]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
14 years agoClean up some printing stuff so that we can have a bit finer control
mjacob [Fri, 26 Mar 2010 15:13:31 +0000 (15:13 +0000)]
Clean up some printing stuff so that we can have a bit finer control
on debug output. Add a new platform function requirement to allow
for printing based upon the ITL nexus instead of the isp unit plus
channel, target and lun. This allows some printouts and error messages
from the core code to appear in the same format as the platform's
subsystem (in FreeBSD's case, CAM path).

MFC after: 1 week

14 years agoRevert r205683 to resolve some code quality issues which do not affect the
netchild [Fri, 26 Mar 2010 14:36:16 +0000 (14:36 +0000)]
Revert r205683 to resolve some code quality issues which do not affect the
build or use of linprocfs, before committing the reworked patch again.

Requested by: des

14 years agoHandling SIGPIPE will cause deadlock/crash.
fabient [Fri, 26 Mar 2010 14:35:48 +0000 (14:35 +0000)]
Handling SIGPIPE will cause deadlock/crash.
Return an error immediatly in case of hard shutdown.

MFC after: 3days

14 years agoDo not overflow the term in the case of multi-line display.
fabient [Fri, 26 Mar 2010 14:29:57 +0000 (14:29 +0000)]
Do not overflow the term in the case of multi-line display.

MFC after: 3days

14 years agoTighten my grip on pseudofs, procfs, linprocfs.
des [Fri, 26 Mar 2010 12:59:15 +0000 (12:59 +0000)]
Tighten my grip on pseudofs, procfs, linprocfs.

14 years agoFix some bogus values in linprocfs.
netchild [Fri, 26 Mar 2010 11:43:15 +0000 (11:43 +0000)]
Fix some bogus values in linprocfs.

Submitted by: Petr Salinger <Petr.Salinger@seznam.cz>
Verified on: GNU/kFreeBSD debian 8.0-1-686 (by submitter)
PR: 144584

14 years agoSupport only LOOKUP operation for "/" in relookup() because lookup()
jh [Fri, 26 Mar 2010 11:33:12 +0000 (11:33 +0000)]
Support only LOOKUP operation for "/" in relookup() because lookup()
can't succeed for CREATE, DELETE and RENAME.

Discussed with: bde

14 years ago- add some usb devices (scanner, printer, usb storage)
netchild [Fri, 26 Mar 2010 11:02:01 +0000 (11:02 +0000)]
- add some usb devices (scanner, printer, usb storage)
- add quirks for the usb storage

Reviewed by: hselasky

14 years agoUse last 16 bytes of serial number in metadata instead of first ones,
mav [Fri, 26 Mar 2010 10:18:19 +0000 (10:18 +0000)]
Use last 16 bytes of serial number in metadata instead of first ones,
same as Intel MatrixRAID does.

PR: kern/124064

14 years agoFix a typo in a comment.
netchild [Fri, 26 Mar 2010 08:47:07 +0000 (08:47 +0000)]
Fix a typo in a comment.

14 years agoFix some problems which may lead to a panic:
netchild [Fri, 26 Mar 2010 08:42:11 +0000 (08:42 +0000)]
Fix some problems which may lead to a panic:
 - right order of src and dst in memcpy
 - NULL out the clips after freeing to prevent an accident

Noticed by: hselasky

14 years agoReplace sb_store64()/sb_load64() with mips3_sd()/mips3_ld() respectively.
neel [Fri, 26 Mar 2010 07:15:27 +0000 (07:15 +0000)]
Replace sb_store64()/sb_load64() with mips3_sd()/mips3_ld() respectively.

Obtained from NetBSD.

Suggested by: jmallett@

14 years agoo Fix typo.
maxim [Fri, 26 Mar 2010 05:35:25 +0000 (05:35 +0000)]
o Fix typo.

PR: docs/145031
Submitted by: olgeni
MFC after: 1 week

14 years agoo Trim leading w/space.
maxim [Fri, 26 Mar 2010 05:34:13 +0000 (05:34 +0000)]
o Trim leading w/space.

PR: docs/145030
Submitted by: olgeni
MFC after: 1 week

14 years agoOnly use the interval timer for clock interrupts on the BSP and
marcel [Fri, 26 Mar 2010 02:29:15 +0000 (02:29 +0000)]
Only use the interval timer for clock interrupts on the BSP and
have the BSP use IPIs to trigger clock interrupts on the APs.
This allows us to run on hardware configurations for which the
ITC has non-uniform frequencies across CPUs.

While here, change the clock XIV to type IPI so as to protect
the interrupt delivery against CPU re-balancing once that's
implemented.

14 years agoPatch the experimental NFS server in a manner analagous to r205661
rmacklem [Fri, 26 Mar 2010 01:35:19 +0000 (01:35 +0000)]
Patch the experimental NFS server in a manner analagous to r205661
for the regular NFS server, to ensure that ESTALE is
returned to the client for all errors returned by VFS_FHTOVP().

MFC after: 2 weeks

14 years agoOur boot loader is capable of booting both i386 and amd64 kernels so
delphij [Fri, 26 Mar 2010 01:30:53 +0000 (01:30 +0000)]
Our boot loader is capable of booting both i386 and amd64 kernels so
call it "x86" instead of "i386".

Suggested by: jhb in response to Alexander Best's loader proposal
MFC after: 1 month

14 years agoPatch the regular NFS server so that it returns ESTALE to the client
rmacklem [Fri, 26 Mar 2010 01:19:29 +0000 (01:19 +0000)]
Patch the regular NFS server so that it returns ESTALE to the client
for all errors returned by VFS_FHTOVP(). This is required to ensure
that EIO doesn't get returned to the client when ZFS is used as the
server file system.

Tested by: korvus AT comcast.net
Reviewed by: jhb
MFC after: 2 weeks

14 years agoFix the ia64 build.
nwhitehorn [Fri, 26 Mar 2010 00:53:13 +0000 (00:53 +0000)]
Fix the ia64 build.

Pointy hat to: me

14 years ago- update zfs and zpool manual pages to match the current state of the source
danger [Fri, 26 Mar 2010 00:49:06 +0000 (00:49 +0000)]
- update zfs and zpool manual pages to match the current state of the source

PR: 144984
Submitted by: mm@
Approved by: pjd@
Obtained from: OpenSolaris
MFC after: 4 days

14 years agoCheck that gl_pathc is bigger than zero before derefencing gl_pathv.
delphij [Thu, 25 Mar 2010 22:41:01 +0000 (22:41 +0000)]
Check that gl_pathc is bigger than zero before derefencing gl_pathv.
When gl_pathc == 0, the content of gl_pathv is undefined.

PR: bin/144761
Submitted by: David BERARD <contact davidberard fr>
Obtained from: OpenBSD
MFC after: 1 week

14 years agoThe rmt client in GNU cpio could have a heap overflow when a malicious
delphij [Thu, 25 Mar 2010 20:02:54 +0000 (20:02 +0000)]
The rmt client in GNU cpio could have a heap overflow when a malicious
remote tape service returns deliberately crafted packets containing
more data than requested.

Fix this by checking the returned amount of data and bail out when it
is more than what we requested.

PR: gnu/145010
Submitted by: naddy
Reviewed by: imp
MFC after: immediately
Security: CVE-2010-0624

14 years agoDo not penalize correct or correctable VESA mode tables by calling another
jkim [Thu, 25 Mar 2010 17:51:05 +0000 (17:51 +0000)]
Do not penalize correct or correctable VESA mode tables by calling another
VBE function.  Most problems should be corrected by the mode table sanity
check and we only need the paranoid in extremely rare cases.

14 years agoA ptrace(2) by one processor may trigger a promotion in the address space
alc [Thu, 25 Mar 2010 17:24:03 +0000 (17:24 +0000)]
A ptrace(2) by one processor may trigger a promotion in the address space
of another process.  Modify pmap_promote_pde() to handle this.  (This is
not a problem on amd64 due to implementation differences.)

Reported by: jh@
MFC after: 1 week

14 years agoUse pci_get_max_read_req() and pci_set_max_read_req() to set maximim
yongari [Thu, 25 Mar 2010 17:17:35 +0000 (17:17 +0000)]
Use pci_get_max_read_req() and pci_set_max_read_req() to set maximim
read request size.

14 years agoRevert accidentally committed initial real mode %sp change of r205347.
jkim [Thu, 25 Mar 2010 17:14:47 +0000 (17:14 +0000)]
Revert accidentally committed initial real mode %sp change of r205347.
Note I am keeping %ds change because X.org int10 handler does it and
it seems reasonable.

14 years agoOptimize real mode page table lookup.
jkim [Thu, 25 Mar 2010 17:03:52 +0000 (17:03 +0000)]
Optimize real mode page table lookup.

14 years agoFix stupid typos. Some VESA BIOSes directly call BIOS interrupt handlers
jkim [Thu, 25 Mar 2010 15:56:04 +0000 (15:56 +0000)]
Fix stupid typos.  Some VESA BIOSes directly call BIOS interrupt handlers
within the VBE interrupt handler.  Unfortunately it was causing real mode
page faults because we were fetching instructions from bogus addresses.
Pass me the pointyhat, please.

PR: kern/144654
MFC after: 3 days

14 years agoAdd the ELF relocation base to struct image_params. This will be
nwhitehorn [Thu, 25 Mar 2010 14:31:26 +0000 (14:31 +0000)]
Add the ELF relocation base to struct image_params. This will be
required to correctly relocate the executable entry point's function
descriptor on powerpc64.

14 years agoChange the arguments of exec_setregs() so that it receives a pointer
nwhitehorn [Thu, 25 Mar 2010 14:24:00 +0000 (14:24 +0000)]
Change the arguments of exec_setregs() so that it receives a pointer
to the image_params struct instead of several members of that struct
individually. This makes it easier to expand its arguments in the future
without touching all platforms.

Reviewed by: jhb

14 years agoChange the way text_addr and data_addr are computed to use the
nwhitehorn [Thu, 25 Mar 2010 14:21:22 +0000 (14:21 +0000)]
Change the way text_addr and data_addr are computed to use the
executable status of segments instead of detecting the main text segment
by which segment contains the program entry point. This affects
obreak() and is required for correct operation of that function
on 64-bit PowerPC systems. The previous behavior was apparently
required only for the Alpha, which is no longer supported.

Reviewed by: jhb
Tested on: amd64, sparc64, powerpc

14 years agoPropagate CONF_CFLAGS (from makeoptions) to the module build too.
netchild [Thu, 25 Mar 2010 13:47:21 +0000 (13:47 +0000)]
Propagate CONF_CFLAGS (from makeoptions) to the module build too.

Discussed with: jhb (on arch@)

14 years agoWe are holding a write lock here so avoid aquiring it twice calling
bz [Thu, 25 Mar 2010 10:29:00 +0000 (10:29 +0000)]
We are holding a write lock here so avoid aquiring it twice calling
the "locked" version rather than the wrapper function.

MFC after: 6 days

14 years agoRemove disabled code. In 99% cases exports are send to ng_ksocket(4), which
glebius [Thu, 25 Mar 2010 10:13:21 +0000 (10:13 +0000)]
Remove disabled code. In 99% cases exports are send to ng_ksocket(4), which
already forces queued mode, so what was suggested in disabled code is already
done.

14 years agofix another bug in "ipfw set N ..."
luigi [Wed, 24 Mar 2010 23:06:16 +0000 (23:06 +0000)]
fix another bug in "ipfw set N ..."

Submitted by: Marcin Wisnicki

14 years agoThis broke when we went to gnu99 as the default standard. Fix the build
imp [Wed, 24 Mar 2010 20:20:28 +0000 (20:20 +0000)]
This broke when we went to gnu99 as the default standard.  Fix the build
by reverting to the gnu89 standard.

14 years agoAdds the option of keeping per-cpu statistics in SCTP. This
rrs [Wed, 24 Mar 2010 20:02:40 +0000 (20:02 +0000)]
Adds the option of keeping per-cpu statistics in SCTP. This
may be useful since it gets rid of atomics but I want it to
remain an option until I can do further testing on if it really
speeds things up.

14 years agolagging file I forgot to commit with my nr-sack fixes... opps
rrs [Wed, 24 Mar 2010 20:01:14 +0000 (20:01 +0000)]
lagging file I forgot to commit with my nr-sack fixes... opps

Reviewed by: tuexen@freebsd.org

14 years agoFix for NR-Sack code. The code was NOT working properly when
rrs [Wed, 24 Mar 2010 19:45:36 +0000 (19:45 +0000)]
Fix for NR-Sack code. The code was NOT working properly when
enabled. Basically most of the operations were incorrect causing
bad sacks when you enabled nr-sack. The fixes range across
4 files and unifiy most of the processing so that we only test
nr_sack flags to decide which type of sack to generate.

Optimization left for this is to combine the sack generation
code and make it capable of generating either sack thus shrinking
out a routine.

Reviewed by: tuexen@freebsd.org

14 years agoPrint the pointer to the lock with the panic message. The previous
bz [Wed, 24 Mar 2010 19:21:26 +0000 (19:21 +0000)]
Print the pointer to the lock with the panic message. The previous
panic: rw lock not unlocked
was not really helpful for debugging. Now one can at least call
show lock <ptr>
form ddb to learn more about the lock.

MFC after: 3 days

14 years agoExpose MACHINE_CPU while building lib32 target.
delphij [Wed, 24 Mar 2010 18:40:57 +0000 (18:40 +0000)]
Expose MACHINE_CPU while building lib32 target.

14 years agomake the module loadable
luigi [Wed, 24 Mar 2010 18:35:31 +0000 (18:35 +0000)]
make the module loadable

Submitted by: Marcin Wisnicki

14 years agoDo not fetch precise time of request start when stats collection disabled.
mav [Wed, 24 Mar 2010 18:04:25 +0000 (18:04 +0000)]
Do not fetch precise time of request start when stats collection disabled.

Reviewed by: pjd, phk

14 years agoNow actually implement reading/refreshing/returning data from the pfTablesAddrTable
syrinx [Wed, 24 Mar 2010 16:07:33 +0000 (16:07 +0000)]
Now actually implement reading/refreshing/returning data from the pfTablesAddrTable
and modify the BEGEMOT-PF-MIB to add support for IPV6 address' statistics in the PF
tables via pfTablesAddrNetType and pfTablesAddrNet. While here, upgrade the
pf_tree.def file to the new format that includes enumerated values. Also make sure
to return SNMP_ERR_NOSUCHNAME for ALTQ objects, if ALTQ is disabled, so that the agent
will know to skip the pfAltq subtree when servicing GETNEXT requests from SNMP clients
(otherwise snmpwalk on begemotPf would stop at the pfAltq subtree with bsnmpd returning
SNMP_ERR_GENERR).

14 years ago- Remove const'ness from dlerror(3) prototype, for consistency with POSIX.
gahr [Wed, 24 Mar 2010 15:59:51 +0000 (15:59 +0000)]
- Remove const'ness from dlerror(3) prototype, for consistency with POSIX.

Approved by: cognet
MFC after: 1 week

14 years agoAlign memory access of 24-bit pixel renderer to word boundary.
jkim [Wed, 24 Mar 2010 15:40:18 +0000 (15:40 +0000)]
Align memory access of 24-bit pixel renderer to word boundary.

14 years agoTeach VGA framebuffer about 8-bit palette format for VESA.
jkim [Wed, 24 Mar 2010 15:37:47 +0000 (15:37 +0000)]
Teach VGA framebuffer about 8-bit palette format for VESA.

14 years agoHonor ip.fw.one_pass when a packet comes out of a pipe without being delayed.
luigi [Wed, 24 Mar 2010 15:16:59 +0000 (15:16 +0000)]
Honor ip.fw.one_pass when a packet comes out of a pipe without being delayed.
I forgot to handle this case when i did the mtag cleanup three months ago.

PR: 145004

14 years agoPrune empty directories.
ed [Wed, 24 Mar 2010 15:16:05 +0000 (15:16 +0000)]
Prune empty directories.

14 years agoAdd missing Giant locking for the vfsconf list.
jhb [Wed, 24 Mar 2010 14:20:37 +0000 (14:20 +0000)]
Add missing Giant locking for the vfsconf list.

Submitted by: kib

14 years agoUse gpart(8) to set the active boot partition.
phk [Wed, 24 Mar 2010 11:21:33 +0000 (11:21 +0000)]
Use gpart(8) to set the active boot partition.

14 years agoo FreeBSD 7.3 added.
maxim [Wed, 24 Mar 2010 06:08:51 +0000 (06:08 +0000)]
o FreeBSD 7.3 added.

14 years agoFix periodic "t_delta 16.01359db7eb5eb3c0 too long" messages on the console by
neel [Wed, 24 Mar 2010 04:52:15 +0000 (04:52 +0000)]
Fix periodic "t_delta 16.01359db7eb5eb3c0 too long" messages on the console by
accounting for the "lost time" between when the timer interrupt fired
and when clock_intr() actually started executing.

14 years agoThe nargvstr and nenvstr properties of arginfo are ints, not longs,
nwhitehorn [Wed, 24 Mar 2010 03:13:24 +0000 (03:13 +0000)]
The nargvstr and nenvstr properties of arginfo are ints, not longs,
so should be copied to userspace with suword32() instead of suword().
This alleviates problems on 64-bit big-endian architectures, and is a
no-op on all 32-bit architectures.

Tested on: amd64, sparc64, powerpc64

14 years agoAdapt r204907 and r205402, the amd64 implementation of the workaround for
alc [Wed, 24 Mar 2010 03:07:35 +0000 (03:07 +0000)]
Adapt r204907 and r205402, the amd64 implementation of the workaround for
AMD Family 10h Erratum 383, to i386.

Enable machine check exceptions by default, just like r204913 for amd64.

Enable superpage promotion only if the processor actually supports large
pages, i.e., PG_PS.

MFC after: 2 weeks

14 years agoFix the experimental NFS subsystem so that it uses the correct
rmacklem [Wed, 24 Mar 2010 02:02:02 +0000 (02:02 +0000)]
Fix the experimental NFS subsystem so that it uses the correct
preprocessor macro name for not requiring strict data alignment.

Suggested by: marius
MFC after: 2 weeks

14 years agoUpdate metadata information as well as upgrade instructions.
delphij [Tue, 23 Mar 2010 23:56:22 +0000 (23:56 +0000)]
Update metadata information as well as upgrade instructions.

14 years agoFix an off-by-one bug for the number of slots on a PCI/PCI-X bus.
marcel [Tue, 23 Mar 2010 23:46:28 +0000 (23:46 +0000)]
Fix an off-by-one bug for the number of slots on a PCI/PCI-X bus.
We failed to setup PCI devices on slot 31 and that's where the
SATA controller is for the P2020 eval board.

14 years agoAdd my copyright here. It seems I have contributed enough code. :-)
jkim [Tue, 23 Mar 2010 23:19:23 +0000 (23:19 +0000)]
Add my copyright here.  It seems I have contributed enough code. :-)

14 years agoBe extremely careful when we determine bytes per scan line information.
jkim [Tue, 23 Mar 2010 23:10:17 +0000 (23:10 +0000)]
Be extremely careful when we determine bytes per scan line information.
First, we compare mode table data against minimum value.  If the mode table
does not make sense, we set the minimum in the mode info.  When we actually
set the mode, we try VESA BIOS function and compare it against the previous
value.  If it makes more sense, update the information.

14 years agoWhen the regular NFS server replied to a UDP client out of the replay
rmacklem [Tue, 23 Mar 2010 23:03:30 +0000 (23:03 +0000)]
When the regular NFS server replied to a UDP client out of the replay
cache, it did not free the request argument mbuf list, resulting in a leak.
This patch fixes that leak.

Tested by: danny AT cs.huji.ac.il
PR: kern/144330
Submitted by: to.my.trociny AT gmail.com (earlier version)
Reviewed by: dfr
MFC after: 2 weeks

14 years agoMFV: Update nc to the version from OpenBSD 4.7.
delphij [Tue, 23 Mar 2010 23:00:35 +0000 (23:00 +0000)]
MFV: Update nc to the version from OpenBSD 4.7.

MFC after: 1 month

14 years agoFall back to VGA palette functions if VESA function failed and DAC is still
jkim [Tue, 23 Mar 2010 22:50:22 +0000 (22:50 +0000)]
Fall back to VGA palette functions if VESA function failed and DAC is still
in 6-bit mode.  Although we have to check non-VGA compatibility bit here,
it seems there are too many broken VESA BIOSes out to rely on it.

14 years agoMap entire video memory again. This is a partial backout of r203535.
jkim [Tue, 23 Mar 2010 22:35:52 +0000 (22:35 +0000)]
Map entire video memory again.  This is a partial backout of r203535.
Although we do not use them all directly, it seems VGA render may access
unmapped memory region and cause kernel panic.

14 years agoSeparate 24-bit pixel draw from 32-bit case. Although it is slower, we do
jkim [Tue, 23 Mar 2010 22:16:57 +0000 (22:16 +0000)]
Separate 24-bit pixel draw from 32-bit case.  Although it is slower, we do
not want to write a useless zero to inaccessible memory region.

14 years agoImplement /proc/filesystems.
jhb [Tue, 23 Mar 2010 21:49:33 +0000 (21:49 +0000)]
Implement /proc/filesystems.

Submitted by: Fernando Apesteguia fernando.apesteguia (gmail)

14 years agoReject attempts to create a MAP_ANON mapping with a non-zero offset.
jhb [Tue, 23 Mar 2010 21:08:07 +0000 (21:08 +0000)]
Reject attempts to create a MAP_ANON mapping with a non-zero offset.

PR: kern/71258
Submitted by: Alexander Best
MFC after: 2 weeks

14 years agoAdd definitions for a 4th PCI host controller. No Freescale processor
marcel [Tue, 23 Mar 2010 20:12:53 +0000 (20:12 +0000)]
Add definitions for a 4th PCI host controller. No Freescale processor
has all 4 implemented, but across the processors we now support all the
combinations. For example, the MPC8533 doesn't have a PCI controller
at 0xA0000, but does at 0xB0000.

14 years agoAdd PCI ID for MCS9901's parallel port.
delphij [Tue, 23 Mar 2010 20:08:18 +0000 (20:08 +0000)]
Add PCI ID for MCS9901's parallel port.

PR: kern/144713
Submitted by: gcooper
MFC after: 2 weeks

14 years agoEnable power management for E500 cores. Use "doze" for now to make
marcel [Tue, 23 Mar 2010 19:30:56 +0000 (19:30 +0000)]
Enable power management for E500 cores. Use "doze" for now to make
sure the caches remain coherent. For single-core configurations and
with busdma changes we could eventually switch to "nap" and force
a D-cache invalidation as part of the DMA completion. To this end,
clear PSL_WE until after we handled the decrementer or external
interrupt as it tells us whether we just woke up or not.

14 years agoCorrect cross reference.
delphij [Tue, 23 Mar 2010 18:00:50 +0000 (18:00 +0000)]
Correct cross reference.

MFC after: 2 weeks

14 years agoWhen receiving a management frame, pass the mbuf to bpf before calling
rpaulo [Tue, 23 Mar 2010 14:31:31 +0000 (14:31 +0000)]
When receiving a management frame, pass the mbuf to bpf before calling
iv_recv_mgmt(). iv_recv_mgmt() will generate management frame responses
and pass them to bpf before the management frame that triggered the
response.

PR: 144323
Submitted by: Alexander Egorenkov <egorenar at gmail.com>
MFC after: 2 weeks
Sponsored by: iXsystems, inc.

14 years agoAdd MCS to the list of media types.
rpaulo [Tue, 23 Mar 2010 13:15:11 +0000 (13:15 +0000)]
Add MCS to the list of media types.

Sponsored by: iXsystems, inc.

14 years agoAdd a missing LINE_BREAK() after printing the roaming parameters in
rpaulo [Tue, 23 Mar 2010 12:05:25 +0000 (12:05 +0000)]
Add a missing LINE_BREAK() after printing the roaming parameters in
verbose mode.

Sponsored by: iXsystems, inc.
MFC after: 2 weeks

14 years agoAdd a new field for extended HT capabilities.
rpaulo [Tue, 23 Mar 2010 12:03:09 +0000 (12:03 +0000)]
Add a new field for extended HT capabilities.

Submitted by: Alexander Egorenkov <egorenar at gmail.com>
MFC after: 2 weeks
Sponsored by: iXsystems, inc.

14 years agoFinish the much belated Intel XScale hwpmc(4) man page.
rpaulo [Tue, 23 Mar 2010 11:33:08 +0000 (11:33 +0000)]
Finish the much belated Intel XScale hwpmc(4) man page.

14 years agoAdd .snap to daily_clean_tmps_ignore; /tmp/.snap ist not supposed to
joerg [Tue, 23 Mar 2010 06:19:44 +0000 (06:19 +0000)]
Add .snap to daily_clean_tmps_ignore; /tmp/.snap ist not supposed to
be auto-removed (and /tmp is a filesystem of its own now by default).

MFC after: 3 days

14 years agoGet nexus(4) out of the RTC business. The interface used by nexus(4)
nwhitehorn [Tue, 23 Mar 2010 03:14:44 +0000 (03:14 +0000)]
Get nexus(4) out of the RTC business. The interface used by nexus(4)
in Open Firmware was Apple-specific, and we have complete coverage of Apple
system controllers, so move RTC responsibilities into the system controller
drivers. This avoids interesting problems from manipulating these devices
through Open Firmware behind the backs of their drivers.

Obtained from: NetBSD
MFC after: 2 weeks

14 years agoFixes a bug where SACKs in the face of
rrs [Tue, 23 Mar 2010 01:36:50 +0000 (01:36 +0000)]
Fixes a bug where SACKs in the face of
mapping_array expansion would break. Basically
once we expanded the array we no longer had both
mapping arrays in sync which the sack processing code depends on.
This would mean we were randomly referring to memory that was probably
not there. This mostly just gave us bad sack results going back to the peer.
If INVARIENTS was on of course we would hit the panic routine in the sack_check
call.

We also add a print routine for the place where one would panic in
invarients so one can see what the main mapping array holds.

Reviewed by: tuexen@freebsd.org
MFC after: 2 weeks

14 years agoOpen Firmware on powerpc is generally non-reetrant, so serialize all
nwhitehorn [Tue, 23 Mar 2010 01:11:10 +0000 (01:11 +0000)]
Open Firmware on powerpc is generally non-reetrant, so serialize all
OF calls with a mutex.

14 years agoDo not declare the various OFW command buffers static. It does not
nwhitehorn [Tue, 23 Mar 2010 01:09:45 +0000 (01:09 +0000)]
Do not declare the various OFW command buffers static. It does not
appear to be necessary on either sparc64 or powerpc, and is a
concurrency nightmare.

Reviewed by: marius

14 years agoActually pass a pointer to the trapframe to powerpc_extr_interrupt().
marcel [Tue, 23 Mar 2010 01:07:30 +0000 (01:07 +0000)]
Actually pass a pointer to the trapframe to powerpc_extr_interrupt().

14 years ago- boot-time size the ipv4 flowtable and the maximum number of flows
kmacy [Mon, 22 Mar 2010 23:04:12 +0000 (23:04 +0000)]
- boot-time size the ipv4 flowtable and the maximum number of flows
- increase flow cleaning frequency and decrease flow caching time
  when near the flow limit
- stop allocating new flows when within 3% of maxflows don't start
  allocating again until below 12.5%

MFC after: 7 days

14 years ago- enable alignment on amd64 only
kmacy [Mon, 22 Mar 2010 22:39:32 +0000 (22:39 +0000)]
- enable alignment on amd64 only
- only align pcpu caches and the volatile portion of uma_zone

14 years agoExpand $FreeBSD$.
delphij [Mon, 22 Mar 2010 22:12:27 +0000 (22:12 +0000)]
Expand $FreeBSD$.

14 years agoMFV of tzdata2010f:
edwin [Mon, 22 Mar 2010 21:27:51 +0000 (21:27 +0000)]
MFV of tzdata2010f:

The Australian Antartic Division:
- Macquarie Island will stay on UTC+11 for winter and not switch back from DST.
- Casey station reverted to its normal time of UTC+8 on 5 March 2010.
- Davis station will revert to its normal time of UTC+7 at 10 March 2010
- Mawson station stays on UTC+5.

Syria will start DST on Thursday 1 April 2010 at midnight.

Correct Samao DST start date (26 Sep vs 24 Oct)

14 years agoEnable mmap for minigzip(1).
delphij [Mon, 22 Mar 2010 21:19:17 +0000 (21:19 +0000)]
Enable mmap for minigzip(1).

14 years agoUpdate to zlib 1.2.4 and add versioned symbols to the
delphij [Mon, 22 Mar 2010 21:11:55 +0000 (21:11 +0000)]
Update to zlib 1.2.4 and add versioned symbols to the
library.

Sponsored by: iXsystems, Inc.

14 years agoVersioned symbols for libz.
delphij [Mon, 22 Mar 2010 20:24:00 +0000 (20:24 +0000)]
Versioned symbols for libz.

14 years agoSupport memory wraparound instead of high memory as VM86 mode does.
jkim [Mon, 22 Mar 2010 18:43:36 +0000 (18:43 +0000)]
Support memory wraparound instead of high memory as VM86 mode does.

Suggested by: delphij

14 years agoo Remove the pmap argument to pmap_invalidate_all() as it's not used
marcel [Mon, 22 Mar 2010 18:24:42 +0000 (18:24 +0000)]
o   Remove the pmap argument to pmap_invalidate_all() as it's not used
    other than in a potentially dangerous KASSERT.
o   Hand-inline pmap_remove_page() as it's only called from 1 place and
    the abstraction that pmap_remove_page() provides is not enough to
    warrant the obfuscation. Eliminate the dangerous KASSERT in the
    process.
o   In pmap_remove_pte(), remove the KASSERT for pmap being the current
    one as it's not safe in the face of CPU migration.

14 years agoFix i386 PAE kernel build.
jkim [Mon, 22 Mar 2010 17:30:34 +0000 (17:30 +0000)]
Fix i386 PAE kernel build.

Reported by: tinderbox

14 years agoRemove unneeded type specifiers from 64-bit constants. The compiler
jhb [Mon, 22 Mar 2010 15:08:26 +0000 (15:08 +0000)]
Remove unneeded type specifiers from 64-bit constants.  The compiler
infers their natural type from the constants' values.

Submitted by: bde
MFC after: 3 days

14 years agoMerge r197455 from amd64:
emaste [Mon, 22 Mar 2010 11:52:53 +0000 (11:52 +0000)]
Merge r197455 from amd64:

  Add a backtrace to the "fpudna in kernel mode!" case, to help track down
  where this comes from.

  Reviewed by: bde

14 years agoForced commit to register MFC reminder.
des [Mon, 22 Mar 2010 11:32:19 +0000 (11:32 +0000)]
Forced commit to register MFC reminder.

MFC after: 2 weeks

14 years agoMerge upstream r432:
des [Mon, 22 Mar 2010 11:00:57 +0000 (11:00 +0000)]
Merge upstream r432:

pam_end() already contains a NULL check, and it is not unreasonable to
call it with a NULL pamh in a cleanup / error-handling situation.  Remove
OPENPAM_NONNULL, which may cause gcc to optimize away the NULL check.

This fixes a potential NULL pointer dereference in error-handling code
in passwd(1).

14 years agoDrop the pmap argument to pmap_invalidate_page(). It's not used other
marcel [Mon, 22 Mar 2010 04:24:19 +0000 (04:24 +0000)]
Drop the pmap argument to pmap_invalidate_page(). It's not used other
than in a KASSERT. The KASSERT is broken in that it's done outside the
critical section and as such isn't protected against CPU migration.
Improve pmap_invalidate_page() as follows:
o   calculate vhpt_ofs inside the critical region for exactly the same
    reason.
o   calculate the tag outside the FOREACH loop, as it's loop-invariant.
    This is more efficient.
o   Replace the test and set with an atomic cmpset operation because we
    are changing other CPU's VHPT tables and this avoids invalidating
    after the entry got modified. Not necessarily a problem, but better
    safe than sorry.

14 years agoWith preemption, the high FP registers may get enabled by cpu_switch()
marcel [Mon, 22 Mar 2010 04:01:45 +0000 (04:01 +0000)]
With preemption, the high FP registers may get enabled by cpu_switch()
before we grab the mutex. Don't assert that they must be disabled at
that point. We pretty much bypass all logic in that case anyway and
leave immediately, so there's no harm.

14 years agoFix interrupt handling by extending the critical region so that
marcel [Mon, 22 Mar 2010 03:55:18 +0000 (03:55 +0000)]
Fix interrupt handling by extending the critical region so that
preemption doesn't happen until after all pending interrupt have
been services.
While here again, simplify the EOI handling by doing it after we
call the XIV-specific handlers, rather than in each of them. The
original thought was that we may want to do an EOI first and the
actual IPI handling next, but that's mostly a micro-optimization.

14 years agoDisable interrupts when calling into SAL for PCI configuration
marcel [Mon, 22 Mar 2010 03:06:11 +0000 (03:06 +0000)]
Disable interrupts when calling into SAL for PCI configuration
cycles. This serves 2 purposes:
1.  It prevents preemption and CPU migration while running SAL code.
2.  It reduces the chance of stack overflows: we're supposed to enter
    SAL with at least 16KB of either memory- or register stack space,
    which we can't do without switching to a different stack.