Commit Graph

10987 Commits

Author SHA1 Message Date
Bruce Evans
67a8a10b9c Updated to BSD4.4lite2. Fixes PR836. `echo abcd | tr a-d A-BC-D' now
works.
1995-11-28 13:18:47 +00:00
Julian Elischer
b2b028bab9 oops forgot one.. 1995-11-28 09:44:42 +00:00
Julian Elischer
376c314995 the second set of changes in a move towards getting devices to be
totally dynamic. (the first was about 7 weeeks ago)

this is only the devices in i386/isa
I'll do more tomorrow.
they're completely masked by #ifdef JREMOD at this stage...
the eventual aim is that every driver will do a SYSINIT
at startup BEFORE the probes, which will effectively
link it into the devsw tables etc.

If I'd thought about it more I'd have put that in in this set (damn)
The ioconf lines generated by config will also end up in the
device's own scope as well, so ioconf.c will eventually be gutted
the SYSINIT call to the driver will include a phase where the
driver links it's ioconf line into a chain of such. when this phase is done
then the user can modify them with the boot: -c
config menu if he wants, just like now..
config will put the config lines out in the .h file
(e.g. in aha.h will be the addresses for the aha driver to look.)
as I said this is a very small first step..
the aim of THIS set of edits is to not have to edit conf.c at all when
adding a new device.. the tabe will be a simple skeleton..

when this is done, it will allow other changes to be made,
all teh time still having a fully working kernel tree,
but the logical outcome is the complete REMOVAL of the devsw tables.

By the end of this, linked in drivers will be exactly the same as
run-time loaded drivers, except they JUST HAPPEN to already be linked
and present at startup..
the SYSINIT calls will be the equivalent of the "init" call
made to a newly loaded driver in every respect.

For this edit,
each of the files has the following code inserted into it:

obviously, tailored to suit..
----------------------somewhere at the top:
#ifdef JREMOD
#include <sys/conf.h>
#define CDEV_MAJOR 13
#define BDEV_MAJOR 4
static void 	sd_devsw_install();
#endif /*JREMOD */
---------------------somewhere that's run during bootup: EVENTUALLY a SYSINIT
#ifdef JREMOD
        sd_devsw_install();
#endif /*JREMOD*/
-----------------------at the bottom:
#ifdef JREMOD
struct bdevsw sd_bdevsw =
	{ sdopen,	sdclose,	sdstrategy,	sdioctl,	/*4*/
	  sddump,	sdsize,		0 };

struct cdevsw sd_cdevsw =
	{ sdopen,	sdclose,	rawread,	rawwrite,	/*13*/
	  sdioctl,	nostop,		nullreset,	nodevtotty,/* sd */
	  seltrue,	nommap,		sdstrategy };

static sd_devsw_installed = 0;

static void 	sd_devsw_install()
{
	dev_t descript;
	if( ! sd_devsw_installed ) {
		descript = makedev(CDEV_MAJOR,0);
		cdevsw_add(&descript,&sd_cdevsw,NULL);
#if defined(BDEV_MAJOR)
		descript = makedev(BDEV_MAJOR,0);
		bdevsw_add(&descript,&sd_bdevsw,NULL);
#endif /*BDEV_MAJOR*/
		sd_devsw_installed = 1;
	}
}
#endif /* JREMOD */
1995-11-28 09:43:45 +00:00
Julian Elischer
7146c13e43 the second set of changes in a move towards getting devices to be
totally dynamic.

this is only the devices in i386/isa
I'll do more tomorrow.
they're completely masked by #ifdef JREMOD at this stage...
the eventual aim is that every driver will do a SYSINIT
at startup BEFORE the probes, which will effectively
link it into the devsw tables etc.

If I'd thought about it more I'd have put that in in this set (damn)
The ioconf lines generated by config will also end up in the
device's own scope as well, so ioconf.c will eventually be gutted
the SYSINIT call to the driver will include a phase where the
driver links it's ioconf line into a chain of such. when this phase is done
then the user can modify them with the boot: -c
config menu if he wants, just like now..
config will put the config lines out in the .h file
(e.g. in aha.h will be the addresses for the aha driver to look.)
as I said this is a very small first step..
the aim of THIS set of edits is to not have to edit conf.c at all when
adding a new device.. the tabe will be a simple skeleton..

when this is done, it will allow other changes to be made,
all teh time still having a fully working kernel tree,
but the logical outcome is the complete REMOVAL of the devsw tables.

By the end of this, linked in drivers will be exactly the same as
run-time loaded drivers, except they JUST HAPPEN to already be linked
and present at startup..
the SYSINIT calls will be the equivalent of the "init" call
made to a newly loaded driver in every respect.

For this edit,
each of the files has the following code inserted into it:

obviously, tailored to suit..
----------------------somewhere at the top:
#ifdef JREMOD
#include <sys/conf.h>
#define CDEV_MAJOR 13
#define BDEV_MAJOR 4
static void 	sd_devsw_install();
#endif /*JREMOD */
---------------------somewhere that's run during bootup: EVENTUALLY a SYSINIT
#ifdef JREMOD
        sd_devsw_install();
#endif /*JREMOD*/
-----------------------at the bottom:
#ifdef JREMOD
struct bdevsw sd_bdevsw =
	{ sdopen,	sdclose,	sdstrategy,	sdioctl,	/*4*/
	  sddump,	sdsize,		0 };

struct cdevsw sd_cdevsw =
	{ sdopen,	sdclose,	rawread,	rawwrite,	/*13*/
	  sdioctl,	nostop,		nullreset,	nodevtotty,/* sd */
	  seltrue,	nommap,		sdstrategy };

static sd_devsw_installed = 0;

static void 	sd_devsw_install()
{
	dev_t descript;
	if( ! sd_devsw_installed ) {
		descript = makedev(CDEV_MAJOR,0);
		cdevsw_add(&descript,&sd_cdevsw,NULL);
#if defined(BDEV_MAJOR)
		descript = makedev(BDEV_MAJOR,0);
		bdevsw_add(&descript,&sd_bdevsw,NULL);
#endif /*BDEV_MAJOR*/
		sd_devsw_installed = 1;
	}
}
#endif /* JREMOD */
1995-11-28 09:42:06 +00:00
Bruce Evans
43cbfcb357 Removed all #includes of the unused file <sys/device.h>. 1995-11-28 07:29:59 +00:00
Bruce Evans
a349554809 Removed bogus __BEGIN_DECS/__END_DECLS. 1995-11-28 07:23:09 +00:00
Peter Wemm
018309ece7 After having put on my Asbestos suit, complete the MFS_ROOT part of Terry's
mountroot changes.  This means that the mfs_initminiroot functionality
into the root mfs_mount....
1995-11-28 03:15:58 +00:00
Peter Wemm
6884d2aae7 Implement read/write to kernel space - I use this for a self-loading/
self-decompressing ram disk that I'm fiddling with..

(Note, this depends on the various syscalls having correctly set uio_segflag
 before calling physio - I've checked and they look correct.)
1995-11-28 02:40:38 +00:00
Peter Wemm
59a4c16a68 Attempt to solve the busy-buffers-on-shutdown caused by MFS once and for all.
What was happening, was that the main mfs loop was sleeping, and when it was
being awoken by a wakeup when it was supposed to process some IO requests.

The problem was that if it was being woken out of the tsleep() by a signal
at shutdown, it was going straight into dounmount() without servicing any
pending IO requests, causing dounmount() to fail because there were busy
buffers (and they could not be "processed" because the processing loop was
trying to unmount rather than dispatching into mfs_doio()).

This (dare I say it :-) appears to be a layering problem....
1995-11-28 02:15:29 +00:00
Peter Wemm
070fc2ce12 Mainly cosmetic cleanups... It now uses more consistant message reporting
on the console, and no longer uses "SLXOS" which I suspect may be a
trademark... (I'm not sure, but this is not really a SLXOS driver anyway)
1995-11-28 02:07:36 +00:00
Peter Wemm
968757cbee Implement some rudimentry IPX support... 1995-11-28 01:59:19 +00:00
Andrey A. Chernov
00265e7d4d Separate colors & attributes as Terry points
Reviewed by: soren
1995-11-28 00:17:32 +00:00
Peter Wemm
761b45035b port 22 is the officially assigned "ssh" port... 1995-11-27 09:11:03 +00:00
John Fieber
9b1ae6a7f3 Typo. 1995-11-27 01:46:51 +00:00
Poul-Henning Kamp
dc22f7a98e Make tip recognize EOF in more cases. 1995-11-26 21:08:36 +00:00
Bruce Evans
84478e7f42 Fixed setting of speed B0 - don't output a bogus divisor of 0 and a
random prescaler, just hang up.  This may fix hangup problems with
mgetty.
1995-11-26 17:13:23 +00:00
Joerg Wunsch
8807593b7e Make the default tape device match dump(8). 1995-11-26 16:57:37 +00:00
Satoshi Asami
a33f5158b1 Add Donald Burr <d_burr@ix.netcom.com> for his contribution of the
workman port.
1995-11-26 13:44:48 +00:00
Satoshi Asami
a72b1b79b2 Make the {FETCH,BUILD,RUN}_DEPEND targets work with non-executable files
too.  Basically, if the name starts with a "/", it's tested with "test -e";
otherwise, it's tested with "witch -s".

Reviewed by:	the ports list (well at least nobody complained)
1995-11-26 12:35:49 +00:00
Bruce Evans
89e36521ab Fixed beforeinstall rule. .CURDIR was spelled .SRCDIR.
Changed beforeinstall rule to use `install -C' instead of `cmp -s'
and `install -c'.  `install -C' has exactly the right semantics
for installing headers and should be used elsewhere.
1995-11-26 09:51:42 +00:00
Jordan K. Hubbard
c2e6431816 Bring forward libkadm change from 2.1 1995-11-26 07:31:13 +00:00
Jordan K. Hubbard
3b34e61a0a My reorganization of chapter 9.
Reviewed by:	jfieber
1995-11-25 20:00:49 +00:00
Peter Wemm
4620010dfc Add -DNOCLEAN option which totally and utterly disables any form of
cleaning during a make.  This may give you more rope to hang yourself if
you are caught with some subtle dependency on installed binaries in your
build, but if you are doing daily 'make -DNOCLEAN world' it's not too bad
at all.  It could take as little as 30 minutes to do an entire sync-up of
your binaries if everything's up to date, especially if you are using
'INSTALL=install -C' in /etc/make.conf (highly recomended!).

Also, add a "reinstall" target. You can do a 'make DESTDIR=/mnt reinstall'
where /mnt is the nfs root of a machine and you get the install parts of
the make world run on it.

I saw this on -hackers quite some time ago and included it in my Makefile
and have been using it on and off for a while. Alas, I cannot find the
actual message with the author's name...
1995-11-25 01:28:07 +00:00
Peter Wemm
3c64c6ee8d Part two of a repository operation to sort out the libmp/libgmp builds.
After this commit, you should be able to build libmp and libgmp independently
and without being forced to do a make depend first..
1995-11-25 00:14:03 +00:00
Peter Wemm
30a2d5501c Connect fsdb to /sbin makefile... 1995-11-24 23:17:58 +00:00
Peter Wemm
e10719ee6a Adjust relative paths in Makefile to ../sbin/fsck and ../../sys/ufs/ffs 1995-11-24 23:12:00 +00:00
Peter Wemm
a92a00527e Disconnect fsdb in preperation for move to /sbin 1995-11-24 23:00:02 +00:00
Bruce Evans
9c68e470a2 Oops, the previous change was missing the declaration of `struct
buf_queue_head'.  It isn't forward declared in <sys/types.h> like
`struct buf'.
1995-11-24 15:59:11 +00:00
Bruce Evans
fec104de77 Completed function declarations and/or added prototypes. 1995-11-24 15:15:30 +00:00
Bruce Evans
b8a89c9ada Added bogusly placed extern prototypes for functions that should probably
be static.
1995-11-24 15:08:07 +00:00
Bruce Evans
d818a9cc1f Fixed a comment. 1995-11-24 14:56:00 +00:00
Bruce Evans
88c2c2b3d2 Declared tqdisksort(). <sys/disklabel.h> is the wrong place, but
<sys.disk.h> isn't used, so the declaration there isn't seen.
1995-11-24 14:50:39 +00:00
Bruce Evans
058284fceb Completed function declarations and/or added prototypes and/or #includes
to get the prototypes.

Changed some `int's to `boolean_t's.  boolean_t's are ints so they are
hard to distinguish from ints.

Converted function headers to old-style.  ddb is written in K&R1 C
except where we broke it.
1995-11-24 14:13:42 +00:00
Bruce Evans
4753168fae Completed function declarations and/or added prototypes.
Removed `extern' from prototypes.
Sorted prototypes.
Uniformized idempotency ifdefs.
1995-11-24 13:53:05 +00:00
Bruce Evans
7f0e0625f6 Staticized. Moved some ero-initialized values to the bss.
Added prototypes.
1995-11-24 13:27:24 +00:00
Bruce Evans
68857518ce Cleaned up prototypes:
- don't #include other headers just to get struct names.
- don't use __BEGIN_DECLS/__END_DECLS for system prototypes.  It is for
  user prototypes.
- don't use extern.
- don't use lines longer than 80 columns.
- use alphabetical order.
- use tabs.

Uniformized idempotency ifdefs.
1995-11-24 12:25:13 +00:00
Bruce Evans
aa0bd366d5 Fixed a bogus name (ifn_en) that was introduced when a type mismatch
was fixed.
1995-11-24 12:07:33 +00:00
Bruce Evans
530e1829ef Added #include <sys/queue.h>. This will be required when I move
the (inline) implementations of insque() and remque() from
<machine/cpufunc.h> to <sys/queue.h>.
1995-11-24 12:01:08 +00:00
Bruce Evans
e316debe42 Undid bogus cleanups. 0 was mistyped as NULL. 1995-11-24 11:43:55 +00:00
Joerg Wunsch
68b8d35346 Add a short man page for cfree(3), in order to hint people about its life in libcompat. 1995-11-24 08:36:51 +00:00
Jordan K. Hubbard
f2e2018809 Various tweaks to things I noticed while reading these docs. 1995-11-24 07:55:27 +00:00
Andrey A. Chernov
07e8877db1 Merge with current. Back out all trailing spaces changes, they
hits again and again.
1995-11-23 20:19:56 +00:00
Andrey A. Chernov
ec1450fb63 This commit was generated by cvs2svn to compensate for changes in r12461,
which included commits to RCS files with non-trunk default branches.
1995-11-23 19:28:56 +00:00
Andrey A. Chernov
39a9eb721a Import of 1.9.5, one my fix included to mainstream 1995-11-23 19:28:56 +00:00
John Dyson
dc4a0cee58 Update the wd.c driver to use the new TAILQ scheme for device
buffer queue.  Also, create a new subroutine 'tqdisksort' that
is an improved version of the original disksort that also uses
TAILQs.
1995-11-23 07:24:41 +00:00
Stefan Eßer
09a67ffa0e Increase width of Network column from 11 to 13 for the AF_INET case.
This seems to have been missed, when the recent IPX changes went in ...
1995-11-22 22:21:04 +00:00
Bruce Evans
1f3dad5a8d Completed function declarations and added prototypes.
Removed some unnecessary #includes.

Fixed warnings about nested externs.
1995-11-22 07:43:53 +00:00
Bruce Evans
55054f3540 Completed function declarations, added prototypes and removed redundant
declarations.
1995-11-21 15:51:39 +00:00
Bruce Evans
f30e535f00 Completed function declarations, added prototypes and removed redundant
declarations.
1995-11-21 15:14:28 +00:00
Bruce Evans
dc288b6fc5 Completed function declarations, added prototypes and removed redundant
declarations.
1995-11-21 14:56:02 +00:00