]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
15 years agoDecrease arc4_count only when needed and with proper bytes amount.
Andrey A. Chernov [Mon, 21 Jul 2008 12:44:47 +0000 (12:44 +0000)]
Decrease arc4_count only when needed and with proper bytes amount.

Obtained from:  OpenBSD

15 years ago1) Set stired flag after forced initialization.
Andrey A. Chernov [Mon, 21 Jul 2008 10:31:28 +0000 (10:31 +0000)]
1) Set stired flag after forced initialization.
2) Increase arc4_count to the limit OpenBSD use.

Submitted by:   Thorsten Glaser <tg@mirbsd.de> (1)
Obtained from:  OpenBSD (2)

15 years agoo Document "show conifhk", wording from r180610 commit log.
Maxim Konovalov [Mon, 21 Jul 2008 10:03:02 +0000 (10:03 +0000)]
o Document "show conifhk", wording from r180610 commit log.

Reviewed by: rwatson

15 years agoWe want to check new options given, not the current ones.
Pawel Jakub Dawidek [Mon, 21 Jul 2008 09:45:44 +0000 (09:45 +0000)]
We want to check new options given, not the current ones.
This fixes 'zpool import -o <mntopt> <name>' not working properly.

15 years agoremove unneeded declarations
Kip Macy [Mon, 21 Jul 2008 02:34:52 +0000 (02:34 +0000)]
remove unneeded declarations

15 years agoremove local version of tcp_offload_* functions
Kip Macy [Mon, 21 Jul 2008 02:29:40 +0000 (02:29 +0000)]
remove local version of tcp_offload_* functions

15 years agoupdate syncache function names
Kip Macy [Mon, 21 Jul 2008 02:26:49 +0000 (02:26 +0000)]
update syncache function names

15 years agoAdd versions of tcp_twstart, tcp_close, and tcp_drop that hide the acquisition the...
Kip Macy [Mon, 21 Jul 2008 02:23:02 +0000 (02:23 +0000)]
Add versions of tcp_twstart, tcp_close, and tcp_drop that hide the acquisition the tcbinfo lock.

MFC after: 1 week

15 years agoremove cxgb local definition of locked syncache_expand
Kip Macy [Mon, 21 Jul 2008 02:17:27 +0000 (02:17 +0000)]
remove cxgb local definition of locked syncache_expand

15 years agoAfter the commit of SVN rev 180236, wilko@ noticed that the approach
Edwin Groothuis [Mon, 21 Jul 2008 02:13:14 +0000 (02:13 +0000)]
After the commit of SVN rev 180236, wilko@ noticed that the approach
doesn't work on the Alpha platform: machine/elf.h doesn't include
sys/elf32.h there.

PR: related to bin/124906
Approved by: bde@
MFC after: 1 week

15 years agoadd interface for external consumers to syncache_expand - rename syncache_add in...
Kip Macy [Mon, 21 Jul 2008 02:11:06 +0000 (02:11 +0000)]
add interface for external consumers to syncache_expand - rename syncache_add in a manner consistent with other bits intended for offload

15 years agoremove cxgb local definitions of socket accessor functions
Kip Macy [Mon, 21 Jul 2008 01:23:19 +0000 (01:23 +0000)]
remove cxgb local definitions of socket accessor functions

15 years agoAdd accessor functions for socket fields.
Kip Macy [Mon, 21 Jul 2008 00:49:34 +0000 (00:49 +0000)]
Add accessor functions for socket fields.

MFC after: 1 week

15 years agoadd inpcb accessor functions for fields needed by TOE devices
Kip Macy [Mon, 21 Jul 2008 00:08:34 +0000 (00:08 +0000)]
add inpcb accessor functions for fields needed by TOE devices

15 years agoAdd support for actually sending WCCP return packets via GRE.
Julian Elischer [Sun, 20 Jul 2008 21:45:15 +0000 (21:45 +0000)]
Add support for actually sending WCCP return packets via GRE.
This MAY be combined by a clever person with the 'key' code recently
added, however a cursary glance suggest that it would be safer to just keep
the patches as it is unlikely that the two modes would be used together
and the separate patch has been extensively tested.

Obtained from:  here and there
MFC after: 1 week

15 years agoClear passphrase buffer after use.
Pawel Jakub Dawidek [Sun, 20 Jul 2008 19:56:13 +0000 (19:56 +0000)]
Clear passphrase buffer after use.

Submitted by: Fabian Keil <fk@fabiankeil.de> (a bit different version)

15 years agotcsh: fix abort in cleanup_until(), caused by internal xgetpass()
Giorgos Keramidas [Sun, 20 Jul 2008 18:05:18 +0000 (18:05 +0000)]
tcsh: fix abort in cleanup_until(), caused by internal xgetpass()

The xgetpass() function pushes a cleanup entry for &osa, but then
attempts to flush the stack until an entry matching &sa.  The two
object pointers do not match, so the stack is popped too much,
and then cleanup_until calls abort() when it unexpectedly hits
the bottom of the cleanup stack.

Reported by: Paul B. Mahol, onemda at gmail.com,
Pietro Cerutti, gahr at gahr.ch
PR: bin/124191
Reviewed by: mp, kris
MFC after: 2 days

15 years agoDocument a few sysctls. While here, remove dead code
Tom Rhodes [Sun, 20 Jul 2008 17:51:58 +0000 (17:51 +0000)]
Document a few sysctls.  While here, remove dead code
related to ip4_esp_randpad.

Reviewed by: gnn, bz (older version)
Approved by: gnn
Tested with: make universe

15 years agoWhile spin-waiting for the mailbox semaphore to update, do flushing reads of
Scott Long [Sun, 20 Jul 2008 16:50:14 +0000 (16:50 +0000)]
While spin-waiting for the mailbox semaphore to update, do flushing reads of
PCI bus so that we don't have to wait more than needed.

15 years agoDocument a few sysctls.
Tom Rhodes [Sun, 20 Jul 2008 15:29:58 +0000 (15:29 +0000)]
Document a few sysctls.

Reviewed by: rwatson

15 years agoDocument a few sysctls.
Tom Rhodes [Sun, 20 Jul 2008 15:25:20 +0000 (15:25 +0000)]
Document a few sysctls.

Reviewed by: rwatson

15 years agoia is a pointer thus use NULL rather then 0 for initialization and
Bjoern A. Zeeb [Sun, 20 Jul 2008 12:31:36 +0000 (12:31 +0000)]
ia is a pointer thus use NULL rather then 0 for initialization and
in comparisons to make this more obvious.

MFC after: 5 days

15 years agoProperly underline force_depend's parameter.
Lars Engels [Sun, 20 Jul 2008 10:06:14 +0000 (10:06 +0000)]
Properly underline force_depend's parameter.

Reviewed by: brueffer

15 years agoo Add cddl, nlm and xdr to CSCOPEDIRS.
Maxim Konovalov [Sun, 20 Jul 2008 09:16:00 +0000 (09:16 +0000)]
o Add cddl, nlm and xdr to CSCOPEDIRS.
o Add pc98 and sun4v to CSCOPE_ARCHDIR.

PR: misc/125792
Submitted by: KOIE Hidetaka
MFC after: 1 week

15 years agoConvert lock that protects sessions list from a mutex to a rwlock.
Pawel Jakub Dawidek [Sun, 20 Jul 2008 07:34:00 +0000 (07:34 +0000)]
Convert lock that protects sessions list from a mutex to a rwlock.
Now we can use read lock in fast path (padlock_process()).

15 years agoEliminate dead code. (The commit message for revision 1.287 explains why
Alan Cox [Sun, 20 Jul 2008 04:13:51 +0000 (04:13 +0000)]
Eliminate dead code.  (The commit message for revision 1.287 explains why
this code is dead.)

15 years agoremove unused toedev functions and add comments for rest
Kip Macy [Sun, 20 Jul 2008 02:02:50 +0000 (02:02 +0000)]
remove unused toedev functions and add comments for rest

15 years agoIncrease the ceiling on the size of the buffer map.
Alan Cox [Sat, 19 Jul 2008 23:42:38 +0000 (23:42 +0000)]
Increase the ceiling on the size of the buffer map.

15 years agoCorrect the description of v_cache_count. Eliminate an unused function.
Alan Cox [Sat, 19 Jul 2008 22:35:13 +0000 (22:35 +0000)]
Correct the description of v_cache_count.  Eliminate an unused function.

15 years agoSay hi to svn, by simplifing ffs_vget() function a bit - there is no need for
Pawel Jakub Dawidek [Sat, 19 Jul 2008 22:29:44 +0000 (22:29 +0000)]
Say hi to svn, by simplifing ffs_vget() function a bit - there is no need for
a variable that is used only once.

15 years agoWith uart(4) default, change sio# to uart# so that
Marcel Moolenaar [Sat, 19 Jul 2008 20:12:33 +0000 (20:12 +0000)]
With uart(4) default, change sio# to uart# so that
out-of-the-box FreeBSD is consistent.

15 years agoWith uart(4) default, change /dev/cuad# to /dev/cuau# and
Marcel Moolenaar [Sat, 19 Jul 2008 20:12:02 +0000 (20:12 +0000)]
With uart(4) default, change /dev/cuad# to /dev/cuau# and
sio# to uart# so that out-of-the-box FreeBSD is consistent.

15 years agoWith uart(4) default, change /dev/cuad# to /dev/cuau# and
Marcel Moolenaar [Sat, 19 Jul 2008 20:11:33 +0000 (20:11 +0000)]
With uart(4) default, change /dev/cuad# to /dev/cuau# and
sio# to uart# so that out-of-the-box FreeBSD is consistent.

15 years agoWith uart(4) default, change /dev/cuad# to /dev/cuau# so that
Marcel Moolenaar [Sat, 19 Jul 2008 20:00:18 +0000 (20:00 +0000)]
With uart(4) default, change /dev/cuad# to /dev/cuau# so that
out-of-the-box FreeBSD is consistent.

15 years agoRather than simply waiting silently and indefinitely for all
Robert Watson [Sat, 19 Jul 2008 19:08:35 +0000 (19:08 +0000)]
Rather than simply waiting silently and indefinitely for all
interrupt-driven configuration handlers to complete, print out a
diagnostic message every 60 second indicating which handlers are
still running.  Do this at most 5 times per run so as to avoid
scrolling out any useful information from the kernel message
buffer.

The interval of 60 seconds was selected based on a best guess as
to the nature of "long enough" and may want to be tuned higher
or lower depending on real-world tolerances.

MFC after: 3 days
Discussed with: scottl

15 years agoWith uart(4) default, change /dev/cuad# to /dev/cuau# so that
Marcel Moolenaar [Sat, 19 Jul 2008 19:08:22 +0000 (19:08 +0000)]
With uart(4) default, change /dev/cuad# to /dev/cuau# so that
out-of-the-box FreeBSD is consistent.  Add uart[0-7] as a
fingerfriendly shortcut alongside sio[0-7] and com[1-8].

15 years agoRemove sade(8) on the following platforms:
Marcel Moolenaar [Sat, 19 Jul 2008 18:21:52 +0000 (18:21 +0000)]
Remove sade(8) on the following platforms:
o  arm, mips & powerpc: libdisk is non-functional.
o  ia64: libdisk is insufficient.

15 years agowitness_addgraph() is required even if DDB isn't compiled into the kernel,
Robert Watson [Sat, 19 Jul 2008 17:47:23 +0000 (17:47 +0000)]
witness_addgraph() is required even if DDB isn't compiled into the kernel,
so exclude it from #ifdef DDB.

Submitted by: attilio

15 years ago- When renaming a drive, also set the drive name in the gvinum header.
Ulf Lilleengen [Sat, 19 Jul 2008 13:53:11 +0000 (13:53 +0000)]
- When renaming a drive, also set the drive name in the gvinum header.

PR: kern/125632
Approved by: pjd (mentor)
MFC after: 3 days

15 years agoAdd DDB "show conifhk" command, which lists hooks currently waiting
Robert Watson [Sat, 19 Jul 2008 12:12:54 +0000 (12:12 +0000)]
Add DDB "show conifhk" command, which lists hooks currently waiting
for completion in run_interrupt_driven_config_hooks().  This is
helpful when trying to figure out which device drivers have gone
into la-la land during boot-time autoconfiguration.

MFC after: 3 days

15 years agoFix a race which could result in some timeout buckets being skipped.
Jeff Roberson [Sat, 19 Jul 2008 05:18:29 +0000 (05:18 +0000)]
Fix a race which could result in some timeout buckets being skipped.
 - When a tick occurs on a cpu, iterate from cs_softticks until ticks.
   The per-cpu tick processing happens asynchronously with the actual
   adjustment of the 'ticks' variable.  Sometimes the results may
   be visible before the local call and sometimes after.  Previously this
   could cause a one tick window where we didn't evaluate the bucket.
 - In softclock fetch curticks before incrementing cc_softticks so we
   don't skip insertions which were made for the current time.

Sponsored by: Nokia

15 years ago - Check whether we've recorded this tick in ts_ticks on another cpu in
Jeff Roberson [Sat, 19 Jul 2008 05:13:47 +0000 (05:13 +0000)]
 - Check whether we've recorded this tick in ts_ticks on another cpu in
   sched_tick() to prevent multiple increments for one tick.  This pushes
   the value out of range and breaks priority calculation.

Reviewed by: kib
Found by: pho/nokia
Sponsored by: Nokia
MFC after: 3 days

15 years agoDisable SSP on arm for the time being.
Olivier Houchard [Sat, 19 Jul 2008 00:19:16 +0000 (00:19 +0000)]
Disable SSP on arm for the time being.
The segfaults when using SSP seem to be a gcc bug, a patch is available
in the gcc bugzilla, and will be imported once it's committed
into the official gcc tree.

15 years agoStaticify symbols that will not be used in other
Xin LI [Sat, 19 Jul 2008 00:13:26 +0000 (00:13 +0000)]
Staticify symbols that will not be used in other
places.

15 years agoUse %zd for size_t. With this gcore(1) is WARNS=6 clean.
Xin LI [Fri, 18 Jul 2008 23:37:05 +0000 (23:37 +0000)]
Use %zd for size_t.  With this gcore(1) is WARNS=6 clean.

15 years agoIndent the else path, reduce diff against OpenBSD.
Xin LI [Fri, 18 Jul 2008 23:35:31 +0000 (23:35 +0000)]
Indent the else path, reduce diff against OpenBSD.

15 years agoCorrect an error in pmap_change_attr()'s initial loop that verifies that the
Alan Cox [Fri, 18 Jul 2008 22:05:51 +0000 (22:05 +0000)]
Correct an error in pmap_change_attr()'s initial loop that verifies that the
given range of addresses are mapped.  Previously, the loop was testing the
same address every time.

Submitted by: Magesh Dhasayyan

15 years agoSimplify pmap_extract()'s control flow, making it more like the related
Alan Cox [Fri, 18 Jul 2008 20:07:50 +0000 (20:07 +0000)]
Simplify pmap_extract()'s control flow, making it more like the related
functions pmap_extract_and_hold() and pmap_kextract().

15 years agoEnhance arena_chunk_map_t to directly support run coalescing, and use
Jason Evans [Fri, 18 Jul 2008 19:35:44 +0000 (19:35 +0000)]
Enhance arena_chunk_map_t to directly support run coalescing, and use
the chunk map instead of red-black trees where possible.  Remove the
red-black trees and node objects that are obsoleted by this change.  The
net result is a ~1-2% memory savings, and a substantial allocation speed
improvement.

15 years agoEliminate stale comments from kmem_malloc().
Alan Cox [Fri, 18 Jul 2008 17:41:31 +0000 (17:41 +0000)]
Eliminate stale comments from kmem_malloc().

15 years agoFix a longstanding bug, from Otto Moerbeck:
Kevin Lo [Fri, 18 Jul 2008 15:05:14 +0000 (15:05 +0000)]
Fix a longstanding bug, from Otto Moerbeck:
if we're reducing a rule that has an empty
right hand side and the yacc stackpointer is pointing at the very
end of the allocated stack, we end up accessing the stack out of
bounds by the implicit $$ = $1 action

Obtained from: OpenBSD

15 years agoRemove unnessasary cast
Kevin Lo [Fri, 18 Jul 2008 14:55:22 +0000 (14:55 +0000)]
Remove unnessasary cast

15 years agoAdd a man page for the DNS accept filter.
David Malone [Fri, 18 Jul 2008 14:46:01 +0000 (14:46 +0000)]
Add a man page for the DNS accept filter.

15 years agoAdd an accept filter for TCP based DNS requests. It waits until the
David Malone [Fri, 18 Jul 2008 14:44:51 +0000 (14:44 +0000)]
Add an accept filter for TCP based DNS requests. It waits until the
whole first request is present before returning from accept.

15 years agoEliminate use of the global ripsrc which was being used to pass address
Robert Watson [Fri, 18 Jul 2008 10:47:07 +0000 (10:47 +0000)]
Eliminate use of the global ripsrc which was being used to pass address
information from rip_input() to rip_append().  Instead, pass the source
address for an IP datagram to rip_append() using a stack-allocated
sockaddr_in, similar to udp_input() and udp_append().

Prior to the move to rwlocks for inpcbinfo, this was not a problem, as
use of the global was synchronized using the ripcbinfo mutex, but with
read-locking there is the potential for a race during concurrent
receive.

This problem is not present in the IPv6 raw IP socket code, which
already used a stack variable for the address.

Spotted by: mav
MFC after: 1 week (before inpcbinfo rwlock changes)

15 years agorevert local change
Kip Macy [Fri, 18 Jul 2008 07:10:33 +0000 (07:10 +0000)]
revert local change

15 years agorevert change from local tree
Kip Macy [Fri, 18 Jul 2008 07:07:57 +0000 (07:07 +0000)]
revert change from local tree

15 years agonew vendor PHY support
Kip Macy [Fri, 18 Jul 2008 07:01:51 +0000 (07:01 +0000)]
new vendor PHY support

15 years agorevert changes accidentally included in last commit
Kip Macy [Fri, 18 Jul 2008 06:22:57 +0000 (06:22 +0000)]
revert changes accidentally included in last commit

15 years agoEliminate unused global variables. (These global variables became fields of
Alan Cox [Fri, 18 Jul 2008 06:14:36 +0000 (06:14 +0000)]
Eliminate unused global variables.  (These global variables became fields of
struct kva_md_info many years ago.)

15 years agoimport vendor fixes to cxgb
Kip Macy [Fri, 18 Jul 2008 06:12:31 +0000 (06:12 +0000)]
import vendor fixes to cxgb

15 years agoCorrect 1000Mbps link handling logic for JMC250. This should make
Pyun YongHyeon [Fri, 18 Jul 2008 04:20:48 +0000 (04:20 +0000)]
Correct 1000Mbps link handling logic for JMC250. This should make
jme(4) run on 1000Mbps link.

15 years agoSort the .PATH entries to give a more reasonable order of precedence:
David Schultz [Fri, 18 Jul 2008 02:18:34 +0000 (02:18 +0000)]
Sort the .PATH entries to give a more reasonable order of precedence:
     1. architecture-specific files
     2. long double format-specific files
     3. bsdsrc
     4. src
     5. man
The original order was virtually the opposite of this.

This should not cause any functional changes at this time. The
difference is only significant when one wants to override, say, a
generic foo.c with a more specialized foo.c (as opposed to foo.S).

15 years agoUse DELAY() instead of pause if waiting time is less than 1ms.
Pyun YongHyeon [Fri, 18 Jul 2008 01:00:54 +0000 (01:00 +0000)]
Use DELAY() instead of pause if waiting time is less than 1ms.
This will fix driver hang if hz < 1000.

Pointed out by: thompsa

15 years agoFix a benign typo that would give out an incorrect warning message.
Luoqi Chen [Thu, 17 Jul 2008 22:40:23 +0000 (22:40 +0000)]
Fix a benign typo that would give out an incorrect warning message.
Change a get-or-set sequence on OHCI_COMMAND_STATUS register which
is "write to set" to a simple set.

15 years agoMake it more clear what it means that SA_RESTART is set for signal
Simon L. B. Nielsen [Thu, 17 Jul 2008 21:54:23 +0000 (21:54 +0000)]
Make it more clear what it means that SA_RESTART is set for signal
handlers added with signal(3).

Submitted by: gnn (slightly modified by me)
MFC after: 1 week

15 years ago- dns queries might go also over TCP, so allow it.
Daniel Gerzo [Thu, 17 Jul 2008 20:00:18 +0000 (20:00 +0000)]
- dns queries might go also over TCP, so allow it.

Approved by: rink
MFC after: 1 week

15 years agoPair the VOP_OPEN call from do_execve() with the reciprocal VOP_CLOSE.
Konstantin Belousov [Thu, 17 Jul 2008 16:44:07 +0000 (16:44 +0000)]
Pair the VOP_OPEN call from do_execve() with the reciprocal VOP_CLOSE.
This was unnoticed because local filesystems usually do nothing
non-trivial in the close vop.

Reported and tested by: Rick Macklem
MFC after: 2 weeks

15 years agoClean up mxge's use of callouts as pointed out by jhb,
Andrew Gallatin [Thu, 17 Jul 2008 15:46:35 +0000 (15:46 +0000)]
Clean up mxge's use of callouts as pointed out by jhb,
and handle NIC hardware watchdog resets.

- remove buggy code at the top of mxge_tick() which tried
  to detect a race which is already detected in the kernel's
  callout code.

- move callout_stop() and callout_reset() into mxge_close()
  mxge_open() rather than doing the callout manipulation
  all over the place.

- use callout_drain(), rather than callout_stop() to prevent
  a potential race between mxge_tick() and mxge_detach()
  which could lead to softclock using a destroyed mutex

- restructure the mxge_tick() and mxge_watchdog_reset()
  routines to avoid resetting a callout, and then
  immediately stopping it if the watchdog reset routine
  is called, and fails.

- enable the driver to handle NIC hardware watchdog
  resets by restoring the NIC's PCI config space, which is
  lost when the NIC hardware watchdog triggers.

Reviewed by: jhb (previus version)

15 years agoDon't set PACKAGE_BUILDING while installing packages.
Florent Thoumie [Thu, 17 Jul 2008 15:08:26 +0000 (15:08 +0000)]
Don't set PACKAGE_BUILDING while installing packages.

Submitted by: erwin
Discussed on: http://lists.freebsd.org/pipermail/freebsd-ports/2008-June/049074.html
MFC after: 1 week

15 years agoBefore updating the password database, the pw(8) utility first performs a
John Baldwin [Thu, 17 Jul 2008 13:47:59 +0000 (13:47 +0000)]
Before updating the password database, the pw(8) utility first performs a
sanity check by invoking "pwd_mkdb -C".  However, if this failed it
silently returned success.  Fix this so it fails the update operation
instead.

MFC after: 1 week

15 years agoAdd the shutdown KEYWORD to those scripts that start persistent services
Doug Barton [Wed, 16 Jul 2008 19:50:29 +0000 (19:50 +0000)]
Add the shutdown KEYWORD to those scripts that start persistent services
to allow them to do a "clean" shutdown.

I purposely avoided making changes to network-related stuff since the
system shutting down is pretty conclusive, and there may be complicated
dependencies on the network that I would rather not try to unravel.

I also skipped kerberos-related stuff for the reasons above, and
because I have no way to test it.

15 years agoAs previously discussed, add the svn:executable property to all scripts
Doug Barton [Wed, 16 Jul 2008 19:22:48 +0000 (19:22 +0000)]
As previously discussed, add the svn:executable property to all scripts

15 years agoMove the TCSA* definitions out of _KERNEL. They are processed in libc.
Ed Schouten [Wed, 16 Jul 2008 12:36:39 +0000 (12:36 +0000)]
Move the TCSA* definitions out of _KERNEL. They are processed in libc.

The tcsetattr() routine already converts the TCSA* arguments to their
respective TIOCSETA* ioctl's in the C library. There is no need to have
these definitions inside the kernel.

Approved by: philip (mentor, implicit)

15 years agoSort the ioctl's in <sys/ttycom.h> by number.
Ed Schouten [Wed, 16 Jul 2008 11:23:15 +0000 (11:23 +0000)]
Sort the ioctl's in <sys/ttycom.h> by number.

I think one of the reasons why we have so many conflicts in the TTY
ioctl category, is because the ioctl's aren't ordered logically. This
commit only sorts them by number. The comments may still be inaccurate.

Approved by: philip (mentor)

15 years agoRemove OTTYDISC, NETLDISC and NTTYDISC definitions.
Ed Schouten [Wed, 16 Jul 2008 11:20:04 +0000 (11:20 +0000)]
Remove OTTYDISC, NETLDISC and NTTYDISC definitions.

When I ported most applications away from <sgtty.h>, I noticed none of
them were actually using these definitions. I kept them in place,
because I didn't want to touch tools like pstat(8) and stty(1).

In preparation for the MPSAFE TTY layer, remove these definitions. This
doesn't have any impact with respect to binary compatibility (see
tty_conf.c).

We couldn now add an #error to <sys/ioctl_compat.h> when included
outside the kernel. Unfortunately, kdump's mkioctls includes this file
unconditionally.

Approved by: philip (mentor)

15 years agoFix error in comment.
Robert Watson [Wed, 16 Jul 2008 10:55:50 +0000 (10:55 +0000)]
Fix error in comment.

MFC after: 3 weeks

15 years agoFix a multicast handling regression on VT6105M introduced in
Pyun YongHyeon [Wed, 16 Jul 2008 08:35:29 +0000 (08:35 +0000)]
Fix a multicast handling regression on VT6105M introduced in
vr(4) overhauling(r177050).

It seems that filtering multicast addresses with multicast CAM
entries require accessing 'CAM enable bit' for each CAM entry.
Subsequent accessing multicast CAM control register without
toggling the 'CAM enable bit' seem to no effects.
In order to fix that separate CAM setup from CAM mask configuration
and CAM entry modification. While I'm here add VLAN CAM filtering
feature which will be enabled in future(FreeBSD now can receive
VLAN id insertion/removal event from vlan(4) on the fly).

For VT6105M hardware, explicitly disable VLAN hardware tag
insertion/stripping and enable VLAN CAM filtering for VLAN id 0.
This shall make non-VLAN frames set VR_RXSTAT_VIDHIT bit in Rx
status word.

Added multicast/VLAN CAM address definition to header file.

PR: kern/125010, kern/125024
MFC after: 1 week

15 years agoFix VR_RXSTAT_RX_OK bit definition which lasted for more than 9
Pyun YongHyeon [Wed, 16 Jul 2008 08:02:23 +0000 (08:02 +0000)]
Fix VR_RXSTAT_RX_OK bit definition which lasted for more than 9
years. All datasheet I have indicates the bit 15 is the
VR_RXSTAT_RX_OK. The bit 14 is reserved for all Rhine family
except VT6105M. VT6105M uses that bit to indicate a VLAN frame
with matching CAM VLAN id.
Use the VR_RXSTAT_RX_OK instead of VR_RXSTAT_RXERR when vr(4)
checks the validity of received frame.
This should fix occasional dropping frames on VT6105M.

Tested by: Goran Lowkrantz ( goran.lowkrantz at ismobile dot com )
MFC after: 1 week

15 years agoInvoke err() with a format string rather than directly with a passed
Robert Watson [Tue, 15 Jul 2008 16:07:34 +0000 (16:07 +0000)]
Invoke err() with a format string rather than directly with a passed
command line argument.

Submitted by: Alfredo Ortega <alfredo at coresecurity dot com>
Thanks to: Core Security Technologies
MFC after: 3 days

15 years agoMerge last of a series of rwlock conversion changes to UDP, which
Robert Watson [Tue, 15 Jul 2008 15:38:47 +0000 (15:38 +0000)]
Merge last of a series of rwlock conversion changes to UDP, which
completes the move to a fully parallel UDP transmit path by using
global read, rather than write, locking of inpcbinfo in further
semi-connected cases:

- Add macros to allow try-locking of inpcb and inpcbinfo.
- Always acquire an incpcb read lock in udp_output(), which stablizes the
  local inpcb address and port bindings in order to determine what further
  locking is required:
  - If the inpcb is currently not bound (at all) and are implicitly
    connecting, we require inpcbinfo and inpcb write locks, so drop the
    read lock and re-acquire.
  - If the inpcb is bound for at least one of the port or address, but an
    explicit source or destination is requested, trylock the inpcbinfo
    lock, and if that fails, drop the inpcb lock, lock the global lock,
    and relock the inpcb lock.
  - Otherwise, no further locking is required (common case).
- Update comments.

In practice, this means that the vast majority of consumers of UDP sockets
will not acquire any exclusive locks at the socket or UDP levels of the
network stack.  This leads to a marked performance improvement in several
important workloads, including BIND, nsd, and memcached over UDP, as well
as significant improvements in pps microbenchmarks.

The plan is to MFC all of the rwlock changes to RELENG_7 once they have
settled for a weeks in the tree.

Tested by: ps, kris (older revision), bde
MFC after: 3 weeks

15 years agoFix commment in typo.
Rui Paulo [Tue, 15 Jul 2008 10:32:35 +0000 (10:32 +0000)]
Fix commment in typo.

M    tcp_output.c

15 years agoUpdate the definition of modspecific structure
Daniel Gerzo [Tue, 15 Jul 2008 10:06:37 +0000 (10:06 +0000)]
Update the definition of modspecific structure

PR: docs/125630

15 years agoUpdate bus_dmamem_alloc()'s first call to malloc() such that M_WAITOK is
Alan Cox [Tue, 15 Jul 2008 03:34:49 +0000 (03:34 +0000)]
Update bus_dmamem_alloc()'s first call to malloc() such that M_WAITOK is
specified when appropriate.

Reviewed by: scottl

15 years agoAdd quirk for Dell D630 laptops.
Xin LI [Tue, 15 Jul 2008 02:34:44 +0000 (02:34 +0000)]
Add quirk for Dell D630 laptops.

Tested by: Quake Lee <quakelee geekcn org>,
Robert Noland <rnoland 2hip net>
MFC after: 1 week
Approved by: ariff

15 years agoAllow injecting big packets via bpf(4) up to min(MTU, 16K-byte).
Jung-uk Kim [Mon, 14 Jul 2008 22:41:48 +0000 (22:41 +0000)]
Allow injecting big packets via bpf(4) up to min(MTU, 16K-byte).

MFC after: 1 week

15 years agoMatch the implementation of the inline function from libkern.h.
David E. O'Brien [Mon, 14 Jul 2008 21:36:02 +0000 (21:36 +0000)]
Match the implementation of the inline function from libkern.h.

15 years agoFix carp(4) panics that can occur during carp interface configuration.
Ermal Luçi [Mon, 14 Jul 2008 20:11:51 +0000 (20:11 +0000)]
Fix carp(4) panics that can occur during carp interface configuration.

Approved by: mlaier (mentor)
Reported by: Scott Ullrich
MFC after: 1 week

15 years agoUnmangle an E.W.Dijkstra quote.
Poul-Henning Kamp [Mon, 14 Jul 2008 18:45:06 +0000 (18:45 +0000)]
Unmangle an E.W.Dijkstra quote.

15 years agoAdd event notification at attach/detach so the NIC
Jack F Vogel [Mon, 14 Jul 2008 18:40:21 +0000 (18:40 +0000)]
Add event notification at attach/detach so the NIC
is able to detect it and do hardware filtering.

15 years agoAdd an event handler to the vlan driver so the NIC driver
Jack F Vogel [Mon, 14 Jul 2008 18:38:52 +0000 (18:38 +0000)]
Add an event handler to the vlan driver so the NIC driver
becomes aware of it, and gets the VLAN ID. This will allow
the easy use of VLAN hardware filtering by adapters that
support it.

15 years agoManpage for the et(4) driver.
Christian Brueffer [Mon, 14 Jul 2008 18:15:43 +0000 (18:15 +0000)]
Manpage for the et(4) driver.

Obtained from: DragonFly

15 years agoFill in the string portion of the bluetooth stack version sysctl.
Tom Rhodes [Mon, 14 Jul 2008 13:45:05 +0000 (13:45 +0000)]
Fill in the string portion of the bluetooth stack version sysctl.

Approved by: emax

15 years ago- Enhance volume handling
Pietro Cerutti [Mon, 14 Jul 2008 13:22:09 +0000 (13:22 +0000)]
- Enhance volume handling

PR: 125054
Submitted by: gahr
Approved by: cperciva

15 years agoChange the character prefixed to the svn version to "r" since that seems
Doug Barton [Sun, 13 Jul 2008 20:08:38 +0000 (20:08 +0000)]
Change the character prefixed to the svn version to "r" since that seems
to be how they are commonly referred to.

15 years agoHandle a race between pmap_kextract() and pmap_promote_pde(). This race
Alan Cox [Sun, 13 Jul 2008 18:19:53 +0000 (18:19 +0000)]
Handle a race between pmap_kextract() and pmap_promote_pde().  This race
caused ZFS to crash when restoring a snapshot with superpage promotion
enabled.

Reported by: kris

15 years agoAdd an obsolete header after pecoff removal.
Antoine Brodin [Sun, 13 Jul 2008 17:17:35 +0000 (17:17 +0000)]
Add an obsolete header after pecoff removal.

Approved by: rwatson (mentor)

15 years agoStaticize M_STACK.
Antoine Brodin [Sun, 13 Jul 2008 17:15:05 +0000 (17:15 +0000)]
Staticize M_STACK.

Approved by: rwatson (mentor)
MFC after: 1 month

15 years agoThe libc acl_valid(3) function validates the contents of a POSIX.1e ACL.
Robert Watson [Sun, 13 Jul 2008 16:37:51 +0000 (16:37 +0000)]
The libc acl_valid(3) function validates the contents of a POSIX.1e ACL.
This change removes the requirement that an ACL contain no ACL_USER
entries with a uid the same as those of a file, or ACL_GROUP entries
with a gid the same as those of a file.  This requirement is not in the
specification, and not enforced by the kernel's ACL implementation.

Reported by: Iustin Pop <iusty at k1024 dot org>
MFC after: 1 week

15 years agoMore explicitly mention that we support the 2920C with the ahc driver.
Remko Lodder [Sun, 13 Jul 2008 13:56:21 +0000 (13:56 +0000)]
More explicitly mention that we support the 2920C with the ahc driver.

PR: 125535
Submitted by: Bob Bishop <rb at gid dot co dot uk>
MFC after: 3 days