]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
23 years agoRemoved a wrong comment.
Munechika SUMIKAWA [Thu, 20 Sep 2001 08:04:21 +0000 (08:04 +0000)]
Removed a wrong comment.

Obtained from: KAME
MFC after: 1 week

23 years agoFixed comment: ipip_input -> mroute_encapcheck.
Munechika SUMIKAWA [Thu, 20 Sep 2001 07:59:45 +0000 (07:59 +0000)]
Fixed comment: ipip_input -> mroute_encapcheck.

Reported by: bde

23 years agoCorrect comments to fit current realities of exporting crypto.
Ruslan Ermilov [Thu, 20 Sep 2001 06:45:44 +0000 (06:45 +0000)]
Correct comments to fit current realities of exporting crypto.

MFC after: 3 days

23 years agoFix alpha gdb -k on "live" kernels. Use offsetof() instead of some evil
Peter Wemm [Thu, 20 Sep 2001 06:31:23 +0000 (06:31 +0000)]
Fix alpha gdb -k on "live" kernels.  Use offsetof() instead of some evil
hand-rolled macros to do the same thing.

23 years agoShared libraries from 4.4-FreeBSD needed for proper binary compatibility.
David E. O'Brien [Thu, 20 Sep 2001 06:22:09 +0000 (06:22 +0000)]
Shared libraries from 4.4-FreeBSD needed for proper binary compatibility.

23 years agodecrement the dumping variable after use so we can call it several times
Peter Wemm [Thu, 20 Sep 2001 06:08:53 +0000 (06:08 +0000)]
decrement the dumping variable after use so we can call it several times
if needed.

23 years agoAdd nfs_lock.c to the build list. *blush*
Peter Wemm [Thu, 20 Sep 2001 05:32:34 +0000 (05:32 +0000)]
Add nfs_lock.c to the build list. *blush*

23 years agoFinish KSE-ification updates.
Peter Wemm [Thu, 20 Sep 2001 05:15:04 +0000 (05:15 +0000)]
Finish KSE-ification updates.

23 years agoWrap a module around the init code so that we have somethign do do a
Peter Wemm [Thu, 20 Sep 2001 05:13:43 +0000 (05:13 +0000)]
Wrap a module around the init code so that we have somethign do do a
modfind(2) on, and declare a version so that loader/kldload etc
can locate it (using kldxref's linker.hints file if needed).

23 years agoAdd probe line for the AMD 761 northbridge chip. At least it detects now,
Coleman Kane [Thu, 20 Sep 2001 05:13:12 +0000 (05:13 +0000)]
Add probe line for the AMD 761 northbridge chip. At least it detects now,
seems to set up memory spaces correctly. This change actually did work for
me using -STABLE, XFree86 4.0.3 ~ some snapshot of DRI awhile back. I sent
mail to dfr to no avail, perhaps someone else would like to test it with
DRI.

Anyway, people have been nagging me about this change for awhile, so here's
the commit.

23 years agoRegenerate. This seems to have been missed in the last commit.
Peter Wemm [Thu, 20 Sep 2001 05:01:08 +0000 (05:01 +0000)]
Regenerate.  This seems to have been missed in the last commit.

23 years agoAdd the magic marker so that loader and kldload(2) can find this in
Peter Wemm [Thu, 20 Sep 2001 04:57:34 +0000 (04:57 +0000)]
Add the magic marker so that loader and kldload(2) can find this in
module form automagically.

23 years agoPass 1: update for KSE and NFS changes. Somehow this got missed.
Peter Wemm [Thu, 20 Sep 2001 04:56:42 +0000 (04:56 +0000)]
Pass 1: update for KSE and NFS changes.  Somehow this got missed.
osf1_mount.c isn't complete yet.

23 years agoMake this compile again. I would have sworn that I did this already in
Peter Wemm [Thu, 20 Sep 2001 04:55:06 +0000 (04:55 +0000)]
Make this compile again. I would have sworn that I did this already in
the KSE tree and thought it had been merged.

23 years agoOops. Fix a missing indirection level. gcc didn't complain about it on
Peter Wemm [Thu, 20 Sep 2001 03:45:51 +0000 (03:45 +0000)]
Oops.  Fix a missing indirection level.  gcc didn't complain about it on
x86, but did complain about it on alpha (since int and pointer are
different sizes)

23 years agoMake the script aware of mdconfig, so it can hopefully run on -CURRENT
Luigi Rizzo [Thu, 20 Sep 2001 02:52:02 +0000 (02:52 +0000)]
Make the script aware of mdconfig, so it can hopefully run on -CURRENT
as well. This works by selecting "md" or "vn" depending on "uname -r"
output, so we can use the same script on -CURRENT and -STABLE.
Also included minor bugfixes and code cleanup.

Testers welcome, as this code has only been tested on -STABLE
(and for this reason I am doing an immediate MFC).

23 years agoReplicate a change from alpha/genassym.c to other arches. This should
Peter Wemm [Thu, 20 Sep 2001 02:49:59 +0000 (02:49 +0000)]
Replicate a change from alpha/genassym.c to other arches. This should
fix nfs-related build breakage.

23 years agoMake alpha/genassym.c compile again. *oops*
Peter Wemm [Thu, 20 Sep 2001 02:48:03 +0000 (02:48 +0000)]
Make alpha/genassym.c compile again. *oops*

23 years agoDeal with nfs server module changes for autoloading.
Peter Wemm [Thu, 20 Sep 2001 02:18:06 +0000 (02:18 +0000)]
Deal with nfs server module changes for autoloading.

23 years agoDeal with module name changes and autoloading.
Peter Wemm [Thu, 20 Sep 2001 02:15:17 +0000 (02:15 +0000)]
Deal with module name changes and autoloading.

23 years agoWelcome 4.4-FreeBSD to the family.
David E. O'Brien [Thu, 20 Sep 2001 02:13:51 +0000 (02:13 +0000)]
Welcome 4.4-FreeBSD to the family.

23 years agoI forgot to rm the nfs module directory.
Peter Wemm [Thu, 20 Sep 2001 01:11:18 +0000 (01:11 +0000)]
I forgot to rm the nfs module directory.

Submitted by:  mux@qualys.com

23 years agoAdd minimal support for "o"-type print-file actions in lpd control files.
Garance A Drosehn [Thu, 20 Sep 2001 01:04:50 +0000 (01:04 +0000)]
Add minimal support for "o"-type print-file actions in lpd control files.
This was described in the original RFC wrt lpr, but most lpr's do not
actually implement it.  There is some indication that MacOS 10.1 will
be using this when sending postscript files to print servers (that is
what "o"-type was supposed to signify -- postscript files).

MFC after: 1 week

23 years agoUse BATL/BATU macros instead of hardcoded hex constants.
Mark Peek [Thu, 20 Sep 2001 00:48:30 +0000 (00:48 +0000)]
Use BATL/BATU macros instead of hardcoded hex constants.

23 years agoUpdate PowerPC MD code to compile and do initial bootstrap based on
Mark Peek [Thu, 20 Sep 2001 00:47:17 +0000 (00:47 +0000)]
Update PowerPC MD code to compile and do initial bootstrap based on
recent changes (KSE and VM requiring physmem to be setup).

Reviewed by: benno, jhb, julian

23 years agoFix a bug in propagate priority: the kse group pointer wasn't being
John Baldwin [Wed, 19 Sep 2001 22:52:59 +0000 (22:52 +0000)]
Fix a bug in propagate priority: the kse group pointer wasn't being
updated in the loop so the new thread always seemd to have the same
priority as the original thread and no actual priorities were changed.

23 years agoHandle the RX FIFO overflow condition with the rxeof handler instead
Bill Paul [Wed, 19 Sep 2001 21:39:26 +0000 (21:39 +0000)]
Handle the RX FIFO overflow condition with the rxeof handler instead
of the rxeoc handler for now. The rxeoc handler will reset the link,
and the NatSemi chip's RX FIFO will overflow on a 32-bit bus once you
start hitting it with 500Mbps or more of traffic.

Also increase the size of the RX ring to 128 descriptors (was 64).

23 years agoAdd a new rc.conf variable, cloned_interfaces, to create cloned
Brooks Davis [Wed, 19 Sep 2001 21:27:27 +0000 (21:27 +0000)]
Add a new rc.conf variable, cloned_interfaces, to create cloned
interfaces at boot.

23 years agoDon't check uninitialized memory for having the shell control character
Tor Egge [Wed, 19 Sep 2001 20:16:38 +0000 (20:16 +0000)]
Don't check uninitialized memory for having the shell control character
value CTLARI since this might break expansion of arithmetic expressions.

Don't access memory below start of stackblock.

Problem analyzed by hunt@iprg.nokia.com, slightly different patch applied.

PR: 24443
Submitted by: hunt@iprg.nokia.com

23 years agoBASESYNTAX, DQSYNTAX, SQSYNTAX and ARISYNTAX handles negative indexes.
Tor Egge [Wed, 19 Sep 2001 20:07:47 +0000 (20:07 +0000)]
BASESYNTAX, DQSYNTAX, SQSYNTAX and ARISYNTAX handles negative indexes.
Allow those to be used to properly quote characters in the shell
control character range.

23 years agoAdjust some type checks to include CTLQUOTEMARK in the range of
Tor Egge [Wed, 19 Sep 2001 19:58:09 +0000 (19:58 +0000)]
Adjust some type checks to include CTLQUOTEMARK in the range of
shell control characters.

23 years agoDon't try to dereference a kernel pointer in userland; use offsetof()
Bill Fenner [Wed, 19 Sep 2001 18:42:19 +0000 (18:42 +0000)]
Don't try to dereference a kernel pointer in userland; use offsetof()
 to get the right address.

This fixes kernel GDB after KSE2.

Reviewed by: jhb, jake

23 years agoSolve our braindead sourcetree dependency on GCC in another way.
Jeroen Ruigrok van der Werven [Wed, 19 Sep 2001 17:51:23 +0000 (17:51 +0000)]
Solve our braindead sourcetree dependency on GCC in another way.

Discussed with: bde

23 years agoNon-decimal ``skipto'' rule numbers are meaningless.
Ruslan Ermilov [Wed, 19 Sep 2001 15:12:14 +0000 (15:12 +0000)]
Non-decimal ``skipto'' rule numbers are meaningless.

Noticed by: "Marc G. Fournier" <scrappy@hub.org>
MFC after: 3 days

23 years agoo Clarification of securelevel_{ge,gt} comment.
Robert Watson [Wed, 19 Sep 2001 14:09:13 +0000 (14:09 +0000)]
o Clarification of securelevel_{ge,gt} comment.

Submitted by: dd

23 years agoFix abuse of vtagtype. In addition, after this the linux programs will be
Maxim Sobolev [Wed, 19 Sep 2001 12:35:51 +0000 (12:35 +0000)]
Fix abuse of vtagtype. In addition, after this the linux programs will be
able correctly distinguish ext2fs from the ufs filesystem (previously ext2fs
was indistinguishable from the ufs).

Reviewed by: phk, marcel

23 years agoImplement time().
Doug Rabson [Wed, 19 Sep 2001 12:12:12 +0000 (12:12 +0000)]
Implement time().

23 years agoVarious fixes and improvements:
Maxim Sobolev [Wed, 19 Sep 2001 08:06:48 +0000 (08:06 +0000)]
Various fixes and improvements:
- fix harmless compiler's warnings (unused variables and missed prototype);
- before refusing to delete package because "there are packages installed
  that require this package" check that packages in question is actually
  installed;
- add new `-r' option to pkg_delete(8), which instructs it to delete not only
  packages specified at command line, but all packages that depend on
  specified packages as well.

MFC after: 2 weeks

23 years agoIn FreeBSD, -Z is reserved for zgrep(1), and only
Ruslan Ermilov [Wed, 19 Sep 2001 07:22:30 +0000 (07:22 +0000)]
In FreeBSD, -Z is reserved for zgrep(1), and only
a long version of the --null option is supported.

PR: gnu/30644

23 years agoReserve an extra 16 bytes in case we have to grow the trapframe into
Peter Wemm [Wed, 19 Sep 2001 05:44:12 +0000 (05:44 +0000)]
Reserve an extra 16 bytes in case we have to grow the trapframe into
a vm86trapframe for switching to vm86 [unlikely] while exiting.
I lost this when doing the pcb move that went in with the KSE commit.

Reviewed by: jake

23 years agoFix a mistake I made with the pcb movement relative to the stack in the
Peter Wemm [Wed, 19 Sep 2001 05:34:12 +0000 (05:34 +0000)]
Fix a mistake I made with the pcb movement relative to the stack in the
KSE patch.  We need to leave the 16 bytes here for enabling the trapframe
to be converted to a vm86trapframe if we're switching *to* a vm86 context.

23 years agoAdd missing ; in last commit
Peter Wemm [Wed, 19 Sep 2001 02:53:59 +0000 (02:53 +0000)]
Add missing ; in last commit

Pointy-hat-to: jhb
23 years agoAdd a quirk entry for the Sony CLIE memory stick device. This will become
Scott Long [Wed, 19 Sep 2001 00:35:09 +0000 (00:35 +0000)]
Add a quirk entry for the Sony CLIE memory stick device.  This will become
useful once ATAPI support is turned on in the umass driver.

23 years agoThe vfs.nfs.bufpackets sysctl is in the client, not the server. Move it
Peter Wemm [Wed, 19 Sep 2001 00:22:26 +0000 (00:22 +0000)]
The vfs.nfs.bufpackets sysctl is in the client, not the server.  Move it
to the client section.  Turn off nfsiod, it no longer exists (now just
kthreads).  I need revisit nfsiod so that we have an argument passthrough.

23 years agoMake stf a clonable device.
Brooks Davis [Wed, 19 Sep 2001 00:13:00 +0000 (00:13 +0000)]
Make stf a clonable device.

Yes this really is rather silly and the implementation is overkill given
that you are only allowed one of them, but NetBSD implements cloning on
this device and it's a less cluttered example of cloning then most.

23 years agoSigh, Last minute pre-merge typo. (missing quotes)
Peter Wemm [Tue, 18 Sep 2001 23:49:33 +0000 (23:49 +0000)]
Sigh, Last minute pre-merge typo.  (missing quotes)

23 years agoUserland part of nfs client/server split and cleanup.
Peter Wemm [Tue, 18 Sep 2001 23:34:44 +0000 (23:34 +0000)]
Userland part of nfs client/server split and cleanup.

23 years agoRegenerate
Peter Wemm [Tue, 18 Sep 2001 23:33:33 +0000 (23:33 +0000)]
Regenerate

23 years agoCleanup and split of nfs client and server code.
Peter Wemm [Tue, 18 Sep 2001 23:32:09 +0000 (23:32 +0000)]
Cleanup and split of nfs client and server code.
This builds on the top of several repo-copies.

23 years agoUse a 'p' variable instead of repetitively indirecting td->td_proc for
John Baldwin [Tue, 18 Sep 2001 23:27:06 +0000 (23:27 +0000)]
Use a 'p' variable instead of repetitively indirecting td->td_proc for
signal things that are still per-process and won't be per-thread.

23 years agoRemove accidentally included debug stuff.
Julian Elischer [Tue, 18 Sep 2001 22:52:42 +0000 (22:52 +0000)]
Remove accidentally included debug stuff.

Noticed by: jhb

23 years agoRegen.
Nick Hibma [Tue, 18 Sep 2001 22:30:02 +0000 (22:30 +0000)]
Regen.

23 years agoUpdate the header of the file with a comment on how to regenerate
Nick Hibma [Tue, 18 Sep 2001 22:27:41 +0000 (22:27 +0000)]
Update the header of the file with a comment on how to regenerate
the usbdevs* file correctly.

Suggested by: phantom

23 years agoDon't initialize proc0's mutex twice. It is already done earlier on in the
John Baldwin [Tue, 18 Sep 2001 22:09:47 +0000 (22:09 +0000)]
Don't initialize proc0's mutex twice.  It is already done earlier on in the
MD startup code.

23 years agoGC obsolete cruft from this file.
John Baldwin [Tue, 18 Sep 2001 21:53:25 +0000 (21:53 +0000)]
GC obsolete cruft from this file.

23 years agoWhitespace fixes.
John Baldwin [Tue, 18 Sep 2001 21:52:15 +0000 (21:52 +0000)]
Whitespace fixes.

23 years ago- If we ever do the per-cpu KTR stuff, the index won't be volatile as it
John Baldwin [Tue, 18 Sep 2001 21:46:26 +0000 (21:46 +0000)]
- If we ever do the per-cpu KTR stuff, the index won't be volatile as it
  will be private to each CPU.
- Re-style(9) the globaldata structures.  There really needs to be a MI
  struct pcpu that has a MD struct mdpcpu member at some point.

23 years ago- Fix a missed idleproc -> idlethread conversion.
John Baldwin [Tue, 18 Sep 2001 21:37:09 +0000 (21:37 +0000)]
- Fix a missed idleproc -> idlethread conversion.
- Remove redundany fpucurproc (fpucurthread already existed)

23 years agoIf we have a p variable pointing to td->td_proc already, use it instead of
John Baldwin [Tue, 18 Sep 2001 21:20:58 +0000 (21:20 +0000)]
If we have a p variable pointing to td->td_proc already, use it instead of
needlessly repeating the indirection in several places.  Half of the places
used td->td_proc, and half used p.  They are now consistent with each other
and all use p.

23 years agoWhitespace fixes.
John Baldwin [Tue, 18 Sep 2001 21:05:04 +0000 (21:05 +0000)]
Whitespace fixes.

23 years agoo Introduce two new calls, securelevel_gt() and securelevel_ge(), which
Robert Watson [Tue, 18 Sep 2001 21:03:53 +0000 (21:03 +0000)]
o Introduce two new calls, securelevel_gt() and securelevel_ge(), which
  abstract the securelevel implementation details from the checking
  code.  The call in -CURRENT accepts a struct ucred--in -STABLE, it
  will accept struct proc.  This facilitates the upcoming commit of
  per-jail securelevel support.  The calls will also generate a
  kernel printf if the calls are made with NULL ucred/proc pointers:
  generally speaking, there are few instances of this, and they should
  be fixed.
o Update p_candebug() to use securelevel_gt(); future updates to the
  remainder of the kernel tree will be committed soon.

Obtained from: TrustedBSD Project

23 years agoHave the driver announce its rxcsum capabilities.
Jonathan Lemon [Tue, 18 Sep 2001 20:19:28 +0000 (20:19 +0000)]
Have the driver announce its rxcsum capabilities.

23 years agoSplit hwcsum into rxcsum and txcsum components.
Jonathan Lemon [Tue, 18 Sep 2001 20:13:48 +0000 (20:13 +0000)]
Split hwcsum into rxcsum and txcsum components.

23 years agoSplit HWCSUM into two components: RX and TX, for the benefit of drivers
Jonathan Lemon [Tue, 18 Sep 2001 20:13:03 +0000 (20:13 +0000)]
Split HWCSUM into two components: RX and TX, for the benefit of drivers
which can only do checksum offloading in one direction.

23 years agoo Remove redundant securelevel/pid1 check in procfs_rw() -- this
Robert Watson [Tue, 18 Sep 2001 19:53:10 +0000 (19:53 +0000)]
o Remove redundant securelevel/pid1 check in procfs_rw() -- this
  protection is enforced at the invidual method layer using
  p_candebug().

Obtained from: TrustedBSD Project

23 years agoSet debug information on the process being traced, not the current (debugger)
Mark Peek [Tue, 18 Sep 2001 19:06:11 +0000 (19:06 +0000)]
Set debug information on the process being traced, not the current (debugger)
process. This should allow gdb to function correctly on post-KSE kernels.

23 years agoHave the driver to announce what capabilitis it supports.
Jonathan Lemon [Tue, 18 Sep 2001 18:53:41 +0000 (18:53 +0000)]
Have the driver to announce what capabilitis it supports.

23 years agoHave the driver announce what capabilities it supports. These are
Jonathan Lemon [Tue, 18 Sep 2001 18:41:39 +0000 (18:41 +0000)]
Have the driver announce what capabilities it supports.  These are
currently not under user control.

23 years agoHave the driver advertise that it is capable of hardware checksums, and
Jonathan Lemon [Tue, 18 Sep 2001 18:40:22 +0000 (18:40 +0000)]
Have the driver advertise that it is capable of hardware checksums, and
allow the user to control the setting.  (checksums are enabled by default)

23 years agonfs_strategy calls nfs_asyncio with td as NULL. So add a bandaid that
Warner Losh [Tue, 18 Sep 2001 18:37:52 +0000 (18:37 +0000)]
nfs_strategy calls nfs_asyncio with td as NULL.  So add a bandaid that
will pass NULL as the struct proc when td is NULL.  This has stopped
crashing on my machine.

Note: The passing of NULL may be bogus, but I'll let others fix that
problem.

Reviewed by: jhb

23 years agoDrop maintainership.
Bruce A. Mah [Tue, 18 Sep 2001 17:58:10 +0000 (17:58 +0000)]
Drop maintainership.

23 years agoSupport the use of ${PORTSDIR} to find a ports tree INDEX file in
Bruce A. Mah [Tue, 18 Sep 2001 17:56:44 +0000 (17:56 +0000)]
Support the use of ${PORTSDIR} to find a ports tree INDEX file in
an other-than-default location.

PR: bin/30350
Submitted by: "NINOMIYA Hideyuki <NINOMIYA Hideyuki" <nin@jp.FreeBSD.org>
MFC after: 3 days

23 years agoTeach ifconfig about the new interface capability words.
Jonathan Lemon [Tue, 18 Sep 2001 17:43:30 +0000 (17:43 +0000)]
Teach ifconfig about the new interface capability words.

23 years agoAdd SIO{CG}IFCAP entries. Fix formatting of some other entries.
Jonathan Lemon [Tue, 18 Sep 2001 17:42:25 +0000 (17:42 +0000)]
Add SIO{CG}IFCAP entries.  Fix formatting of some other entries.

23 years agoAdd two fields to the ifnet structure indicating what extra capabilities
Jonathan Lemon [Tue, 18 Sep 2001 17:41:42 +0000 (17:41 +0000)]
Add two fields to the ifnet structure indicating what extra capabilities
a network device has, and which ones are enabled.

23 years agoTry to work off backlog of CURRENT release notes: vm.zeroidle_enable,
Bruce A. Mah [Tue, 18 Sep 2001 15:25:00 +0000 (15:25 +0000)]
Try to work off backlog of CURRENT release notes:  vm.zeroidle_enable,
an(4)/ancontrol(8) -M monitor mode, burncd(8) -d, cat(1) UNIX-domain
sockets.

23 years agoAdd ia64_get_cpuid().
Doug Rabson [Tue, 18 Sep 2001 15:21:18 +0000 (15:21 +0000)]
Add ia64_get_cpuid().

23 years agowhen setting blocksize, fix blockcount to 2 to reduce latency with smaller
Cameron Grant [Tue, 18 Sep 2001 15:19:44 +0000 (15:19 +0000)]
when setting blocksize, fix blockcount to 2 to reduce latency with smaller
block sizes.

23 years agomake buffer size adjustable.
Cameron Grant [Tue, 18 Sep 2001 15:12:35 +0000 (15:12 +0000)]
make buffer size adjustable.

23 years agodon't enable the mic record channel unless the codec supports it.
Cameron Grant [Tue, 18 Sep 2001 14:53:01 +0000 (14:53 +0000)]
don't enable the mic record channel unless the codec supports it.

allow hardware buffer size to be specified using hints.

23 years agoAdd support for loading bzip2-compressed kernels and modules. This support
Maxim Sobolev [Tue, 18 Sep 2001 14:52:36 +0000 (14:52 +0000)]
Add support for loading bzip2-compressed kernels and modules. This support
is turned off by default and could be enabled by defining LOADER_BZIP2_SUPPORT
make variable. Also make gzip support optional (turned on by default) -
it could be turned off via LOADER_NO_GZIP_SUPPORT make variable.

Please note, that due to limit on the amount of memory available to the
loader(8), it is possible to load modules/kernels compressed with the smallest
block size supported by the bzip2 - 100k (`-1' bzip2(1) option), however
even in this mode bzip2(1) usually provides better compression ratio than
gzip(1) in its best compression mode.

MFC after: 1 month

23 years agoRemoved ipip_input(). No codes calls it anymore due to ip_encap.c's
Munechika SUMIKAWA [Tue, 18 Sep 2001 14:52:20 +0000 (14:52 +0000)]
Removed ipip_input(). No codes calls it anymore due to ip_encap.c's
encapsulation support.

23 years agochange tracking of channel counts.
Cameron Grant [Tue, 18 Sep 2001 14:50:52 +0000 (14:50 +0000)]
change tracking of channel counts.

add method for retrieving "buffersize" hints for pcm devices, adjusted for
specified minimum, maximum and default values.

23 years agoadd a tunable/sysctl, hw.snd.targetirqrate, to modify the default target
Cameron Grant [Tue, 18 Sep 2001 14:45:09 +0000 (14:45 +0000)]
add a tunable/sysctl, hw.snd.targetirqrate, to modify the default target
irq rate for apps that do not set a blocksize.

23 years agoadd a method for retrieving codec capabilities
Cameron Grant [Tue, 18 Sep 2001 14:34:59 +0000 (14:34 +0000)]
add a method for retrieving codec capabilities

23 years agoFlesh out identifycpu().
Doug Rabson [Tue, 18 Sep 2001 13:26:30 +0000 (13:26 +0000)]
Flesh out identifycpu().

23 years agoAdd support for loading bzip2-compressed filesystems. Among other things
Maxim Sobolev [Tue, 18 Sep 2001 13:01:12 +0000 (13:01 +0000)]
Add support for loading bzip2-compressed filesystems. Among other things
this would allow to load bzip2-compressed kernels/modules from the loader(8)
(support for that will be committer separately).

MFC after: 1 month

23 years agoStyle.
Ruslan Ermilov [Tue, 18 Sep 2001 12:28:22 +0000 (12:28 +0000)]
Style.

23 years agoStyle.
Ruslan Ermilov [Tue, 18 Sep 2001 12:03:42 +0000 (12:03 +0000)]
Style.

23 years agoKLDMOD is not used since share/mk/bsd.kmod.mk,v 1.60.
Ruslan Ermilov [Tue, 18 Sep 2001 11:03:35 +0000 (11:03 +0000)]
KLDMOD is not used since share/mk/bsd.kmod.mk,v 1.60.

Forgotten by: peter

23 years agoAdd release note entry about the amdpm(4) driver.
Murray Stokely [Tue, 18 Sep 2001 04:55:25 +0000 (04:55 +0000)]
Add release note entry about the amdpm(4) driver.

23 years agoChange p into ke->ke_proc, this was hidden behind INVARIANTS.
Jonathan Lemon [Tue, 18 Sep 2001 03:36:21 +0000 (03:36 +0000)]
Change p into ke->ke_proc, this was hidden behind INVARIANTS.

23 years agoAdd 'device smb' to the synopsis, as this is required to do anything
Murray Stokely [Tue, 18 Sep 2001 02:50:24 +0000 (02:50 +0000)]
Add 'device smb' to the synopsis, as this is required to do anything
useful with these devices.

23 years agoRemove hard coded magic load address. Now to change the load address,
Peter Wemm [Tue, 18 Sep 2001 01:12:43 +0000 (01:12 +0000)]
Remove hard coded magic load address.  Now to change the load address,
we just have to change the pmap.h constants and ld will automatically
adapt based on the "kernbase" symbol.

23 years agoTweak the autoneg kickoff code to that it more closely resembles the
Bill Paul [Tue, 18 Sep 2001 00:31:19 +0000 (00:31 +0000)]
Tweak the autoneg kickoff code to that it more closely resembles the
method uses in the nsgphy driver.

23 years agoHandle the absence of net.inet.ip.fw.verbose_limit better, just like
Bill Fenner [Tue, 18 Sep 2001 00:03:48 +0000 (00:03 +0000)]
Handle the absence of net.inet.ip.fw.verbose_limit better, just like
 brian's fix for v6 in rev 1.56.

23 years agoFix a fatal type mismatch (char *static_env; vs char static_env[]).
Peter Wemm [Mon, 17 Sep 2001 21:27:41 +0000 (21:27 +0000)]
Fix a fatal type mismatch (char *static_env; vs char static_env[]).

Submitted by: bde

23 years agoUse includes to get prototypes for hints and env arrays.
Peter Wemm [Mon, 17 Sep 2001 21:25:48 +0000 (21:25 +0000)]
Use includes to get prototypes for hints and env arrays.

Submitted by: bde

23 years agoUse NULL instead of __FILE__ in the !LOCK_DEBUG case in the locking code
John Baldwin [Mon, 17 Sep 2001 21:21:02 +0000 (21:21 +0000)]
Use NULL instead of __FILE__ in the !LOCK_DEBUG case in the locking code
since the filenames are only used in the LOCK_DEBUG case and are just bloat
in the !LOCK_DEBUG case.

23 years agoFix multiple uses of commas to separate countries. This construct is
Greg Lehey [Mon, 17 Sep 2001 21:19:28 +0000 (21:19 +0000)]
Fix multiple uses of commas to separate countries.  This construct is
commonly used in the United States of America to represent a
hierarchical relationship between city and state or country
("Evacuation Day in Suffolk County, Massachusetts"), but it was also
being abused to enumerate unrelated places ("Independence Day in
Albania, Mauritania").

Remove the list of countries observing All Souls' Day.  This is a
widely observed holiday, and the list only included South American
contries.

There's more to do here; the list is rather patchy.