]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
22 years ago - Return two shared locks to exclusive locks. This was premature.
Jeff Roberson [Thu, 22 Aug 2002 07:26:18 +0000 (07:26 +0000)]
 - Return two shared locks to exclusive locks.  This was premature.
 - Document the problems that prevent us from using shared locks.

22 years ago(forced commit to get the correct log message)
Jeff Roberson [Thu, 22 Aug 2002 07:01:52 +0000 (07:01 +0000)]
(forced commit to get the correct log message)
 - Remove unsafe and dead code from getnewvnode().
 - Fixup vclean() so that it grabs the VOP_LOCK before calling VOP_GETATTR()
   and grabs interlock afterwards.

22 years ago - Fix interlock handling in vn_lock(). Previously, vn_lock() could return
Jeff Roberson [Thu, 22 Aug 2002 06:58:11 +0000 (06:58 +0000)]
 - Fix interlock handling in vn_lock().  Previously, vn_lock() could return
   with interlock held in error conditions when the caller did not specify
   LK_INTERLOCK.
 - Add several comments to vn_lock() describing the rational behind the code
   flow since it was not immediately obvious.

22 years ago - Fix interlock handling in vn_lock(). Previously, vn_lock() could return
Jeff Roberson [Thu, 22 Aug 2002 06:51:06 +0000 (06:51 +0000)]
 - Fix interlock handling in vn_lock().  Previously, vn_lock() could return
   with interlock held in error conditions when the caller did not specify
   LK_INTERLOCK.
 - Add several comments to vn_lock() describing the rational behind the code
   flow since it was not immediately obvious.

22 years agoEncrypted strings (after hex decoding) aren't null terminated, because
Nick Sayer [Thu, 22 Aug 2002 06:19:07 +0000 (06:19 +0000)]
Encrypted strings (after hex decoding) aren't null terminated, because
0 might simply be part of the ciphertext.

PR: bin/40266
Submitted by: andr@dgap.mipt.ru
MFC after: 3 days

22 years agoFix a nasty memory corruption bug caused by having a bogus pointer
Marcel Moolenaar [Thu, 22 Aug 2002 03:56:57 +0000 (03:56 +0000)]
Fix a nasty memory corruption bug caused by having a bogus pointer
for the DT_IA64_PLT_RESERVE dynamic table entry. When a shared object
does not have any PLT relocations, the linker apparently doesn't find
it necessary to actually reserve the space for the BOR (Bind On
Reference) entries as pointed to by the DTE. As a result, relocatable
data in the PLT was overwritten, causing some unexpected control flow
with annoyingly predictable outcome: coredump.
To reproduce:
% echo 'int main() { return 0; }' > foo.c
% cc -o foo foo.c -lxpg4

22 years agoUse correct document date for the -a option.
Johan Karlsson [Thu, 22 Aug 2002 01:54:18 +0000 (01:54 +0000)]
Use correct document date for the -a option.

22 years agoAdd the -a option to report all matches instead of only the
Johan Karlsson [Thu, 22 Aug 2002 01:50:51 +0000 (01:50 +0000)]
Add the -a option to report all matches instead of only the
first of each requested type.

Approved by: joerg, sheldonh (mentor)

22 years agoRemoved extra $FreeBSD$. Moved #if 0 for the vendor id to the correct
Bruce Evans [Thu, 22 Aug 2002 01:07:16 +0000 (01:07 +0000)]
Removed extra $FreeBSD$.  Moved #if 0 for the vendor id to the correct
place.  Restored indent protection of copyright comment.

22 years agoDon't use "NULL" when "0" is really meant.
Archie Cobbs [Thu, 22 Aug 2002 00:30:03 +0000 (00:30 +0000)]
Don't use "NULL" when "0" is really meant.

22 years agoMake __clock_t use `unsigned long' rather than a fixed 32-bit integer
Mike Barcroft [Thu, 22 Aug 2002 00:28:24 +0000 (00:28 +0000)]
Make __clock_t use `unsigned long' rather than a fixed 32-bit integer
so that it changes to the correct size in IP32L64 mode.  Other
architectures don't have this issue.

22 years agoTX/RX descriptor/buffer management changes.
Prafulla Deuskar [Wed, 21 Aug 2002 23:44:15 +0000 (23:44 +0000)]
TX/RX descriptor/buffer management changes.
  Use array based scheme instead of queueing macros.

Submitted by: Luigi Rizzo (rizzo@icir.org)
MFC after: 3 days

22 years agoDon't use "NULL" when "0" is really meant.
Archie Cobbs [Wed, 21 Aug 2002 23:39:52 +0000 (23:39 +0000)]
Don't use "NULL" when "0" is really meant.

22 years agoJump in.
Thomas Quinot [Wed, 21 Aug 2002 21:18:14 +0000 (21:18 +0000)]
Jump in.

Reviewed by: roberto

22 years agoFor scripts that are called in non-login environments (in particular
Joerg Wunsch [Wed, 21 Aug 2002 20:54:45 +0000 (20:54 +0000)]
For scripts that are called in non-login environments (in particular
mail filters), the /usr/bin/perl dummy fails since it assumes that the
installed package could be reached via the existing $PATH.  Also,
since we are obviously looking for a perl installed by a package, we
should make sure that at least the default package bindir is being
searched.

22 years agoRevert some suspension/sleep/signal code from KSE-III
Julian Elischer [Wed, 21 Aug 2002 20:03:55 +0000 (20:03 +0000)]
Revert some suspension/sleep/signal code from KSE-III
We need to rethink a bit of this and it doesn't matter if
we break the KSE test program for now as long
as non-KSE programs act as expected.

Submitted by: David Xu <bsddiy@yahoo.com>
(this guy's just asking to get hit with a commit bit..)

22 years agoFixed getaddrinfo to honor sortlist in /etc/resolv.conf
Jim Pirzyk [Wed, 21 Aug 2002 19:57:53 +0000 (19:57 +0000)]
Fixed getaddrinfo to honor sortlist in /etc/resolv.conf

PR: bin/27939
Reviewed by: ru, sheldonh (about a year ago)
Obtained from: ume (via KAME, I think)
MFC after: 1 month

22 years agoMake the V2 errno translation more resistent to new errnos.
Poul-Henning Kamp [Wed, 21 Aug 2002 19:28:44 +0000 (19:28 +0000)]
Make the V2 errno translation more resistent to new errnos.

22 years agoInclude stddef.h for NULL definition, rather than rolling our own here.
Warner Losh [Wed, 21 Aug 2002 19:03:26 +0000 (19:03 +0000)]
Include stddef.h for NULL definition, rather than rolling our own here.

Reviewed by: jdp

22 years agoDon't use NULL when we really mean 0 for call to pw_edit.
Warner Losh [Wed, 21 Aug 2002 19:01:50 +0000 (19:01 +0000)]
Don't use NULL when we really mean 0 for call to pw_edit.

22 years agoUse '0' instead of NULL when we mean the integer constant 0 and not a
Warner Losh [Wed, 21 Aug 2002 18:52:32 +0000 (18:52 +0000)]
Use '0' instead of NULL when we mean the integer constant 0 and not a
"null pointer expression".

22 years agos/filesystem/file system/g as discussed on -developers
Tom Rhodes [Wed, 21 Aug 2002 18:11:48 +0000 (18:11 +0000)]
s/filesystem/file system/g as discussed on -developers

22 years agos/filesystem/file system/ as discussed on -developers
Tom Rhodes [Wed, 21 Aug 2002 17:32:44 +0000 (17:32 +0000)]
s/filesystem/file system/  as discussed on -developers

22 years agos/EDOFUS/EDOOFUS/
Poul-Henning Kamp [Wed, 21 Aug 2002 17:11:00 +0000 (17:11 +0000)]
s/EDOFUS/EDOOFUS/

Persuaded by: Google

22 years agoo Merge <machine/ansi.h> and <machine/types.h> into a new header
Mike Barcroft [Wed, 21 Aug 2002 16:20:02 +0000 (16:20 +0000)]
o Merge <machine/ansi.h> and <machine/types.h> into a new header
  called <machine/_types.h>.
o <machine/ansi.h> will continue to live so it can define MD clock
  macros, which are only MD because of gratuitous differences between
  architectures.
o Change all headers to make use of this.  This mainly involves
  changing:
    #ifdef _BSD_FOO_T_
    typedef _BSD_FOO_T_ foo_t;
    #undef _BSD_FOO_T_
    #endif
  to:
    #ifndef _FOO_T_DECLARED
    typedef __foo_t foo_t;
    #define _FOO_T_DECLARED
    #endif

Concept by: bde
Reviewed by: jake, obrien

22 years ago - Do not pretend to compile a kernel and remove the definition
Robert Drehmel [Wed, 21 Aug 2002 15:52:23 +0000 (15:52 +0000)]
 - Do not pretend to compile a kernel and remove the definition
   of the _KERNEL macro.
 - Do not include <sys/pcpu.h> for no reason.

Suggested by: jake

22 years agoDon't use NULL where you really mean 0 to sysctlbyname.
Warner Losh [Wed, 21 Aug 2002 15:15:15 +0000 (15:15 +0000)]
Don't use NULL where you really mean 0 to sysctlbyname.

22 years agoFix a few typos, among them s/builtin/built-in/ (except for the ref
Jens Schweikhardt [Wed, 21 Aug 2002 15:05:55 +0000 (15:05 +0000)]
Fix a few typos, among them s/builtin/built-in/ (except for the ref
to the builtin(1) man page.) This is for consistency with the spelling
both proposed by ispell as well as IEEE Std 1003.1-2001.

MFC after: 3 days

22 years agoUse the _STA method of a battery control method to see if a battery is
John Baldwin [Wed, 21 Aug 2002 15:01:53 +0000 (15:01 +0000)]
Use the _STA method of a battery control method to see if a battery is
present or not.  My laptop now properly notices when a battery is removed.

Reviewed by: iwasaki

22 years agoAdd myself.
Oliver Braun [Wed, 21 Aug 2002 14:25:48 +0000 (14:25 +0000)]
Add myself.

Approved by: dwcjr (mentor)

22 years ago - Define the macro _KERNEL to pretend we are compiling a kernel.
Robert Drehmel [Wed, 21 Aug 2002 12:45:36 +0000 (12:45 +0000)]
 - Define the macro _KERNEL to pretend we are compiling a kernel.
   This is required by recent changes to <sys/pcpu.h>, which uses
   the #error preprocessor directive to keep non-kernel
   applications from using it.
   _KERNEL is defined below the #include <stand.h>, because <stand.h>
   removes the definition of _KERNEL.
 - Move the inclusion of <sys/queue.h> above the inclusion of
   <sys/linker.h> to avoid syntax errors.

22 years agoCreate new functions in_sockaddr(), in6_sockaddr(), and
Don Lewis [Wed, 21 Aug 2002 11:57:12 +0000 (11:57 +0000)]
Create new functions in_sockaddr(), in6_sockaddr(), and
in6_v4mapsin6_sockaddr() which allocate the appropriate sockaddr_in*
structure and initialize it with the address and port information passed
as arguments.  Use calls to these new functions to replace code that is
replicated multiple times in in_setsockaddr(), in_setpeeraddr(),
in6_setsockaddr(), in6_setpeeraddr(), in6_mapped_sockaddr(), and
in6_mapped_peeraddr().  Inline COMMON_END in tcp_usr_accept() so that
we can call in_sockaddr() with temporary copies of the address and port
after the PCB is unlocked.

Fix the lock violation in tcp6_usr_accept() (caused by calling MALLOC()
inside in6_mapped_peeraddr() while the PCB is locked) by changing
the implementation of tcp6_usr_accept() to match tcp_usr_accept().

Reviewed by: suz

22 years agoFix a typo in #if 0 code
Scott Long [Wed, 21 Aug 2002 09:30:45 +0000 (09:30 +0000)]
Fix a typo in #if 0 code

22 years agoFix for stand-alone compiling
Scott Long [Wed, 21 Aug 2002 09:28:00 +0000 (09:28 +0000)]
Fix for stand-alone compiling

Reviewed by: mini

22 years ago - Document two cases, one in vget and the other in vn_lock, where the state
Jeff Roberson [Wed, 21 Aug 2002 08:34:48 +0000 (08:34 +0000)]
 - Document two cases, one in vget and the other in vn_lock, where the state
   of interlock on exit is not consistent.  There are probably several bugs
   relating to this.

22 years agoFix typo and ispell while I'm here.
Jens Schweikhardt [Wed, 21 Aug 2002 07:51:45 +0000 (07:51 +0000)]
Fix typo and ispell while I'm here.

PR: bin/41819
Submitted by: Christophe Juniet <cjuniet@entreview.com>
MFC after: 3 days

22 years agoSuggest that function prototypes in kernel headers be alphabetical,
Warner Losh [Wed, 21 Aug 2002 06:58:21 +0000 (06:58 +0000)]
Suggest that function prototypes in kernel headers be alphabetical,
unless there's a compelling reason to deviate.

Submitted by: Don Lewis
Suggestion not objected to by: developers@

22 years ago - If vn_lock fails with the LK_INTERLOCK flag set, interlock will not be
Jeff Roberson [Wed, 21 Aug 2002 06:40:34 +0000 (06:40 +0000)]
 - If vn_lock fails with the LK_INTERLOCK flag set, interlock will not be
   released.  vcanrecycle() failed to unlock interlock under this condition.
 - Remove an extra VOP_UNLOCK from a failure case in vcanrecycle().

Pointed out by: rwatson

22 years ago - Add two new debugging macros: ASSERT_VI_LOCKED and ASSERT_VI_UNLOCKED
Jeff Roberson [Wed, 21 Aug 2002 06:19:29 +0000 (06:19 +0000)]
 - Add two new debugging macros: ASSERT_VI_LOCKED and ASSERT_VI_UNLOCKED
 - Use the new VI asserts in place of the old mtx_assert checks.
 - Add the VI asserts to the automated lock checking in the VOP calls.  The
   interlock should not be held across vops with a few exceptions.
 - Add the vop_(un)lock_{pre,post} functions to assert that interlock is held
   when LK_INTERLOCK is set.

22 years agoWhoops, the manpage lied... ipfw2 has always accepted addr:mask
Luigi Rizzo [Wed, 21 Aug 2002 05:57:41 +0000 (05:57 +0000)]
Whoops, the manpage lied... ipfw2 has always accepted addr:mask
specifications.

22 years agoWhen setting the WI_RID_ENCRYPTION mode, we have to conditionally
Warner Losh [Wed, 21 Aug 2002 05:57:37 +0000 (05:57 +0000)]
When setting the WI_RID_ENCRYPTION mode, we have to conditionally
set EXCLUDE_UNENCRYPTED if we're not in OPEN mode (or if we are a
symbol card).

Obtained from: OpenBSD (looks like a millert@ special)

22 years agocomment about the magic number -149
Warner Losh [Wed, 21 Aug 2002 05:35:51 +0000 (05:35 +0000)]
comment about the magic number -149

22 years agoTreat IBSS the same as ADHOC in wi_get_cur_ssid. This makes ifconfig
Warner Losh [Wed, 21 Aug 2002 05:31:04 +0000 (05:31 +0000)]
Treat IBSS the same as ADHOC in wi_get_cur_ssid.  This makes ifconfig
work better, but might not make ibss actually work on the prism2 cards
that I have (this is the first time I've tried ibss in a long time).

22 years ago - Hold the vnode lock across unlink() so that the v_vflag check is safe.
Jeff Roberson [Wed, 21 Aug 2002 03:55:35 +0000 (03:55 +0000)]
 - Hold the vnode lock across unlink() so that the v_vflag check is safe.
 - Fix the long broken error handling for VV_ROOT and VDIR.

22 years agoKeep subdirectory list sorted.
Archie Cobbs [Tue, 20 Aug 2002 23:58:49 +0000 (23:58 +0000)]
Keep subdirectory list sorted.

22 years agoA chunk of cleanup, both stylistic and substantive.
Matt Jacob [Tue, 20 Aug 2002 23:04:08 +0000 (23:04 +0000)]
A chunk of cleanup, both stylistic and substantive.

We now also read configuration information for the SCSI cards- this allows
us to try and say what the speed settings now are.

Start, but not yet complete, the process of reorgs && #defines so that we
can backport to RELENG_4 pretty soon.

22 years agoRestrict visibility of wcwidth() and wcswidth(); they are XSI extensions.
Tim J. Robbins [Tue, 20 Aug 2002 22:44:40 +0000 (22:44 +0000)]
Restrict visibility of wcwidth() and wcswidth(); they are XSI extensions.

22 years agoAdd ng_l2tp module.
Archie Cobbs [Tue, 20 Aug 2002 22:03:57 +0000 (22:03 +0000)]
Add ng_l2tp module.

22 years agoMan page for new L2TP netgraph node type.
Archie Cobbs [Tue, 20 Aug 2002 22:00:56 +0000 (22:00 +0000)]
Man page for new L2TP netgraph node type.

Obtained from: Packet Design

22 years agoNew L2TP netgraph node type.
Archie Cobbs [Tue, 20 Aug 2002 21:59:50 +0000 (21:59 +0000)]
New L2TP netgraph node type.

Obtained from: Packet Design

22 years agoFix url for wordlists.
Alfred Perlstein [Tue, 20 Aug 2002 21:41:14 +0000 (21:41 +0000)]
Fix url for wordlists.

PR: docs/41830

22 years agoTake care of WARNS=3 warnings by using unsigned/signed variable
Johan Karlsson [Tue, 20 Aug 2002 20:54:03 +0000 (20:54 +0000)]
Take care of WARNS=3 warnings by using unsigned/signed variable
as needed.

Approved by: ken, sheldonh (mentor)

22 years agoWhitespace and style fixes.
John Baldwin [Tue, 20 Aug 2002 19:50:30 +0000 (19:50 +0000)]
Whitespace and style fixes.

Submitted by: bde

22 years agoWhen declaring local variables in macros, always use "_name" instead
Archie Cobbs [Tue, 20 Aug 2002 18:57:11 +0000 (18:57 +0000)]
When declaring local variables in macros, always use "_name" instead
of "name" to avoid ugly problems when the containing code already has
a variable named "name".

22 years agoPrevious commit of adding ut to xterm entry broke kterm-color entry.
Hajimu UMEMOTO [Tue, 20 Aug 2002 16:38:05 +0000 (16:38 +0000)]
Previous commit of adding ut to xterm entry broke kterm-color entry.
3 bytes (ut:) seems too long for kterm-color.  There is a limitation
of buffer size within 1024 bytes in our ncurses.

Submitted by: mistral@imasy.or.jp
Reviewed by: matusita
MFC after: 1 day

22 years agoFix buildworld breakage.
Ollivier Robert [Tue, 20 Aug 2002 15:58:44 +0000 (15:58 +0000)]
Fix buildworld breakage.

Pointy hat: phk

22 years agoDocument SWIDTH*
Andrey A. Chernov [Tue, 20 Aug 2002 15:19:34 +0000 (15:19 +0000)]
Document SWIDTH*

22 years agoClean up hostname and hostinfo handling in inithosts():
Yaroslav Tykhiy [Tue, 20 Aug 2002 14:56:06 +0000 (14:56 +0000)]
Clean up hostname and hostinfo handling in inithosts():
o check getaddrinfo(3) return value, not result pointer
o getaddrinfo(3) returns int, not pointer
o don't leak memory allocated for hostnames and hostinfo structures
o initialize pointers that will be checked for NULL somewhere

MFC after: 1 week

22 years agoInsert missing 'are' to fix grammar bogon.
Jens Schweikhardt [Tue, 20 Aug 2002 14:43:10 +0000 (14:43 +0000)]
Insert missing 'are' to fix grammar bogon.
MFC after: 3 days

22 years agoRemove duplicate vinum entry.
Yoshihiro Takahashi [Tue, 20 Aug 2002 14:41:44 +0000 (14:41 +0000)]
Remove duplicate vinum entry.

22 years agoMerged from sys/dev/syscons/syscons.c revisions 1.386 and 1.387.
Yoshihiro Takahashi [Tue, 20 Aug 2002 14:39:29 +0000 (14:39 +0000)]
Merged from sys/dev/syscons/syscons.c revisions 1.386 and 1.387.

22 years agoUse the __BUS_ACCESSOR macro for NEXUS_ACCESSOR
Maxime Henrion [Tue, 20 Aug 2002 13:50:17 +0000 (13:50 +0000)]
Use the __BUS_ACCESSOR macro for NEXUS_ACCESSOR
instead of rolling our own implementation.

Reviewed by: tmm

22 years agoAllow embedded `:' and `!' in target names.
Ruslan Ermilov [Tue, 20 Aug 2002 12:50:32 +0000 (12:50 +0000)]
Allow embedded `:' and `!' in target names.

PR: bin/6612
Obtained from: OpenBSD
MFC after: 1 week

22 years ago- Use the __BUS_ACCESSOR macro for EISA_ACCESSOR instead
Maxime Henrion [Tue, 20 Aug 2002 12:32:06 +0000 (12:32 +0000)]
- Use the __BUS_ACCESSOR macro for EISA_ACCESSOR instead
  of reimplementing it.
- #undef EISA_ACCESSOR after use like we do for other accessors.

Reviewed by: tmm

22 years agoImplement list of EA return functionality.
Poul-Henning Kamp [Tue, 20 Aug 2002 11:34:58 +0000 (11:34 +0000)]
Implement list of EA return functionality.
Correctly delete EA's when the content length is set to zero.

Sponsored by: DARPA & NAI Labs.

22 years agoBe consistent about quoting things.
Poul-Henning Kamp [Tue, 20 Aug 2002 11:33:25 +0000 (11:33 +0000)]
Be consistent about quoting things.
Be (somewhat) prepared for things to change size under us.
Recognize a empty attribute name as magic and print the list of attributes.
Use <err.h> for code clarity.
Deal with zero length returns.

Sponsored by: DARPA & NAI Labs.

22 years agoAllow one to override ${MINSTALL} in /etc/make.conf.
Ruslan Ermilov [Tue, 20 Aug 2002 11:11:37 +0000 (11:11 +0000)]
Allow one to override ${MINSTALL} in /etc/make.conf.

Prompted by: Andre Albsmeier <andre.albsmeier@mchp.siemens.de>

22 years agoBack out last commit. Needs slightly more subtle handling.
Orion Hodson [Tue, 20 Aug 2002 08:02:56 +0000 (08:02 +0000)]
Back out last commit.  Needs slightly more subtle handling.

22 years agoRemove the possibility of a race condition when reading the . and ..
Scott Long [Tue, 20 Aug 2002 04:46:59 +0000 (04:46 +0000)]
Remove the possibility of a race condition when reading the . and ..
entries.

22 years agoDon't abuse the stack when translating names.
Scott Long [Tue, 20 Aug 2002 04:46:04 +0000 (04:46 +0000)]
Don't abuse the stack when translating names.

22 years agoMerge changes in GENERIC to reduce difference (mostly cosmetic).
Jun Kuriyama [Tue, 20 Aug 2002 04:36:31 +0000 (04:36 +0000)]
Merge changes in GENERIC to reduce difference (mostly cosmetic).

22 years agoAdd a manual page for wcwidth().
Tim J. Robbins [Tue, 20 Aug 2002 03:42:21 +0000 (03:42 +0000)]
Add a manual page for wcwidth().

22 years agoAdd a manual page for wcswidth().
Tim J. Robbins [Tue, 20 Aug 2002 03:40:45 +0000 (03:40 +0000)]
Add a manual page for wcswidth().

22 years agoUse mv -f. Otherwise if you accidently build as root, and then as a user,
David E. O'Brien [Tue, 20 Aug 2002 03:06:30 +0000 (03:06 +0000)]
Use mv -f.  Otherwise if you accidently build as root, and then as a user,
it keeps asking you if you want to overwrite it or not.

Submitted by: peter

22 years agoPrefer NULL to 0 when passing a NULL pointer.
Robert Watson [Tue, 20 Aug 2002 02:54:09 +0000 (02:54 +0000)]
Prefer NULL to 0 when passing a NULL pointer.

22 years agoProvide stub mpo_syscall() implementations for mac_none and mac_test.
Robert Watson [Tue, 20 Aug 2002 02:53:35 +0000 (02:53 +0000)]
Provide stub mpo_syscall() implementations for mac_none and mac_test.

Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs

22 years agoCorrect one more errant whitespace nit that crept in during changes
Robert Watson [Tue, 20 Aug 2002 02:21:38 +0000 (02:21 +0000)]
Correct one more errant whitespace nit that crept in during changes
in the arguments to vn_rdwr().  Hopefully the last.

22 years agoHandle one more case of a fifofs filetmp: set filetmp.f_cred to
Robert Watson [Tue, 20 Aug 2002 02:17:59 +0000 (02:17 +0000)]
Handle one more case of a fifofs filetmp: set filetmp.f_cred to
ap->a_cred, and pass in ap->a_td->td_ucred as the active_cred to
soo_poll().

Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs

22 years agoMove just committed version of wcswidth.c here (from "locale"),
Andrey A. Chernov [Tue, 20 Aug 2002 02:06:28 +0000 (02:06 +0000)]
Move just committed version of wcswidth.c here (from "locale"),
a bit optimized now.

22 years agoRemove wcswidth.c from here (and move it to "string")
Andrey A. Chernov [Tue, 20 Aug 2002 01:59:26 +0000 (01:59 +0000)]
Remove wcswidth.c from here (and move it to "string")

22 years agoRemove space at the end of continuation line in prev. commit
Andrey A. Chernov [Tue, 20 Aug 2002 01:16:06 +0000 (01:16 +0000)]
Remove space at the end of continuation line in prev. commit

22 years agoUntangle this warning a bit:
Peter Wemm [Tue, 20 Aug 2002 00:29:16 +0000 (00:29 +0000)]
Untangle this warning a bit:
COMPAT_SVR4 is broken and usage is, until fixed, not recommended
BTW; does anybody remember why this is here?

22 years agoAdd support for the R_IA64_IPLTLSB relocation in non-PLT context.
Marcel Moolenaar [Tue, 20 Aug 2002 00:24:33 +0000 (00:24 +0000)]
Add support for the R_IA64_IPLTLSB relocation in non-PLT context.
This relocation creates a function descriptor at the specified
address and is commonly used for C++ to create virtual function
tables.

22 years agoPrint out a carriage return to make the screen output make more sense
Gordon Tetlow [Tue, 20 Aug 2002 00:14:11 +0000 (00:14 +0000)]
Print out a carriage return to make the screen output make more sense

Submitted by: mike@

22 years agoremove unit counts from atkbdc, pckbd, sc
Peter Wemm [Tue, 20 Aug 2002 00:10:22 +0000 (00:10 +0000)]
remove unit counts from atkbdc, pckbd, sc

22 years agode-count pcbkd
Peter Wemm [Tue, 20 Aug 2002 00:09:16 +0000 (00:09 +0000)]
de-count pcbkd

22 years agode-count pckbd for pc98. This file is only compiled if NPCKBD was 1,
Peter Wemm [Tue, 20 Aug 2002 00:05:41 +0000 (00:05 +0000)]
de-count pckbd for pc98.  This file is only compiled if NPCKBD was 1,
so the conditional compile should never have been used.

22 years agode-count atkbdc and sc. Folks, remove the '1' from 'device sc 1' and
Peter Wemm [Mon, 19 Aug 2002 23:59:21 +0000 (23:59 +0000)]
de-count atkbdc and sc.  Folks, remove the '1' from 'device sc 1' and
'device atkbdc 1'.

22 years agode-count atkbdc. I have more extensive patches to make properly dynamic,
Peter Wemm [Mon, 19 Aug 2002 23:58:37 +0000 (23:58 +0000)]
de-count atkbdc.  I have more extensive patches to make properly dynamic,
but since pc hardware only allows one AT-style keyboard controller, this
doesn't seem particularly urgent.  (I do not know what the old sunriver
remote keyboard/mouse/vga cards do, that might be an exception).

22 years agode-count schistory.c. The handling of NSC was rather bogus here and was
Peter Wemm [Mon, 19 Aug 2002 23:56:01 +0000 (23:56 +0000)]
de-count schistory.c.  The handling of NSC was rather bogus here and was
little more than a place holder, because nothing actually counted the
number of 'sc' units to compare it against NSC.  A bit more work here
is needed so that the scaling of SC_MAX_HISTORY_SIZE and extra_history_size
goes up when more sc units are added.  But, it does not appear that we can
have more than one console yet, so it does not seem particularly urgent.

22 years agoxdr.3 is added to MAN in ../xdr/Makefile.inc where it belongs, so don't
Bruce Evans [Mon, 19 Aug 2002 23:06:00 +0000 (23:06 +0000)]
xdr.3 is added to MAN in ../xdr/Makefile.inc where it belongs, so don't
add it here.

22 years agoTypo fix.
Maxime Henrion [Mon, 19 Aug 2002 22:57:32 +0000 (22:57 +0000)]
Typo fix.

Reviewed by: tmm

22 years agostyle(9) nit.
Maxime Henrion [Mon, 19 Aug 2002 22:51:23 +0000 (22:51 +0000)]
style(9) nit.

22 years agoUse the __BUS_ACCESSOR macro for PCIB_ACCESSOR instead of
Maxime Henrion [Mon, 19 Aug 2002 22:50:08 +0000 (22:50 +0000)]
Use the __BUS_ACCESSOR macro for PCIB_ACCESSOR instead of
reimplementing it.

Reviewed by: tmm

22 years agoSort unsorted prototypes
Andrey A. Chernov [Mon, 19 Aug 2002 21:00:13 +0000 (21:00 +0000)]
Sort unsorted prototypes

22 years agoWhen login tries to do the chmod/chflags on a read only file system,
Warner Losh [Mon, 19 Aug 2002 20:54:00 +0000 (20:54 +0000)]
When login tries to do the chmod/chflags on a read only file system,
it complains that it can't do it because the filesystem is readonly.
Assume that when the user has a readonly /dev that they don't care if
login can't change the permissions/flags.  While this does break a few
things like msgs, we'll assume that the user setting up the read only
system knows what they are doing.

All this change does is to stop the complaint when the file system is
read only.  It also adds comments as to why EROFS and EOPNOTSUPP are
ignored.

This allows one to have a read-only / w/o a /dev MFS and have a
relatively warning-free existence.  /etc/rc still complains when it
can't chown/chflags/chmod things, but that's easy to ignore/tweak.

Reviewed by: roberto, phk
Sponsored by: Timing Solutions

22 years agoActivate (uncomment) wcwidth() and wcswidth() now implemented
Andrey A. Chernov [Mon, 19 Aug 2002 20:48:18 +0000 (20:48 +0000)]
Activate (uncomment) wcwidth() and wcswidth() now implemented

22 years agoImplement wcswidth()
Andrey A. Chernov [Mon, 19 Aug 2002 20:46:10 +0000 (20:46 +0000)]
Implement wcswidth()

22 years agoStop trying to align malloc()ed memory. Rely on malloc() instead, just like
Semen Ustimenko [Mon, 19 Aug 2002 20:36:08 +0000 (20:36 +0000)]
Stop trying to align malloc()ed memory. Rely on malloc() instead, just like
others.

22 years agoUse modern-style arguments declaration
Andrey A. Chernov [Mon, 19 Aug 2002 20:32:27 +0000 (20:32 +0000)]
Use modern-style arguments declaration