Garrett Wollman [Fri, 14 Mar 1997 16:51:28 +0000 (16:51 +0000)]
Add protoswitch entries for shortcut send/receive. Correct
a few misleading comments, and move allthe struct tag
forward declarations to be in one place.
Peter Wemm [Fri, 14 Mar 1997 14:45:30 +0000 (14:45 +0000)]
Tone down the paranoia a bit in from the previous commit. setusercontext()
automatically Does The Right Thing when lc == NULL, which just happens to
be what the extra code in cron was trying to do. Simplify.
Peter Wemm [Fri, 14 Mar 1997 13:48:04 +0000 (13:48 +0000)]
For some reason, the old login class code didn't seem to be working here.
I suspect it was because the child exec code's parent was doing the
initial lookups, then forking, then doing other things (possibly trashing
the static data in the getpw*() buffer), then attempting to dereference
*pwd and *lc. Also, no error checking appeared to be done - I've allowed
it to fall back to the old "become user" code on critical failure rather than
risk running a user's cron jobs as root.
Mike Pritchard [Fri, 14 Mar 1997 05:10:47 +0000 (05:10 +0000)]
Remove the text that says that freebsd-bugs can't be subscribed
to by anyone.
While I was at it, I changed all of the mailing list addresses to be
consistent with how they are referenced in the handbook. E.g
freebsd-bugs@FreeBSD.ORG instead of bugs@FreeBSD.ORG.
Joerg Wunsch [Fri, 14 Mar 1997 01:27:02 +0000 (01:27 +0000)]
Save and restore the state of the variable Initialized inside die().
Otherwise, the shutdown message will print out with 15 NUL bytes in
front, due to a missing timestamp.
Bruce Evans [Thu, 13 Mar 1997 19:23:27 +0000 (19:23 +0000)]
Updated comment about npx0.
Added obsolete option GATEWAY so that kern_opt.c gets tested.
Added undocumented options LOCKF_DEBUG and SIMPLELOCK_DEBUG so
that these options get tested. The addition of LOCKF_DEBUG shows
that all of kern/kern_lockf.c shouldn't have been moved from ufs.
The debugging parts are very fs-dependent.
Joerg Wunsch [Thu, 13 Mar 1997 18:04:05 +0000 (18:04 +0000)]
Various stylistic improvements regarding num_eisa_slots & co.:
. properly declare the variable in in a .h file, as opposed to
using a private extern declaration in userconfig.c;
. move the definition of EISA_SLOTS and therefore the inclusion of
opt_eisa.h into eisaconf.c.
Peter Wemm [Thu, 13 Mar 1997 15:37:23 +0000 (15:37 +0000)]
Kill the Lite2 early "filesystem clean abort" check and go back to
something closer to how we used to do it. The Lite2 way is to check the
"fsclean" flag in the superblock and stop there if so (during preen).
We now do the various superblock sanity checks that we used to do before
since it's cheap. We now get the filesystem state summary again instead
of "FILESYSTEM CLEAN; CHECKING SKIPPED" (or whatever).
Accept SIGHUP as a "re-open logfile" signal. As ppp
doesn't set it's serial line to it's controlling terminal,
we can use HUP :)
This is a candidate for 2.2. The log.[ch] changes won't
conflict, but the main.c changes will. We just want to change the
kill(...,SIGHUP) to a SIGTERM and change the signal(SIGHUP,Hangup)
to a pending_signal(SIGHUP,LogReOpen).
Brian Somers [Thu, 13 Mar 1997 12:45:35 +0000 (12:45 +0000)]
Reviewed by: ache@freebsd.org
These changes should fix the signal "problems" in ppp.
The signal changes should really be put into 2.2 too !
The following patches should do it. There were some other
changes made by Andrey recently that havn't been brought
into 2.2, it may be worth doing them now.
Peter Wemm [Thu, 13 Mar 1997 06:58:46 +0000 (06:58 +0000)]
Part 2 of a failed commit (cvs broke). Original message:
Back out a dubious Lite2 change to "optimise" getcwd() to look at $PWD
because it's potentially dangerous (think: symlink races). Move
realpath() back to it's original location, and remove getcwd_physical()
by renaming it back to getcwd() and zapping the original getcwd wrapper.
Noticed by: bde
The following commits already happened but the log message got lost:
Peter Wemm [Thu, 13 Mar 1997 06:45:38 +0000 (06:45 +0000)]
Back out a dubious Lite2 change to "optimise" getcwd() to look at $PWD
because it's potentially dangerous (think: symlink races). Move
realpath() back to it's original location, and remove getcwd_physical()
by renaming it back to getcwd() and zapping the original getcwd wrapper.
Joerg Wunsch [Wed, 12 Mar 1997 17:41:35 +0000 (17:41 +0000)]
Since i don't see that anybody is implementing a more correct EISA
probing anytime soon, make EISA_SLOTS a fully supported option. It's
required for the HP NetServer LC series machines.
Bruce Evans [Wed, 12 Mar 1997 17:16:22 +0000 (17:16 +0000)]
Make this man page really match the current <sys/signal.h> again and
the 2.x <sys/signal.h> for the first time :-). The type of ss_sp
was lexically different.
Mike Pritchard [Wed, 12 Mar 1997 16:23:36 +0000 (16:23 +0000)]
Make this man page match the current <sys/signal.h> again :-).
Bruce says that since NetBSD, OpenBSD and Linux currently
use ss_sp, we won't be changing it to the lite2 ss_base.
The type may change at some later date.
Bruce Evans [Wed, 12 Mar 1997 14:54:22 +0000 (14:54 +0000)]
Fixed merging error. Lite2 fixed premature failure and didn't
touch duplicate group suppression, but the merge blew away our
duplicate group suppression.
The merge also blew away the -Wall cleanup in rev.1.5, but that
was misformatted, so I didn't restore it.
Mike Pritchard [Wed, 12 Mar 1997 14:49:41 +0000 (14:49 +0000)]
Cleanup some of the Lite2 merge. Most of it is mdoc cleanup,
but in one case the Lite2 changes were flat out wrong and
caused the man page to disagree with a header file.
There are still some *roff macro calls that were added that
I have yet to figure out what to do with in some of the man pages.
Mike Pritchard [Wed, 12 Mar 1997 13:46:16 +0000 (13:46 +0000)]
Improve the wording of the section on comments in the group and passwd
files. Also mention that this feature is not enabled by default
and that they must be enabled by compile-time options.
Fixed another typo or two I noticed while I was at it.
Bruce Evans [Wed, 12 Mar 1997 12:35:44 +0000 (12:35 +0000)]
Fixed errors in the Lite2 merge. Some style changes were mismerged.
My changes to preserve errno across free() and close() and to report
fstat() errors properly were blown away.
Updated the FreeBSD changes to match the Lite2 style fixes.
Peter Wemm [Wed, 12 Mar 1997 11:10:54 +0000 (11:10 +0000)]
Fix problem with FD_SET* overflow reporting.. Perror() didn't have enough
args, and errno hasn't actually been set so it probably doesn't make sense
to report it via strerror().
Peter Wemm [Wed, 12 Mar 1997 08:29:44 +0000 (08:29 +0000)]
Make amd actually work instead of just compiling. The new Lite2 code
was not setting the version number in the nfs_args data to mount(2), so it
was returning EFAULT. Perhaps the nfs_args version number was something
we added at some point?
Peter Wemm [Wed, 12 Mar 1997 07:14:44 +0000 (07:14 +0000)]
makeinfo can fail but still leave behind zero length files which cause
interesting problems because the resulting file is newer than the source
and this stops 'make' from rebuilding it. Go via an intermediate file
and rename to make sure this doesn't happen.
Peter Wemm [Wed, 12 Mar 1997 06:24:52 +0000 (06:24 +0000)]
Move the conditional declarations of ICOMPRESS_CMD and ICOMPRESS_EXT
to before they are used in .for loops and .SUFFIXES lists, or it
doesn't work! Without this, 'make all' doesn't gzip the result as the
generated 'all' target doesn't seem to be defined completely.
Stefan Eßer [Tue, 11 Mar 1997 23:17:28 +0000 (23:17 +0000)]
Add support for the buggy CMD640B PCI EIDE controller chip, which
can't perform overlapping commands on both of its channels.
To enable the CMD640B work-around, the kernel must be compiled with
"options CMD640". Without that option there should be no difference
in the code produced compared to the previous revision of wd.c.
Submitted by: Wolfgang Helbig <helbig@ba-stuttgart.de>