]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
21 years agoThis commit was generated by cvs2svn to compensate for changes in r103449,
Alexander Kabaev [Tue, 17 Sep 2002 04:04:15 +0000 (04:04 +0000)]
This commit was generated by cvs2svn to compensate for changes in r103449,
which included commits to RCS files with non-trunk default branches.

21 years agoGcc 3.2.1-prerelease libf2c bits from the FSF anoncvs repo gcc-3_2-branch on 16-Sep...
Alexander Kabaev [Tue, 17 Sep 2002 04:04:15 +0000 (04:04 +0000)]
Gcc 3.2.1-prerelease libf2c bits from the FSF anoncvs repo gcc-3_2-branch on 16-Sep-2002 13:23:11 EDT.

21 years agoThis commit was generated by cvs2svn to compensate for changes in r103447,
Alexander Kabaev [Tue, 17 Sep 2002 04:04:04 +0000 (04:04 +0000)]
This commit was generated by cvs2svn to compensate for changes in r103447,
which included commits to RCS files with non-trunk default branches.

21 years agoGcc 3.2.1-prerelease C++ support bits from the FSF anoncvs repo gcc-3_2-branch on...
Alexander Kabaev [Tue, 17 Sep 2002 04:04:04 +0000 (04:04 +0000)]
Gcc 3.2.1-prerelease C++ support bits from the FSF anoncvs repo gcc-3_2-branch on 16-Sep-2002 13:23:11 EDT.

21 years agoThis commit was generated by cvs2svn to compensate for changes in r103445,
Alexander Kabaev [Tue, 17 Sep 2002 04:03:37 +0000 (04:03 +0000)]
This commit was generated by cvs2svn to compensate for changes in r103445,
which included commits to RCS files with non-trunk default branches.

21 years agoGcc 3.2.1-prerelease from the FSF anoncvs repo gcc-3_2-branch on 16-Sep-2002 13:23...
Alexander Kabaev [Tue, 17 Sep 2002 04:03:37 +0000 (04:03 +0000)]
Gcc 3.2.1-prerelease from the FSF anoncvs repo gcc-3_2-branch on 16-Sep-2002 13:23:11 EDT.

21 years agoTempary fix for inet6. The final fix is to change in6_pcbnotify to take pcbinfo instead
Jennifer Yang [Tue, 17 Sep 2002 03:19:43 +0000 (03:19 +0000)]
Tempary fix for inet6. The final fix is to change in6_pcbnotify to take pcbinfo instead
of pcbhead. It is on the way.

21 years agoUse a weak symbol for signalcontext().
Jonathan Mini [Tue, 17 Sep 2002 03:15:47 +0000 (03:15 +0000)]
Use a weak symbol for signalcontext().

21 years agoAdd signalcontext(), which adds a signal frame to a ucontext_t.
Jonathan Mini [Tue, 17 Sep 2002 03:11:07 +0000 (03:11 +0000)]
Add signalcontext(), which adds a signal frame to a ucontext_t.

21 years agoConglomerate printing of ps_pgtok'd data into a PLONG type. I couldn't think
Juli Mallett [Tue, 17 Sep 2002 02:52:44 +0000 (02:52 +0000)]
Conglomerate printing of ps_pgtok'd data into a PLONG type.  I couldn't think
of a better name, except PINT, but I decided to go with assuming LONG to
be safe, rather than assuming INT.

21 years agoBANG! BANG! BANG! Put these bits out of their misery.
David E. O'Brien [Tue, 17 Sep 2002 01:54:56 +0000 (01:54 +0000)]
BANG! BANG! BANG!  Put these bits out of their misery.
Murdered by members of: a.out.die.die.die

ELF is the 1 true path now.  So make good on the src/Makefile threat that
building a.out will not be supported post 4.x.  These bits should either
resurface as a port, or a new port using the latest Binutils bits.  The
later will not support our SunOS-style shared a.out libs; but we shouldn't
need such support by this point in time.

21 years agoInitiate deorbit burn for the i386-only a.out related support. Moves are
Peter Wemm [Tue, 17 Sep 2002 01:49:00 +0000 (01:49 +0000)]
Initiate deorbit burn for the i386-only a.out related support.  Moves are
under way to move the remnants of the a.out toolchain to ports.  As the
comment in src/Makefile said, this stuff is deprecated and one should not
expect this to remain beyond 4.0-REL.  It has already lasted WAY beyond
that.

Notable exceptions:
gcc - I have not touched the a.out generation stuff there.
ldd/ldconfig - still have some code to interface with a.out rtld.
old as/ld/etc - I have not removed these yet, pending their move to ports.
some includes - necessary for ldd/ldconfig for now.

Tested on: i386 (extensively), alpha

21 years agoDisable building of the a.out gas and ld. They will become ports.
David E. O'Brien [Tue, 17 Sep 2002 01:43:21 +0000 (01:43 +0000)]
Disable building of the a.out gas and ld.  They will become ports.

21 years agoNew release note: SA-02:39.
Bruce A. Mah [Mon, 16 Sep 2002 23:02:37 +0000 (23:02 +0000)]
New release note:  SA-02:39.

21 years agoFix documentation of clnt_control()'s CL{GET|SET}_{VERS|XID} to
Bill Fenner [Mon, 16 Sep 2002 21:45:37 +0000 (21:45 +0000)]
Fix documentation of clnt_control()'s CL{GET|SET}_{VERS|XID} to
 reflect that they actually require a u_int32_t *, which is not
 necessarily the same as an unsigned long *.

21 years agoxids are u_int32_ts, which are not necessarily the same size as time_ts or
Bill Fenner [Mon, 16 Sep 2002 21:39:42 +0000 (21:39 +0000)]
xids are u_int32_ts, which are not necessarily the same size as time_ts or
 unsigned longs.  This fixes "ypbind -S ... -m" on sparc64.

21 years agoUpdate build infrastructure after import of Heimdal Kerberos 2002/09/16.
Jacques Vidrine [Mon, 16 Sep 2002 21:10:29 +0000 (21:10 +0000)]
Update build infrastructure after import of Heimdal Kerberos 2002/09/16.

21 years agoResolve conflicts.
Jacques Vidrine [Mon, 16 Sep 2002 21:07:11 +0000 (21:07 +0000)]
Resolve conflicts.

21 years agoThis commit was generated by cvs2svn to compensate for changes in r103423,
Jacques Vidrine [Mon, 16 Sep 2002 21:04:40 +0000 (21:04 +0000)]
This commit was generated by cvs2svn to compensate for changes in r103423,
which included commits to RCS files with non-trunk default branches.

21 years agoImport of Heimdal Kerberos from KTH repository circa 2002/09/16.
Jacques Vidrine [Mon, 16 Sep 2002 21:04:40 +0000 (21:04 +0000)]
Import of Heimdal Kerberos from KTH repository circa 2002/09/16.

21 years agoPerform keyword.c:1.27 properly, implement -orss in the New World Order of
Juli Mallett [Mon, 16 Sep 2002 20:19:51 +0000 (20:19 +0000)]
Perform keyword.c:1.27 properly, implement -orss in the New World Order of
ps(1) formatting, using pgtok() to get the value in K, rather than printing
it in pages.  This is consistent with behaviour before keyword.c:1.26 (et al)
which exists in STABLE today, and which uses the same metric as VSZ.

Submitted by: bde

21 years agoRemove some NOTINUSE stuff. Good housekeeping.
Juli Mallett [Mon, 16 Sep 2002 20:06:04 +0000 (20:06 +0000)]
Remove some NOTINUSE stuff.  Good housekeeping.

21 years agoStart new sentence with newline.
Nate Lawson [Mon, 16 Sep 2002 19:53:51 +0000 (19:53 +0000)]
Start new sentence with newline.

Suggested by: jhb

21 years agoMake libpthread KSE aware.
Jonathan Mini [Mon, 16 Sep 2002 19:52:52 +0000 (19:52 +0000)]
Make libpthread KSE aware.

Reviewed by: deischen, julian
Approved by: -arch

21 years agoDon't depend on <sys/types.h> pollution in <fcntl.h>. Sort includes.
Mike Barcroft [Mon, 16 Sep 2002 19:46:20 +0000 (19:46 +0000)]
Don't depend on <sys/types.h> pollution in <fcntl.h>.  Sort includes.

21 years agoMechanically change all libc_r references to libpthread.
Jonathan Mini [Mon, 16 Sep 2002 19:29:34 +0000 (19:29 +0000)]
Mechanically change all libc_r references to libpthread.

21 years agoAdd kernel support needed for the KSE-aware libpthread:
Jonathan Mini [Mon, 16 Sep 2002 19:28:01 +0000 (19:28 +0000)]
Add kernel support needed for the KSE-aware libpthread:
- Use ucontext_t's to store KSE thread state.
- Synthesize state for the UTS upon each upcall, rather than
  saving and copying a trapframe.
- Save and restore FPU state properly in ucontext_t's.
- Deliver signals to KSE-aware processes via upcall.
- Rename kse mailbox structure fields to be more BSD-like.
- Store the UTS's stack in struct proc in a stack_t.

Reviewed by: bde, deischen, julian
Approved by: -arch

21 years agoAdd kernel support needed for the KSE-aware libpthread:
Jonathan Mini [Mon, 16 Sep 2002 19:26:48 +0000 (19:26 +0000)]
Add kernel support needed for the KSE-aware libpthread:
- Use ucontext_t's to store KSE thread state.
- Synthesize state for the UTS upon each upcall, rather than
  saving and copying a trapframe.
- Deliver signals to KSE-aware processes via upcall.
- Rename kse mailbox structure fields to be more BSD-like.
- Store the UTS's stack in struct proc in a stack_t.

Reviewed by: bde, deischen, julian
Approved by: -arch

21 years agoAdd kernel support needed for the KSE-aware libpthread:
Jonathan Mini [Mon, 16 Sep 2002 19:25:59 +0000 (19:25 +0000)]
Add kernel support needed for the KSE-aware libpthread:
- Maintain fpu state across signals.
- Save and restore FPU state properly in ucontext_t's.

Reviewed by: bde, deischen, julian
Approved by: -arch

21 years agoAdd kernel support needed for the KSE-aware libpthread:
Jonathan Mini [Mon, 16 Sep 2002 19:25:41 +0000 (19:25 +0000)]
Add kernel support needed for the KSE-aware libpthread:
- Maintain fpu state across signals.
- Save and restore FPU state properly in ucontext_t's.

Reviewed by: deischen, julian
Approved by: -arch

21 years agoAdd kernel support needed for the KSE-aware libpthread:
Jonathan Mini [Mon, 16 Sep 2002 19:25:08 +0000 (19:25 +0000)]
Add kernel support needed for the KSE-aware libpthread:
- Maintain fpu state across signals.
- Use ucontext_t's to store KSE thread state.
- Synthesize state for the UTS upon each upcall, rather than
  saving and copying a trapframe.
- Save and restore FPU state properly in ucontext_t's.

Reviewed by: deischen, julian
Approved by: -arch

21 years agoSave and restore FPU state properly in ucontext_t's.
Jonathan Mini [Mon, 16 Sep 2002 19:24:31 +0000 (19:24 +0000)]
Save and restore FPU state properly in ucontext_t's.

Reviewed by: deischen, julian
Approved by: -arch

21 years agoAdd signalcontext(), which lays down a signal frame onto a ucontext_t.
Jonathan Mini [Mon, 16 Sep 2002 19:23:35 +0000 (19:23 +0000)]
Add signalcontext(), which lays down a signal frame onto a ucontext_t.

Reviewed by: deischen, julian
Approved by: -arch

21 years agoAdd a cast to make this file compile in userland on sparc64 without
Poul-Henning Kamp [Mon, 16 Sep 2002 18:45:18 +0000 (18:45 +0000)]
Add a cast to make this file compile in userland on sparc64 without
warnings.

21 years agoUpdate man page for enum -> const char * change for v_tag
Nate Lawson [Mon, 16 Sep 2002 18:16:07 +0000 (18:16 +0000)]
Update man page for enum -> const char * change for v_tag

21 years agoI forgot fsdb was still sharing files with fsck.
Poul-Henning Kamp [Mon, 16 Sep 2002 14:18:43 +0000 (14:18 +0000)]
I forgot fsdb was still sharing files with fsck.

21 years agoAllocate 64K recieve buffer for DNS responses.
Hajimu UMEMOTO [Mon, 16 Sep 2002 13:19:47 +0000 (13:19 +0000)]
Allocate 64K recieve buffer for DNS responses.

21 years agoAdd sleep state explanation.
Takanori Watanabe [Mon, 16 Sep 2002 12:11:25 +0000 (12:11 +0000)]
Add sleep state explanation.
Requested by: murray at Cafe in Tokyo.

21 years agoConvert eight space indents to tabs in the "*" format handling code.
Tim J. Robbins [Mon, 16 Sep 2002 12:07:13 +0000 (12:07 +0000)]
Convert eight space indents to tabs in the "*" format handling code.

21 years agoAdd a source file where EXTATTR checks will happen and hook it in even
Poul-Henning Kamp [Mon, 16 Sep 2002 11:27:47 +0000 (11:27 +0000)]
Add a source file where EXTATTR checks will happen and hook it in even
if it doesn't do anything yet.

Sponsored by: DARPA & NAI Labs.

21 years agoReserve a transport parameter name and number for Darwin.
Boris Popov [Mon, 16 Sep 2002 10:57:13 +0000 (10:57 +0000)]
Reserve a transport parameter name and number for Darwin.

Obtained from: Darwin
MFC after: 2 weeks

21 years agoImplement support for mixed case passwords.
Boris Popov [Mon, 16 Sep 2002 10:50:38 +0000 (10:50 +0000)]
Implement support for mixed case passwords.

Obtained from: Darwin
MFC after: 2 weeks

21 years agoAdd support for large readx and writex functions if server supports them.
Boris Popov [Mon, 16 Sep 2002 10:18:34 +0000 (10:18 +0000)]
Add support for large readx and writex functions if server supports them.

Obtained from: Darwin
MFC after: 2 weeks

21 years agoInclude include "opt_atalk.h" so that the NETATALK support can work.
Bruce Evans [Mon, 16 Sep 2002 10:10:18 +0000 (10:10 +0000)]
Include include "opt_atalk.h" so that the NETATALK support can work.

Removed unused includes.

Removed used includes of <sys/queue.h> and <sys/time.h>, since these are
standard pollution (especially the latter).

Reviewed by: sobomax

21 years agoBacked out revs 1.32-1.33. The problem has been fixed better by
Bruce Evans [Mon, 16 Sep 2002 10:06:02 +0000 (10:06 +0000)]
Backed out revs 1.32-1.33.  The problem has been fixed better by
depolluting <netinet/if_gre.h> in the !_KERNEL case.

Reviewed by: sobomax

21 years agoPick up modules from the compile directory instead of /boot/kernel.
Poul-Henning Kamp [Mon, 16 Sep 2002 09:57:10 +0000 (09:57 +0000)]
Pick up modules from the compile directory instead of /boot/kernel.

Allow the script to operate on any convenient kernel.

Submitted by: "Matthew Emmerton" <matt@gsicomp.on.ca>

21 years agoEnable browsing of NetApp servers (use ascii mode).
Boris Popov [Mon, 16 Sep 2002 09:51:58 +0000 (09:51 +0000)]
Enable browsing of NetApp servers (use ascii mode).

Obtained from: Darwin (PR-3002667)
MFC after: 2 weeks

21 years agoFix the clockprobe test on the Sil 680
Søren Schmidt [Mon, 16 Sep 2002 09:37:26 +0000 (09:37 +0000)]
Fix the clockprobe test on the Sil 680

21 years agoAdd more constants for future use in kernel and userland.
Boris Popov [Mon, 16 Sep 2002 09:26:07 +0000 (09:26 +0000)]
Add more constants for future use in kernel and userland.

Obtained from: Darwin

21 years agoMake the changes needed for libpthread to compile in its new home.
Jonathan Mini [Mon, 16 Sep 2002 08:45:36 +0000 (08:45 +0000)]
Make the changes needed for libpthread to compile in its new home.
The new libpthread will provide POSIX threading support using KSE.
These files were previously repo-copied from src/lib/libc_r.

Reviewed by: deischen
Approved by: -arch

21 years agoDon't depend on <sys/types.h> pollution in <fcntl.h>. Sort includes.
Mike Barcroft [Mon, 16 Sep 2002 08:34:08 +0000 (08:34 +0000)]
Don't depend on <sys/types.h> pollution in <fcntl.h>.  Sort includes.

21 years agoDon't build unneeded modules for pc98.
Yoshihiro Takahashi [Mon, 16 Sep 2002 08:32:48 +0000 (08:32 +0000)]
Don't build unneeded modules for pc98.

21 years agoFixed messes involving $FreeBSD$ starting with one left in the copyright
Bruce Evans [Mon, 16 Sep 2002 08:22:57 +0000 (08:22 +0000)]
Fixed messes involving $FreeBSD$ starting with one left in the copyright
after adding __FBSDID().

Garbage-collected kvm_readswap().  This was once used by kvm_uread(), but
kvm_uread() now just reads /proc/<pid>/mem and procfs hopefully handles
swapped out pages.

21 years agoMerged from sys/isa/fd.c revisions 1.234, 1.235 and 1.236.
Yoshihiro Takahashi [Mon, 16 Sep 2002 07:39:47 +0000 (07:39 +0000)]
Merged from sys/isa/fd.c revisions 1.234, 1.235 and 1.236.

21 years agoMFi386: revision 1.178.
Yoshihiro Takahashi [Mon, 16 Sep 2002 07:19:43 +0000 (07:19 +0000)]
MFi386: revision 1.178.

21 years agoDon't depend on <sys/types.h> pollution in <fcntl.h>.
Mike Barcroft [Mon, 16 Sep 2002 07:18:27 +0000 (07:18 +0000)]
Don't depend on <sys/types.h> pollution in <fcntl.h>.

21 years agoMFi386: revisions 1.417 and 1.418.
Yoshihiro Takahashi [Mon, 16 Sep 2002 07:17:43 +0000 (07:17 +0000)]
MFi386: revisions 1.417 and 1.418.

21 years agoMerged from sys/dev/sio/sio.c revision 1.378.
Yoshihiro Takahashi [Mon, 16 Sep 2002 07:14:02 +0000 (07:14 +0000)]
Merged from sys/dev/sio/sio.c revision 1.378.

21 years agoMerged from sys/isa/syscons_isa.c revision 1.18.
Yoshihiro Takahashi [Mon, 16 Sep 2002 07:11:43 +0000 (07:11 +0000)]
Merged from sys/isa/syscons_isa.c revision 1.18.

21 years agoUpdate for version 3.39.
David E. O'Brien [Mon, 16 Sep 2002 06:31:49 +0000 (06:31 +0000)]
Update for version 3.39.

21 years agoUpdate for 3.39.
David E. O'Brien [Mon, 16 Sep 2002 06:22:04 +0000 (06:22 +0000)]
Update for 3.39.

21 years agoMake current.
David E. O'Brien [Mon, 16 Sep 2002 06:02:55 +0000 (06:02 +0000)]
Make current.

21 years agoThis commit was generated by cvs2svn to compensate for changes in r103373,
David E. O'Brien [Mon, 16 Sep 2002 05:54:26 +0000 (05:54 +0000)]
This commit was generated by cvs2svn to compensate for changes in r103373,
which included commits to RCS files with non-trunk default branches.

21 years agoVirgin import of Christos Zoulas's FILE 3.39.
David E. O'Brien [Mon, 16 Sep 2002 05:54:26 +0000 (05:54 +0000)]
Virgin import of Christos Zoulas's FILE 3.39.

21 years agoCreate bzgrep.
David E. O'Brien [Mon, 16 Sep 2002 04:27:29 +0000 (04:27 +0000)]
Create bzgrep.

21 years agogeom_aes.c requires rijndael crypto
Peter Grehan [Mon, 16 Sep 2002 04:13:48 +0000 (04:13 +0000)]
geom_aes.c requires rijndael crypto

Approved by: phk, benno

21 years agoReformat an overly-wide comment.
Garrett Wollman [Mon, 16 Sep 2002 03:22:06 +0000 (03:22 +0000)]
Reformat an overly-wide comment.

21 years agofcntl(..., F_SETLKW, ...) takes a pointer to a struct flock just like
Thomas Moestl [Mon, 16 Sep 2002 01:05:15 +0000 (01:05 +0000)]
fcntl(..., F_SETLKW, ...) takes a pointer to a struct flock just like
F_SETLK does, so it also needs this structure copied in in fnctl() before
calling kern_fcntl().

21 years agoAllocate KSEs and KSEGRPs separatly and remove them from the proc structure.
Julian Elischer [Sun, 15 Sep 2002 23:52:25 +0000 (23:52 +0000)]
Allocate KSEs and KSEGRPs separatly and remove them from the proc structure.
next step is to allow > 1 to be allocated per process. This would give
multi-processor threads. (when the rest of the infrastructure is
in place)

While doing this I noticed libkvm and sys/kern/kern_proc.c:fill_kinfo_proc
are diverging more than they should.. corrective action needed soon.

21 years agoDeprecate the USB_SETDEBUG IOCTL as debug levels can now be set
Josef Karthauser [Sun, 15 Sep 2002 22:35:58 +0000 (22:35 +0000)]
Deprecate the USB_SETDEBUG IOCTL as debug levels can now be set
using sysctl.

21 years agoGarbage-collected __ELF__ ifdefs.
Bruce Evans [Sun, 15 Sep 2002 22:28:39 +0000 (22:28 +0000)]
Garbage-collected __ELF__ ifdefs.

Fixed some style bugs (mainly unused includes).

21 years agoDon't use the ELF symbol type to summarily reject symbols in
Bruce Evans [Sun, 15 Sep 2002 22:17:40 +0000 (22:17 +0000)]
Don't use the ELF symbol type to summarily reject symbols in
X_db_search_symbol().  Otherwise we don't see important symbols in
non-verbosely written assembler code.

NetBSD already has this.  The kld version already has a stronger form
of it without really trying -- linker_ddb_search_symbol() doesn't
support ddb's symbol search strategy parameter, so the kld
X_db_search_symbol() doesn't pass the parameter to linker_ddb...() and
linker_ddb...() doesn't make distinctions based on the symbol type.

db_elf.c now works better than db_kld.c when it works (which is essentially
when there are no modules except the kernel).  It works after booting
with -d.  db_kld.c doesn't work until lots of SYSINIT()s have run.

21 years agoMade this work on i386's at least. It wants ELF section headers for
Bruce Evans [Sun, 15 Sep 2002 21:49:13 +0000 (21:49 +0000)]
Made this work on i386's at least.  It wants ELF section headers for
symbol table sections.  Reconstruct the necessary section headers from
(ksym_start, ksym_end).  This was much easier than converting to use
module metadata, and just works for static symbols, unlike db_kld when
there is no module metadata.  Initialize (ksym_start, ksym_end) from
bootinfo on i386's only.

The boot loader should load section headers for all sections that it
loads, and apparently did this for at least the symbol table sections
when this file last worked under FreeBSD (on alphas only) and always
did this under NetBSD (where this file was obtained from).  At least
on i386's, boot2 discards the section headers (except for converting
them to (bootinfo.bi_symtab, bootinfo.bi_esymtab), and as far as I can
tell, loader(8) discards them apart from converting them to the bootinfo
values and module metadata.

21 years agoSuitably shrink a comment so it applies again.
Poul-Henning Kamp [Sun, 15 Sep 2002 21:25:55 +0000 (21:25 +0000)]
Suitably shrink a comment so it applies again.

Submitted by: bde

21 years agoMade this compile (but not work). This involved mainly const poisoning
Bruce Evans [Sun, 15 Sep 2002 20:48:08 +0000 (20:48 +0000)]
Made this compile (but not work).  This involved mainly const poisoning
and renaming ALIGNED_POINTER() to _ALIGNED_POINTER() plus the following
hacks for i386's:
- define _ALIGNED_POINTER() if it is not already defined.  Most non-i386
  arches define it <machine/param.h> define it in <machine/param.h>,
  although none actually used it in the kernel.
- define ksym_start and ksym_end.  Most non-i386 arches still define and
  initialize these in machdep.c although they didn't used them.  Here is
  a better place to define them but not to initialize them.

21 years agoAllocate 64K recieve buffer for DNS responses.
Hajimu UMEMOTO [Sun, 15 Sep 2002 20:36:38 +0000 (20:36 +0000)]
Allocate 64K recieve buffer for DNS responses.
KAME did the modification only to _dns_getaddrinfo().  However,
it is not sufficient, and res_queryN() should be modified, too.
So, I did same modification to res_queryN().

Obtained from: KAME

21 years agos/protocole/protocol/
Ceri Davies [Sun, 15 Sep 2002 20:33:05 +0000 (20:33 +0000)]
s/protocole/protocol/

PR: docs/42815
Submitted by: Nicola Vitale <nivit@libero.it>
MFC after: 21 days

21 years agoDocument intptr_t and uintptr_t.
Mike Barcroft [Sun, 15 Sep 2002 20:12:20 +0000 (20:12 +0000)]
Document intptr_t and uintptr_t.

21 years agomakeLINT.sed replaced makeLINT.sh about 4 months ago.
Mike Barcroft [Sun, 15 Sep 2002 19:04:23 +0000 (19:04 +0000)]
makeLINT.sed replaced makeLINT.sh about 4 months ago.

21 years agos/Danglish/English/
Poul-Henning Kamp [Sun, 15 Sep 2002 17:52:35 +0000 (17:52 +0000)]
s/Danglish/English/
Some style issues.
Change the timeout to be hz/10 instead of hz.

Brucification by: bde.

21 years agoRemoved unused includes. Sorted includes. This is part of removing
Bruce Evans [Sun, 15 Sep 2002 17:45:10 +0000 (17:45 +0000)]
Removed unused includes.  Sorted includes.  This is part of removing
includes of <sys/user.h> for its pollution only.  <sys/user.h> wasn't
even used for its pollution here.

21 years agoGarbage-collected splmem.
Bruce Evans [Sun, 15 Sep 2002 17:10:18 +0000 (17:10 +0000)]
Garbage-collected splmem.

Moved the declaration of malloc_mtx to be with the other extern declarations
and not exposed to userland.

Fixed some minor style bugs.

21 years agoCheck for truncation in calls to res_send/res_query/res_search.
Jacques Vidrine [Sun, 15 Sep 2002 16:51:09 +0000 (16:51 +0000)]
Check for truncation in calls to res_send/res_query/res_search.
Fail when it is detected.

21 years agoRemoved most namespace pollution in this header: don't include <vm/uma.h>;
Bruce Evans [Sun, 15 Sep 2002 16:48:25 +0000 (16:48 +0000)]
Removed most namespace pollution in this header: don't include <vm/uma.h>;
include <sys/_mutex.h> and its prerequisites instead of <sys/mutex.h> and
its prerequisite.

21 years agoTry to pick up disk geometry with specific DIOC* ioctls, rather than
Poul-Henning Kamp [Sun, 15 Sep 2002 16:08:52 +0000 (16:08 +0000)]
Try to pick up disk geometry with specific DIOC* ioctls, rather than
expecting a bogo-disklabel to contain them, if possible.

This makes fdisk work with GEOM.

21 years agoAdd "-t" to useage message and comment. (The -t option was added
David Malone [Sun, 15 Sep 2002 15:37:50 +0000 (15:37 +0000)]
Add "-t" to useage message and comment. (The -t option was added
independently by Robert, but also proposed in the PR below).

PR: 38126
Submitted by: Josh Elsasser <jre@vineyard.net>
MFC after: 1 month

21 years agoSome BIOSs are using MTRR values that are only documented under NDA
David Malone [Sun, 15 Sep 2002 15:07:55 +0000 (15:07 +0000)]
Some BIOSs are using MTRR values that are only documented under NDA
to control the mapping of things like the ACPI and APM into memory.

The problem is that starting X changes these values, so if something
was using the bits of BIOS mapped into memory (say ACPI or APM),
then next time they access this memory the machine would hang.

This patch refuse to change MTRR values it doesn't understand,
unless a new "force" option is given. This means X doesn't change
them by accident but someone can override that if they really want
to.

PR: 28418
Tested by: Christopher Masto <chris@netmonger.net>,
David Bushong <david@bushong.net>,
Santos <casd@myrealbox.com>
MFC after: 1 week

21 years agoDefine __LONG_LONG_SUPPORTED if the compiler claims to support C99.
Bruce Evans [Sun, 15 Sep 2002 14:17:29 +0000 (14:17 +0000)]
Define __LONG_LONG_SUPPORTED if the compiler claims to support C99.
Don't set __LONG_LONG_SUPPORTED for gcc-1.  I didn't check exactly
when gcc started to support long long, but it was in the version
of gcc in FreeBSD-1.0 (gcc-2.4.5).

Other C99 features should be ifdefed similarly, but the ifdefs for
old versions of gcc will be more complicated since the features
weren't in all versions of gcc-2.

21 years agoInclude <sys/systm.h> instead of depending on namespace pollution 2
Bruce Evans [Sun, 15 Sep 2002 13:54:12 +0000 (13:54 +0000)]
Include <sys/systm.h> instead of depending on namespace pollution 2
layers deep in <sys/malloc.h> or 1 layer deep in <net/if_var.h>.

21 years agoSome change in src/share/mk seems to have made it possible to remove
Poul-Henning Kamp [Sun, 15 Sep 2002 13:05:42 +0000 (13:05 +0000)]
Some change in src/share/mk seems to have made it possible to remove
50% of the active ingredients in these Makefiles.

21 years agoRemove the unused _bio_buf field. I can't even remember if this ever got
Poul-Henning Kamp [Sun, 15 Sep 2002 12:50:37 +0000 (12:50 +0000)]
Remove the unused _bio_buf field.  I can't even remember if this ever got
to be used in the first place.

Spotted by: bde

21 years agoFix i810 after i830 commit.
Eric Anholt [Sun, 15 Sep 2002 08:43:23 +0000 (08:43 +0000)]
Fix i810 after i830 commit.

Submitted by: David Dawes <dawes@XFree86.Org>

21 years agoAdd wcstod() as a wrapper around strtod(). It does not handle any characters
Tim J. Robbins [Sun, 15 Sep 2002 08:38:51 +0000 (08:38 +0000)]
Add wcstod() as a wrapper around strtod(). It does not handle any characters
that strtod() does not (alternate digit characters, etc. are not handled).

21 years agoFixed breakage from removing the used include of <sys/buf.h> in the
Bruce Evans [Sun, 15 Sep 2002 08:21:14 +0000 (08:21 +0000)]
Fixed breakage from removing the used include of <sys/buf.h> in the
previous commit:
Include <sys/mutex.h> and its prerequisite <sys/lock.h> instead of
depending on namespace pollution in <sys/buf.h>.  Only do this in the
__FreeBSD_version > 500000 case although the __FreeBSD_version < 500000
case has already rotted.

21 years agoUse the heap instead of the stack to store temporary multibyte string
Tim J. Robbins [Sun, 15 Sep 2002 08:06:17 +0000 (08:06 +0000)]
Use the heap instead of the stack to store temporary multibyte string
buffers; this is slower but safer for threaded programs where threads
often have relatively low stack size limits.

21 years agoMake a comment reflect less of a lie, NOCPU is used to generally mean that we
Juli Mallett [Sun, 15 Sep 2002 04:45:56 +0000 (04:45 +0000)]
Make a comment reflect less of a lie, NOCPU is used to generally mean that we
are not on a specific CPU in a distributed system, certainly not just for
<struct proc>.p_oncpu, which is of course now <struct kse>.ke_oncpu.

21 years agoBackout the increase of MAXPACKET from 1024 to 65536: it
Jacques Vidrine [Sun, 15 Sep 2002 04:23:20 +0000 (04:23 +0000)]
Backout the increase of MAXPACKET from 1024 to 65536: it
broke pthreads.

Reported by: mbr, tjr

21 years agoUn-inline the non-trivial "trivial" bio* functions.
Poul-Henning Kamp [Sat, 14 Sep 2002 19:34:11 +0000 (19:34 +0000)]
Un-inline the non-trivial "trivial" bio* functions.
Untangle devstat_end_transaction_bio()

21 years agoAdd additional information returned to userland by the device querying
Warner Losh [Sat, 14 Sep 2002 19:15:12 +0000 (19:15 +0000)]
Add additional information returned to userland by the device querying
functions.  We add pnpinfo, locationinfo, devflags (the newbus flags
on the device), flags (the flags that device_get_flags returns) and
state to the list of things we return.

pnpinfo and locationinfo are place holders at the moment that will be
filled in by the device's parent (optionally).  Userland programs will
likely use this information from time to time and take appropriate
actions.

Improvements to devinfo to follow.

21 years agoremove #includes of <sys/bio.h> where not needed.
Poul-Henning Kamp [Sat, 14 Sep 2002 18:59:32 +0000 (18:59 +0000)]
remove #includes of <sys/bio.h> where not needed.