]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
13 years ago- Added systcls for header splitting, RX/TX buffer count, interrupt
David Christensen [Mon, 7 Feb 2011 23:00:24 +0000 (23:00 +0000)]
- Added systcls for header splitting, RX/TX buffer count, interrupt
  coalescing, strict RX MTU, verbose output, and shared memory debug.
- Added additional debug counters (VLAN tags and split header frames).
- Updated debug counters to 64 bit definitions.
- Updated l2fhdr bit definitions.
- Combined RX buffer sizing into a single function.
- Added buffer size and interrupt coalescing settings to adapter info
  printout.

Submitted by: davidch
MFC after: 2 weeks

13 years agoCuritel UM150 needs a quirk to stop it from detaching straight after
Nick Hibma [Mon, 7 Feb 2011 22:37:27 +0000 (22:37 +0000)]
Curitel UM150 needs a quirk to stop it from detaching straight after
attach (resetting actually).

Submitted by: Oleg Nauman
MFC after: 1 week

13 years agoUpdate manpage to remove CRT reference.
Glen Barber [Mon, 7 Feb 2011 22:33:39 +0000 (22:33 +0000)]
Update manpage to remove CRT reference.

PR: 154441
Submitted by: Ben Haga (bhaga of absoludicrous com)
Suggestion by: pluknet
Approved by: keramida (mentor)
MFC after: 3 days

13 years agoAdd in some AR9280 specific board configuration options.
Adrian Chadd [Mon, 7 Feb 2011 22:00:31 +0000 (22:00 +0000)]
Add in some AR9280 specific board configuration options.

* The existing radio config code was for the AR5416/AR9160 and missed out
  on some of the AR9280 specific stuff. Include said stuff from ath9k.

* Refactor out the gain control settings into a new function, again pilfered
  from ath9k.

* Use the analog register RMW macro when touching analog registers.

Obtained from: Linux ath9k

13 years agoBring over some AR9280-specific v14 additions that exist in ath9k.
Adrian Chadd [Mon, 7 Feb 2011 21:48:26 +0000 (21:48 +0000)]
Bring over some AR9280-specific v14 additions that exist in ath9k.

Obtained from: Linux ath9k

13 years agoFix the previous commit of pthread_getthreadid_np(3).
Jung-uk Kim [Mon, 7 Feb 2011 21:31:21 +0000 (21:31 +0000)]
Fix the previous commit of pthread_getthreadid_np(3).

13 years agoUse analog delay macro for modifying an analog register.
Adrian Chadd [Mon, 7 Feb 2011 21:30:56 +0000 (21:30 +0000)]
Use analog delay macro for modifying an analog register.

13 years agoAdd a new RMW macro for analog register writes which implements the needed
Adrian Chadd [Mon, 7 Feb 2011 21:30:13 +0000 (21:30 +0000)]
Add a new RMW macro for analog register writes which implements the needed
wait period between operations.

13 years agoIntroduce a non-portable function pthread_getthreadid_np(3) to retrieve
Jung-uk Kim [Mon, 7 Feb 2011 21:26:46 +0000 (21:26 +0000)]
Introduce a non-portable function pthread_getthreadid_np(3) to retrieve
calling thread's unique integral ID, which is similar to AIX function of
the same name.  Bump __FreeBSD_version to note its introduction.

Reviewed by: kib

13 years ago- Use LINE_MAX from limits.h as the maximum line length instead of
Jaakko Heinonen [Mon, 7 Feb 2011 18:10:18 +0000 (18:10 +0000)]
- Use LINE_MAX from limits.h as the maximum line length instead of
  BUFSIZ. Use LINE_MAX * 2 as the buffer size (BSIZE).
- Error out if we encounter a line longer than LINE_MAX. The previous
  behavior was to silently split long lines and produce corrupted
  output.

PR: bin/151384

13 years agoHandle EOF when skipping lines.
Jaakko Heinonen [Mon, 7 Feb 2011 18:05:56 +0000 (18:05 +0000)]
Handle EOF when skipping lines.

13 years agoFix typo in SIFS setup
Adrian Chadd [Mon, 7 Feb 2011 17:04:31 +0000 (17:04 +0000)]
Fix typo in SIFS setup

13 years agoAdd in a per phy error sysctl.
Adrian Chadd [Mon, 7 Feb 2011 15:09:13 +0000 (15:09 +0000)]
Add in a per phy error sysctl.

13 years agoFix bugs related to M_FLOWID:
Michael Tuexen [Mon, 7 Feb 2011 15:04:23 +0000 (15:04 +0000)]
Fix bugs related to M_FLOWID:
* Store the flowid when receiving an SCTP/IPv6 packet.
* Store the flowid when receiving an SCTP packet with wrong CRC.
* Initilize flowid correctly.
* Put test code under INVARIANTS.
MFC after: 3 months.

13 years agoMinor grammar fix. s/cores'/core's/
Glen Barber [Mon, 7 Feb 2011 11:20:58 +0000 (11:20 +0000)]
Minor grammar fix.  s/cores'/core's/

Approved by: keramida (mentor)
MFC after: 3 days

13 years agoAdd "iface name" and "iface description" commands.
Brian Somers [Mon, 7 Feb 2011 11:18:18 +0000 (11:18 +0000)]
Add "iface name" and "iface description" commands.

PR: 151400
Submitted by: Aragon Gouveia - aragon at phat dot za dot net with minor fixes
MFC after: 3 weeks

13 years agoIf not set (due to some error Michael is working on
Randall Stewart [Mon, 7 Feb 2011 08:12:24 +0000 (08:12 +0000)]
If not set (due to some error Michael is working on
fixing) set it for the net.

MFC after: 3 months

13 years ago1) Track when flowid does get set.
Randall Stewart [Mon, 7 Feb 2011 08:10:29 +0000 (08:10 +0000)]
1) Track when flowid does get set.
MFC after: 3 months

13 years agoMFi386: revision 218329
Yoshihiro Takahashi [Mon, 7 Feb 2011 07:52:53 +0000 (07:52 +0000)]
MFi386: revision 218329

  Fix linking of the kernel without device npx.

13 years agoMFi386: revision 218327
Yoshihiro Takahashi [Mon, 7 Feb 2011 07:51:31 +0000 (07:51 +0000)]
MFi386: revision 218327

  Clear the padding when returning context to the usermode, for
  MI ucontext_t and x86 MD parts.
  Kernel allocates the structures on the stack, and not clearing
  reserved fields and paddings causes leakage.

13 years agoMFi386: revision 217886
Yoshihiro Takahashi [Mon, 7 Feb 2011 07:39:09 +0000 (07:39 +0000)]
MFi386: revision 217886

  Set td_kstack_pages for thread0.

13 years agoRemove use_high from the softc and simply check the number of GPIO
Marcel Moolenaar [Mon, 7 Feb 2011 05:36:32 +0000 (05:36 +0000)]
Remove use_high from the softc and simply check the number of GPIO
pins to determine whether there's a high register set or not. This
allows platform_gpio_init() to work without duplicating the work
done in the attach method.

13 years agounbreak mutlicast hash creation for the second time.
Bernd Walter [Mon, 7 Feb 2011 01:13:57 +0000 (01:13 +0000)]
unbreak mutlicast hash creation for the second time.
at91_emac hardware is *not* using ether_crc32_be algorithm!

13 years agoMake it impossible to clear the MNT_NFS4ACLS flag on ZFS filesystem
Edward Tomasz Napierala [Sun, 6 Feb 2011 23:34:09 +0000 (23:34 +0000)]
Make it impossible to clear the MNT_NFS4ACLS flag on ZFS filesystem
by using "mount -uw".

Reviewed by: pjd
MFC after: 2 weeks

13 years agorfork_thread(3): Mark deprecated in favor of pthread_create(3).
Jilles Tjoelker [Sun, 6 Feb 2011 23:26:14 +0000 (23:26 +0000)]
rfork_thread(3): Mark deprecated in favor of pthread_create(3).

13 years agoUpdate to BIND 9.6.3, the latest from ISC on the 9.6 branch.
Doug Barton [Sun, 6 Feb 2011 22:46:07 +0000 (22:46 +0000)]
Update to BIND 9.6.3, the latest from ISC on the 9.6 branch.

All 9.6 users with DNSSEC validation enabled should upgrade to this
version, or the latest version in the 9.7 branch, prior to 2011-03-31
in order to avoid validation failures for names in .COM as described
here:

https://www.isc.org/announcement/bind-9-dnssec-validation-fails-new-ds-record

In addition the fixes for this and other bugs, there are also the
following:

  * Various fixes to kerberos support, including GSS-TSIG
  * Various fixes to avoid leaking memory, and to problems that could
    prevent a clean shutdown of named

13 years agoo) Cavium Octeon doesn't need nop barriers.
Juli Mallett [Sun, 6 Feb 2011 22:21:18 +0000 (22:21 +0000)]
o) Cavium Octeon doesn't need nop barriers.
o) Have mips_wblush just do syncw, not sync on Cavium Octeon.
o) Add support for reading and writing some Octeon-specific registers.
   NB: Some of these are not entirely Octeon-specific.

Submitted by: Bhanu Prakash

13 years agoComment-out FLOWTABLE. It causes a kernel panic due to a misaligned memory
Marcel Moolenaar [Sun, 6 Feb 2011 22:18:37 +0000 (22:18 +0000)]
Comment-out FLOWTABLE. It causes a kernel panic due to a misaligned memory
access related to an IPv6 route update.

PR: kern/148018

13 years agoJust tag all RX packets as needing reorder processing for now.
Adrian Chadd [Sun, 6 Feb 2011 20:11:24 +0000 (20:11 +0000)]
Just tag all RX packets as needing reorder processing for now.

This fixes two problems -

* All packets need to be processed here, not just aggregate ones - as any
  received frames (AMPDU or otherwise) in the given TID (traffic class id)
  will update the sequence number and, implied with that, update the window;
* It seems there's situations where packets aren't matching a current node but
  somehow need to be tracked. Thus just tag them all for now; I'll figure out
  the why later.

Whilst I'm here, bump the stats counters whilst I'm at it.

This fixes AMPDU RX in my tests; the main problems now stem from what look
like PHY level error/retransmits which are impeding general throughput, incl.
AMPDU.

13 years agoOnly tag packets with the A-MPDU bit if they were part of an A-MPDU RX.
Adrian Chadd [Sun, 6 Feb 2011 17:40:27 +0000 (17:40 +0000)]
Only tag packets with the A-MPDU bit if they were part of an A-MPDU RX.

Whilst I'm here, add a counter to count said packets.

13 years agoNow that we break the loop on fstat(2) failure we no longer need to satisfy
Pawel Jakub Dawidek [Sun, 6 Feb 2011 14:17:08 +0000 (14:17 +0000)]
Now that we break the loop on fstat(2) failure we no longer need to satisfy
gcc's imperfections.

MFC after: 1 week

13 years agoAdd (void) cast before snprintf(3)s for which we are not interested in return
Pawel Jakub Dawidek [Sun, 6 Feb 2011 14:09:19 +0000 (14:09 +0000)]
Add (void) cast before snprintf(3)s for which we are not interested in return
values.

MFC after: 1 week

13 years agoTreat fstat(2) failure (different than EBADF) as fatal error.
Pawel Jakub Dawidek [Sun, 6 Feb 2011 14:07:58 +0000 (14:07 +0000)]
Treat fstat(2) failure (different than EBADF) as fatal error.

Reported by: Mikolaj Golub <to.my.trociny@gmail.com>
MFC after: 1 week

13 years agoOpen syslog when logging sysconf(3) failure.
Pawel Jakub Dawidek [Sun, 6 Feb 2011 14:06:37 +0000 (14:06 +0000)]
Open syslog when logging sysconf(3) failure.

Reported by: Mikolaj Golub <to.my.trociny@gmail.com>
MFC after: 1 week

13 years ago1) Use same scheme Michael and I discussed for a selected for a flowid
Randall Stewart [Sun, 6 Feb 2011 13:17:40 +0000 (13:17 +0000)]
1) Use same scheme Michael and I discussed for a selected for a flowid
2) If flowid is not set, arrange so it is stored.
3) If flowid is set by lower layer, use it.

MFC after: 3 Months

13 years agoClose more descriptors that can be open if the worker process for the given
Pawel Jakub Dawidek [Sun, 6 Feb 2011 12:21:29 +0000 (12:21 +0000)]
Close more descriptors that can be open if the worker process for the given
resource is already running.

Submitted by: Mikolaj Golub <to.my.trociny@gmail.com>
MFC after: 1 week

13 years ago- Use tabs not spaces
Martin Wilke [Sun, 6 Feb 2011 11:39:51 +0000 (11:39 +0000)]
- Use tabs not spaces

Reported by: bschmidt

13 years agoFix cut&paste mistake.
Bernhard Schmidt [Sun, 6 Feb 2011 10:25:11 +0000 (10:25 +0000)]
Fix cut&paste mistake.

MFC after: 1 week

13 years ago- Add Dell Wireless 1390 BCM4311 Mini PCI to supported list.
Martin Wilke [Sun, 6 Feb 2011 09:49:55 +0000 (09:49 +0000)]
- Add Dell Wireless 1390 BCM4311 Mini PCI to supported list.

PR: 154500
Submitted by: Ruslan Mahmatkhanov <cvs-src@yandex.ru>

13 years agocorrect the 'output_time' of packets generated by dummynet.
Luigi Rizzo [Sat, 5 Feb 2011 23:32:17 +0000 (23:32 +0000)]
correct the 'output_time' of packets generated by dummynet.
In the dec.2009 rewrite I introduced a bug, using for the
computation the arrival time instead of the time the packet
has exited from the queue.
The bandwidth computation was still correct because it is
computed elsewhere, but traffic was sent out in bursts.

The bug is also present in RELENG_8 after dec.2009

Thanks to Daikichi Osuga for investingating, finding and fixing the
bug with detailed graphs of the behaviour before and after the fix.

Submitted by: Daikichi Osuga
MFC after: 2 weeks

13 years agofollowing the recent (218130) changes to Makefile and Makefile.inc1,
Luigi Rizzo [Sat, 5 Feb 2011 23:03:40 +0000 (23:03 +0000)]
following the recent (218130) changes to Makefile and Makefile.inc1,
we need to set TARGET and TARGET_ARCH to get a correct WMAKEENV.
I am setting both to i386 since this is what picobsd is used for,
though there might be a better fix.

Add initial support for parallel make. This is disabled right now,
because there are incorrect dependencies somewhere which require
to run picobsd 2-3 times to complete a build.

MFC after: 2 weeks

13 years agosh: Weaken some tests to allow /rescue/sh to pass everything.
Jilles Tjoelker [Sat, 5 Feb 2011 23:00:24 +0000 (23:00 +0000)]
sh: Weaken some tests to allow /rescue/sh to pass everything.

/rescue/sh has a different _PATH_STDPATH which affects command -p.

13 years agoAdd a temporary workaround so the 11n rate scenario setup code sets a useful
Adrian Chadd [Sat, 5 Feb 2011 22:54:37 +0000 (22:54 +0000)]
Add a temporary workaround so the 11n rate scenario setup code sets a useful
TX chainmask.

since the upper layers don't (yet) know about the active TX/RX chainmasks,
it can't tell the rate scenario functions what to use. I'll eventually sort
this out; this restores functionality in the meantime.

13 years agoCatch up with reality and references from the latest RFCs
Doug Barton [Sat, 5 Feb 2011 21:43:37 +0000 (21:43 +0000)]
Catch up with reality and references from the latest RFCs
(especially 5735) for our default empty zones.

13 years agoUnless "cnt" exceeds MAX_COMMIT_COUNT, nfsrv_commit() and nfsvno_fsync() are
Alan Cox [Sat, 5 Feb 2011 21:21:27 +0000 (21:21 +0000)]
Unless "cnt" exceeds MAX_COMMIT_COUNT, nfsrv_commit() and nfsvno_fsync() are
incorrectly calling vm_object_page_clean().  They are passing the length of
the range rather than the ending offset of the range.

Perform the OFF_TO_IDX() conversion in vm_object_page_clean() rather than the
callers.

Reviewed by: kib
MFC after: 3 weeks

13 years agoAdd support for M_FLOWID.
Michael Tuexen [Sat, 5 Feb 2011 19:13:38 +0000 (19:13 +0000)]
Add support for M_FLOWID.

13 years agoEnable wide character support in libdialog.
Nathan Whitehorn [Sat, 5 Feb 2011 16:23:32 +0000 (16:23 +0000)]
Enable wide character support in libdialog.

13 years agoFix linking of the kernel without device npx.
Konstantin Belousov [Sat, 5 Feb 2011 15:37:10 +0000 (15:37 +0000)]
Fix linking of the kernel without device npx.

MFC after: 2 weeks

13 years agoClear the padding when returning context to the usermode, for
Konstantin Belousov [Sat, 5 Feb 2011 15:10:27 +0000 (15:10 +0000)]
Clear the padding when returning context to the usermode, for
MI ucontext_t and x86 MD parts.
Kernel allocates the structures on the stack, and not clearing
reserved fields and paddings causes leakage.

Noted and discussed with: bde
MFC after: 2 weeks

13 years agosh: Fix two things about {(...)} <redir:
Jilles Tjoelker [Sat, 5 Feb 2011 15:02:19 +0000 (15:02 +0000)]
sh: Fix two things about {(...)} <redir:

* In {(...) <redir1;} <redir2, do not drop redir1.
* Maintain the difference between (...) <redir and {(...)} <redir:
  In (...) <redir, the redirection is performed in the child, while in
  {(...)} <redir it should be performed in the parent (like {(...); :;}
  <redir)

13 years agosh: Remove clearcmdentry()'s now unused argument.
Jilles Tjoelker [Sat, 5 Feb 2011 14:08:51 +0000 (14:08 +0000)]
sh: Remove clearcmdentry()'s now unused argument.

13 years agosh: Forget all cached command locations on any PATH change.
Jilles Tjoelker [Sat, 5 Feb 2011 14:01:46 +0000 (14:01 +0000)]
sh: Forget all cached command locations on any PATH change.

POSIX requires this and it is simpler than the previous code that remembered
command locations when appending directories to PATH.

In particular,
  PATH=$PATH
is no longer a no-op but discards all cached command locations.

13 years agosh: Do not try to execute binary files as scripts.
Jilles Tjoelker [Sat, 5 Feb 2011 12:54:59 +0000 (12:54 +0000)]
sh: Do not try to execute binary files as scripts.

If execve() returns an [ENOEXEC] error, check if the file is binary before
trying to execute it using sh. A file is considered binary if at least one
of the first 256 bytes is '\0'.

In particular, trying to execute ELF binaries for the wrong architecture now
fails with an "Exec format error" message instead of syntax errors and
potentially strange results.

13 years ago1) Typo correction in comments and one spacing change.
Randall Stewart [Sat, 5 Feb 2011 12:12:51 +0000 (12:12 +0000)]
1) Typo correction in comments and one spacing change.
2) Mass update to all copyrights.
MFC after: 3 Months

13 years agophys_addr is a PA not a VA so declare it as a vm_paddr_t not a vm_offset_t.
Warner Losh [Sat, 5 Feb 2011 03:36:34 +0000 (03:36 +0000)]
phys_addr is a PA not a VA so declare it as a vm_paddr_t not a vm_offset_t.

13 years agoMake md_tp a register_t not a void *. This will keep us from
Warner Losh [Sat, 5 Feb 2011 03:30:29 +0000 (03:30 +0000)]
Make md_tp a register_t not a void *.  This will keep us from
accidentally dereferencng it and might be one fewer things to change
if arm64 happens...

Submitted by: rwatson's question on irc...

13 years agoVendor import of BIND 9.6.3
Doug Barton [Sat, 5 Feb 2011 03:14:55 +0000 (03:14 +0000)]
Vendor import of BIND 9.6.3

13 years agosh: Remove special code for shell scripts without magic number.
Jilles Tjoelker [Fri, 4 Feb 2011 22:47:55 +0000 (22:47 +0000)]
sh: Remove special code for shell scripts without magic number.

These are called "shell procedures" in the source.

If execve() failed with [ENOEXEC], the shell would reinitialize itself
and execute the program as a script. This requires a fair amount of code
which is not frequently used (most scripts have a #! magic number).
Therefore just execute a new instance of sh (_PATH_BSHELL) to run the
script.

13 years agoRemove duplicate .note.GNU-stack section declaration.
Konstantin Belousov [Fri, 4 Feb 2011 21:54:06 +0000 (21:54 +0000)]
Remove duplicate .note.GNU-stack section declaration.

Reported by: arundel

13 years agoSince the last parameter to vm_object_shadow() is a vm_size_t and not a
Alan Cox [Fri, 4 Feb 2011 21:49:24 +0000 (21:49 +0000)]
Since the last parameter to vm_object_shadow() is a vm_size_t and not a
vm_pindex_t, it makes no sense for its callers to perform atop().  Let
vm_object_shadow() do that instead.

13 years agoRemove duplicate .note.GNU-stack section declaration. bcopy already
Konstantin Belousov [Fri, 4 Feb 2011 21:04:00 +0000 (21:04 +0000)]
Remove duplicate .note.GNU-stack section declaration. bcopy already
made the neccessary provisions.

Reported by: arundel

13 years agoDo not let pw.conf(5) or -M option affect creation of basehome, e.g., /home.
Jung-uk Kim [Fri, 4 Feb 2011 19:49:02 +0000 (19:49 +0000)]
Do not let pw.conf(5) or -M option affect creation of basehome, e.g., /home.
When the basehome does not exist, it creates all intermediate directories as
required, which is logically equivalent to mkdir(1) with -m and -p options.
However, it modifies all intermediate directories, not just the final home
directory unlike mkdir.  This problem was introduced in two revisions, i.e.,
r1.59 (SVN r167919) and r1.60 (SVN r168044).

MFC after: 1 month

13 years agoCorrect include path.
John Baldwin [Fri, 4 Feb 2011 18:36:09 +0000 (18:36 +0000)]
Correct include path.

Submitted by: arundel
MFC after: 1 week

13 years agoDisable TX IP checksum offloading for RTL8168C controllers. The
Pyun YongHyeon [Fri, 4 Feb 2011 17:49:55 +0000 (17:49 +0000)]
Disable TX IP checksum offloading for RTL8168C controllers.  The
controller in question generates frames with bad IP checksum value
if packets contain IP options.  For instance, packets generated by
ping(8) with record route option have wrong IP checksum value. The
controller correctly computes checksum for normal TCP/UDP packets
though.
There are two known RTL8168/8111C variants in market and the issue
I observed happened on RL_HWREV_8168C_SPIN2. I'm not sure
RL_HWREV_8168C also has the same issue but it would be better to
assume it has the same issue since they shall share same core.
RTL8102E which is supposed to be released at the time of
RTL8168/8111C announcement does not have the issue.

Tested by: Konstantin V. Krotov ( kkv <> insysnet dot ru )

13 years agoMake sys_signame upper case.
Jilles Tjoelker [Fri, 4 Feb 2011 16:40:50 +0000 (16:40 +0000)]
Make sys_signame upper case.

This matches the constants from <signal.h> with 'SIG' removed, which POSIX
requires kill and trap to accept and 'kill -l' to write.

'kill -l', 'trap', 'trap -l' output is now upper case.

In Turkish locales, signal names with an upper case 'I' are now accepted,
while signal names with a lower case 'i' are no longer accepted, and the
output of 'killall -l' now contains proper capital 'I' without dot instead
of a dotted capital 'I'.

13 years agoWe can pass a format string and args to panic(), so instead of using
Ed Maste [Fri, 4 Feb 2011 15:45:48 +0000 (15:45 +0000)]
We can pass a format string and args to panic(), so instead of using
printf() to output some information before a panic, just include that
information in the panic.

Submitted by: bde
Reviewed by: bde

13 years agovdev's sectorsize should not be greater than 8 Kbytes and also
Andrey V. Elsukov [Fri, 4 Feb 2011 15:22:56 +0000 (15:22 +0000)]
vdev's sectorsize should not be greater than 8 Kbytes and also
it should be power of 2. This prevents non-aligned access while
probing vdev's labels.

PR: kern/147852
Reviewed by: pjd
MFC after: 1 week

13 years agoCollapse duplicate definitions of EXT2_SB().
John Baldwin [Fri, 4 Feb 2011 14:20:27 +0000 (14:20 +0000)]
Collapse duplicate definitions of EXT2_SB().

Submitted by: Pedro F. Giffuni  giffunip at yahoo

13 years agoAlways assert that the turnstile chain lock is held in turnstile_wait()
John Baldwin [Fri, 4 Feb 2011 14:16:41 +0000 (14:16 +0000)]
Always assert that the turnstile chain lock is held in turnstile_wait()
and remove a duplicate hash lookup.

MFC after: 1 week

13 years agoWhen turning off TCP_NOPUSH, only call tcp_output() to immediately flush
John Baldwin [Fri, 4 Feb 2011 14:13:15 +0000 (14:13 +0000)]
When turning off TCP_NOPUSH, only call tcp_output() to immediately flush
any pending data if the connection is established.

Submitted by: csjp
Reviewed by: lstewart
MFC after: 1 week

13 years agoUse M_WAITOK rather than M_NOWAIT when creating taskqueues via the
John Baldwin [Fri, 4 Feb 2011 14:06:57 +0000 (14:06 +0000)]
Use M_WAITOK rather than M_NOWAIT when creating taskqueues via the
TASKQUEUE_DEFINE macros.  All the places that use these macros to create
taskqueues assume that the operation succeeds.

MFC after: 1 week

13 years ago1) Fix cpu mapping per JB's suggestions
Randall Stewart [Fri, 4 Feb 2011 13:50:30 +0000 (13:50 +0000)]
1) Fix cpu mapping per JB's suggestions
2) Fix it so INIT's don't always end up on CPU0

MFC after: 3 months

13 years agoReplace __LP64__ with __mips_n64. This partly reverts r217147.
Tijl Coosemans [Fri, 4 Feb 2011 13:09:46 +0000 (13:09 +0000)]
Replace __LP64__ with __mips_n64. This partly reverts r217147.

Requested by: jmallett, imp
Approved by: kib (mentor)

13 years agoFix typo (Tuneable -> Tunable).
Rebecca Cran [Fri, 4 Feb 2011 12:03:48 +0000 (12:03 +0000)]
Fix typo (Tuneable -> Tunable).

13 years agoAdd PEX0 and select the first serial port as console.
Marcel Moolenaar [Fri, 4 Feb 2011 01:09:02 +0000 (01:09 +0000)]
Add PEX0 and select the first serial port as console.

13 years agoOops, fix newbie mistake that breaks the normal build.
Adrian Chadd [Fri, 4 Feb 2011 00:25:18 +0000 (00:25 +0000)]
Oops, fix newbie mistake that breaks the normal build.

13 years agosh: Return only 126 or 127 for execve() failures.
Jilles Tjoelker [Thu, 3 Feb 2011 23:38:11 +0000 (23:38 +0000)]
sh: Return only 126 or 127 for execve() failures.

Do not return 2 for errors other than [EACCES] or [ENOENT].

13 years agoFix several bugs in the stream schedulers.
Michael Tuexen [Thu, 3 Feb 2011 20:44:49 +0000 (20:44 +0000)]
Fix several bugs in the stream schedulers.
From Robin Seggelmann.

MFC after: 3 months.

13 years agoModify the TX path to set and use the 11n rate scenario bits.
Adrian Chadd [Thu, 3 Feb 2011 20:30:17 +0000 (20:30 +0000)]
Modify the TX path to set and use the 11n rate scenario bits.

This isn't strictly required to TX (at least non-agg and non-HT40,
non-short-GI) frames; but as it needs to be done anyway, just get
it done.

Linux ath9k uses the rate scenario style path for -all- packets,
legacy or otherwise. This code does much the same.

Beacon TX still uses the legacy, non-rate-scenario TX descriptor
setup. Ath9k also does this.

This 11n rate scenario path is only called for chips in the AR5416
HAL; legacy chips use the previous interface for TX'ing.

13 years agoDisable the code I previously added from Rui's 802.11n branch.
Adrian Chadd [Thu, 3 Feb 2011 20:26:26 +0000 (20:26 +0000)]
Disable the code I previously added from Rui's 802.11n branch.

A-MPDU RX interferes with packet retransmission/reordering.
In local testing, I was seeing A-MPDU being negotiated and then
not used by the AP sending frames to the STA; the STA would then
treat non A-MPDU frames that are retransmits as out of the window
and get plain confused.

The hardware RX status descriptor has a "I'm part of an aggregate"
bit; so this should eventually be tested and then punted to the
A-MPDU reorder handling only if it has this bit set.

13 years agoMake sure that changing the ECN sysctl does not affect
Michael Tuexen [Thu, 3 Feb 2011 19:59:00 +0000 (19:59 +0000)]
Make sure that changing the ECN sysctl does not affect
exisiting associations and endpoints.

MFC after: 3 months.

13 years ago1) Move per John Baldwin to mp_maxid
Randall Stewart [Thu, 3 Feb 2011 19:22:21 +0000 (19:22 +0000)]
1) Move per John Baldwin to mp_maxid
2) Some signed/unsigned errors found by Mac OS compiler (from Michael)
3) a couple of copyright updates on the effected files.

MFC after: 3 months

13 years agoFix for detection of MTK 3329 GPS USB devices.
Hans Petter Selasky [Thu, 3 Feb 2011 18:25:55 +0000 (18:25 +0000)]
Fix for detection of MTK 3329 GPS USB devices.

Submitted by: Mykhaylo Yehorov
PR: usb/153929
Approved by: thompsa (mentor)

13 years agoThe FDT describes the host controller directly. There's no need to
Marcel Moolenaar [Thu, 3 Feb 2011 18:07:05 +0000 (18:07 +0000)]
The FDT describes the host controller directly. There's no need to
get properties from the parent. The parent is in fact the FDT bus
itself and will therefore not have the properties we're looking
for.

Sponsored by: Juniper Networks

13 years agoAccept r1 as having the metadata pointer argument if r0 is 0.
Marcel Moolenaar [Thu, 3 Feb 2011 17:35:16 +0000 (17:35 +0000)]
Accept r1 as having the metadata pointer argument if r0 is 0.
This provides backward compatibility with Juniper loaders.

Sponsored by: Juniper Networks

13 years agoBring the list of OIDs up-to-date to prevent conflicts.
Hartmut Brandt [Thu, 3 Feb 2011 15:19:18 +0000 (15:19 +0000)]
Bring the list of OIDs up-to-date to prevent conflicts.

13 years agoEliminate unnecessary page hold_count checks. These checks predate
Alan Cox [Thu, 3 Feb 2011 14:42:46 +0000 (14:42 +0000)]
Eliminate unnecessary page hold_count checks.  These checks predate
r90944, which introduced a general mechanism for handling the freeing
of held pages.

Reviewed by: kib@

13 years agoUse a dedicated taskqueue with a thread that runs at a software-interrupt
John Baldwin [Thu, 3 Feb 2011 13:09:22 +0000 (13:09 +0000)]
Use a dedicated taskqueue with a thread that runs at a software-interrupt
priority for the periodic polling of the machine check registers.

13 years agoFix the per CPU stats so that:
Randall Stewart [Thu, 3 Feb 2011 11:52:22 +0000 (11:52 +0000)]
Fix the per CPU stats so that:
1) They don't use the giant "MAX_CPU" define and instead
   are allocated dynamically based on mp_ncpus
2) Will zero with the netstat -z -s -p sctp
3) Will be properly handled by both the sctp_init and finish
   (the multi-net stuff was incorrectly bzero'ing in sctp_init
    the wrong size.. the bzero is now moved to the right places).
    And of course the free is put in at the very end.

MFC after: 3 Months

13 years agoSetup another socketpair between parent and child, so that primary sandboxed
Pawel Jakub Dawidek [Thu, 3 Feb 2011 11:39:49 +0000 (11:39 +0000)]
Setup another socketpair between parent and child, so that primary sandboxed
worker can ask the main privileged process to connect in worker's behalf
and then we can migrate descriptor using this socketpair to worker.
This is not really needed now, but will be needed once we start to use
capsicum for sandboxing.

MFC after: 1 week

13 years agoAdd missing locking after moving keepalive_send() to remote send thread
Pawel Jakub Dawidek [Thu, 3 Feb 2011 11:33:32 +0000 (11:33 +0000)]
Add missing locking after moving keepalive_send() to remote send thread
in r214692.

MFC after: 1 week

13 years agoDrop privileges after connecting to hastd, but before sending or receiving
Pawel Jakub Dawidek [Thu, 3 Feb 2011 10:44:40 +0000 (10:44 +0000)]
Drop privileges after connecting to hastd, but before sending or receiving
anything.

MFC after: 1 week

13 years agoLet the caller log info about successful privilege drop.
Pawel Jakub Dawidek [Thu, 3 Feb 2011 10:37:44 +0000 (10:37 +0000)]
Let the caller log info about successful privilege drop.
We don't want to log this in hastctl.

MFC after: 1 week

13 years agoAdds an experimental option to create a pool of
Randall Stewart [Thu, 3 Feb 2011 10:05:30 +0000 (10:05 +0000)]
Adds an experimental option to create a pool of
threads. These serve as input threads and are queued
packets based on the V-tag number. This is similar to
what a modern card can do with queue's for TCP... but
alas modern cards know nothing about SCTP.

MFC after: 3 months (maybe)

13 years agoInclude driver name in panic string, to make it easier to find these should
Ed Maste [Thu, 3 Feb 2011 03:07:11 +0000 (03:07 +0000)]
Include driver name in panic string, to make it easier to find these should
the panic ever occur.

13 years agoRevert part of r173264. Both aac_ioctl_sendfib and aac_ioctl_send_raw_srb
Ed Maste [Thu, 3 Feb 2011 02:14:53 +0000 (02:14 +0000)]
Revert part of r173264.  Both aac_ioctl_sendfib and aac_ioctl_send_raw_srb
make use of the aac_ioctl_event callback, if aac_alloc_command fails.  This
can end up in an infinite loop in the while loop in aac_release_command.

Further investigation into the issue mentioned by Scott Long [1] will be
necessary.

[1] http://lists.freebsd.org/pipermail/freebsd-current/2007-October/078740.html

13 years agoSetting TARGET and TARGET_ARCH needs to be done in _MAKE, not in the
Warner Losh [Wed, 2 Feb 2011 23:59:24 +0000 (23:59 +0000)]
Setting TARGET and TARGET_ARCH needs to be done in _MAKE, not in the
TGTS rule as _MAKE is used elsewhere.  This should fix make world.

13 years agosh: Add test for shell script without '#!'.
Jilles Tjoelker [Wed, 2 Feb 2011 22:03:18 +0000 (22:03 +0000)]
sh: Add test for shell script without '#!'.

13 years agosh: Remove comment mentioning herefd, which is gone.
Jilles Tjoelker [Wed, 2 Feb 2011 21:48:53 +0000 (21:48 +0000)]
sh: Remove comment mentioning herefd, which is gone.

13 years agoAdd some obsolete manpages.
Ulrich Spörlein [Wed, 2 Feb 2011 21:09:30 +0000 (21:09 +0000)]
Add some obsolete manpages.