]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
19 years agoAdd power profile support so that the LCD changes brightness levels based
njl [Sun, 7 Nov 2004 23:18:23 +0000 (23:18 +0000)]
Add power profile support so that the LCD changes brightness levels based
on the AC line state.

Submitted by: OGAWA Takaya <t-ogawa@triaez.kaisei.org>
MFC after: 1 week

19 years agoAdd basic critical section tracing to KTR using event type KTR_CRITICAL.
rwatson [Sun, 7 Nov 2004 23:11:32 +0000 (23:11 +0000)]
Add basic critical section tracing to KTR using event type KTR_CRITICAL.
This generates a KTR event for each critical section entered and exited.

It would be desirable to also log the filename and line number of the
source entering or exiting the critical section, but this requires
hacking up the critical section API, so I've not done that yet.

19 years agoAdd device mem in config files.
cognet [Sun, 7 Nov 2004 23:02:15 +0000 (23:02 +0000)]
Add device mem in config files.

19 years agoImport md bits for mem(4) on arm.
cognet [Sun, 7 Nov 2004 23:01:36 +0000 (23:01 +0000)]
Import md bits for mem(4) on arm.
While I'm there, cleanup a bit pmap.h.

19 years agoRemove trailing whitespace.
kan [Sun, 7 Nov 2004 22:49:20 +0000 (22:49 +0000)]
Remove trailing whitespace.

Pointed out by: ru

19 years agoIntroduce fdclose() which will clean an entry in a filedesc.
phk [Sun, 7 Nov 2004 22:16:07 +0000 (22:16 +0000)]
Introduce fdclose() which will clean an entry in a filedesc.

Replace homerolled versions with call to fdclose().

Make fdunused() static to kern_descrip.c

19 years agofiledesc0 is an internal detail of the kernel, don't look at it.
phk [Sun, 7 Nov 2004 20:38:29 +0000 (20:38 +0000)]
filedesc0 is an internal detail of the kernel, don't look at it.

19 years agoDo some re-sorting of TCP pcbinfo locking and assertions: make sure to
rwatson [Sun, 7 Nov 2004 19:19:35 +0000 (19:19 +0000)]
Do some re-sorting of TCP pcbinfo locking and assertions: make sure to
retain the pcbinfo lock until we're done using a pcb in the in-bound
path, as the pcbinfo lock acts as a pseuo-reference to prevent the pcb
from potentially being recycled.  Clean up assertions and make sure to
assert that the pcbinfo is locked at the head of code subsections where
it is needed.  Free the mbuf at the end of tcp_input after releasing
any held locks to reduce the time the locks are held.

MFC after: 3 weeks

19 years agoTrim the relnotes items added before 5.3-RELEASE.
hrs [Sun, 7 Nov 2004 18:17:38 +0000 (18:17 +0000)]
Trim the relnotes items added before 5.3-RELEASE.

19 years agoBump &release.manpath.netbsd; to 1.6.2.
hrs [Sun, 7 Nov 2004 18:12:57 +0000 (18:12 +0000)]
Bump &release.manpath.netbsd; to 1.6.2.

19 years agoDocument "loader_color" variable.
krion [Sun, 7 Nov 2004 17:01:08 +0000 (17:01 +0000)]
Document "loader_color" variable.

PR: conf/73396
Submitted by: Aleksander Fafula <alex at fafula dot com>
Approved by: scottl

19 years agoMFKernel: Implement ffs with clz on Xscale.
cognet [Sun, 7 Nov 2004 16:54:54 +0000 (16:54 +0000)]
MFKernel: Implement ffs with clz on Xscale.

19 years agoImplement ffs with clz for Xscale.
cognet [Sun, 7 Nov 2004 16:53:29 +0000 (16:53 +0000)]
Implement ffs with clz for Xscale.
Idea taken from: NetBSD

19 years agoRemove useless code.
cognet [Sun, 7 Nov 2004 16:51:33 +0000 (16:51 +0000)]
Remove useless code.

19 years agoUse fget_locked() instead of homerolled
phk [Sun, 7 Nov 2004 16:09:56 +0000 (16:09 +0000)]
Use fget_locked() instead of homerolled

19 years agoMore sensible FILEDESC_ locking.
phk [Sun, 7 Nov 2004 15:59:27 +0000 (15:59 +0000)]
More sensible FILEDESC_ locking.

19 years agoMove fdinit() related stuff from .h to .c
phk [Sun, 7 Nov 2004 15:34:45 +0000 (15:34 +0000)]
Move fdinit() related stuff from .h to .c

19 years agoDon't abuse tp->t_sc in sl(4) either.
cognet [Sun, 7 Nov 2004 14:36:47 +0000 (14:36 +0000)]
Don't abuse tp->t_sc in sl(4) either.

19 years agoDon't abuse tp->t_sc, as it is now used by tty drivers.
cognet [Sun, 7 Nov 2004 14:35:53 +0000 (14:35 +0000)]
Don't abuse tp->t_sc, as it is now used by tty drivers.
This fixes the panic that occurs when using ppp(4)

Reported and tested by: Yann Berthier (yb at sainte-barbe dot org)

19 years agoReconnect c++filt to the build.
kan [Sun, 7 Nov 2004 13:46:56 +0000 (13:46 +0000)]
Reconnect c++filt to the build.

19 years agoUnbreak c++filt. It was disconnected from the build and never fixed during
kan [Sun, 7 Nov 2004 13:46:16 +0000 (13:46 +0000)]
Unbreak c++filt. It was disconnected from the build and never fixed during
GCC 3.4.2 work unintentioanlly.

Reminded by: bms

19 years agoAvoid second person usage.
brueffer [Sun, 7 Nov 2004 13:40:27 +0000 (13:40 +0000)]
Avoid second person usage.

Obtained from: OpenBSD
MFC after: 1 week

19 years agoAllow fdinit() to be called with a NULL fdp argument so we can use
phk [Sun, 7 Nov 2004 12:39:28 +0000 (12:39 +0000)]
Allow fdinit() to be called with a NULL fdp argument so we can use
it when setting up init.

Make fdinit() lock the fdp argument as needed.

19 years agoInstall in /usr/local/bin.
des [Sun, 7 Nov 2004 11:09:44 +0000 (11:09 +0000)]
Install in /usr/local/bin.

19 years agoAdd comments to clarify why we need to run shutdown code on the BSP, update
njl [Sun, 7 Nov 2004 06:58:45 +0000 (06:58 +0000)]
Add comments to clarify why we need to run shutdown code on the BSP, update
an old comment about boot() being MI, and note that splhigh() no longer
disables interrupts.

19 years agoReflect that -h takes an argument and belatedly bump .Dd for addition of -h
dd [Sun, 7 Nov 2004 04:34:37 +0000 (04:34 +0000)]
Reflect that -h takes an argument and belatedly bump .Dd for addition of -h

19 years agoMake the usage message match reality about -h and -w.
dd [Sun, 7 Nov 2004 04:32:51 +0000 (04:32 +0000)]
Make the usage message match reality about -h and -w.

19 years agoFix the release date of 5.3-RELEASE.
hrs [Sun, 7 Nov 2004 03:55:47 +0000 (03:55 +0000)]
Fix the release date of 5.3-RELEASE.

19 years agoWhen we open /dev/null for stdin/out/err for safety reasons, do it right:
phk [Sat, 6 Nov 2004 23:36:09 +0000 (23:36 +0000)]
When we open /dev/null for stdin/out/err for safety reasons, do it right:
we should preserve f_data and f_ops if they are already set.

19 years agoEliminate an unnecessary atomic operation. Articulate the rationale in
alc [Sat, 6 Nov 2004 21:48:45 +0000 (21:48 +0000)]
Eliminate an unnecessary atomic operation.  Articulate the rationale in
a comment.

19 years agoFreeBSD does have uintmax_t, so it can correctly
kientzle [Sat, 6 Nov 2004 18:38:13 +0000 (18:38 +0000)]
FreeBSD does have uintmax_t, so it can correctly
print out very large file sizes.

Thanks to: Yar Tikhiy

19 years agoGet even closer to not crashing ext2fs
phk [Sat, 6 Nov 2004 18:24:33 +0000 (18:24 +0000)]
Get even closer to not crashing ext2fs

19 years agoGet closer to unbreaking ext2fs
phk [Sat, 6 Nov 2004 18:18:58 +0000 (18:18 +0000)]
Get closer to unbreaking ext2fs

19 years agoANSI'fy nfsd(8) and some minor changes to make it WARNS=6 clean.
delphij [Sat, 6 Nov 2004 18:15:44 +0000 (18:15 +0000)]
ANSI'fy nfsd(8) and some minor changes to make it WARNS=6 clean.

This commit does not affect the code generated, as proven by md5'ing
resulting binaries.

Bump WARNS accordingly.

Compiled on: sparc64, ia64, i386
Reviewed by: alfred (but blame me if anything goes wrong :-)

19 years agoAdd description of all available options.
pjd [Sat, 6 Nov 2004 17:19:55 +0000 (17:19 +0000)]
Add description of all available options.

19 years agoAllow to write random data while testing RAID device, so we can also
pjd [Sat, 6 Nov 2004 17:10:24 +0000 (17:10 +0000)]
Allow to write random data while testing RAID device, so we can also
test data integrity.

19 years agoAdd ftp8.de.freebsd.org
josef [Sat, 6 Nov 2004 14:38:54 +0000 (14:38 +0000)]
Add ftp8.de.freebsd.org

Requested by: vs
Approved by: ssouhlal

19 years agoBump .Dd, replace missing letter, and prefer to use a verb that
dd [Sat, 6 Nov 2004 14:03:37 +0000 (14:03 +0000)]
Bump .Dd, replace missing letter, and prefer to use a verb that
matches the option letter.

Submitted by: ru
Should know better by now: dd

19 years agoBack out recent TTYDEF_LFLAG_ECHO invention.
ache [Sat, 6 Nov 2004 13:56:18 +0000 (13:56 +0000)]
Back out recent TTYDEF_LFLAG_ECHO invention.

This change is NOP, because TTYDEF_LFLAG = TTYDEF_LFLAG_ECHO now,
but to minimize diffs with other BSDs.

19 years agoFirst there was an -m option.
ru [Sat, 6 Nov 2004 13:38:45 +0000 (13:38 +0000)]
First there was an -m option.
Then Peter killed it and made it the default.
Then Brooks and Poul-Henning added it again.
Then Ruslan fixed the manpage...

19 years agoSync description of IP portrange sysctls with reality.
mtm [Sat, 6 Nov 2004 13:24:53 +0000 (13:24 +0000)]
Sync description of IP portrange sysctls with reality.

MFC after: 2 weeks

19 years agoBe consistent and use 'if (error != 0)' instead of 'if (error)' everywhere.
pjd [Sat, 6 Nov 2004 13:16:35 +0000 (13:16 +0000)]
Be consistent and use 'if (error != 0)' instead of 'if (error)' everywhere.

19 years agoFor file backed md(4) devices output their source file via
pjd [Sat, 6 Nov 2004 13:07:02 +0000 (13:07 +0000)]
For file backed md(4) devices output their source file via
'mdconfig -l -u <unit>'.
Bump version number, as this change breaks ABI/API.

19 years agoAbstract the logic to look up the uma_bucket_zone given a desired
rwatson [Sat, 6 Nov 2004 11:43:30 +0000 (11:43 +0000)]
Abstract the logic to look up the uma_bucket_zone given a desired
number of entries into bucket_zone_lookup(), which helps make more
clear the logic of consumers of bucket zones.

Annotate the behavior of bucket_init() with a comment indicating
how the various data structures, including the bucket lookup tables,
are initialized.

19 years agoProperly implement a default version of VOP_GETWRITEMOUNT.
phk [Sat, 6 Nov 2004 11:41:22 +0000 (11:41 +0000)]
Properly implement a default version of VOP_GETWRITEMOUNT.

Remove improper access to vop_stdgetwritemount() which should and
will instead rely on the VOP default path.

19 years agoIncrease default HZ for i386 to 1000
phk [Sat, 6 Nov 2004 11:33:43 +0000 (11:33 +0000)]
Increase default HZ for i386 to 1000

19 years agoRemove dangling variable
phk [Sat, 6 Nov 2004 11:33:11 +0000 (11:33 +0000)]
Remove dangling variable

19 years agoAnnotate what bucket_size[] array does; staticize since it's used only
rwatson [Sat, 6 Nov 2004 11:24:40 +0000 (11:24 +0000)]
Annotate what bucket_size[] array does; staticize since it's used only
in uma_core.c.

19 years agoRemove a useless use of cat, strip overzealous punctuation, and
dd [Sat, 6 Nov 2004 11:16:56 +0000 (11:16 +0000)]
Remove a useless use of cat, strip overzealous punctuation, and
clarify some text.

19 years agoParse file mode with setmode(3) for consistency with other programs
dd [Sat, 6 Nov 2004 11:03:59 +0000 (11:03 +0000)]
Parse file mode with setmode(3) for consistency with other programs
that do this kind of thing.

19 years agoFix a double-free in the 'm->m_len < sizeof (struct ip)' sanity check.
andre [Sat, 6 Nov 2004 10:47:36 +0000 (10:47 +0000)]
Fix a double-free in the 'm->m_len < sizeof (struct ip)' sanity check.

Bug report by: <james@towardex.com>
MFC after: 2 weeks

19 years agoUpdate the description of -l to reflect reality. After mdconfig.c
dd [Sat, 6 Nov 2004 10:21:12 +0000 (10:21 +0000)]
Update the description of -l to reflect reality. After mdconfig.c
1.25, -l without -u only lists the names, so specifying -u does more
than just limit the output to one device.

19 years agoIf there aren't any devices to list, output nothing instead of an
dd [Sat, 6 Nov 2004 09:56:27 +0000 (09:56 +0000)]
If there aren't any devices to list, output nothing instead of an
empty line. This is consistent with other utilities.

While I'm here, remove artifacts of the previous list implementation.

19 years agoFix the last known race in swapoff(), which could lead to a spurious panic:
das [Sat, 6 Nov 2004 07:17:50 +0000 (07:17 +0000)]
Fix the last known race in swapoff(), which could lead to a spurious panic:

swapoff: failed to locate %d swap blocks

The race occurred because putpages() can block between the time it
allocates swap space and the time it updates the swap metadata to
associate that space with a vm_object, so swapoff() would complain
about the temporary inconsistency.  I hoped to fix this by making
swp_pager_getswapspace() and swp_pager_meta_build() a single atomic
operation, but that proved to be inconvenient.  With this change,
swapoff() simply doesn't attempt to be so clever about detecting when
all the pageout activity to the target device should have drained.

19 years agoClean up the comments, rewording, adding missing punctuation
keramida [Sat, 6 Nov 2004 06:13:13 +0000 (06:13 +0000)]
Clean up the comments, rewording, adding missing punctuation
and other minor nits.

Approved by: jhb

19 years agoMove a call to wakeup() from vm_object_terminate() to vnode_pager_dealloc()
alc [Sat, 6 Nov 2004 05:33:02 +0000 (05:33 +0000)]
Move a call to wakeup() from vm_object_terminate() to vnode_pager_dealloc()
because this call is only needed to wake threads that slept when they
discovered a dead object connected to a vnode.  To eliminate unnecessary
calls to wakeup() by vnode_pager_dealloc(), introduce a new flag,
OBJ_DISCONNECTWNT.

Reviewed by: tegge@

19 years agoOoops. ssize_t != int. <sigh>
kientzle [Sat, 6 Nov 2004 05:25:53 +0000 (05:25 +0000)]
Ooops.  ssize_t != int.  <sigh>

Thanks to: Oliver Lehmann and Peter Wemm

19 years agoi386_set_ldt() is not available when running 32 bit binaries on amd64
peter [Sat, 6 Nov 2004 03:35:51 +0000 (03:35 +0000)]
i386_set_ldt() is not available when running 32 bit binaries on amd64
kernels.  Use the recently exposed direct-set routines instead.  This is
only activated for when we compile i386 support libraries on amd64.

19 years agoCosmetic tweaks to reduce diffs to the i386 counterpart.
peter [Sat, 6 Nov 2004 03:33:19 +0000 (03:33 +0000)]
Cosmetic tweaks to reduce diffs to the i386 counterpart.

19 years agoThe 32 bit compatability ld-elf32.so.1 cannot use i386_set_ldt() when
peter [Sat, 6 Nov 2004 03:32:07 +0000 (03:32 +0000)]
The 32 bit compatability ld-elf32.so.1 cannot use i386_set_ldt() when
running on an amd64 kernel.  Use the recently exposed direct %fs/%gs set
routines instead for the TLS setup of 32 bit binaries.

19 years agoUse the recently exposed fs/gs set functions when compiling libthr to
peter [Sat, 6 Nov 2004 03:30:53 +0000 (03:30 +0000)]
Use the recently exposed fs/gs set functions when compiling libthr to
run as a 32 bit support library for an amd64 kernel.  32 bit consumers of
libthr have zero chance of running on an amd64 kernel since we don't
implement the i386_set_ldt() family of functions.  Note that this commit
doesn't make it actually work, it just removes one more obstacle.

19 years agoWhen building the custom i386 libraries for use on amd64 kernels, we
peter [Sat, 6 Nov 2004 03:28:26 +0000 (03:28 +0000)]
When building the custom i386 libraries for use on amd64 kernels, we
can't use the i386_set_ldt() family of routines, because they are not
implemented.  Instead, use the recently exposed direct access sysarch
routines for setting what %fs and %gs point to.

Use this for the i386 TLS _set_tp() routine, but only when compiling to
run as a 32 bit support binary for amd64 kernels.

19 years agoBegin an invasion of i386-land by amd64.
peter [Sat, 6 Nov 2004 03:23:36 +0000 (03:23 +0000)]
Begin an invasion of i386-land by amd64.

Expose some of the amd64-specific sysarch functions to allow alternative
implementations of the %fs/%gs code for TLS, threads, etc.  USER_LDT does
not exist on the amd64 kernel, so we have to implement things other ways.

19 years agoConvert tools/lib32/build.sh into world connectable hooks. This still
peter [Sat, 6 Nov 2004 03:14:26 +0000 (03:14 +0000)]
Convert tools/lib32/build.sh into world connectable hooks.  This still
rates pretty high on the "hack!" scale, but it works for me.  Adding
-DWANT_LIB32 to the world build command line, or 'WANT_LIB32=yes' to
/etc/make.conf will include the 32 bit libraries with the build.

I have not made this default behavior.  Cross compiling this stuff is an
adventure I have not investigated.

This is still a WIP.  We needed this at work so that we could install from
a readonly obj tree - lib32/build.sh wasn't up to that.

19 years agoDon't define SOFTFLOAT directly in source files, it's now done in the Makefile.
cognet [Fri, 5 Nov 2004 23:53:54 +0000 (23:53 +0000)]
Don't define SOFTFLOAT directly in source files, it's now done in the Makefile.

19 years agoFix signalcontext and makecontext.
cognet [Fri, 5 Nov 2004 23:53:02 +0000 (23:53 +0000)]
Fix signalcontext and makecontext.

19 years agoRemove getcontext.S, it is not needed.
cognet [Fri, 5 Nov 2004 23:52:05 +0000 (23:52 +0000)]
Remove getcontext.S, it is not needed.

19 years agoImport a Makefile for arm.
cognet [Fri, 5 Nov 2004 23:51:24 +0000 (23:51 +0000)]
Import a Makefile for arm.

19 years agoPartial support of KSE for arm.
cognet [Fri, 5 Nov 2004 23:49:21 +0000 (23:49 +0000)]
Partial support of KSE for arm.

19 years agoDisable interrupts for atomic_cmpset_32, this one is just not atomic.
cognet [Fri, 5 Nov 2004 23:48:12 +0000 (23:48 +0000)]
Disable interrupts for atomic_cmpset_32, this one is just not atomic.
Don't export it to userland.

19 years agoRespect TDF_SINTR, don't suspend uninterruptible thread.
davidxu [Fri, 5 Nov 2004 22:40:33 +0000 (22:40 +0000)]
Respect TDF_SINTR, don't suspend uninterruptible thread.

19 years agoNote that sysctls documentation is missing.
pjd [Fri, 5 Nov 2004 22:38:14 +0000 (22:38 +0000)]
Note that sysctls documentation is missing.

19 years agoBackout previous commit, the P_STOPPED_BOUNDARY flag was already
davidxu [Fri, 5 Nov 2004 22:31:20 +0000 (22:31 +0000)]
Backout previous commit, the P_STOPPED_BOUNDARY flag was already
cleared at the begin of thread_single() when needed.

19 years ago- Store threads on sleep queues in FIFO order rather than sorted by
jhb [Fri, 5 Nov 2004 20:19:58 +0000 (20:19 +0000)]
- Store threads on sleep queues in FIFO order rather than sorted by
  priority.  The sleep queues don't get updated when the priority of
  threads changes, so sleepq_signal() might not always wakeup the
  highest priority thread.  Updating the queues when thread priorities
  change cannot be easily done due to lock orders, so instead we do an
  O(n) walk of the queue for a sleepq_signal() operation instead of O(1).
  On the other hand, adding a thread to a sleep queue now goes from O(n)
  to O(1) so it ends up as an even tradeoff.  The correctness here with
  regards to priorities is actually fairly important.  msleep() gives
  interactive threads their priority "boost" after they are placed on the
  queue, but before this fix that "boost" wasn't used to determine the
  highest priority thread that sleepq_signal() awoke.
- Fix up some comments.

Inspired by: ups, bde

19 years agoDon't change the priority to PUSER when sleeping, just keep the current
jhb [Fri, 5 Nov 2004 20:15:06 +0000 (20:15 +0000)]
Don't change the priority to PUSER when sleeping, just keep the current
priority.

19 years agoCopy the syscall args in a tmp variable instead of directly using the
cognet [Fri, 5 Nov 2004 19:57:10 +0000 (19:57 +0000)]
Copy the syscall args in a tmp variable instead of directly using the
trapframe, as it can be modified in the syscall.
Call thread_user_enter() when appropriate.

19 years agoSave a few cycles in context switch.
cognet [Fri, 5 Nov 2004 19:54:13 +0000 (19:54 +0000)]
Save a few cycles in context switch.
Update comments to reflect reality.

19 years agoIf we're still running at the physical address, jump to the virtual address
cognet [Fri, 5 Nov 2004 19:52:55 +0000 (19:52 +0000)]
If we're still running at the physical address, jump to the virtual address
instead before calling initarm().
This removes the need to map virtual == physical in initarm().

19 years agoBe more verbose about cache capacities.
cognet [Fri, 5 Nov 2004 19:51:23 +0000 (19:51 +0000)]
Be more verbose about cache capacities.

19 years agoImplement casuptr.
cognet [Fri, 5 Nov 2004 19:50:48 +0000 (19:50 +0000)]
Implement casuptr.

19 years agoCall pmap_pte_init_arm9 instead of pmap_pte_init_generic if
cognet [Fri, 5 Nov 2004 19:48:40 +0000 (19:48 +0000)]
Call pmap_pte_init_arm9 instead of pmap_pte_init_generic if
ARM9_CACHE_WRITE_THROUGH is defined.

19 years ago- Add a new MD per-CPU field for Alpha 'pal_id' which is the PAL ID
jhb [Fri, 5 Nov 2004 19:16:44 +0000 (19:16 +0000)]
- Add a new MD per-CPU field for Alpha 'pal_id' which is the PAL ID
  associated with each processor.  This ID is inferred from the index
  of the pcs structure in the hwprb.
- Give Alpha CPUs FreeBSD CPU IDs more like other architectures where the
  boot processor is always CPU 0 and the other processors are numbered
  1 ... N.  List active CPUs in the system in cpu_mp_announce() as well.

Silence on: alpha@

19 years ago- Set the priority of the page zeroing thread using sched_prio() when the
jhb [Fri, 5 Nov 2004 19:14:02 +0000 (19:14 +0000)]
- Set the priority of the page zeroing thread using sched_prio() when the
  thread is created rather than adjusting the priority in the main
  function.  (kthread_create() should probably take the initial priority
  as an argument.)
- Only yield the CPU in the !PREEMPTION case if there are any other
  runnable threads.  Yielding when there isn't anything else better to do
  just wastes time in pointless context switches (albeit while the system
  is idle.)

19 years ago- Make setting of IT_ENTROPY a bit simpler in ithread_update().
jhb [Fri, 5 Nov 2004 19:11:24 +0000 (19:11 +0000)]
- Make setting of IT_ENTROPY a bit simpler in ithread_update().
- Tweak the updating of the ithread name in ithread_update() so that the
  '+' and '*' characters for device names that were too short only get
  added at the end after as many device names as possible were fit into
  the allocated space.  Prior to this, some long devices would result
  in '+' chars showing up between two different devices rather than at the
  end.

19 years agoRestrict the sched_bind to cpu 0 to i386 and amd64 for now. I forgot that
peter [Fri, 5 Nov 2004 19:00:23 +0000 (19:00 +0000)]
Restrict the sched_bind to cpu 0 to i386 and amd64 for now.  I forgot that
alpha still doesn't use logical cpu id's.

19 years agoWalk through the snp softc list instead of abusing tp->ts_c to find the
cognet [Fri, 5 Nov 2004 18:32:14 +0000 (18:32 +0000)]
Walk through the snp softc list instead of abusing tp->ts_c to find the
snp attached to a tty.
This fixes the panic that happens when using snp.

19 years agoIn cpu_critical_fork_exit(), make sure to set td_md.md_critnest so that
cognet [Fri, 5 Nov 2004 18:29:45 +0000 (18:29 +0000)]
In cpu_critical_fork_exit(), make sure to set td_md.md_critnest so that
interrupts will be enabled.

Spotted out by: jhb

19 years agoBind to cpu0 for boot() processing. (Note this is reboot, not startup)
peter [Fri, 5 Nov 2004 18:29:10 +0000 (18:29 +0000)]
Bind to cpu0 for boot() processing. (Note this is reboot, not startup)
This means we'll always call the event hooks, device_shutdown etc on the
BSP and theoretically means we can de-cruftify the cpu_reset_proxy stuff.

19 years agoMFi386 1.238 (jhb): Allow hints to disable cpus
peter [Fri, 5 Nov 2004 18:25:22 +0000 (18:25 +0000)]
MFi386 1.238 (jhb): Allow hints to disable cpus

19 years agoMFi386:
peter [Fri, 5 Nov 2004 18:24:01 +0000 (18:24 +0000)]
MFi386:
rev 1.61 (scottl):  Add KTR tracing
rev 1.62 (scottl):  Optimize (td->pmap, inlines, etc)

19 years agoDon't forget to make sure that there are no not-finished requests before
pjd [Fri, 5 Nov 2004 17:18:39 +0000 (17:18 +0000)]
Don't forget to make sure that there are no not-finished requests before
marking components as clean.

Pointed out by: scottl

19 years ago- Mark all raid3 components as clean after kern.geom.raid3.idletime seconds.
pjd [Fri, 5 Nov 2004 13:12:58 +0000 (13:12 +0000)]
- Mark all raid3 components as clean after kern.geom.raid3.idletime seconds.
- Make kern.geom.raid3.timeout variable tunable.

19 years agoMark raid3 devices as clean on shutdown (after all file systems are
pjd [Fri, 5 Nov 2004 13:01:25 +0000 (13:01 +0000)]
Mark raid3 devices as clean on shutdown (after all file systems are
unmounted).

Suggested by: scottl

19 years ago- Use ->index consumer's field to track number of in-flight requests.
pjd [Fri, 5 Nov 2004 12:42:16 +0000 (12:42 +0000)]
- Use ->index consumer's field to track number of in-flight requests.
- Remove unused #include.

19 years agoStop method for swap1 script was introduced, because gmirror needed it.
pjd [Fri, 5 Nov 2004 12:38:27 +0000 (12:38 +0000)]
Stop method for swap1 script was introduced, because gmirror needed it.
Now gmirror use shutdown hooks to mark mirrors as clean on shutdown,
so this is not needed anymore.

19 years agoUse shutdown hooks to mark mirrors as clean after all file systems are
pjd [Fri, 5 Nov 2004 12:35:21 +0000 (12:35 +0000)]
Use shutdown hooks to mark mirrors as clean after all file systems are
unmounted.

Suggested by: scottl

19 years agoRemove unused #include.
pjd [Fri, 5 Nov 2004 12:31:32 +0000 (12:31 +0000)]
Remove unused #include.

19 years agoNow with the remote job support removed there can be no remote jobs and
harti [Fri, 5 Nov 2004 11:41:36 +0000 (11:41 +0000)]
Now with the remote job support removed there can be no remote jobs and
no jobs that need to be remigrated. Remove the flags and the associated
code.

19 years ago- Add a sysctl kern.geom.mirror.idletime, so one can specify after how many
pjd [Fri, 5 Nov 2004 10:55:04 +0000 (10:55 +0000)]
- Add a sysctl kern.geom.mirror.idletime, so one can specify after how many
  seconds of idling, DRITY flags are removed.
- If mirror is in idle state or is not open for writing, sleep without
  timeout when waiting for I/O requests.
- Don't use atomic operations, for now sysctls are protected by Giant.
- Update debugs.

19 years agoWhen the last line of a file is missing a newline in -f mode, pass a
tjr [Fri, 5 Nov 2004 10:45:23 +0000 (10:45 +0000)]
When the last line of a file is missing a newline in -f mode, pass a
length argument to mbrtowc() that accounts for the terminating newline
character we add automatically. Failing to do this caused the loop to
unexpectedly run out of characters and incorrectly signal an "Illegal byte
sequence" error.