17246 Commits

Author SHA1 Message Date
joerg
16296391cd Merge from the 2.2 branch: describe MNT_FORCE for mount(2). 1996-11-10 09:28:20 +00:00
bde
d6f309e9f6 Initialize interrupt counters. The boot time values were displayed as ***.
Should be in 2.2.

Pointed out by:	/etc/malloc.conf -> AJ
1996-11-10 08:30:18 +00:00
fenner
3227402145 Re-enable the TCP SYN-attack protection code. I was the one who didn't
understand the socket state flag.

2.2 candidate.
1996-11-10 07:37:24 +00:00
hsu
5b38eecfbf Remove unneeded #include <sys/proc.h>. 1996-11-10 06:13:12 +00:00
obrien
7d9f6cd5a3 Added to MNT_FORCE option description after seeing Bruce's commit message
dealing w/the fixit floppy.
Also added the MNT_RELOAD, MNT_WANTRDWR, MNT_ASYNC, MNT_NOATIME,
MOUNT_UNION flags.  Someone might want to check my description of MNT_RELOAD.

2.2-R candidate.  Not a 2.1.6-R candidate -- some current flags aren't in
2.1.5-R's version.
1996-11-10 05:56:43 +00:00
wosch
e15ba7df33 Add sendmail option '-oi'
-oi	Do not take dots on a line by themselves as a
		message terminator.

Now this crontab entry works:
* * * *	*	echo foo; echo .; echo blah
1996-11-10 04:46:58 +00:00
ache
29308755dd pcfs -> msdos
2.2 candidate
1996-11-10 02:23:33 +00:00
ache
3cd081b175 Add mount_msdos(8) reference
2.2 candidate
1996-11-10 02:21:05 +00:00
jkh
050dc7a285 Disgusting kludge to make depend rule work again in the fact of
Joerg's keymap.[ch] changes.  If there's a better way, I'd like to know.
1996-11-09 20:50:23 +00:00
jdp
0118326e5f Document that "ldconfig -m" rescans all directories previously
entered into the hints file, in addition to the directories named
on the command line.

2.2 Candidate.
1996-11-09 20:26:19 +00:00
jkh
8cbeefe2ba Whups, we need the VGA16 server by default too. 1996-11-09 19:47:24 +00:00
jkh
1efb28c052 1. Saner ldconfig path.
2. Choose DIST_XF86_CFG in defaults; XF86Setup isn't happy otherwise.
1996-11-09 19:26:01 +00:00
jkh
d21e76e0ba 1. Take out console setup in options now that Joerg has moved it into
the main menu.
2. Conditionalized a few small things which needed it.
3. Put PC98 X servers in their own menu, there are so many of them now.
4. Rampaged on the menus.c file in general, reformatting and cleaning up.
1996-11-09 18:12:17 +00:00
joerg
d30fd798ce Merge from the 2.2 branch: avoid including the API stuff when building
vi for the fixit floppy.  Perhaps we'll find a more elegant solution
some day, by now it has proven to work reasonably.
1996-11-09 16:54:33 +00:00
joerg
88d8c0377c Sysinstall now offers the installation of a keymap early in the game!
Not all mappings are supported, most languages come only with one
encoding since this should be sufficient to get up & running in using
sysinstall, and we are already pretty tight on space.  (My previous
commit has already bumped the boot MFS size by another 50 KB for
this.)

This feature requires the `kbdcontrol -L' i've just committed.  Plain
text keymaps and the entire scanner are overkill for sysinstall.

Also updated the list of available keymaps while i was at it.

Reviewed by:	jkh
1996-11-09 16:47:08 +00:00
joerg
ce411f960d Several improvements to the `fixit' configuration:
. Don't gzip the crunched binary by now; it just fits, and execution is
  a lot faster this way (it's truly demand-paged again).

. Add more(1), ft(8), protocols(5), a stripped down services(5).

. Improve the .profile, and make sysinstall actually use it again.

Still no go for a 4 MB configuration though. :-(
1996-11-09 16:25:47 +00:00
joerg
5e95f3ceb1 Implement a -L option that dumps the compiled keymap (as C code) to
stdout.  The next commit to sysinstall will use this...
1996-11-09 16:16:19 +00:00
joerg
3f3155c6db Merge from the 2.2 branch: call our `vi' by its right name.
Detected by:	jkh
1996-11-09 12:08:57 +00:00
joerg
99f986b1f6 If mounting the fixit floppy fails in the first place, warn the user
but make a second attempt using MNT_FORCE, just in case it has been
unclean from a previous crash.  That's dangerous, but far better than
keeping the despaired user standing in the rain...

(Experienced admins can still fsck it then, and remount.  Others will
either totally crash, or incidentally succeed, without much further
help possible...)

Btw., mount(2) misses the description of MNT_FORCE for the mount
syscall.
1996-11-09 11:57:40 +00:00
bde
45a0793694 Fixed lookup of ".." in checkpath. It always failed, so renames of
directories to a different parent directory always failed.  This bug
was caused by 4.4Lite2 changing the directory format and ext2fs not
keeping up.

Should be in 2.2.
1996-11-09 10:25:04 +00:00
jkh
e0d772d463 Purely cosmetic cleanup. 1996-11-09 09:24:30 +00:00
jkh
b80f28ff3d Add an entry for minimum - that's a lot easier. 1996-11-09 09:23:09 +00:00
dyson
6f0ff19196 Fix an ordering bug -- pmap_remove_pages should be called BEFORE
vm_map_remove, not after...

2.2-RELEASE candidate.
1996-11-09 03:54:25 +00:00
joerg
a88e311da4 Fix a syntax error in the Belgian keymap. It caused a wrong mapping
and a superfluous `i' on stdout (lex' default rule...).
1996-11-09 03:03:47 +00:00
max
dba4c1691b Back out my previous change.
It probably broke the support for the device.

Pointed-Out by:	joerg
1996-11-09 02:44:11 +00:00
jfieber
2eb6d57072 A skeletal, but functional Docbook to HTML translation spec.
The Docbook DTD will follow once I investigate how "contrib" works and
if it should be imported there.
1996-11-09 02:04:05 +00:00
asami
89988533ea Re-sync with -current. Should be in 2.2.
Submitted by:	The FreeBSD(98) Development Team
1996-11-09 00:39:59 +00:00
asami
bf6d735fd0 (1) Update
(2) Don't depend on BOOTSEG

(3) Change BOOTSEG from 0x9000 to 0x1000

Should be in 2.2.

Submitted by:	FreeBSD(98) Development Team
1996-11-09 00:18:25 +00:00
obrien
88aa304193 Added Oly (oly@world.std.com) for his port of lout and xripple 1996-11-08 23:57:04 +00:00
se
b006cd5501 Only access the specific interrupt status registers if required.
This follows more closely the suggestions in the latest NCR docs, and has
been running on my system for weeks with no problem. It does improve the
quality of diagnostic messages and does allow to better understand the
sequence of events in case of an error.

This should go into 2.2 and 2.1.6.
1996-11-08 23:46:04 +00:00
guido
cc0b422ef3 Add add Naoki Hamada <nao@sbl.cl.nec.co.jp> to the contributors, because of his
work on the vx driver.
1996-11-08 21:56:21 +00:00
jkh
3fca91d40c Remove a mkdir in doTARBALL which, as far as I can tell, is not needed
for anything.
1996-11-08 19:09:45 +00:00
bde
7cae5544c3 Fixed spacefree calculation in ext2_direnter(). This bug sometimes caused
panics.

This should be in 2.2, of course.

Submitted by:	davidg
Obtained from:	bouyer@antioche.ibp.fr (Manuel BOUYER) (fix for NetBSD)
1996-11-08 19:06:34 +00:00
bde
5580ec0a03 Removed gratuitous differences between ext2_readwrite.c and ufs_readwrite.c.
This fixes several bugs and one missing feature:
- cluster_read() was needlessly used for reading files of size exactly 1
  block.
- EFAULT errors for read didn't terminate the loop.  This was probably
  harmless.
- IO_VMIO handling was missing near line 275.  I don't know what this does.
- B_CLUSTEROK was only set if (doclusterwrite) nead line 293.  This was
  harmless, if only because another bug prevents doclusterwrite from being
  0.
- MNT_NOATIME wasn't implemented.

This should be in 2.2, of course.

Reviewed by:	davidg
1996-11-08 18:50:09 +00:00
jkh
6c6d893c19 Merge from 2_1_0 1996-11-08 13:27:00 +00:00
jkh
b37d914c07 Merge with 2_1_0 1996-11-08 13:09:27 +00:00
tg
d74c3626c5 Add Mats Lofkvist for his elk port update (ages ago). 1996-11-08 08:10:57 +00:00
jkh
f8952e421c 1. Eliminate the sendmail.cw bogon again. Peter fixed it correctly.
2. Preload the ldconfig cache before calling X setup.
1996-11-08 05:38:27 +00:00
peter
e0450ac1fb make the /etc/sendmail.cw file optional by default.
Definately should go into 2.2..

Requested by: lots of people
1996-11-08 05:22:50 +00:00
asami
e068ad8a7c Remove option I586_FAST_BCOPY. The code will be included by default
if I586_CPU is defined.  Note there is a runtime check so the code
won't be run for non-Pentium CPUs anyway.

2.2 candidate, this code has been tested for almost half year in -current.
1996-11-08 02:38:44 +00:00
jdp
81bba450d5 Fixed a bug in the handling of the directories in the search path
that is stored in the hints file.  If that search path contained
a non-existent directory (one, say, that had been removed), and
"ldconfig -m /a/perfectly/good/directory" was run, ldconfig returned
an error status without printing an error message.  This caused
some confusing bombs when installing ports, in particular.

I changed it so that non-existent directories from the stored search
path are silently ignored.  Only non-existent directories named
explicitly on the command line are treated as errors.  Also, a
diagnostic is printed if and only if an error status is returned.

In an unrelated fix, ldconfig now silently ignores any directories
named on the command line when the "-r" option is given.  Formerly,
these directories incorrectly made their way into the "search
directories" line of the listing.  It really should be an error to
specify directories together with "-r", but I don't have time to
fix the manual page in that way right now.

2.2 Candidate.
1996-11-08 02:12:40 +00:00
wpaul
844872c49d - Avoid possible SEGVs: never call clnt_destroy() without checking
for NULL RPC client handles. This should hopefully fix the problems
  Satoshi reported on -current.

- Add socket descriptor sanity checks to _yp_unbind().

- Fix yp_order() so that it handles the RPC_PROCUNAVAIL error gracefully.
  NIS+ in YP compat mode doesn't support the YPPROC_ORDER procedure.

This is a 2.2 candidate with bells on.
1996-11-08 01:42:02 +00:00
jkh
80280d6037 More cosmetic tweaks, get the prefixes adjusted properly to XFree86 3.2's
new packaging scheme.
1996-11-07 18:30:59 +00:00
max
7dddc49bcd Typo, MATSHITA -> MATSUSHITA.
2.2 & 2.1.6 candidate.
1996-11-07 18:13:57 +00:00
jkh
e570bb4c0c Make sure current_chunk is 0 each time we enter the editor. 1996-11-07 16:40:10 +00:00
jkh
a216a996d5 This warning is too annoying now. Kill it. 1996-11-07 16:34:12 +00:00
jkh
5e17097f62 Install an evil work-around for /etc/sendmail.cw
[same as from 2.2]
1996-11-07 15:45:59 +00:00
jkh
cc06811526 Complete switch-over to XF86 3.2. 1996-11-07 15:35:48 +00:00
max
7ff854abda Since we use &dollar; in our sample Makefiles, we can just say
&dollar;Id&dollar;.

Suggested by:	asami
1996-11-07 15:07:00 +00:00
joerg
7ab469589f Merge my vi hack^H^H^H^Hframework into HEAD, so the various revisions
should be in sync again.  It's an env variable only, hence it's
harmless and ignored without the actual changes in vi's Makefile.

Suggested by:	jkh
1996-11-07 14:50:28 +00:00