00:00
|
|
|
int
|
1994-08-18 05:09:36 +00:00
|
|
|
haveseen_isadev(dvp, checkbits)
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
struct isa_device *dvp;
|
1994-08-18 05:09:36 +00:00
|
|
|
u_int checkbits;
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
{
|
1997-09-19 15:20:25 +00:00
|
|
|
#if NPNP > 0
|
|
|
|
|
struct pnp_dlist_node *nod;
|
|
|
|
|
#endif
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
struct isa_device *tmpdvp;
|
|
|
|
|
int status = 0;
|
1993-06-12 14:58:17 +00:00
|
|
|
|
1994-10-17 21:16:41 +00:00
|
|
|
for (tmpdvp = isa_devtab_tty; tmpdvp->id_driver; tmpdvp++) {
|
1994-08-18 05:09:36 +00:00
|
|
|
status |= haveseen(dvp, tmpdvp, checkbits);
|
1994-10-17 21:16:41 +00:00
|
|
|
if (status)
|
|
|
|
|
return status;
|
|
|
|
|
}
|
|
|
|
|
for (tmpdvp = isa_devtab_bio; tmpdvp->id_driver; tmpdvp++) {
|
1994-08-18 05:09:36 +00:00
|
|
|
status |= haveseen(dvp, tmpdvp, checkbits);
|
1994-10-17 21:16:41 +00:00
|
|
|
if (status)
|
|
|
|
|
return status;
|
|
|
|
|
}
|
|
|
|
|
for (tmpdvp = isa_devtab_net; tmpdvp->id_driver; tmpdvp++) {
|
1994-08-18 05:09:36 +00:00
|
|
|
status |= haveseen(dvp, tmpdvp, checkbits);
|
1994-10-17 21:16:41 +00:00
|
|
|
if (status)
|
|
|
|
|
return status;
|
|
|
|
|
}
|
1997-09-21 21:41:49 +00:00
|
|
|
for (tmpdvp = isa_devtab_cam; tmpdvp->id_driver; tmpdvp++) {
|
|
|
|
|
status |= haveseen(dvp, tmpdvp, checkbits);
|
|
|
|
|
if (status)
|
|
|
|
|
return status;
|
|
|
|
|
}
|
1994-10-17 21:16:41 +00:00
|
|
|
for (tmpdvp = isa_devtab_null; tmpdvp->id_driver; tmpdvp++) {
|
1994-08-18 05:09:36 +00:00
|
|
|
status |= haveseen(dvp, tmpdvp, checkbits);
|
1994-10-17 21:16:41 +00:00
|
|
|
if (status)
|
|
|
|
|
return status;
|
|
|
|
|
}
|
1997-09-19 15:20:25 +00:00
|
|
|
#if NPNP > 0
|
|
|
|
|
for (nod = pnp_device_list; nod != NULL; nod = nod->next)
|
|
|
|
|
if (status |= haveseen(dvp, &(nod->dev), checkbits))
|
|
|
|
|
return status;
|
|
|
|
|
#endif
|
1997-05-26 14:42:24 +00:00
|
|
|
#ifdef RESOURCE_CHECK
|
1997-11-30 09:44:28 +00:00
|
|
|
if (!dvp->id_conflicts)
|
1997-05-26 14:42:24 +00:00
|
|
|
status = check_pciconflict(dvp, checkbits);
|
1997-11-30 09:44:28 +00:00
|
|
|
else if (bootverbose)
|
1997-05-26 14:42:24 +00:00
|
|
|
printf("\tnot checking for resource conflicts ...\n");
|
|
|
|
|
#endif /* RESOURCE_CHECK */
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
return(status);
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
|
1993-06-12 14:58:17 +00:00
|
|
|
/*
|
|
|
|
|
* Configure all ISA devices
|
|
|
|
|
*/
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
void
|
1997-11-30 09:44:28 +00:00
|
|
|
isa_configure()
|
|
|
|
|
{
|
1993-06-12 14:58:17 +00:00
|
|
|
struct isa_device *dvp;
|
|
|
|
|
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
printf("Probing for devices on the ISA bus:\n");
|
1994-10-17 21:16:41 +00:00
|
|
|
/* First probe all the sensitive probes */
|
1994-08-18 05:09:36 +00:00
|
|
|
for (dvp = isa_devtab_tty; dvp->id_driver; dvp++)
|
1994-10-17 21:16:41 +00:00
|
|
|
if (dvp->id_driver->sensitive_hw)
|
|
|
|
|
config_isadev(dvp, &tty_imask);
|
1994-08-18 05:09:36 +00:00
|
|
|
for (dvp = isa_devtab_bio; dvp->id_driver; dvp++)
|
1994-10-17 21:16:41 +00:00
|
|
|
if (dvp->id_driver->sensitive_hw)
|
|
|
|
|
config_isadev(dvp, &bio_imask);
|
1994-08-18 05:09:36 +00:00
|
|
|
for (dvp = isa_devtab_net; dvp->id_driver; dvp++)
|
1994-10-17 21:16:41 +00:00
|
|
|
if (dvp->id_driver->sensitive_hw)
|
|
|
|
|
config_isadev(dvp, &net_imask);
|
1997-09-21 21:41:49 +00:00
|
|
|
for (dvp = isa_devtab_cam; dvp->id_driver; dvp++)
|
|
|
|
|
if (dvp->id_driver->sensitive_hw)
|
|
|
|
|
config_isadev(dvp, &cam_imask);
|
1994-08-18 05:09:36 +00:00
|
|
|
for (dvp = isa_devtab_null; dvp->id_driver; dvp++)
|
1994-10-17 21:16:41 +00:00
|
|
|
if (dvp->id_driver->sensitive_hw)
|
|
|
|
|
config_isadev(dvp, (u_int *)NULL);
|
|
|
|
|
|
|
|
|
|
/* Then all the bad ones */
|
|
|
|
|
for (dvp = isa_devtab_tty; dvp->id_driver; dvp++)
|
|
|
|
|
if (!dvp->id_driver->sensitive_hw)
|
|
|
|
|
config_isadev(dvp, &tty_imask);
|
|
|
|
|
for (dvp = isa_devtab_bio; dvp->id_driver; dvp++)
|
|
|
|
|
if (!dvp->id_driver->sensitive_hw)
|
|
|
|
|
config_isadev(dvp, &bio_imask);
|
|
|
|
|
for (dvp = isa_devtab_net; dvp->id_driver; dvp++)
|
|
|
|
|
if (!dvp->id_driver->sensitive_hw)
|
|
|
|
|
config_isadev(dvp, &net_imask);
|
1997-09-21 21:41:49 +00:00
|
|
|
for (dvp = isa_devtab_cam; dvp->id_driver; dvp++)
|
|
|
|
|
if (!dvp->id_driver->sensitive_hw)
|
|
|
|
|
config_isadev(dvp, &cam_imask);
|
1994-10-17 21:16:41 +00:00
|
|
|
for (dvp = isa_devtab_null; dvp->id_driver; dvp++)
|
|
|
|
|
if (!dvp->id_driver->sensitive_hw)
|
|
|
|
|
config_isadev(dvp, (u_int *)NULL);
|
|
|
|
|
|
New interrupt code from Bruce Evans. In additional to Bruce's attached
list of changes, I've made the following additional changes:
1) i386/include/ipl.h renamed to spl.h as the name conflicts with the
file of the same name in i386/isa/ipl.h.
2) changed all use of *mask (i.e. netmask, biomask, ttymask, etc) to
*_imask (net_imask, etc).
3) changed vestige of splnet use in if_is to splimp.
4) got rid of "impmask" completely (Bruce had gotten rid of netmask),
and are now using net_imask instead.
5) dozens of minor cruft to glue in Bruce's changes.
These require changes I made to config(8) as well, and thus it must
be rebuilt.
-DG
from Bruce Evans:
sio:
o No diff is supplied. Remove the define of setsofttty(). I hope
that is enough.
*.s:
o i386/isa/debug.h no longer exists. The event counters became too
much trouble to maintain. All function call entry and exception
entry counters can be recovered by using profiling kernel (the new
profiling supports all entry points; however, it is too slow to
leave enabled all the time; it also). Only BDBTRAP() from debug.h
is now used. That is moved to exception.s. It might be worth
preserving SHOW_BITS() and calling it from _mcount() (if enabled).
o T_ASTFLT is now only set just before calling trap().
o All exception handlers set SWI_AST_MASK in cpl as soon as possible
after entry and arrange for _doreti to restore it atomically with
exiting. It is not possible to set it atomically with entering
the kernel, so it must be checked against the user mode bits in
the trap frame before committing to using it. There is no place
to store the old value of cpl for syscalls or traps, so there are
some complications restoring it.
Profiling stuff (mostly in *.s):
o Changes to kern/subr_mcount.c, gcc and gprof are not supplied yet.
o All interesting labels `foo' are renamed `_foo' and all
uninteresting labels `_bar' are renamed `bar'. A small change
to gprof allows ignoring labels not starting with underscores.
o MCOUNT_LABEL() is to provide names for counters for times spent
in exception handlers.
o FAKE_MCOUNT() is a version of MCOUNT() suitable for exception
handlers. Its arg is the pc where the exception occurred. The
new mcount() pretends that this was a call from that pc to a
suitable MCOUNT_LABEL().
o MEXITCOUNT is to turn off any timer started by MCOUNT().
/usr/src/sys/i386/i386/exception.s:
o The non-BDB BPTTRAP() macros were doing a sti even when interrupts
were disabled when the trap occurred. The sti (fixed) sti is
actually a no-op unless you have my changes to machdep.c that make
the debugger trap gates interrupt gates, but fixing that would
make the ifdefs messier. ddb seems to be unharmed by both
interrupts always disabled and always enabled (I had the branch in
the fix back to front for some time :-().
o There is no known pushal bug.
o tf_err can be left as garbage for syscalls.
/usr/src/sys/i386/i386/locore.s:
o Fix and update BDE_DEBUGGER support.
o ENTRY(btext) before initialization was dangerous.
o Warm boot shot was longer than intended.
/usr/src/sys/i386/i386/machdep.c:
o DON'T APPLY ALL OF THIS DIFF. It's what I'm using, but may require
other changes.
Use the following:
o Remove aston() and setsoftclock().
Maybe use the following:
o No netisr.h.
o Spelling fix.
o Delay to read the Rebooting message.
o Fix for vm system unmapping a reduced area of memory
after bounds_check_with_label() reduces the size of
a physical i/o for a partition boundary. A similar
fix is required in kern_physio.c.
o Correct use of __CONCAT. It never worked here for non-
ANSI cpp's. Is it time to drop support for non-ANSI?
o gdt_segs init. 0xffffffffUL is bogus because ssd_limit
is not 32 bits. The replacement may have the same
value :-), but is more natural.
o physmem was one page too low. Confusing variable names.
Don't use the following:
o Better numbers of buffers. Each 8K page requires up to
16 buffer headers. On my system, this results in 5576
buffers containing [up to] 2854912 bytes of memory.
The usual allocation of about 384 buffers only holds
192K of disk if you use it on an fs with a block size
of 512.
o gdt changes for bdb.
o *TGT -> *IDT changes for bdb.
o #ifdefed changes for bdb.
/usr/src/sys/i386/i386/microtime.s:
o Use the correct asm macros. I think asm.h was copied from Mach
just for microtime and isn't used now. It certainly doesn't
belong in <sys>. Various macros are also duplicated in
sys/i386/boot.h and libc/i386/*.h.
o Don't switch to and from the IRR; it is guaranteed to be selected
(default after ICU init and explicitly selected in isa.c too, and
never changed until the old microtime clobbered it).
/usr/src/sys/i386/i386/support.s:
o Non-essential changes (none related to spls or profiling).
o Removed slow loads of %gs again. The LDT support may require
not relying on %gs, but loading it is not the way to fix it!
Some places (copyin ...) forgot to load it. Loading it clobbers
the user %gs. trap() still loads it after certain types of
faults so that fuword() etc can rely on it without loading it
explicitly. Exception handlers don't restore it. If we want
to preserve the user %gs, then the fastest method is to not
touch it except for context switches. Comparing with
VM_MAXUSER_ADDRESS and branching takes only 2 or 4 cycles on
a 486, while loading %gs takes 9 cycles and using it takes
another.
o Fixed a signed branch to unsigned.
/usr/src/sys/i386/i386/swtch.s:
o Move spl0() outside of idle loop.
o Remove cli/sti from idle loop. sw1 does a cli, and in the
unlikely event of an interrupt occurring and whichqs becoming
zero, sw1 will just jump back to _idle.
o There's no spl0() function in asm any more, so use splz().
o swtch() doesn't need to be superaligned, at least with the
new mcounting.
o Fixed a signed branch to unsigned.
o Removed astoff().
/usr/src/sys/i386/i386/trap.c:
o The decentralized extern decls were inconsistent, of course.
o Fixed typo MATH_EMULTATE in comments. */
o Removed unused variables.
o Old netmask is now impmask; print it instead. Perhaps we
should print some of the new masks.
o BTW, trap() should not print anything for normal debugger
traps.
/usr/src/sys/i386/include/asmacros.h:
o DON'T APPLY ALL OF THIS DIFF. Just use some of the null macros
as necessary.
/usr/src/sys/i386/include/cpu.h:
o CLKF_BASEPRI() changes since cpl == SWI_AST_MASK is now normal
while the kernel is running.
o Don't use var++ to set boolean variables. It fails after a mere
4G times :-) and is slower than storing a constant on [3-4]86s.
/usr/src/sys/i386/include/cpufunc.h:
o DON'T APPLY ALL OF THIS DIFF. You need mainly the include of
<machine/ipl.h>. Unfortunately, <machine/ipl.h> is needed by
almost everything for the inlines.
/usr/src/sys/i386/include/ipl.h:
o New file. Defines spl inlines and SWI macros and declares most
variables related to hard and soft interrupt masks.
/usr/src/sys/i386/isa/icu.h:
o Moved definitions to <machine/ipl.h>
/usr/src/sys/i386/isa/icu.s:
o Software interrupts (SWIs) and delayed hardware interrupts (HWIs)
are now handled uniformally, and dispatching them from splx() is
more like dispatching them from _doreti. The dispatcher is
essentially *(handler[ffs(ipending & ~cpl)]().
o More care (not quite enough) is taken to avoid unbounded nesting
of interrupts.
o The interface to softclock() is changed so that a trap frame is
not required.
o Fast interrupt handlers are now handled more uniformally.
Configuration is still too early (new handlers would require
bits in <machine/ipl.h> and functions to vector.s).
o splnnn() and splx() are no longer here; they are inline functions
(could be macros for other compilers). splz() is the nontrivial
part of the old splx().
/usr/src/sys/i386/isa/ipl.h
o New file. Supposed to have only bus-dependent stuff. Perhaps
the h/w masks should be declared here.
/usr/src/sys/i386/isa/isa.c:
o DON'T APPLY ALL OF THIS DIFF. You need only things involving
*mask and *MASK and comments about them. netmask is now a pure
software mask. It works like the softclock mask.
/usr/src/sys/i386/isa/vector.s:
o Reorganize AUTO_EOI* macros.
o Option FAST_INTR_HANDLER_USERS_ES for people who don't trust
fastintr handlers.
o fastintr handlers need to metamorphose into ordinary interrupt
handlers if their SWI bit has become set. Previously, sio had
unintended latency for handling output completions and input
of SLIP framing characters because this was not done.
/usr/src/sys/net/netisr.h:
o The machine-dependent stuff is now imported from <machine/ipl.h>.
/usr/src/sys/sys/systm.h
o DON'T APPLY ALL OF THIS DIFF. You need mainly the different
splx() prototype. The spl*() prototypes are duplicated as
inlines in <machine/ipl.h> but they need to be duplicated here
in case there are no inlines. I sent systm.h and cpufunc.h
to Garrett. We agree that spl0 should be replaced by splnone
and not the other way around like I've done.
/usr/src/sys/kern/kern_clock.c
o splsoftclock() now lowers cpl so the direct call to softclock()
works as intended.
o softclock() interface changed to avoid passing the whole frame
(some machines may need another change for profile_tick()).
o profiling renamed _profiling to avoid ANSI namespace pollution.
(I had to improve the mcount() interface and may as well fix it.)
The GUPROF variant doesn't actually reference profiling here,
but the 'U' in GUPROF should mean to select the microtimer
mcount() and not change the interface.
1994-04-02 07:00:53 +00:00
|
|
|
bio_imask |= SWI_CLOCK_MASK;
|
|
|
|
|
net_imask |= SWI_NET_MASK;
|
|
|
|
|
tty_imask |= SWI_TTY_MASK;
|
|
|
|
|
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
/*
|
New interrupt code from Bruce Evans. In additional to Bruce's attached
list of changes, I've made the following additional changes:
1) i386/include/ipl.h renamed to spl.h as the name conflicts with the
file of the same name in i386/isa/ipl.h.
2) changed all use of *mask (i.e. netmask, biomask, ttymask, etc) to
*_imask (net_imask, etc).
3) changed vestige of splnet use in if_is to splimp.
4) got rid of "impmask" completely (Bruce had gotten rid of netmask),
and are now using net_imask instead.
5) dozens of minor cruft to glue in Bruce's changes.
These require changes I made to config(8) as well, and thus it must
be rebuilt.
-DG
from Bruce Evans:
sio:
o No diff is supplied. Remove the define of setsofttty(). I hope
that is enough.
*.s:
o i386/isa/debug.h no longer exists. The event counters became too
much trouble to maintain. All function call entry and exception
entry counters can be recovered by using profiling kernel (the new
profiling supports all entry points; however, it is too slow to
leave enabled all the time; it also). Only BDBTRAP() from debug.h
is now used. That is moved to exception.s. It might be worth
preserving SHOW_BITS() and calling it from _mcount() (if enabled).
o T_ASTFLT is now only set just before calling trap().
o All exception handlers set SWI_AST_MASK in cpl as soon as possible
after entry and arrange for _doreti to restore it atomically with
exiting. It is not possible to set it atomically with entering
the kernel, so it must be checked against the user mode bits in
the trap frame before committing to using it. There is no place
to store the old value of cpl for syscalls or traps, so there are
some complications restoring it.
Profiling stuff (mostly in *.s):
o Changes to kern/subr_mcount.c, gcc and gprof are not supplied yet.
o All interesting labels `foo' are renamed `_foo' and all
uninteresting labels `_bar' are renamed `bar'. A small change
to gprof allows ignoring labels not starting with underscores.
o MCOUNT_LABEL() is to provide names for counters for times spent
in exception handlers.
o FAKE_MCOUNT() is a version of MCOUNT() suitable for exception
handlers. Its arg is the pc where the exception occurred. The
new mcount() pretends that this was a call from that pc to a
suitable MCOUNT_LABEL().
o MEXITCOUNT is to turn off any timer started by MCOUNT().
/usr/src/sys/i386/i386/exception.s:
o The non-BDB BPTTRAP() macros were doing a sti even when interrupts
were disabled when the trap occurred. The sti (fixed) sti is
actually a no-op unless you have my changes to machdep.c that make
the debugger trap gates interrupt gates, but fixing that would
make the ifdefs messier. ddb seems to be unharmed by both
interrupts always disabled and always enabled (I had the branch in
the fix back to front for some time :-().
o There is no known pushal bug.
o tf_err can be left as garbage for syscalls.
/usr/src/sys/i386/i386/locore.s:
o Fix and update BDE_DEBUGGER support.
o ENTRY(btext) before initialization was dangerous.
o Warm boot shot was longer than intended.
/usr/src/sys/i386/i386/machdep.c:
o DON'T APPLY ALL OF THIS DIFF. It's what I'm using, but may require
other changes.
Use the following:
o Remove aston() and setsoftclock().
Maybe use the following:
o No netisr.h.
o Spelling fix.
o Delay to read the Rebooting message.
o Fix for vm system unmapping a reduced area of memory
after bounds_check_with_label() reduces the size of
a physical i/o for a partition boundary. A similar
fix is required in kern_physio.c.
o Correct use of __CONCAT. It never worked here for non-
ANSI cpp's. Is it time to drop support for non-ANSI?
o gdt_segs init. 0xffffffffUL is bogus because ssd_limit
is not 32 bits. The replacement may have the same
value :-), but is more natural.
o physmem was one page too low. Confusing variable names.
Don't use the following:
o Better numbers of buffers. Each 8K page requires up to
16 buffer headers. On my system, this results in 5576
buffers containing [up to] 2854912 bytes of memory.
The usual allocation of about 384 buffers only holds
192K of disk if you use it on an fs with a block size
of 512.
o gdt changes for bdb.
o *TGT -> *IDT changes for bdb.
o #ifdefed changes for bdb.
/usr/src/sys/i386/i386/microtime.s:
o Use the correct asm macros. I think asm.h was copied from Mach
just for microtime and isn't used now. It certainly doesn't
belong in <sys>. Various macros are also duplicated in
sys/i386/boot.h and libc/i386/*.h.
o Don't switch to and from the IRR; it is guaranteed to be selected
(default after ICU init and explicitly selected in isa.c too, and
never changed until the old microtime clobbered it).
/usr/src/sys/i386/i386/support.s:
o Non-essential changes (none related to spls or profiling).
o Removed slow loads of %gs again. The LDT support may require
not relying on %gs, but loading it is not the way to fix it!
Some places (copyin ...) forgot to load it. Loading it clobbers
the user %gs. trap() still loads it after certain types of
faults so that fuword() etc can rely on it without loading it
explicitly. Exception handlers don't restore it. If we want
to preserve the user %gs, then the fastest method is to not
touch it except for context switches. Comparing with
VM_MAXUSER_ADDRESS and branching takes only 2 or 4 cycles on
a 486, while loading %gs takes 9 cycles and using it takes
another.
o Fixed a signed branch to unsigned.
/usr/src/sys/i386/i386/swtch.s:
o Move spl0() outside of idle loop.
o Remove cli/sti from idle loop. sw1 does a cli, and in the
unlikely event of an interrupt occurring and whichqs becoming
zero, sw1 will just jump back to _idle.
o There's no spl0() function in asm any more, so use splz().
o swtch() doesn't need to be superaligned, at least with the
new mcounting.
o Fixed a signed branch to unsigned.
o Removed astoff().
/usr/src/sys/i386/i386/trap.c:
o The decentralized extern decls were inconsistent, of course.
o Fixed typo MATH_EMULTATE in comments. */
o Removed unused variables.
o Old netmask is now impmask; print it instead. Perhaps we
should print some of the new masks.
o BTW, trap() should not print anything for normal debugger
traps.
/usr/src/sys/i386/include/asmacros.h:
o DON'T APPLY ALL OF THIS DIFF. Just use some of the null macros
as necessary.
/usr/src/sys/i386/include/cpu.h:
o CLKF_BASEPRI() changes since cpl == SWI_AST_MASK is now normal
while the kernel is running.
o Don't use var++ to set boolean variables. It fails after a mere
4G times :-) and is slower than storing a constant on [3-4]86s.
/usr/src/sys/i386/include/cpufunc.h:
o DON'T APPLY ALL OF THIS DIFF. You need mainly the include of
<machine/ipl.h>. Unfortunately, <machine/ipl.h> is needed by
almost everything for the inlines.
/usr/src/sys/i386/include/ipl.h:
o New file. Defines spl inlines and SWI macros and declares most
variables related to hard and soft interrupt masks.
/usr/src/sys/i386/isa/icu.h:
o Moved definitions to <machine/ipl.h>
/usr/src/sys/i386/isa/icu.s:
o Software interrupts (SWIs) and delayed hardware interrupts (HWIs)
are now handled uniformally, and dispatching them from splx() is
more like dispatching them from _doreti. The dispatcher is
essentially *(handler[ffs(ipending & ~cpl)]().
o More care (not quite enough) is taken to avoid unbounded nesting
of interrupts.
o The interface to softclock() is changed so that a trap frame is
not required.
o Fast interrupt handlers are now handled more uniformally.
Configuration is still too early (new handlers would require
bits in <machine/ipl.h> and functions to vector.s).
o splnnn() and splx() are no longer here; they are inline functions
(could be macros for other compilers). splz() is the nontrivial
part of the old splx().
/usr/src/sys/i386/isa/ipl.h
o New file. Supposed to have only bus-dependent stuff. Perhaps
the h/w masks should be declared here.
/usr/src/sys/i386/isa/isa.c:
o DON'T APPLY ALL OF THIS DIFF. You need only things involving
*mask and *MASK and comments about them. netmask is now a pure
software mask. It works like the softclock mask.
/usr/src/sys/i386/isa/vector.s:
o Reorganize AUTO_EOI* macros.
o Option FAST_INTR_HANDLER_USERS_ES for people who don't trust
fastintr handlers.
o fastintr handlers need to metamorphose into ordinary interrupt
handlers if their SWI bit has become set. Previously, sio had
unintended latency for handling output completions and input
of SLIP framing characters because this was not done.
/usr/src/sys/net/netisr.h:
o The machine-dependent stuff is now imported from <machine/ipl.h>.
/usr/src/sys/sys/systm.h
o DON'T APPLY ALL OF THIS DIFF. You need mainly the different
splx() prototype. The spl*() prototypes are duplicated as
inlines in <machine/ipl.h> but they need to be duplicated here
in case there are no inlines. I sent systm.h and cpufunc.h
to Garrett. We agree that spl0 should be replaced by splnone
and not the other way around like I've done.
/usr/src/sys/kern/kern_clock.c
o splsoftclock() now lowers cpl so the direct call to softclock()
works as intended.
o softclock() interface changed to avoid passing the whole frame
(some machines may need another change for profile_tick()).
o profiling renamed _profiling to avoid ANSI namespace pollution.
(I had to improve the mcount() interface and may as well fix it.)
The GUPROF variant doesn't actually reference profiling here,
but the 'U' in GUPROF should mean to select the microtimer
mcount() and not change the interface.
1994-04-02 07:00:53 +00:00
|
|
|
* XXX we should really add the tty device to net_imask when the line is
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
* switched to SLIPDISC, and then remove it when it is switched away from
|
New interrupt code from Bruce Evans. In additional to Bruce's attached
list of changes, I've made the following additional changes:
1) i386/include/ipl.h renamed to spl.h as the name conflicts with the
file of the same name in i386/isa/ipl.h.
2) changed all use of *mask (i.e. netmask, biomask, ttymask, etc) to
*_imask (net_imask, etc).
3) changed vestige of splnet use in if_is to splimp.
4) got rid of "impmask" completely (Bruce had gotten rid of netmask),
and are now using net_imask instead.
5) dozens of minor cruft to glue in Bruce's changes.
These require changes I made to config(8) as well, and thus it must
be rebuilt.
-DG
from Bruce Evans:
sio:
o No diff is supplied. Remove the define of setsofttty(). I hope
that is enough.
*.s:
o i386/isa/debug.h no longer exists. The event counters became too
much trouble to maintain. All function call entry and exception
entry counters can be recovered by using profiling kernel (the new
profiling supports all entry points; however, it is too slow to
leave enabled all the time; it also). Only BDBTRAP() from debug.h
is now used. That is moved to exception.s. It might be worth
preserving SHOW_BITS() and calling it from _mcount() (if enabled).
o T_ASTFLT is now only set just before calling trap().
o All exception handlers set SWI_AST_MASK in cpl as soon as possible
after entry and arrange for _doreti to restore it atomically with
exiting. It is not possible to set it atomically with entering
the kernel, so it must be checked against the user mode bits in
the trap frame before committing to using it. There is no place
to store the old value of cpl for syscalls or traps, so there are
some complications restoring it.
Profiling stuff (mostly in *.s):
o Changes to kern/subr_mcount.c, gcc and gprof are not supplied yet.
o All interesting labels `foo' are renamed `_foo' and all
uninteresting labels `_bar' are renamed `bar'. A small change
to gprof allows ignoring labels not starting with underscores.
o MCOUNT_LABEL() is to provide names for counters for times spent
in exception handlers.
o FAKE_MCOUNT() is a version of MCOUNT() suitable for exception
handlers. Its arg is the pc where the exception occurred. The
new mcount() pretends that this was a call from that pc to a
suitable MCOUNT_LABEL().
o MEXITCOUNT is to turn off any timer started by MCOUNT().
/usr/src/sys/i386/i386/exception.s:
o The non-BDB BPTTRAP() macros were doing a sti even when interrupts
were disabled when the trap occurred. The sti (fixed) sti is
actually a no-op unless you have my changes to machdep.c that make
the debugger trap gates interrupt gates, but fixing that would
make the ifdefs messier. ddb seems to be unharmed by both
interrupts always disabled and always enabled (I had the branch in
the fix back to front for some time :-().
o There is no known pushal bug.
o tf_err can be left as garbage for syscalls.
/usr/src/sys/i386/i386/locore.s:
o Fix and update BDE_DEBUGGER support.
o ENTRY(btext) before initialization was dangerous.
o Warm boot shot was longer than intended.
/usr/src/sys/i386/i386/machdep.c:
o DON'T APPLY ALL OF THIS DIFF. It's what I'm using, but may require
other changes.
Use the following:
o Remove aston() and setsoftclock().
Maybe use the following:
o No netisr.h.
o Spelling fix.
o Delay to read the Rebooting message.
o Fix for vm system unmapping a reduced area of memory
after bounds_check_with_label() reduces the size of
a physical i/o for a partition boundary. A similar
fix is required in kern_physio.c.
o Correct use of __CONCAT. It never worked here for non-
ANSI cpp's. Is it time to drop support for non-ANSI?
o gdt_segs init. 0xffffffffUL is bogus because ssd_limit
is not 32 bits. The replacement may have the same
value :-), but is more natural.
o physmem was one page too low. Confusing variable names.
Don't use the following:
o Better numbers of buffers. Each 8K page requires up to
16 buffer headers. On my system, this results in 5576
buffers containing [up to] 2854912 bytes of memory.
The usual allocation of about 384 buffers only holds
192K of disk if you use it on an fs with a block size
of 512.
o gdt changes for bdb.
o *TGT -> *IDT changes for bdb.
o #ifdefed changes for bdb.
/usr/src/sys/i386/i386/microtime.s:
o Use the correct asm macros. I think asm.h was copied from Mach
just for microtime and isn't used now. It certainly doesn't
belong in <sys>. Various macros are also duplicated in
sys/i386/boot.h and libc/i386/*.h.
o Don't switch to and from the IRR; it is guaranteed to be selected
(default after ICU init and explicitly selected in isa.c too, and
never changed until the old microtime clobbered it).
/usr/src/sys/i386/i386/support.s:
o Non-essential changes (none related to spls or profiling).
o Removed slow loads of %gs again. The LDT support may require
not relying on %gs, but loading it is not the way to fix it!
Some places (copyin ...) forgot to load it. Loading it clobbers
the user %gs. trap() still loads it after certain types of
faults so that fuword() etc can rely on it without loading it
explicitly. Exception handlers don't restore it. If we want
to preserve the user %gs, then the fastest method is to not
touch it except for context switches. Comparing with
VM_MAXUSER_ADDRESS and branching takes only 2 or 4 cycles on
a 486, while loading %gs takes 9 cycles and using it takes
another.
o Fixed a signed branch to unsigned.
/usr/src/sys/i386/i386/swtch.s:
o Move spl0() outside of idle loop.
o Remove cli/sti from idle loop. sw1 does a cli, and in the
unlikely event of an interrupt occurring and whichqs becoming
zero, sw1 will just jump back to _idle.
o There's no spl0() function in asm any more, so use splz().
o swtch() doesn't need to be superaligned, at least with the
new mcounting.
o Fixed a signed branch to unsigned.
o Removed astoff().
/usr/src/sys/i386/i386/trap.c:
o The decentralized extern decls were inconsistent, of course.
o Fixed typo MATH_EMULTATE in comments. */
o Removed unused variables.
o Old netmask is now impmask; print it instead. Perhaps we
should print some of the new masks.
o BTW, trap() should not print anything for normal debugger
traps.
/usr/src/sys/i386/include/asmacros.h:
o DON'T APPLY ALL OF THIS DIFF. Just use some of the null macros
as necessary.
/usr/src/sys/i386/include/cpu.h:
o CLKF_BASEPRI() changes since cpl == SWI_AST_MASK is now normal
while the kernel is running.
o Don't use var++ to set boolean variables. It fails after a mere
4G times :-) and is slower than storing a constant on [3-4]86s.
/usr/src/sys/i386/include/cpufunc.h:
o DON'T APPLY ALL OF THIS DIFF. You need mainly the include of
<machine/ipl.h>. Unfortunately, <machine/ipl.h> is needed by
almost everything for the inlines.
/usr/src/sys/i386/include/ipl.h:
o New file. Defines spl inlines and SWI macros and declares most
variables related to hard and soft interrupt masks.
/usr/src/sys/i386/isa/icu.h:
o Moved definitions to <machine/ipl.h>
/usr/src/sys/i386/isa/icu.s:
o Software interrupts (SWIs) and delayed hardware interrupts (HWIs)
are now handled uniformally, and dispatching them from splx() is
more like dispatching them from _doreti. The dispatcher is
essentially *(handler[ffs(ipending & ~cpl)]().
o More care (not quite enough) is taken to avoid unbounded nesting
of interrupts.
o The interface to softclock() is changed so that a trap frame is
not required.
o Fast interrupt handlers are now handled more uniformally.
Configuration is still too early (new handlers would require
bits in <machine/ipl.h> and functions to vector.s).
o splnnn() and splx() are no longer here; they are inline functions
(could be macros for other compilers). splz() is the nontrivial
part of the old splx().
/usr/src/sys/i386/isa/ipl.h
o New file. Supposed to have only bus-dependent stuff. Perhaps
the h/w masks should be declared here.
/usr/src/sys/i386/isa/isa.c:
o DON'T APPLY ALL OF THIS DIFF. You need only things involving
*mask and *MASK and comments about them. netmask is now a pure
software mask. It works like the softclock mask.
/usr/src/sys/i386/isa/vector.s:
o Reorganize AUTO_EOI* macros.
o Option FAST_INTR_HANDLER_USERS_ES for people who don't trust
fastintr handlers.
o fastintr handlers need to metamorphose into ordinary interrupt
handlers if their SWI bit has become set. Previously, sio had
unintended latency for handling output completions and input
of SLIP framing characters because this was not done.
/usr/src/sys/net/netisr.h:
o The machine-dependent stuff is now imported from <machine/ipl.h>.
/usr/src/sys/sys/systm.h
o DON'T APPLY ALL OF THIS DIFF. You need mainly the different
splx() prototype. The spl*() prototypes are duplicated as
inlines in <machine/ipl.h> but they need to be duplicated here
in case there are no inlines. I sent systm.h and cpufunc.h
to Garrett. We agree that spl0 should be replaced by splnone
and not the other way around like I've done.
/usr/src/sys/kern/kern_clock.c
o splsoftclock() now lowers cpl so the direct call to softclock()
works as intended.
o softclock() interface changed to avoid passing the whole frame
(some machines may need another change for profile_tick()).
o profiling renamed _profiling to avoid ANSI namespace pollution.
(I had to improve the mcount() interface and may as well fix it.)
The GUPROF variant doesn't actually reference profiling here,
but the 'U' in GUPROF should mean to select the microtimer
mcount() and not change the interface.
1994-04-02 07:00:53 +00:00
|
|
|
* SLIPDISC. No need to block out ALL ttys during a splimp when only one
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
* of them is running slip.
|
New interrupt code from Bruce Evans. In additional to Bruce's attached
list of changes, I've made the following additional changes:
1) i386/include/ipl.h renamed to spl.h as the name conflicts with the
file of the same name in i386/isa/ipl.h.
2) changed all use of *mask (i.e. netmask, biomask, ttymask, etc) to
*_imask (net_imask, etc).
3) changed vestige of splnet use in if_is to splimp.
4) got rid of "impmask" completely (Bruce had gotten rid of netmask),
and are now using net_imask instead.
5) dozens of minor cruft to glue in Bruce's changes.
These require changes I made to config(8) as well, and thus it must
be rebuilt.
-DG
from Bruce Evans:
sio:
o No diff is supplied. Remove the define of setsofttty(). I hope
that is enough.
*.s:
o i386/isa/debug.h no longer exists. The event counters became too
much trouble to maintain. All function call entry and exception
entry counters can be recovered by using profiling kernel (the new
profiling supports all entry points; however, it is too slow to
leave enabled all the time; it also). Only BDBTRAP() from debug.h
is now used. That is moved to exception.s. It might be worth
preserving SHOW_BITS() and calling it from _mcount() (if enabled).
o T_ASTFLT is now only set just before calling trap().
o All exception handlers set SWI_AST_MASK in cpl as soon as possible
after entry and arrange for _doreti to restore it atomically with
exiting. It is not possible to set it atomically with entering
the kernel, so it must be checked against the user mode bits in
the trap frame before committing to using it. There is no place
to store the old value of cpl for syscalls or traps, so there are
some complications restoring it.
Profiling stuff (mostly in *.s):
o Changes to kern/subr_mcount.c, gcc and gprof are not supplied yet.
o All interesting labels `foo' are renamed `_foo' and all
uninteresting labels `_bar' are renamed `bar'. A small change
to gprof allows ignoring labels not starting with underscores.
o MCOUNT_LABEL() is to provide names for counters for times spent
in exception handlers.
o FAKE_MCOUNT() is a version of MCOUNT() suitable for exception
handlers. Its arg is the pc where the exception occurred. The
new mcount() pretends that this was a call from that pc to a
suitable MCOUNT_LABEL().
o MEXITCOUNT is to turn off any timer started by MCOUNT().
/usr/src/sys/i386/i386/exception.s:
o The non-BDB BPTTRAP() macros were doing a sti even when interrupts
were disabled when the trap occurred. The sti (fixed) sti is
actually a no-op unless you have my changes to machdep.c that make
the debugger trap gates interrupt gates, but fixing that would
make the ifdefs messier. ddb seems to be unharmed by both
interrupts always disabled and always enabled (I had the branch in
the fix back to front for some time :-().
o There is no known pushal bug.
o tf_err can be left as garbage for syscalls.
/usr/src/sys/i386/i386/locore.s:
o Fix and update BDE_DEBUGGER support.
o ENTRY(btext) before initialization was dangerous.
o Warm boot shot was longer than intended.
/usr/src/sys/i386/i386/machdep.c:
o DON'T APPLY ALL OF THIS DIFF. It's what I'm using, but may require
other changes.
Use the following:
o Remove aston() and setsoftclock().
Maybe use the following:
o No netisr.h.
o Spelling fix.
o Delay to read the Rebooting message.
o Fix for vm system unmapping a reduced area of memory
after bounds_check_with_label() reduces the size of
a physical i/o for a partition boundary. A similar
fix is required in kern_physio.c.
o Correct use of __CONCAT. It never worked here for non-
ANSI cpp's. Is it time to drop support for non-ANSI?
o gdt_segs init. 0xffffffffUL is bogus because ssd_limit
is not 32 bits. The replacement may have the same
value :-), but is more natural.
o physmem was one page too low. Confusing variable names.
Don't use the following:
o Better numbers of buffers. Each 8K page requires up to
16 buffer headers. On my system, this results in 5576
buffers containing [up to] 2854912 bytes of memory.
The usual allocation of about 384 buffers only holds
192K of disk if you use it on an fs with a block size
of 512.
o gdt changes for bdb.
o *TGT -> *IDT changes for bdb.
o #ifdefed changes for bdb.
/usr/src/sys/i386/i386/microtime.s:
o Use the correct asm macros. I think asm.h was copied from Mach
just for microtime and isn't used now. It certainly doesn't
belong in <sys>. Various macros are also duplicated in
sys/i386/boot.h and libc/i386/*.h.
o Don't switch to and from the IRR; it is guaranteed to be selected
(default after ICU init and explicitly selected in isa.c too, and
never changed until the old microtime clobbered it).
/usr/src/sys/i386/i386/support.s:
o Non-essential changes (none related to spls or profiling).
o Removed slow loads of %gs again. The LDT support may require
not relying on %gs, but loading it is not the way to fix it!
Some places (copyin ...) forgot to load it. Loading it clobbers
the user %gs. trap() still loads it after certain types of
faults so that fuword() etc can rely on it without loading it
explicitly. Exception handlers don't restore it. If we want
to preserve the user %gs, then the fastest method is to not
touch it except for context switches. Comparing with
VM_MAXUSER_ADDRESS and branching takes only 2 or 4 cycles on
a 486, while loading %gs takes 9 cycles and using it takes
another.
o Fixed a signed branch to unsigned.
/usr/src/sys/i386/i386/swtch.s:
o Move spl0() outside of idle loop.
o Remove cli/sti from idle loop. sw1 does a cli, and in the
unlikely event of an interrupt occurring and whichqs becoming
zero, sw1 will just jump back to _idle.
o There's no spl0() function in asm any more, so use splz().
o swtch() doesn't need to be superaligned, at least with the
new mcounting.
o Fixed a signed branch to unsigned.
o Removed astoff().
/usr/src/sys/i386/i386/trap.c:
o The decentralized extern decls were inconsistent, of course.
o Fixed typo MATH_EMULTATE in comments. */
o Removed unused variables.
o Old netmask is now impmask; print it instead. Perhaps we
should print some of the new masks.
o BTW, trap() should not print anything for normal debugger
traps.
/usr/src/sys/i386/include/asmacros.h:
o DON'T APPLY ALL OF THIS DIFF. Just use some of the null macros
as necessary.
/usr/src/sys/i386/include/cpu.h:
o CLKF_BASEPRI() changes since cpl == SWI_AST_MASK is now normal
while the kernel is running.
o Don't use var++ to set boolean variables. It fails after a mere
4G times :-) and is slower than storing a constant on [3-4]86s.
/usr/src/sys/i386/include/cpufunc.h:
o DON'T APPLY ALL OF THIS DIFF. You need mainly the include of
<machine/ipl.h>. Unfortunately, <machine/ipl.h> is needed by
almost everything for the inlines.
/usr/src/sys/i386/include/ipl.h:
o New file. Defines spl inlines and SWI macros and declares most
variables related to hard and soft interrupt masks.
/usr/src/sys/i386/isa/icu.h:
o Moved definitions to <machine/ipl.h>
/usr/src/sys/i386/isa/icu.s:
o Software interrupts (SWIs) and delayed hardware interrupts (HWIs)
are now handled uniformally, and dispatching them from splx() is
more like dispatching them from _doreti. The dispatcher is
essentially *(handler[ffs(ipending & ~cpl)]().
o More care (not quite enough) is taken to avoid unbounded nesting
of interrupts.
o The interface to softclock() is changed so that a trap frame is
not required.
o Fast interrupt handlers are now handled more uniformally.
Configuration is still too early (new handlers would require
bits in <machine/ipl.h> and functions to vector.s).
o splnnn() and splx() are no longer here; they are inline functions
(could be macros for other compilers). splz() is the nontrivial
part of the old splx().
/usr/src/sys/i386/isa/ipl.h
o New file. Supposed to have only bus-dependent stuff. Perhaps
the h/w masks should be declared here.
/usr/src/sys/i386/isa/isa.c:
o DON'T APPLY ALL OF THIS DIFF. You need only things involving
*mask and *MASK and comments about them. netmask is now a pure
software mask. It works like the softclock mask.
/usr/src/sys/i386/isa/vector.s:
o Reorganize AUTO_EOI* macros.
o Option FAST_INTR_HANDLER_USERS_ES for people who don't trust
fastintr handlers.
o fastintr handlers need to metamorphose into ordinary interrupt
handlers if their SWI bit has become set. Previously, sio had
unintended latency for handling output completions and input
of SLIP framing characters because this was not done.
/usr/src/sys/net/netisr.h:
o The machine-dependent stuff is now imported from <machine/ipl.h>.
/usr/src/sys/sys/systm.h
o DON'T APPLY ALL OF THIS DIFF. You need mainly the different
splx() prototype. The spl*() prototypes are duplicated as
inlines in <machine/ipl.h> but they need to be duplicated here
in case there are no inlines. I sent systm.h and cpufunc.h
to Garrett. We agree that spl0 should be replaced by splnone
and not the other way around like I've done.
/usr/src/sys/kern/kern_clock.c
o splsoftclock() now lowers cpl so the direct call to softclock()
works as intended.
o softclock() interface changed to avoid passing the whole frame
(some machines may need another change for profile_tick()).
o profiling renamed _profiling to avoid ANSI namespace pollution.
(I had to improve the mcount() interface and may as well fix it.)
The GUPROF variant doesn't actually reference profiling here,
but the 'U' in GUPROF should mean to select the microtimer
mcount() and not change the interface.
1994-04-02 07:00:53 +00:00
|
|
|
*
|
1995-05-30 08:16:23 +00:00
|
|
|
* XXX actually, blocking all ttys during a splimp doesn't matter so much
|
New interrupt code from Bruce Evans. In additional to Bruce's attached
list of changes, I've made the following additional changes:
1) i386/include/ipl.h renamed to spl.h as the name conflicts with the
file of the same name in i386/isa/ipl.h.
2) changed all use of *mask (i.e. netmask, biomask, ttymask, etc) to
*_imask (net_imask, etc).
3) changed vestige of splnet use in if_is to splimp.
4) got rid of "impmask" completely (Bruce had gotten rid of netmask),
and are now using net_imask instead.
5) dozens of minor cruft to glue in Bruce's changes.
These require changes I made to config(8) as well, and thus it must
be rebuilt.
-DG
from Bruce Evans:
sio:
o No diff is supplied. Remove the define of setsofttty(). I hope
that is enough.
*.s:
o i386/isa/debug.h no longer exists. The event counters became too
much trouble to maintain. All function call entry and exception
entry counters can be recovered by using profiling kernel (the new
profiling supports all entry points; however, it is too slow to
leave enabled all the time; it also). Only BDBTRAP() from debug.h
is now used. That is moved to exception.s. It might be worth
preserving SHOW_BITS() and calling it from _mcount() (if enabled).
o T_ASTFLT is now only set just before calling trap().
o All exception handlers set SWI_AST_MASK in cpl as soon as possible
after entry and arrange for _doreti to restore it atomically with
exiting. It is not possible to set it atomically with entering
the kernel, so it must be checked against the user mode bits in
the trap frame before committing to using it. There is no place
to store the old value of cpl for syscalls or traps, so there are
some complications restoring it.
Profiling stuff (mostly in *.s):
o Changes to kern/subr_mcount.c, gcc and gprof are not supplied yet.
o All interesting labels `foo' are renamed `_foo' and all
uninteresting labels `_bar' are renamed `bar'. A small change
to gprof allows ignoring labels not starting with underscores.
o MCOUNT_LABEL() is to provide names for counters for times spent
in exception handlers.
o FAKE_MCOUNT() is a version of MCOUNT() suitable for exception
handlers. Its arg is the pc where the exception occurred. The
new mcount() pretends that this was a call from that pc to a
suitable MCOUNT_LABEL().
o MEXITCOUNT is to turn off any timer started by MCOUNT().
/usr/src/sys/i386/i386/exception.s:
o The non-BDB BPTTRAP() macros were doing a sti even when interrupts
were disabled when the trap occurred. The sti (fixed) sti is
actually a no-op unless you have my changes to machdep.c that make
the debugger trap gates interrupt gates, but fixing that would
make the ifdefs messier. ddb seems to be unharmed by both
interrupts always disabled and always enabled (I had the branch in
the fix back to front for some time :-().
o There is no known pushal bug.
o tf_err can be left as garbage for syscalls.
/usr/src/sys/i386/i386/locore.s:
o Fix and update BDE_DEBUGGER support.
o ENTRY(btext) before initialization was dangerous.
o Warm boot shot was longer than intended.
/usr/src/sys/i386/i386/machdep.c:
o DON'T APPLY ALL OF THIS DIFF. It's what I'm using, but may require
other changes.
Use the following:
o Remove aston() and setsoftclock().
Maybe use the following:
o No netisr.h.
o Spelling fix.
o Delay to read the Rebooting message.
o Fix for vm system unmapping a reduced area of memory
after bounds_check_with_label() reduces the size of
a physical i/o for a partition boundary. A similar
fix is required in kern_physio.c.
o Correct use of __CONCAT. It never worked here for non-
ANSI cpp's. Is it time to drop support for non-ANSI?
o gdt_segs init. 0xffffffffUL is bogus because ssd_limit
is not 32 bits. The replacement may have the same
value :-), but is more natural.
o physmem was one page too low. Confusing variable names.
Don't use the following:
o Better numbers of buffers. Each 8K page requires up to
16 buffer headers. On my system, this results in 5576
buffers containing [up to] 2854912 bytes of memory.
The usual allocation of about 384 buffers only holds
192K of disk if you use it on an fs with a block size
of 512.
o gdt changes for bdb.
o *TGT -> *IDT changes for bdb.
o #ifdefed changes for bdb.
/usr/src/sys/i386/i386/microtime.s:
o Use the correct asm macros. I think asm.h was copied from Mach
just for microtime and isn't used now. It certainly doesn't
belong in <sys>. Various macros are also duplicated in
sys/i386/boot.h and libc/i386/*.h.
o Don't switch to and from the IRR; it is guaranteed to be selected
(default after ICU init and explicitly selected in isa.c too, and
never changed until the old microtime clobbered it).
/usr/src/sys/i386/i386/support.s:
o Non-essential changes (none related to spls or profiling).
o Removed slow loads of %gs again. The LDT support may require
not relying on %gs, but loading it is not the way to fix it!
Some places (copyin ...) forgot to load it. Loading it clobbers
the user %gs. trap() still loads it after certain types of
faults so that fuword() etc can rely on it without loading it
explicitly. Exception handlers don't restore it. If we want
to preserve the user %gs, then the fastest method is to not
touch it except for context switches. Comparing with
VM_MAXUSER_ADDRESS and branching takes only 2 or 4 cycles on
a 486, while loading %gs takes 9 cycles and using it takes
another.
o Fixed a signed branch to unsigned.
/usr/src/sys/i386/i386/swtch.s:
o Move spl0() outside of idle loop.
o Remove cli/sti from idle loop. sw1 does a cli, and in the
unlikely event of an interrupt occurring and whichqs becoming
zero, sw1 will just jump back to _idle.
o There's no spl0() function in asm any more, so use splz().
o swtch() doesn't need to be superaligned, at least with the
new mcounting.
o Fixed a signed branch to unsigned.
o Removed astoff().
/usr/src/sys/i386/i386/trap.c:
o The decentralized extern decls were inconsistent, of course.
o Fixed typo MATH_EMULTATE in comments. */
o Removed unused variables.
o Old netmask is now impmask; print it instead. Perhaps we
should print some of the new masks.
o BTW, trap() should not print anything for normal debugger
traps.
/usr/src/sys/i386/include/asmacros.h:
o DON'T APPLY ALL OF THIS DIFF. Just use some of the null macros
as necessary.
/usr/src/sys/i386/include/cpu.h:
o CLKF_BASEPRI() changes since cpl == SWI_AST_MASK is now normal
while the kernel is running.
o Don't use var++ to set boolean variables. It fails after a mere
4G times :-) and is slower than storing a constant on [3-4]86s.
/usr/src/sys/i386/include/cpufunc.h:
o DON'T APPLY ALL OF THIS DIFF. You need mainly the include of
<machine/ipl.h>. Unfortunately, <machine/ipl.h> is needed by
almost everything for the inlines.
/usr/src/sys/i386/include/ipl.h:
o New file. Defines spl inlines and SWI macros and declares most
variables related to hard and soft interrupt masks.
/usr/src/sys/i386/isa/icu.h:
o Moved definitions to <machine/ipl.h>
/usr/src/sys/i386/isa/icu.s:
o Software interrupts (SWIs) and delayed hardware interrupts (HWIs)
are now handled uniformally, and dispatching them from splx() is
more like dispatching them from _doreti. The dispatcher is
essentially *(handler[ffs(ipending & ~cpl)]().
o More care (not quite enough) is taken to avoid unbounded nesting
of interrupts.
o The interface to softclock() is changed so that a trap frame is
not required.
o Fast interrupt handlers are now handled more uniformally.
Configuration is still too early (new handlers would require
bits in <machine/ipl.h> and functions to vector.s).
o splnnn() and splx() are no longer here; they are inline functions
(could be macros for other compilers). splz() is the nontrivial
part of the old splx().
/usr/src/sys/i386/isa/ipl.h
o New file. Supposed to have only bus-dependent stuff. Perhaps
the h/w masks should be declared here.
/usr/src/sys/i386/isa/isa.c:
o DON'T APPLY ALL OF THIS DIFF. You need only things involving
*mask and *MASK and comments about them. netmask is now a pure
software mask. It works like the softclock mask.
/usr/src/sys/i386/isa/vector.s:
o Reorganize AUTO_EOI* macros.
o Option FAST_INTR_HANDLER_USERS_ES for people who don't trust
fastintr handlers.
o fastintr handlers need to metamorphose into ordinary interrupt
handlers if their SWI bit has become set. Previously, sio had
unintended latency for handling output completions and input
of SLIP framing characters because this was not done.
/usr/src/sys/net/netisr.h:
o The machine-dependent stuff is now imported from <machine/ipl.h>.
/usr/src/sys/sys/systm.h
o DON'T APPLY ALL OF THIS DIFF. You need mainly the different
splx() prototype. The spl*() prototypes are duplicated as
inlines in <machine/ipl.h> but they need to be duplicated here
in case there are no inlines. I sent systm.h and cpufunc.h
to Garrett. We agree that spl0 should be replaced by splnone
and not the other way around like I've done.
/usr/src/sys/kern/kern_clock.c
o splsoftclock() now lowers cpl so the direct call to softclock()
works as intended.
o softclock() interface changed to avoid passing the whole frame
(some machines may need another change for profile_tick()).
o profiling renamed _profiling to avoid ANSI namespace pollution.
(I had to improve the mcount() interface and may as well fix it.)
The GUPROF variant doesn't actually reference profiling here,
but the 'U' in GUPROF should mean to select the microtimer
mcount() and not change the interface.
1994-04-02 07:00:53 +00:00
|
|
|
* with sio because the serial interrupt layer doesn't use tty_imask. Only
|
|
|
|
|
* non-serial ttys suffer. It's more stupid that ALL 'net's are blocked
|
|
|
|
|
* during spltty.
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
*/
|
1993-06-12 14:58:17 +00:00
|
|
|
#include "sl.h"
|
1996-01-27 01:56:30 +00:00
|
|
|
#if NSL > 0
|
New interrupt code from Bruce Evans. In additional to Bruce's attached
list of changes, I've made the following additional changes:
1) i386/include/ipl.h renamed to spl.h as the name conflicts with the
file of the same name in i386/isa/ipl.h.
2) changed all use of *mask (i.e. netmask, biomask, ttymask, etc) to
*_imask (net_imask, etc).
3) changed vestige of splnet use in if_is to splimp.
4) got rid of "impmask" completely (Bruce had gotten rid of netmask),
and are now using net_imask instead.
5) dozens of minor cruft to glue in Bruce's changes.
These require changes I made to config(8) as well, and thus it must
be rebuilt.
-DG
from Bruce Evans:
sio:
o No diff is supplied. Remove the define of setsofttty(). I hope
that is enough.
*.s:
o i386/isa/debug.h no longer exists. The event counters became too
much trouble to maintain. All function call entry and exception
entry counters can be recovered by using profiling kernel (the new
profiling supports all entry points; however, it is too slow to
leave enabled all the time; it also). Only BDBTRAP() from debug.h
is now used. That is moved to exception.s. It might be worth
preserving SHOW_BITS() and calling it from _mcount() (if enabled).
o T_ASTFLT is now only set just before calling trap().
o All exception handlers set SWI_AST_MASK in cpl as soon as possible
after entry and arrange for _doreti to restore it atomically with
exiting. It is not possible to set it atomically with entering
the kernel, so it must be checked against the user mode bits in
the trap frame before committing to using it. There is no place
to store the old value of cpl for syscalls or traps, so there are
some complications restoring it.
Profiling stuff (mostly in *.s):
o Changes to kern/subr_mcount.c, gcc and gprof are not supplied yet.
o All interesting labels `foo' are renamed `_foo' and all
uninteresting labels `_bar' are renamed `bar'. A small change
to gprof allows ignoring labels not starting with underscores.
o MCOUNT_LABEL() is to provide names for counters for times spent
in exception handlers.
o FAKE_MCOUNT() is a version of MCOUNT() suitable for exception
handlers. Its arg is the pc where the exception occurred. The
new mcount() pretends that this was a call from that pc to a
suitable MCOUNT_LABEL().
o MEXITCOUNT is to turn off any timer started by MCOUNT().
/usr/src/sys/i386/i386/exception.s:
o The non-BDB BPTTRAP() macros were doing a sti even when interrupts
were disabled when the trap occurred. The sti (fixed) sti is
actually a no-op unless you have my changes to machdep.c that make
the debugger trap gates interrupt gates, but fixing that would
make the ifdefs messier. ddb seems to be unharmed by both
interrupts always disabled and always enabled (I had the branch in
the fix back to front for some time :-().
o There is no known pushal bug.
o tf_err can be left as garbage for syscalls.
/usr/src/sys/i386/i386/locore.s:
o Fix and update BDE_DEBUGGER support.
o ENTRY(btext) before initialization was dangerous.
o Warm boot shot was longer than intended.
/usr/src/sys/i386/i386/machdep.c:
o DON'T APPLY ALL OF THIS DIFF. It's what I'm using, but may require
other changes.
Use the following:
o Remove aston() and setsoftclock().
Maybe use the following:
o No netisr.h.
o Spelling fix.
o Delay to read the Rebooting message.
o Fix for vm system unmapping a reduced area of memory
after bounds_check_with_label() reduces the size of
a physical i/o for a partition boundary. A similar
fix is required in kern_physio.c.
o Correct use of __CONCAT. It never worked here for non-
ANSI cpp's. Is it time to drop support for non-ANSI?
o gdt_segs init. 0xffffffffUL is bogus because ssd_limit
is not 32 bits. The replacement may have the same
value :-), but is more natural.
o physmem was one page too low. Confusing variable names.
Don't use the following:
o Better numbers of buffers. Each 8K page requires up to
16 buffer headers. On my system, this results in 5576
buffers containing [up to] 2854912 bytes of memory.
The usual allocation of about 384 buffers only holds
192K of disk if you use it on an fs with a block size
of 512.
o gdt changes for bdb.
o *TGT -> *IDT changes for bdb.
o #ifdefed changes for bdb.
/usr/src/sys/i386/i386/microtime.s:
o Use the correct asm macros. I think asm.h was copied from Mach
just for microtime and isn't used now. It certainly doesn't
belong in <sys>. Various macros are also duplicated in
sys/i386/boot.h and libc/i386/*.h.
o Don't switch to and from the IRR; it is guaranteed to be selected
(default after ICU init and explicitly selected in isa.c too, and
never changed until the old microtime clobbered it).
/usr/src/sys/i386/i386/support.s:
o Non-essential changes (none related to spls or profiling).
o Removed slow loads of %gs again. The LDT support may require
not relying on %gs, but loading it is not the way to fix it!
Some places (copyin ...) forgot to load it. Loading it clobbers
the user %gs. trap() still loads it after certain types of
faults so that fuword() etc can rely on it without loading it
explicitly. Exception handlers don't restore it. If we want
to preserve the user %gs, then the fastest method is to not
touch it except for context switches. Comparing with
VM_MAXUSER_ADDRESS and branching takes only 2 or 4 cycles on
a 486, while loading %gs takes 9 cycles and using it takes
another.
o Fixed a signed branch to unsigned.
/usr/src/sys/i386/i386/swtch.s:
o Move spl0() outside of idle loop.
o Remove cli/sti from idle loop. sw1 does a cli, and in the
unlikely event of an interrupt occurring and whichqs becoming
zero, sw1 will just jump back to _idle.
o There's no spl0() function in asm any more, so use splz().
o swtch() doesn't need to be superaligned, at least with the
new mcounting.
o Fixed a signed branch to unsigned.
o Removed astoff().
/usr/src/sys/i386/i386/trap.c:
o The decentralized extern decls were inconsistent, of course.
o Fixed typo MATH_EMULTATE in comments. */
o Removed unused variables.
o Old netmask is now impmask; print it instead. Perhaps we
should print some of the new masks.
o BTW, trap() should not print anything for normal debugger
traps.
/usr/src/sys/i386/include/asmacros.h:
o DON'T APPLY ALL OF THIS DIFF. Just use some of the null macros
as necessary.
/usr/src/sys/i386/include/cpu.h:
o CLKF_BASEPRI() changes since cpl == SWI_AST_MASK is now normal
while the kernel is running.
o Don't use var++ to set boolean variables. It fails after a mere
4G times :-) and is slower than storing a constant on [3-4]86s.
/usr/src/sys/i386/include/cpufunc.h:
o DON'T APPLY ALL OF THIS DIFF. You need mainly the include of
<machine/ipl.h>. Unfortunately, <machine/ipl.h> is needed by
almost everything for the inlines.
/usr/src/sys/i386/include/ipl.h:
o New file. Defines spl inlines and SWI macros and declares most
variables related to hard and soft interrupt masks.
/usr/src/sys/i386/isa/icu.h:
o Moved definitions to <machine/ipl.h>
/usr/src/sys/i386/isa/icu.s:
o Software interrupts (SWIs) and delayed hardware interrupts (HWIs)
are now handled uniformally, and dispatching them from splx() is
more like dispatching them from _doreti. The dispatcher is
essentially *(handler[ffs(ipending & ~cpl)]().
o More care (not quite enough) is taken to avoid unbounded nesting
of interrupts.
o The interface to softclock() is changed so that a trap frame is
not required.
o Fast interrupt handlers are now handled more uniformally.
Configuration is still too early (new handlers would require
bits in <machine/ipl.h> and functions to vector.s).
o splnnn() and splx() are no longer here; they are inline functions
(could be macros for other compilers). splz() is the nontrivial
part of the old splx().
/usr/src/sys/i386/isa/ipl.h
o New file. Supposed to have only bus-dependent stuff. Perhaps
the h/w masks should be declared here.
/usr/src/sys/i386/isa/isa.c:
o DON'T APPLY ALL OF THIS DIFF. You need only things involving
*mask and *MASK and comments about them. netmask is now a pure
software mask. It works like the softclock mask.
/usr/src/sys/i386/isa/vector.s:
o Reorganize AUTO_EOI* macros.
o Option FAST_INTR_HANDLER_USERS_ES for people who don't trust
fastintr handlers.
o fastintr handlers need to metamorphose into ordinary interrupt
handlers if their SWI bit has become set. Previously, sio had
unintended latency for handling output completions and input
of SLIP framing characters because this was not done.
/usr/src/sys/net/netisr.h:
o The machine-dependent stuff is now imported from <machine/ipl.h>.
/usr/src/sys/sys/systm.h
o DON'T APPLY ALL OF THIS DIFF. You need mainly the different
splx() prototype. The spl*() prototypes are duplicated as
inlines in <machine/ipl.h> but they need to be duplicated here
in case there are no inlines. I sent systm.h and cpufunc.h
to Garrett. We agree that spl0 should be replaced by splnone
and not the other way around like I've done.
/usr/src/sys/kern/kern_clock.c
o splsoftclock() now lowers cpl so the direct call to softclock()
works as intended.
o softclock() interface changed to avoid passing the whole frame
(some machines may need another change for profile_tick()).
o profiling renamed _profiling to avoid ANSI namespace pollution.
(I had to improve the mcount() interface and may as well fix it.)
The GUPROF variant doesn't actually reference profiling here,
but the 'U' in GUPROF should mean to select the microtimer
mcount() and not change the interface.
1994-04-02 07:00:53 +00:00
|
|
|
net_imask |= tty_imask;
|
|
|
|
|
tty_imask = net_imask;
|
|
|
|
|
#endif
|
1995-10-31 21:03:57 +00:00
|
|
|
|
New interrupt code from Bruce Evans. In additional to Bruce's attached
list of changes, I've made the following additional changes:
1) i386/include/ipl.h renamed to spl.h as the name conflicts with the
file of the same name in i386/isa/ipl.h.
2) changed all use of *mask (i.e. netmask, biomask, ttymask, etc) to
*_imask (net_imask, etc).
3) changed vestige of splnet use in if_is to splimp.
4) got rid of "impmask" completely (Bruce had gotten rid of netmask),
and are now using net_imask instead.
5) dozens of minor cruft to glue in Bruce's changes.
These require changes I made to config(8) as well, and thus it must
be rebuilt.
-DG
from Bruce Evans:
sio:
o No diff is supplied. Remove the define of setsofttty(). I hope
that is enough.
*.s:
o i386/isa/debug.h no longer exists. The event counters became too
much trouble to maintain. All function call entry and exception
entry counters can be recovered by using profiling kernel (the new
profiling supports all entry points; however, it is too slow to
leave enabled all the time; it also). Only BDBTRAP() from debug.h
is now used. That is moved to exception.s. It might be worth
preserving SHOW_BITS() and calling it from _mcount() (if enabled).
o T_ASTFLT is now only set just before calling trap().
o All exception handlers set SWI_AST_MASK in cpl as soon as possible
after entry and arrange for _doreti to restore it atomically with
exiting. It is not possible to set it atomically with entering
the kernel, so it must be checked against the user mode bits in
the trap frame before committing to using it. There is no place
to store the old value of cpl for syscalls or traps, so there are
some complications restoring it.
Profiling stuff (mostly in *.s):
o Changes to kern/subr_mcount.c, gcc and gprof are not supplied yet.
o All interesting labels `foo' are renamed `_foo' and all
uninteresting labels `_bar' are renamed `bar'. A small change
to gprof allows ignoring labels not starting with underscores.
o MCOUNT_LABEL() is to provide names for counters for times spent
in exception handlers.
o FAKE_MCOUNT() is a version of MCOUNT() suitable for exception
handlers. Its arg is the pc where the exception occurred. The
new mcount() pretends that this was a call from that pc to a
suitable MCOUNT_LABEL().
o MEXITCOUNT is to turn off any timer started by MCOUNT().
/usr/src/sys/i386/i386/exception.s:
o The non-BDB BPTTRAP() macros were doing a sti even when interrupts
were disabled when the trap occurred. The sti (fixed) sti is
actually a no-op unless you have my changes to machdep.c that make
the debugger trap gates interrupt gates, but fixing that would
make the ifdefs messier. ddb seems to be unharmed by both
interrupts always disabled and always enabled (I had the branch in
the fix back to front for some time :-().
o There is no known pushal bug.
o tf_err can be left as garbage for syscalls.
/usr/src/sys/i386/i386/locore.s:
o Fix and update BDE_DEBUGGER support.
o ENTRY(btext) before initialization was dangerous.
o Warm boot shot was longer than intended.
/usr/src/sys/i386/i386/machdep.c:
o DON'T APPLY ALL OF THIS DIFF. It's what I'm using, but may require
other changes.
Use the following:
o Remove aston() and setsoftclock().
Maybe use the following:
o No netisr.h.
o Spelling fix.
o Delay to read the Rebooting message.
o Fix for vm system unmapping a reduced area of memory
after bounds_check_with_label() reduces the size of
a physical i/o for a partition boundary. A similar
fix is required in kern_physio.c.
o Correct use of __CONCAT. It never worked here for non-
ANSI cpp's. Is it time to drop support for non-ANSI?
o gdt_segs init. 0xffffffffUL is bogus because ssd_limit
is not 32 bits. The replacement may have the same
value :-), but is more natural.
o physmem was one page too low. Confusing variable names.
Don't use the following:
o Better numbers of buffers. Each 8K page requires up to
16 buffer headers. On my system, this results in 5576
buffers containing [up to] 2854912 bytes of memory.
The usual allocation of about 384 buffers only holds
192K of disk if you use it on an fs with a block size
of 512.
o gdt changes for bdb.
o *TGT -> *IDT changes for bdb.
o #ifdefed changes for bdb.
/usr/src/sys/i386/i386/microtime.s:
o Use the correct asm macros. I think asm.h was copied from Mach
just for microtime and isn't used now. It certainly doesn't
belong in <sys>. Various macros are also duplicated in
sys/i386/boot.h and libc/i386/*.h.
o Don't switch to and from the IRR; it is guaranteed to be selected
(default after ICU init and explicitly selected in isa.c too, and
never changed until the old microtime clobbered it).
/usr/src/sys/i386/i386/support.s:
o Non-essential changes (none related to spls or profiling).
o Removed slow loads of %gs again. The LDT support may require
not relying on %gs, but loading it is not the way to fix it!
Some places (copyin ...) forgot to load it. Loading it clobbers
the user %gs. trap() still loads it after certain types of
faults so that fuword() etc can rely on it without loading it
explicitly. Exception handlers don't restore it. If we want
to preserve the user %gs, then the fastest method is to not
touch it except for context switches. Comparing with
VM_MAXUSER_ADDRESS and branching takes only 2 or 4 cycles on
a 486, while loading %gs takes 9 cycles and using it takes
another.
o Fixed a signed branch to unsigned.
/usr/src/sys/i386/i386/swtch.s:
o Move spl0() outside of idle loop.
o Remove cli/sti from idle loop. sw1 does a cli, and in the
unlikely event of an interrupt occurring and whichqs becoming
zero, sw1 will just jump back to _idle.
o There's no spl0() function in asm any more, so use splz().
o swtch() doesn't need to be superaligned, at least with the
new mcounting.
o Fixed a signed branch to unsigned.
o Removed astoff().
/usr/src/sys/i386/i386/trap.c:
o The decentralized extern decls were inconsistent, of course.
o Fixed typo MATH_EMULTATE in comments. */
o Removed unused variables.
o Old netmask is now impmask; print it instead. Perhaps we
should print some of the new masks.
o BTW, trap() should not print anything for normal debugger
traps.
/usr/src/sys/i386/include/asmacros.h:
o DON'T APPLY ALL OF THIS DIFF. Just use some of the null macros
as necessary.
/usr/src/sys/i386/include/cpu.h:
o CLKF_BASEPRI() changes since cpl == SWI_AST_MASK is now normal
while the kernel is running.
o Don't use var++ to set boolean variables. It fails after a mere
4G times :-) and is slower than storing a constant on [3-4]86s.
/usr/src/sys/i386/include/cpufunc.h:
o DON'T APPLY ALL OF THIS DIFF. You need mainly the include of
<machine/ipl.h>. Unfortunately, <machine/ipl.h> is needed by
almost everything for the inlines.
/usr/src/sys/i386/include/ipl.h:
o New file. Defines spl inlines and SWI macros and declares most
variables related to hard and soft interrupt masks.
/usr/src/sys/i386/isa/icu.h:
o Moved definitions to <machine/ipl.h>
/usr/src/sys/i386/isa/icu.s:
o Software interrupts (SWIs) and delayed hardware interrupts (HWIs)
are now handled uniformally, and dispatching them from splx() is
more like dispatching them from _doreti. The dispatcher is
essentially *(handler[ffs(ipending & ~cpl)]().
o More care (not quite enough) is taken to avoid unbounded nesting
of interrupts.
o The interface to softclock() is changed so that a trap frame is
not required.
o Fast interrupt handlers are now handled more uniformally.
Configuration is still too early (new handlers would require
bits in <machine/ipl.h> and functions to vector.s).
o splnnn() and splx() are no longer here; they are inline functions
(could be macros for other compilers). splz() is the nontrivial
part of the old splx().
/usr/src/sys/i386/isa/ipl.h
o New file. Supposed to have only bus-dependent stuff. Perhaps
the h/w masks should be declared here.
/usr/src/sys/i386/isa/isa.c:
o DON'T APPLY ALL OF THIS DIFF. You need only things involving
*mask and *MASK and comments about them. netmask is now a pure
software mask. It works like the softclock mask.
/usr/src/sys/i386/isa/vector.s:
o Reorganize AUTO_EOI* macros.
o Option FAST_INTR_HANDLER_USERS_ES for people who don't trust
fastintr handlers.
o fastintr handlers need to metamorphose into ordinary interrupt
handlers if their SWI bit has become set. Previously, sio had
unintended latency for handling output completions and input
of SLIP framing characters because this was not done.
/usr/src/sys/net/netisr.h:
o The machine-dependent stuff is now imported from <machine/ipl.h>.
/usr/src/sys/sys/systm.h
o DON'T APPLY ALL OF THIS DIFF. You need mainly the different
splx() prototype. The spl*() prototypes are duplicated as
inlines in <machine/ipl.h> but they need to be duplicated here
in case there are no inlines. I sent systm.h and cpufunc.h
to Garrett. We agree that spl0 should be replaced by splnone
and not the other way around like I've done.
/usr/src/sys/kern/kern_clock.c
o splsoftclock() now lowers cpl so the direct call to softclock()
works as intended.
o softclock() interface changed to avoid passing the whole frame
(some machines may need another change for profile_tick()).
o profiling renamed _profiling to avoid ANSI namespace pollution.
(I had to improve the mcount() interface and may as well fix it.)
The GUPROF variant doesn't actually reference profiling here,
but the 'U' in GUPROF should mean to select the microtimer
mcount() and not change the interface.
1994-04-02 07:00:53 +00:00
|
|
|
/* bio_imask |= tty_imask ; can some tty devices use buffers? */
|
1995-10-31 21:03:57 +00:00
|
|
|
|
|
|
|
|
if (bootverbose)
|
|
|
|
|
printf("imasks: bio %x, tty %x, net %x\n",
|
1996-01-27 01:56:30 +00:00
|
|
|
bio_imask, tty_imask, net_imask);
|
1995-10-31 21:03:57 +00:00
|
|
|
|
1994-08-18 05:09:36 +00:00
|
|
|
/*
|
|
|
|
|
* Finish initializing intr_mask[]. Note that the partly
|
|
|
|
|
* constructed masks aren't actually used since we're at splhigh.
|
|
|
|
|
* For fully dynamic initialization, register_intr() and
|
1997-05-26 14:42:24 +00:00
|
|
|
* icu_unset() will have to adjust the masks for _all_
|
1994-08-18 05:09:36 +00:00
|
|
|
* interrupts and for tty_imask, etc.
|
|
|
|
|
*/
|
|
|
|
|
for (dvp = isa_devtab_tty; dvp->id_driver; dvp++)
|
|
|
|
|
register_imask(dvp, tty_imask);
|
|
|
|
|
for (dvp = isa_devtab_bio; dvp->id_driver; dvp++)
|
|
|
|
|
register_imask(dvp, bio_imask);
|
|
|
|
|
for (dvp = isa_devtab_net; dvp->id_driver; dvp++)
|
|
|
|
|
register_imask(dvp, net_imask);
|
1997-09-21 21:41:49 +00:00
|
|
|
for (dvp = isa_devtab_cam; dvp->id_driver; dvp++)
|
|
|
|
|
register_imask(dvp, cam_imask);
|
1994-08-18 05:09:36 +00:00
|
|
|
for (dvp = isa_devtab_null; dvp->id_driver; dvp++)
|
|
|
|
|
register_imask(dvp, SWI_CLOCK_MASK);
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Configure an ISA device.
|
|
|
|
|
*/
|
1994-08-18 05:09:36 +00:00
|
|
|
static void
|
1993-06-12 14:58:17 +00:00
|
|
|
config_isadev(isdp, mp)
|
1997-05-26 14:42:24 +00:00
|
|
|
struct isa_device *isdp;
|
|
|
|
|
u_int *mp;
|
1994-10-01 02:56:21 +00:00
|
|
|
{
|
|
|
|
|
config_isadev_c(isdp, mp, 0);
|
|
|
|
|
}
|
1995-05-30 08:16:23 +00:00
|
|
|
|
1994-10-01 02:56:21 +00:00
|
|
|
void
|
|
|
|
|
reconfig_isadev(isdp, mp)
|
|
|
|
|
struct isa_device *isdp;
|
|
|
|
|
u_int *mp;
|
|
|
|
|
{
|
|
|
|
|
config_isadev_c(isdp, mp, 1);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void
|
|
|
|
|
config_isadev_c(isdp, mp, reconfig)
|
1993-06-12 14:58:17 +00:00
|
|
|
struct isa_device *isdp;
|
|
|
|
|
u_int *mp;
|
1994-10-01 02:56:21 +00:00
|
|
|
int reconfig;
|
1993-06-12 14:58:17 +00:00
|
|
|
{
|
1994-08-18 05:09:36 +00:00
|
|
|
u_int checkbits;
|
|
|
|
|
int id_alive;
|
1994-10-01 02:56:21 +00:00
|
|
|
int last_alive;
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
struct isa_driver *dp = isdp->id_driver;
|
1995-05-30 08:16:23 +00:00
|
|
|
|
1994-11-03 04:15:04 +00:00
|
|
|
if (!isdp->id_enabled) {
|
1997-10-12 08:31:41 +00:00
|
|
|
if (bootverbose)
|
|
|
|
|
printf("%s%d: disabled, not probed.\n", dp->name, isdp->id_unit);
|
|
|
|
|
return;
|
1994-11-03 04:15:04 +00:00
|
|
|
}
|
1996-01-27 01:56:30 +00:00
|
|
|
checkbits = CC_DRQ | CC_IOADDR | CC_MEMADDR;
|
1994-10-01 02:56:21 +00:00
|
|
|
if (!reconfig && haveseen_isadev(isdp, checkbits))
|
1994-08-18 05:09:36 +00:00
|
|
|
return;
|
1994-10-01 02:56:21 +00:00
|
|
|
if (!reconfig && isdp->id_maddr) {
|
1997-01-08 05:56:55 +00:00
|
|
|
isdp->id_maddr -= ISA_HOLE_START;
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
isdp->id_maddr += atdevbase;
|
|
|
|
|
}
|
1994-10-01 02:56:21 +00:00
|
|
|
if (reconfig) {
|
|
|
|
|
last_alive = isdp->id_alive;
|
1995-02-17 02:22:57 +00:00
|
|
|
isdp->id_reconfig = 1;
|
1994-10-01 02:56:21 +00:00
|
|
|
}
|
|
|
|
|
else {
|
|
|
|
|
last_alive = 0;
|
1995-02-17 02:22:57 +00:00
|
|
|
isdp->id_reconfig = 0;
|
1994-10-01 02:56:21 +00:00
|
|
|
}
|
1994-08-18 05:09:36 +00:00
|
|
|
id_alive = (*dp->probe)(isdp);
|
|
|
|
|
if (id_alive) {
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
/*
|
|
|
|
|
* Only print the I/O address range if id_alive != -1
|
|
|
|
|
* Right now this is a temporary fix just for the new
|
|
|
|
|
* NPX code so that if it finds a 486 that can use trap
|
|
|
|
|
* 16 it will not report I/O addresses.
|
|
|
|
|
* Rod Grimes 04/26/94
|
|
|
|
|
*/
|
1994-10-01 02:56:21 +00:00
|
|
|
if (!isdp->id_reconfig) {
|
|
|
|
|
printf("%s%d", dp->name, isdp->id_unit);
|
|
|
|
|
if (id_alive != -1) {
|
1997-03-25 03:13:05 +00:00
|
|
|
if (isdp->id_iobase == -1)
|
1998-11-29 15:42:40 +00:00
|
|
|
printf(" at");
|
1997-03-28 01:02:17 +00:00
|
|
|
else {
|
1997-03-25 03:13:05 +00:00
|
|
|
printf(" at 0x%x", isdp->id_iobase);
|
|
|
|
|
if (isdp->id_iobase + id_alive - 1 !=
|
|
|
|
|
isdp->id_iobase) {
|
|
|
|
|
printf("-0x%x",
|
|
|
|
|
isdp->id_iobase + id_alive - 1);
|
|
|
|
|
}
|
1994-10-01 02:56:21 +00:00
|
|
|
}
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
1994-10-01 02:56:21 +00:00
|
|
|
if (isdp->id_irq)
|
|
|
|
|
printf(" irq %d", ffs(isdp->id_irq) - 1);
|
|
|
|
|
if (isdp->id_drq != -1)
|
|
|
|
|
printf(" drq %d", isdp->id_drq);
|
|
|
|
|
if (isdp->id_maddr)
|
|
|
|
|
printf(" maddr 0x%lx", kvtop(isdp->id_maddr));
|
|
|
|
|
if (isdp->id_msize)
|
|
|
|
|
printf(" msize %d", isdp->id_msize);
|
|
|
|
|
if (isdp->id_flags)
|
|
|
|
|
printf(" flags 0x%x", isdp->id_flags);
|
1995-04-06 13:55:56 +00:00
|
|
|
if (isdp->id_iobase && !(isdp->id_iobase & 0xf300)) {
|
1995-04-04 22:48:40 +00:00
|
|
|
printf(" on motherboard");
|
|
|
|
|
} else if (isdp->id_iobase >= 0x1000 &&
|
|
|
|
|
!(isdp->id_iobase & 0x300)) {
|
|
|
|
|
printf (" on eisa slot %d",
|
|
|
|
|
isdp->id_iobase >> 12);
|
|
|
|
|
} else {
|
|
|
|
|
printf (" on isa");
|
1994-01-22 21:52:04 +00:00
|
|
|
}
|
1995-03-25 05:54:33 +00:00
|
|
|
printf("\n");
|
1994-10-01 02:56:21 +00:00
|
|
|
/*
|
1995-05-30 08:16:23 +00:00
|
|
|
* Check for conflicts again. The driver may have
|
|
|
|
|
* changed *dvp. We should weaken the early check
|
|
|
|
|
* since the driver may have been able to change
|
|
|
|
|
* *dvp to avoid conflicts if given a chance. We
|
|
|
|
|
* already skip the early check for IRQs and force
|
1994-10-01 02:56:21 +00:00
|
|
|
* a check for IRQs in the next group of checks.
|
1996-01-27 01:56:30 +00:00
|
|
|
*/
|
1998-10-12 13:12:45 +00:00
|
|
|
checkbits |= CC_ATTACH | CC_IRQ;
|
1997-08-25 22:52:59 +00:00
|
|
|
if (haveseen_isadev(isdp, checkbits))
|
1994-10-01 02:56:21 +00:00
|
|
|
return;
|
|
|
|
|
isdp->id_alive = id_alive;
|
|
|
|
|
}
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
(*dp->attach)(isdp);
|
1998-10-22 05:58:45 +00:00
|
|
|
if (isdp->id_irq != 0 && isdp->id_intr == NULL)
|
|
|
|
|
printf("%s%d: irq with no handler\n",
|
|
|
|
|
dp->name, isdp->id_unit);
|
|
|
|
|
if (isdp->id_irq != 0 && isdp->id_intr != NULL) {
|
1997-06-25 21:00:00 +00:00
|
|
|
#ifdef APIC_IO
|
1997-05-29 04:55:39 +00:00
|
|
|
/*
|
|
|
|
|
* Some motherboards use upper IRQs for traditional
|
|
|
|
|
* ISA INTerrupt sources. In particular we have
|
|
|
|
|
* seen the secondary IDE connected to IRQ20.
|
|
|
|
|
* This code detects and fixes this situation.
|
|
|
|
|
*/
|
1997-05-05 22:56:13 +00:00
|
|
|
u_int apic_mask;
|
|
|
|
|
int rirq;
|
|
|
|
|
|
1997-06-25 21:00:00 +00:00
|
|
|
apic_mask = isa_apic_mask(isdp->id_irq);
|
1997-05-29 04:55:39 +00:00
|
|
|
if (apic_mask != isdp->id_irq) {
|
|
|
|
|
rirq = ffs(isdp->id_irq) - 1;
|
1997-05-05 22:56:13 +00:00
|
|
|
isdp->id_irq = apic_mask;
|
1997-05-29 04:55:39 +00:00
|
|
|
undirect_isa_irq(rirq); /* free for ISA */
|
1997-05-05 22:56:13 +00:00
|
|
|
}
|
1997-06-25 21:00:00 +00:00
|
|
|
#endif /* APIC_IO */
|
1997-07-09 17:58:16 +00:00
|
|
|
register_intr(ffs(isdp->id_irq) - 1, isdp->id_id,
|
|
|
|
|
isdp->id_ri_flags, isdp->id_intr,
|
|
|
|
|
mp, isdp->id_unit);
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
}
|
|
|
|
|
} else {
|
1994-10-01 02:56:21 +00:00
|
|
|
if (isdp->id_reconfig) {
|
|
|
|
|
(*dp->attach)(isdp); /* reconfiguration attach */
|
|
|
|
|
}
|
|
|
|
|
if (!last_alive) {
|
|
|
|
|
if (!isdp->id_reconfig) {
|
1996-01-27 01:56:30 +00:00
|
|
|
printf("%s%d not found",
|
|
|
|
|
dp->name, isdp->id_unit);
|
1997-03-28 01:02:17 +00:00
|
|
|
if (isdp->id_iobase != -1)
|
1994-10-01 02:56:21 +00:00
|
|
|
printf(" at 0x%x", isdp->id_iobase);
|
|
|
|
|
printf("\n");
|
|
|
|
|
}
|
1997-09-19 15:20:25 +00:00
|
|
|
} else {
|
1997-05-26 14:42:24 +00:00
|
|
|
#if 0
|
1994-10-01 02:56:21 +00:00
|
|
|
/* This code has not been tested.... */
|
1998-10-22 05:58:45 +00:00
|
|
|
if (isdp->id_irq != 0 && isdp->id_intr != NULL) {
|
1997-05-26 14:42:24 +00:00
|
|
|
icu_unset(ffs(isdp->id_irq) - 1,
|
1994-10-01 02:56:21 +00:00
|
|
|
isdp->id_intr);
|
|
|
|
|
if (mp)
|
|
|
|
|
INTRUNMASK(*mp, isdp->id_irq);
|
|
|
|
|
}
|
1997-05-26 14:42:24 +00:00
|
|
|
#else
|
|
|
|
|
printf ("icu_unset() not supported here ...\n");
|
|
|
|
|
#endif
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
ALL:
Removed patch kit headers and rcsid strings, add $Id$.
isa.c:
Removed old #ifdef notyet isa_configure code, since it will never be
used, and I have done 90% of what it attempted to.
Add conflict checking code that searchs back through the devtab's looking
for any device that has already been found that may conflict with what
we are about to probe. Checks are mode for I/O address, memory address,
IRQ, and DRQ. This should stop the screwing up of any device that has
alread been found by other device probes.
Print out messages when we are not going to probe a device due to
a conflict so the user knows WHY something was not found. For example:
aha0 not probed due to irq conflict with ahb0 at 11
Now print out a message when a device is not found so the user knows
that it was probed for, but could not be found. For example:
ed1 not found at 0x320
For devices that have I/O address < 0x100 say that they are on the
motherboard, not on isa! The 0x100 magic number is per ISA spec. It
may seem funny that pc0 and sc0 report as being on the motherboard, but
this is due to the fact that the I/O address used is that of the keyboard
controller which IS on the motherboard. We really need to split the
keyboard probe from the display probe. It is completly legal to build
a pc with out one or the other, or even with out both!
npx.c:
Return -1 from the probe routine if we are using the Emulator so
that the i/o addresses are not printed, this is the same trick used
for 486's.
Do not print the ``Errors reported via Exception 16'', and
``Errors reported via IRQ 13'' messages any more, since these just lead
to more user confusion that anything. It still prints the message
``Error reporting broken, using 387 emulator'' so that the person is
aware that there mother board is ill.
1993-10-13 15:59:30 +00:00
|
|
|
}
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
|
|
|
|
|
1996-01-27 02:33:48 +00:00
|
|
|
static caddr_t dma_bouncebuf[8];
|
1996-04-08 19:38:57 +00:00
|
|
|
static u_int dma_bouncebufsize[8];
|
|
|
|
|
static u_int8_t dma_bounced = 0;
|
|
|
|
|
static u_int8_t dma_busy = 0; /* Used in isa_dmastart() */
|
|
|
|
|
static u_int8_t dma_inuse = 0; /* User for acquire/release */
|
1997-08-28 03:36:40 +00:00
|
|
|
static u_int8_t dma_auto_mode = 0;
|
1996-04-08 19:38:57 +00:00
|
|
|
|
|
|
|
|
#define VALID_DMA_MASK (7)
|
1993-06-12 14:58:17 +00:00
|
|
|
|
|
|
|
|
/* high byte of address is stored in this port for i-th dma channel */
|
1996-10-08 21:08:18 +00:00
|
|
|
static int dmapageport[8] = { 0x87, 0x83, 0x81, 0x82, 0x8f, 0x8b, 0x89, 0x8a };
|
1993-06-12 14:58:17 +00:00
|
|
|
|
1996-01-27 02:33:48 +00:00
|
|
|
/*
|
1996-04-08 19:38:57 +00:00
|
|
|
* Setup a DMA channel's bounce buffer.
|
1996-01-27 02:33:48 +00:00
|
|
|
*/
|
|
|
|
|
void
|
|
|
|
|
isa_dmainit(chan, bouncebufsize)
|
1996-04-08 19:38:57 +00:00
|
|
|
int chan;
|
|
|
|
|
u_int bouncebufsize;
|
1996-01-27 02:33:48 +00:00
|
|
|
{
|
|
|
|
|
void *buf;
|
|
|
|
|
|
1996-04-08 19:38:57 +00:00
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (chan & ~VALID_DMA_MASK)
|
|
|
|
|
panic("isa_dmainit: channel out of range");
|
|
|
|
|
|
|
|
|
|
if (dma_bouncebuf[chan] != NULL)
|
1996-01-27 02:33:48 +00:00
|
|
|
panic("isa_dmainit: impossible request");
|
1996-04-08 19:38:57 +00:00
|
|
|
#endif
|
|
|
|
|
|
1996-01-27 02:33:48 +00:00
|
|
|
dma_bouncebufsize[chan] = bouncebufsize;
|
|
|
|
|
|
|
|
|
|
/* Try malloc() first. It works better if it works. */
|
|
|
|
|
buf = malloc(bouncebufsize, M_DEVBUF, M_NOWAIT);
|
|
|
|
|
if (buf != NULL) {
|
|
|
|
|
if (isa_dmarangecheck(buf, bouncebufsize, chan) == 0) {
|
|
|
|
|
dma_bouncebuf[chan] = buf;
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
free(buf, M_DEVBUF);
|
|
|
|
|
}
|
|
|
|
|
buf = contigmalloc(bouncebufsize, M_DEVBUF, M_NOWAIT, 0ul, 0xfffffful,
|
|
|
|
|
1ul, chan & 4 ? 0x20000ul : 0x10000ul);
|
|
|
|
|
if (buf == NULL)
|
|
|
|
|
printf("isa_dmainit(%d, %d) failed\n", chan, bouncebufsize);
|
|
|
|
|
else
|
|
|
|
|
dma_bouncebuf[chan] = buf;
|
|
|
|
|
}
|
|
|
|
|
|
1996-04-08 19:38:57 +00:00
|
|
|
/*
|
|
|
|
|
* Register a DMA channel's usage. Usually called from a device driver
|
1998-04-17 22:37:19 +00:00
|
|
|
* in open() or during its initialization.
|
1996-04-08 19:38:57 +00:00
|
|
|
*/
|
|
|
|
|
int
|
|
|
|
|
isa_dma_acquire(chan)
|
|
|
|
|
int chan;
|
|
|
|
|
{
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (chan & ~VALID_DMA_MASK)
|
|
|
|
|
panic("isa_dma_acquire: channel out of range");
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
|
if (dma_inuse & (1 << chan)) {
|
|
|
|
|
printf("isa_dma_acquire: channel %d already in use\n", chan);
|
|
|
|
|
return (EBUSY);
|
|
|
|
|
}
|
|
|
|
|
dma_inuse |= (1 << chan);
|
1997-08-28 03:36:40 +00:00
|
|
|
dma_auto_mode &= ~(1 << chan);
|
1996-04-08 19:38:57 +00:00
|
|
|
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Unregister a DMA channel's usage. Usually called from a device driver
|
1998-04-17 22:37:19 +00:00
|
|
|
* during close() or during its shutdown.
|
1996-04-08 19:38:57 +00:00
|
|
|
*/
|
|
|
|
|
void
|
|
|
|
|
isa_dma_release(chan)
|
|
|
|
|
int chan;
|
|
|
|
|
{
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (chan & ~VALID_DMA_MASK)
|
|
|
|
|
panic("isa_dma_release: channel out of range");
|
|
|
|
|
|
1997-08-13 02:40:06 +00:00
|
|
|
if ((dma_inuse & (1 << chan)) == 0)
|
1996-04-08 19:38:57 +00:00
|
|
|
printf("isa_dma_release: channel %d not in use\n", chan);
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
|
if (dma_busy & (1 << chan)) {
|
|
|
|
|
dma_busy &= ~(1 << chan);
|
|
|
|
|
/*
|
|
|
|
|
* XXX We should also do "dma_bounced &= (1 << chan);"
|
|
|
|
|
* because we are acting on behalf of isa_dmadone() which
|
|
|
|
|
* was not called to end the last DMA operation. This does
|
|
|
|
|
* not matter now, but it may in the future.
|
|
|
|
|
*/
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
dma_inuse &= ~(1 << chan);
|
1997-08-28 03:36:40 +00:00
|
|
|
dma_auto_mode &= ~(1 << chan);
|
1996-04-08 19:38:57 +00:00
|
|
|
}
|
|
|
|
|
|
1993-06-12 14:58:17 +00:00
|
|
|
/*
|
|
|
|
|
* isa_dmacascade(): program 8237 DMA controller channel to accept
|
|
|
|
|
* external dma control by a board.
|
|
|
|
|
*/
|
1997-11-30 09:44:28 +00:00
|
|
|
void
|
|
|
|
|
isa_dmacascade(chan)
|
1996-04-08 19:38:57 +00:00
|
|
|
int chan;
|
1993-06-12 14:58:17 +00:00
|
|
|
{
|
1996-04-08 19:38:57 +00:00
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (chan & ~VALID_DMA_MASK)
|
|
|
|
|
panic("isa_dmacascade: channel out of range");
|
|
|
|
|
#endif
|
1993-06-12 14:58:17 +00:00
|
|
|
|
|
|
|
|
/* set dma channel mode, and set dma channel mode */
|
|
|
|
|
if ((chan & 4) == 0) {
|
|
|
|
|
outb(DMA1_MODE, DMA37MD_CASCADE | chan);
|
|
|
|
|
outb(DMA1_SMSK, chan);
|
|
|
|
|
} else {
|
|
|
|
|
outb(DMA2_MODE, DMA37MD_CASCADE | (chan & 3));
|
|
|
|
|
outb(DMA2_SMSK, chan & 3);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* isa_dmastart(): program 8237 DMA controller channel, avoid page alignment
|
|
|
|
|
* problems by using a bounce buffer.
|
|
|
|
|
*/
|
1997-11-30 09:44:28 +00:00
|
|
|
void
|
|
|
|
|
isa_dmastart(int flags, caddr_t addr, u_int nbytes, int chan)
|
1996-04-08 19:38:57 +00:00
|
|
|
{
|
|
|
|
|
vm_offset_t phys;
|
1993-06-12 14:58:17 +00:00
|
|
|
int waport;
|
|
|
|
|
caddr_t newaddr;
|
|
|
|
|
|
1996-04-08 19:38:57 +00:00
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (chan & ~VALID_DMA_MASK)
|
|
|
|
|
panic("isa_dmastart: channel out of range");
|
|
|
|
|
|
|
|
|
|
if ((chan < 4 && nbytes > (1<<16))
|
1993-06-12 14:58:17 +00:00
|
|
|
|| (chan >= 4 && (nbytes > (1<<17) || (u_int)addr & 1)))
|
1995-05-30 08:16:23 +00:00
|
|
|
panic("isa_dmastart: impossible request");
|
1993-06-12 14:58:17 +00:00
|
|
|
|
1997-08-13 02:40:06 +00:00
|
|
|
if ((dma_inuse & (1 << chan)) == 0)
|
1996-04-08 19:38:57 +00:00
|
|
|
printf("isa_dmastart: channel %d not acquired\n", chan);
|
1996-03-10 02:10:27 +00:00
|
|
|
#endif
|
1996-04-08 19:38:57 +00:00
|
|
|
|
1997-05-05 22:56:13 +00:00
|
|
|
#if 0
|
|
|
|
|
/*
|
|
|
|
|
* XXX This should be checked, but drivers like ad1848 only call
|
|
|
|
|
* isa_dmastart() once because they use Auto DMA mode. If we
|
|
|
|
|
* leave this in, drivers that do this will print this continuously.
|
|
|
|
|
*/
|
1996-04-08 19:38:57 +00:00
|
|
|
if (dma_busy & (1 << chan))
|
|
|
|
|
printf("isa_dmastart: channel %d busy\n", chan);
|
1997-05-05 22:56:13 +00:00
|
|
|
#endif
|
1996-04-08 19:38:57 +00:00
|
|
|
|
|
|
|
|
dma_busy |= (1 << chan);
|
|
|
|
|
|
1993-06-12 14:58:17 +00:00
|
|
|
if (isa_dmarangecheck(addr, nbytes, chan)) {
|
1996-01-27 02:33:48 +00:00
|
|
|
if (dma_bouncebuf[chan] == NULL
|
|
|
|
|
|| dma_bouncebufsize[chan] < nbytes)
|
|
|
|
|
panic("isa_dmastart: bad bounce buffer");
|
1996-04-08 19:38:57 +00:00
|
|
|
dma_bounced |= (1 << chan);
|
1996-01-27 02:33:48 +00:00
|
|
|
newaddr = dma_bouncebuf[chan];
|
1993-06-12 14:58:17 +00:00
|
|
|
|
|
|
|
|
/* copy bounce buffer on write */
|
|
|
|
|
if (!(flags & B_READ))
|
|
|
|
|
bcopy(addr, newaddr, nbytes);
|
|
|
|
|
addr = newaddr;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* translate to physical */
|
|
|
|
|
phys = pmap_extract(pmap_kernel(), (vm_offset_t)addr);
|
|
|
|
|
|
1997-08-28 03:36:40 +00:00
|
|
|
if (flags & B_RAW) {
|
|
|
|
|
dma_auto_mode |= (1 << chan);
|
|
|
|
|
} else {
|
|
|
|
|
dma_auto_mode &= ~(1 << chan);
|
|
|
|
|
}
|
|
|
|
|
|
1993-06-12 14:58:17 +00:00
|
|
|
if ((chan & 4) == 0) {
|
|
|
|
|
/*
|
|
|
|
|
* Program one of DMA channels 0..3. These are
|
|
|
|
|
* byte mode channels.
|
|
|
|
|
*/
|
|
|
|
|
/* set dma channel mode, and reset address ff */
|
1994-11-01 17:26:50 +00:00
|
|
|
|
|
|
|
|
/* If B_RAW flag is set, then use autoinitialise mode */
|
|
|
|
|
if (flags & B_RAW) {
|
|
|
|
|
if (flags & B_READ)
|
|
|
|
|
outb(DMA1_MODE, DMA37MD_AUTO|DMA37MD_WRITE|chan);
|
|
|
|
|
else
|
|
|
|
|
outb(DMA1_MODE, DMA37MD_AUTO|DMA37MD_READ|chan);
|
|
|
|
|
}
|
|
|
|
|
else
|
1993-06-12 14:58:17 +00:00
|
|
|
if (flags & B_READ)
|
|
|
|
|
outb(DMA1_MODE, DMA37MD_SINGLE|DMA37MD_WRITE|chan);
|
|
|
|
|
else
|
|
|
|
|
outb(DMA1_MODE, DMA37MD_SINGLE|DMA37MD_READ|chan);
|
|
|
|
|
outb(DMA1_FFC, 0);
|
|
|
|
|
|
|
|
|
|
/* send start address */
|
|
|
|
|
waport = DMA1_CHN(chan);
|
|
|
|
|
outb(waport, phys);
|
|
|
|
|
outb(waport, phys>>8);
|
|
|
|
|
outb(dmapageport[chan], phys>>16);
|
|
|
|
|
|
|
|
|
|
/* send count */
|
|
|
|
|
outb(waport + 1, --nbytes);
|
|
|
|
|
outb(waport + 1, nbytes>>8);
|
|
|
|
|
|
|
|
|
|
/* unmask channel */
|
|
|
|
|
outb(DMA1_SMSK, chan);
|
|
|
|
|
} else {
|
|
|
|
|
/*
|
|
|
|
|
* Program one of DMA channels 4..7. These are
|
|
|
|
|
* word mode channels.
|
|
|
|
|
*/
|
|
|
|
|
/* set dma channel mode, and reset address ff */
|
1994-11-01 17:26:50 +00:00
|
|
|
|
|
|
|
|
/* If B_RAW flag is set, then use autoinitialise mode */
|
|
|
|
|
if (flags & B_RAW) {
|
|
|
|
|
if (flags & B_READ)
|
|
|
|
|
outb(DMA2_MODE, DMA37MD_AUTO|DMA37MD_WRITE|(chan&3));
|
|
|
|
|
else
|
|
|
|
|
outb(DMA2_MODE, DMA37MD_AUTO|DMA37MD_READ|(chan&3));
|
|
|
|
|
}
|
|
|
|
|
else
|
1993-06-12 14:58:17 +00:00
|
|
|
if (flags & B_READ)
|
|
|
|
|
outb(DMA2_MODE, DMA37MD_SINGLE|DMA37MD_WRITE|(chan&3));
|
|
|
|
|
else
|
|
|
|
|
outb(DMA2_MODE, DMA37MD_SINGLE|DMA37MD_READ|(chan&3));
|
|
|
|
|
outb(DMA2_FFC, 0);
|
|
|
|
|
|
|
|
|
|
/* send start address */
|
|
|
|
|
waport = DMA2_CHN(chan - 4);
|
|
|
|
|
outb(waport, phys>>1);
|
|
|
|
|
outb(waport, phys>>9);
|
|
|
|
|
outb(dmapageport[chan], phys>>16);
|
|
|
|
|
|
|
|
|
|
/* send count */
|
|
|
|
|
nbytes >>= 1;
|
|
|
|
|
outb(waport + 2, --nbytes);
|
|
|
|
|
outb(waport + 2, nbytes>>8);
|
|
|
|
|
|
|
|
|
|
/* unmask channel */
|
|
|
|
|
outb(DMA2_SMSK, chan & 3);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
1997-11-30 09:44:28 +00:00
|
|
|
void
|
|
|
|
|
isa_dmadone(int flags, caddr_t addr, int nbytes, int chan)
|
1996-04-08 19:38:57 +00:00
|
|
|
{
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (chan & ~VALID_DMA_MASK)
|
|
|
|
|
panic("isa_dmadone: channel out of range");
|
|
|
|
|
|
1997-08-13 02:40:06 +00:00
|
|
|
if ((dma_inuse & (1 << chan)) == 0)
|
1996-04-08 19:38:57 +00:00
|
|
|
printf("isa_dmadone: channel %d not acquired\n", chan);
|
|
|
|
|
#endif
|
1993-06-12 14:58:17 +00:00
|
|
|
|
1997-08-28 03:36:40 +00:00
|
|
|
if (((dma_busy & (1 << chan)) == 0) &&
|
|
|
|
|
(dma_auto_mode & (1 << chan)) == 0 )
|
1996-01-27 02:33:48 +00:00
|
|
|
printf("isa_dmadone: channel %d not busy\n", chan);
|
1997-08-28 03:36:40 +00:00
|
|
|
|
1998-07-19 04:22:55 +00:00
|
|
|
if ((dma_auto_mode & (1 << chan)) == 0)
|
|
|
|
|
outb(chan & 4 ? DMA2_SMSK : DMA1_SMSK, (chan & 3) | 4);
|
1996-04-08 19:38:57 +00:00
|
|
|
|
|
|
|
|
if (dma_bounced & (1 << chan)) {
|
1996-01-27 02:33:48 +00:00
|
|
|
/* copy bounce buffer on read */
|
|
|
|
|
if (flags & B_READ)
|
|
|
|
|
bcopy(dma_bouncebuf[chan], addr, nbytes);
|
|
|
|
|
|
1996-04-08 19:38:57 +00:00
|
|
|
dma_bounced &= ~(1 << chan);
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
1996-04-08 19:38:57 +00:00
|
|
|
dma_busy &= ~(1 << chan);
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Check for problems with the address range of a DMA transfer
|
|
|
|
|
* (non-contiguous physical pages, outside of bus address space,
|
|
|
|
|
* crossing DMA page boundaries).
|
|
|
|
|
* Return true if special handling needed.
|
|
|
|
|
*/
|
|
|
|
|
|
1994-08-18 05:09:36 +00:00
|
|
|
static int
|
1997-11-30 09:44:28 +00:00
|
|
|
isa_dmarangecheck(caddr_t va, u_int length, int chan)
|
|
|
|
|
{
|
1993-06-12 14:58:17 +00:00
|
|
|
vm_offset_t phys, priorpage = 0, endva;
|
|
|
|
|
u_int dma_pgmsk = (chan & 4) ? ~(128*1024-1) : ~(64*1024-1);
|
|
|
|
|
|
1998-10-13 08:24:45 +00:00
|
|
|
endva = (vm_offset_t)round_page((vm_offset_t)va + length);
|
1996-05-02 10:43:17 +00:00
|
|
|
for (; va < (caddr_t) endva ; va += PAGE_SIZE) {
|
1993-06-12 14:58:17 +00:00
|
|
|
phys = trunc_page(pmap_extract(pmap_kernel(), (vm_offset_t)va));
|
|
|
|
|
#define ISARAM_END RAM_END
|
|
|
|
|
if (phys == 0)
|
|
|
|
|
panic("isa_dmacheck: no physical page present");
|
1994-04-02 20:43:25 +00:00
|
|
|
if (phys >= ISARAM_END)
|
1993-06-12 14:58:17 +00:00
|
|
|
return (1);
|
|
|
|
|
if (priorpage) {
|
1996-05-02 10:43:17 +00:00
|
|
|
if (priorpage + PAGE_SIZE != phys)
|
1993-06-12 14:58:17 +00:00
|
|
|
return (1);
|
|
|
|
|
/* check if crossing a DMA page boundary */
|
|
|
|
|
if (((u_int)priorpage ^ (u_int)phys) & dma_pgmsk)
|
|
|
|
|
return (1);
|
|
|
|
|
}
|
|
|
|
|
priorpage = phys;
|
|
|
|
|
}
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
1997-07-24 05:27:40 +00:00
|
|
|
/*
|
|
|
|
|
* Query the progress of a transfer on a DMA channel.
|
|
|
|
|
*
|
|
|
|
|
* To avoid having to interrupt a transfer in progress, we sample
|
|
|
|
|
* each of the high and low databytes twice, and apply the following
|
|
|
|
|
* logic to determine the correct count.
|
|
|
|
|
*
|
|
|
|
|
* Reads are performed with interrupts disabled, thus it is to be
|
|
|
|
|
* expected that the time between reads is very small. At most
|
|
|
|
|
* one rollover in the low count byte can be expected within the
|
|
|
|
|
* four reads that are performed.
|
|
|
|
|
*
|
|
|
|
|
* There are three gaps in which a rollover can occur :
|
|
|
|
|
*
|
|
|
|
|
* - read low1
|
|
|
|
|
* gap1
|
|
|
|
|
* - read high1
|
|
|
|
|
* gap2
|
|
|
|
|
* - read low2
|
|
|
|
|
* gap3
|
|
|
|
|
* - read high2
|
|
|
|
|
*
|
|
|
|
|
* If a rollover occurs in gap1 or gap2, the low2 value will be
|
|
|
|
|
* greater than the low1 value. In this case, low2 and high2 are a
|
|
|
|
|
* corresponding pair.
|
|
|
|
|
*
|
|
|
|
|
* In any other case, low1 and high1 can be considered to be correct.
|
|
|
|
|
*
|
|
|
|
|
* The function returns the number of bytes remaining in the transfer,
|
|
|
|
|
* or -1 if the channel requested is not active.
|
|
|
|
|
*
|
|
|
|
|
*/
|
|
|
|
|
int
|
|
|
|
|
isa_dmastatus(int chan)
|
|
|
|
|
{
|
|
|
|
|
u_long cnt = 0;
|
1997-07-28 07:49:40 +00:00
|
|
|
int ffport, waport;
|
1997-07-28 09:13:11 +00:00
|
|
|
u_long low1, high1, low2, high2;
|
1997-07-24 05:27:40 +00:00
|
|
|
|
|
|
|
|
/* channel active? */
|
1997-08-13 02:40:06 +00:00
|
|
|
if ((dma_inuse & (1 << chan)) == 0) {
|
1997-07-24 05:27:40 +00:00
|
|
|
printf("isa_dmastatus: channel %d not active\n", chan);
|
|
|
|
|
return(-1);
|
|
|
|
|
}
|
1997-08-28 03:36:40 +00:00
|
|
|
/* channel busy? */
|
1997-07-24 05:27:40 +00:00
|
|
|
|
1997-08-28 03:36:40 +00:00
|
|
|
if (((dma_busy & (1 << chan)) == 0) &&
|
|
|
|
|
(dma_auto_mode & (1 << chan)) == 0 ) {
|
|
|
|
|
printf("chan %d not busy\n", chan);
|
|
|
|
|
return -2 ;
|
|
|
|
|
}
|
1997-07-24 05:27:40 +00:00
|
|
|
if (chan < 4) { /* low DMA controller */
|
|
|
|
|
ffport = DMA1_FFC;
|
|
|
|
|
waport = DMA1_CHN(chan) + 1;
|
|
|
|
|
} else { /* high DMA controller */
|
|
|
|
|
ffport = DMA2_FFC;
|
|
|
|
|
waport = DMA2_CHN(chan - 4) + 2;
|
|
|
|
|
}
|
|
|
|
|
|
1997-07-28 07:49:40 +00:00
|
|
|
disable_intr(); /* no interrupts Mr Jones! */
|
1997-07-24 05:27:40 +00:00
|
|
|
outb(ffport, 0); /* clear register LSB flipflop */
|
1997-07-28 09:13:11 +00:00
|
|
|
low1 = inb(waport);
|
|
|
|
|
high1 = inb(waport);
|
1997-07-29 05:24:36 +00:00
|
|
|
outb(ffport, 0); /* clear again */
|
1997-07-24 05:27:40 +00:00
|
|
|
low2 = inb(waport);
|
|
|
|
|
high2 = inb(waport);
|
1997-07-29 05:24:36 +00:00
|
|
|
enable_intr(); /* enable interrupts again */
|
1997-07-24 05:27:40 +00:00
|
|
|
|
1997-07-29 05:24:36 +00:00
|
|
|
/*
|
|
|
|
|
* Now decide if a wrap has tried to skew our results.
|
|
|
|
|
* Note that after TC, the count will read 0xffff, while we want
|
|
|
|
|
* to return zero, so we add and then mask to compensate.
|
|
|
|
|
*/
|
1997-07-28 09:13:11 +00:00
|
|
|
if (low1 >= low2) {
|
1997-07-29 05:24:36 +00:00
|
|
|
cnt = (low1 + (high1 << 8) + 1) & 0xffff;
|
1997-07-24 05:27:40 +00:00
|
|
|
} else {
|
1997-07-29 05:24:36 +00:00
|
|
|
cnt = (low2 + (high2 << 8) + 1) & 0xffff;
|
1997-07-24 05:27:40 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (chan >= 4) /* high channels move words */
|
|
|
|
|
cnt *= 2;
|
|
|
|
|
return(cnt);
|
|
|
|
|
}
|
|
|
|
|
|
1997-08-28 03:36:40 +00:00
|
|
|
/*
|
|
|
|
|
* Stop a DMA transfer currently in progress.
|
|
|
|
|
*/
|
|
|
|
|
int
|
|
|
|
|
isa_dmastop(int chan)
|
|
|
|
|
{
|
|
|
|
|
if ((dma_inuse & (1 << chan)) == 0)
|
|
|
|
|
printf("isa_dmastop: channel %d not acquired\n", chan);
|
|
|
|
|
|
|
|
|
|
if (((dma_busy & (1 << chan)) == 0) &&
|
|
|
|
|
((dma_auto_mode & (1 << chan)) == 0)) {
|
|
|
|
|
printf("chan %d not busy\n", chan);
|
|
|
|
|
return -2 ;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if ((chan & 4) == 0) {
|
|
|
|
|
outb(DMA1_SMSK, (chan & 3) | 4 /* disable mask */);
|
|
|
|
|
} else {
|
|
|
|
|
outb(DMA2_SMSK, (chan & 3) | 4 /* disable mask */);
|
|
|
|
|
}
|
|
|
|
|
return(isa_dmastatus(chan));
|
|
|
|
|
}
|
|
|
|
|
|
1995-09-10 21:36:12 +00:00
|
|
|
/*
|
|
|
|
|
* Find the highest priority enabled display device. Since we can't
|
|
|
|
|
* distinguish display devices from ttys, depend on display devices
|
1995-10-22 15:07:43 +00:00
|
|
|
* being sensitive and before sensitive non-display devices (if any)
|
|
|
|
|
* in isa_devtab_tty.
|
|
|
|
|
*
|
|
|
|
|
* XXX we should add capability flags IAMDISPLAY and ISUPPORTCONSOLES.
|
1995-09-10 21:36:12 +00:00
|
|
|
*/
|
|
|
|
|
struct isa_device *
|
|
|
|
|
find_display()
|
|
|
|
|
{
|
|
|
|
|
struct isa_device *dvp;
|
|
|
|
|
|
|
|
|
|
for (dvp = isa_devtab_tty; dvp->id_driver != NULL; dvp++)
|
1995-10-22 15:07:43 +00:00
|
|
|
if (dvp->id_driver->sensitive_hw && dvp->id_enabled)
|
1995-09-10 21:36:12 +00:00
|
|
|
return (dvp);
|
|
|
|
|
return (NULL);
|
|
|
|
|
}
|
|
|
|
|
|
1993-06-12 14:58:17 +00:00
|
|
|
/*
|
|
|
|
|
* find an ISA device in a given isa_devtab_* table, given
|
|
|
|
|
* the table to search, the expected id_driver entry, and the unit number.
|
|
|
|
|
*
|
|
|
|
|
* this function is defined in isa_device.h, and this location is debatable;
|
|
|
|
|
* i put it there because it's useless w/o, and directly operates on
|
|
|
|
|
* the other stuff in that file.
|
|
|
|
|
*
|
|
|
|
|
*/
|
|
|
|
|
|
1997-11-30 09:44:28 +00:00
|
|
|
struct isa_device *
|
|
|
|
|
find_isadev(table, driverp, unit)
|
1997-05-26 14:42:24 +00:00
|
|
|
struct isa_device *table;
|
|
|
|
|
struct isa_driver *driverp;
|
|
|
|
|
int unit;
|
1993-06-12 14:58:17 +00:00
|
|
|
{
|
1997-05-26 14:42:24 +00:00
|
|
|
if (driverp == NULL) /* sanity check */
|
|
|
|
|
return (NULL);
|
1993-06-12 14:58:17 +00:00
|
|
|
|
1997-05-26 14:42:24 +00:00
|
|
|
while ((table->id_driver != driverp) || (table->id_unit != unit)) {
|
|
|
|
|
if (table->id_driver == 0)
|
|
|
|
|
return NULL;
|
1995-05-30 08:16:23 +00:00
|
|
|
|
1997-05-26 14:42:24 +00:00
|
|
|
table++;
|
|
|
|
|
}
|
1993-06-12 14:58:17 +00:00
|
|
|
|
1997-05-26 14:42:24 +00:00
|
|
|
return (table);
|
1993-06-12 14:58:17 +00:00
|
|
|
}
|
