Paul Menage [Wed, 7 May 2008 03:42:41 +0000 (20:42 -0700)]
Fix cpuset sched_relax_domain_level control file
Due to a merge conflict, the sched_relax_domain_level control file was marked
as being handled by cpuset_read/write_u64, but the code to handle it was
actually in cpuset_common_file_read/write.
Since the value being written/read is in fact a signed integer, it should be
treated as such; this patch adds cpuset_read/write_s64 functions, and uses
them to handle the sched_relax_domain_level file.
With this patch, the sched_relax_domain_level can be read and written, and the
correct contents seen/updated.
Signed-off-by: Paul Menage <menage@google.com> Cc: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com> Cc: Paul Jackson <pj@sgi.com> Cc: Ingo Molnar <mingo@elte.hu> Reviewed-by: Li Zefan <lizf@cn.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
any_slab_objects() does an atomic_read on an atomic_long_t, this
fixes it to use atomic_long_read instead.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Christoph Lameter <clameter@sgi.com> Cc: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Andrew Morton [Wed, 7 May 2008 03:42:35 +0000 (20:42 -0700)]
sx.c: fix printk warnings on sparc32
drivers/char/sx.c: In function 'sx_set_real_termios':
drivers/char/sx.c:973: warning: format '%u' expects type 'unsigned int', but argument 2 has type 'long unsigned int'
drivers/char/sx.c:999: warning: format '%x' expects type 'unsigned int', but argument 2 has type 'tcflag_t'
drivers/char/sx.c:1012: warning: format '%x' expects type 'unsigned int', but argument 2 has type 'tcflag_t'
sparc32 seems to use weird types for its tty things.
[ Fine by me but this is ancient debug and most of the debug in sx just
wants deleting eventually. - Alan ]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Alan Cox <alan@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
WANG Cong [Wed, 7 May 2008 03:42:33 +0000 (20:42 -0700)]
uml: fix inconsistence due to tty_operation change
'put_char' of 'struct tty_operations' has changed from 'void' into 'int'.
This can also shut up compiler warnings.
Cc: Jeff Dike <jdike@addtoit.com> Signed-off-by: WANG Cong <wangcong@zeuux.org> Acked-by: Alan Cox <alan@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Harvey Harrison [Wed, 7 May 2008 03:42:32 +0000 (20:42 -0700)]
misc: fix integer as NULL pointer warnings
drivers/md/raid10.c:889:17: warning: Using plain integer as NULL pointer
drivers/media/video/cx18/cx18-driver.c:616:12: warning: Using plain integer as NULL pointer
sound/oss/kahlua.c:70:12: warning: Using plain integer as NULL pointer
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Cc: Neil Brown <neilb@suse.de> Cc: Mauro Carvalho Chehab <mchehab@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Steven Rostedt [Wed, 7 May 2008 03:42:31 +0000 (20:42 -0700)]
fix irq flags for iuu_phoenix.c
The file drivers/usb/serial/iuu_phoenix.c uses "int" for flags. This can
cause hard to find bugs on some architectures. This patch converts the flags
to use "long" instead.
This bug was discovered by doing an allyesconfig make on the -rt kernel where
checks are done to ensure all flags are of size sizeof(long).
Signed-off-by: Steven Rostedt <srostedt@redhat.com> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Steven Rostedt [Wed, 7 May 2008 03:42:30 +0000 (20:42 -0700)]
fix irq flags in rtc-ds1511
The file in drivers/rtc/rtc-ds1551.c uses "int" for flags. This can cause
hard to find bugs on some architectures. This patch converts the flags to use
"long" instead.
This bug was discovered by doing an allyesconfig make on the -rt kernel where
checks are done to ensure all flags are of size sizeof(long).
Signed-off-by: Steven Rostedt <srostedt@redhat.com> Cc: Alessandro Zummo <a.zummo@towertech.it> Cc: David Brownell <david-b@pacbell.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Steven Rostedt [Wed, 7 May 2008 03:42:29 +0000 (20:42 -0700)]
fix irq flags in saa7134
Some files in the drivers/media/video/saa7134 directory uses "int" for flags.
This can cause hard to find bugs on some architectures. This patch converts
the flags to use "long" instead.
This bug was discovered by doing an allyesconfig make on the -rt kernel where
checks are done to ensure all flags are of size sizeof(long).
Signed-off-by: Steven Rostedt <srostedt@redhat.com> Acked-by: Mauro Carvalho Chehab <mchehab@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Steven Rostedt [Wed, 7 May 2008 03:42:28 +0000 (20:42 -0700)]
fix irq flags in mac80211 code
A file in the net/mac80211 directory uses "int" for flags. This can cause
hard to find bugs on some architectures. This patch converts the flags to use
"long" instead.
This bug was discovered by doing an allyesconfig make on the -rt kernel where
checks are done to ensure all flags are of size sizeof(long).
Signed-off-by: Steven Rostedt <srostedt@redhat.com> Cc: "John W. Linville" <linville@tuxdriver.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
RCSR is only present on PXA2xx CPUs, not on PXA3xx CPUs. Therefore,
we should not be unconditionally writing to RCSR from generic code.
Since we now clear the RCSR status from the SoC specific PXA PM code
and before reset in the arch_reset() function, the duplication in
the corgi, poodle, spitz and tosa code can be removed.
Acked-by: Richard Purdie <rpurdie@rpsys.net> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Ingo Molnar [Thu, 8 May 2008 09:53:48 +0000 (11:53 +0200)]
semaphore: fix
Yanmin Zhang reported:
| Comparing with kernel 2.6.25, AIM7 (use tmpfs) has more th
| regression under 2.6.26-rc1 on my 8-core stoakley, 16-core tigerton,
| and Itanium Montecito. Bisect located the patch below:
|
| 64ac24e738823161693bf791f87adc802cf529ff is first bad commit
| commit 64ac24e738823161693bf791f87adc802cf529ff
| Author: Matthew Wilcox <matthew@wil.cx>
| Date: Fri Mar 7 21:55:58 2008 -0500
|
| Generic semaphore implementation
|
| After I manually reverted the patch against 2.6.26-rc1 while fixing
| lots of conflicts/errors, aim7 regression became less than 2%.
i reproduced the AIM7 workload and can confirm Yanmin's findings that
-.26-rc1 regresses over .25 - by over 67% here.
Looking at the workload i found and fixed what i believe to be the real
bug causing the AIM7 regression: it was inefficient wakeup / scheduling
/ locking behavior of the new generic semaphore code, causing suboptimal
performance.
The problem comes from the following code. The new semaphore code does
this on down():
the fastpath looks good and obvious, but note the following property of
the contended path: if there's a task on the ->wait_list, the up() of
the current owner will "pass over" ownership to that waiting task, in a
wake-one manner, via the waiter->up flag and by removing the waiter from
the wait list.
That is all and fine in principle, but as implemented in
kernel/semaphore.c it also creates a nasty, hidden source of contention!
The contention comes from the following property of the new semaphore
code: the new owner owns the semaphore exclusively, even if it is not
running yet.
So if the old owner, even if just a few instructions later, does a
down() [lock_kernel()] again, it will be blocked and will have to wait
on the new owner to eventually be scheduled (possibly on another CPU)!
Or if another task gets to lock_kernel() sooner than the "new owner"
scheduled, it will be blocked unnecessarily and for a very long time
when there are 2000 tasks running.
I.e. the implementation of the new semaphores code does wake-one and
lock ownership in a very restrictive way - it does not allow
opportunistic re-locking of the lock at all and keeps the scheduler from
picking task order intelligently.
This kind of scheduling, with 2000 AIM7 processes running, creates awful
cross-scheduling between those 2000 tasks, causes reduced parallelism, a
throttled runqueue length and a lot of idle time. With increasing number
of CPUs it causes an exponentially worse behavior in AIM7, as the chance
for a newly woken new-owner task to actually run anytime soon is less
and less likely.
Note that it takes just a tiny bit of contention for the 'new-semaphore
catastrophy' to happen: the wakeup latencies get added to whatever small
contention there is, and quickly snowball out of control!
I believe Yanmin's findings and numbers support this analysis too.
The best fix for this problem is to use the same scheduling logic that
the kernel/mutex.c code uses: keep the wake-one behavior (that is OK and
wanted because we do not want to over-schedule), but also allow
opportunistic locking of the lock even if a wakee is already "in
flight".
The patch below implements this new logic. With this patch applied the
AIM7 regression is largely fixed on my quad testbox:
i.e. a 67% speedup. We are now back to within 1% of the v2.6.25
performance levels and have zero idle time during the test, as expected.
Btw., interactivity also improved dramatically with the fix - for
example console-switching became almost instantaneous during this
workload (which after all is running 2000 tasks at once!), without the
patch it was stuck for a minute at times.
There's another nice side-effect of this speedup patch, the new generic
semaphore code got even smaller:
text data bss dec hex filename
1241 0 0 1241 4d9 semaphore.o.before
1207 0 0 1207 4b7 semaphore.o.after
(because the waiter.up complication got removed.)
Longer-term we should look into using the mutex code for the generic
semaphore code as well - but i's not easy due to legacies and it's
outside of the scope of v2.6.26 and outside the scope of this patch as
well.
Thomas Gleixner [Thu, 8 May 2008 07:18:43 +0000 (09:18 +0200)]
x86: cleanup PAT cpu validation
Move the scattered checks for PAT support to a single function. Its
moved to addon_cpuid_features.c as this file is shared between 32 and
64 bit.
Remove the manipulation of the PAT feature bit and just disable PAT in
the PAT layer, based on the PAT bit provided by the CPU and the
current CPU version/model white list.
Change the boot CPU check so it works on Voyager somewhere in the
future as well :) Also panic, when a secondary has PAT disabled but
the primary one has alrady switched to PAT. We have no way to undo
that.
The white list is kept for now to ensure that we can rely on known to
work CPU types and concentrate on the software induced problems
instead of fighthing CPU erratas and subtle wreckage caused by not yet
verified CPUs. Once the PAT code has stabilized enough, we can remove
the white list and open the can of worms.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Andres Salomon [Wed, 7 May 2008 20:07:38 +0000 (13:07 -0700)]
x86: geode: define geode_has_vsa2() even if CONFIG_MGEODE_LX is not set
We want drivers to be able to use geode_has_vsa2 without having to worry
about what model geode is being compiled for. This patch ensures that
geode_has_vsa2 is always defined.
Signed-off-by: Andres Salomon <dilinger@debian.org> Cc: Jordan Crouse <jordan.crouse@amd.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Andres Salomon [Wed, 7 May 2008 20:07:37 +0000 (13:07 -0700)]
x86: GEODE: cache results from geode_has_vsa2() and uninline
This moves geode_has_vsa2 into a .c file, caches the result we get from
the VSA virtual registers, and causes the function to no longer be inline.
[akpm@linux-foundation.org: cleanup]
Signed-off-by: Andres Salomon <dilinger@debian.org> Cc: Jordan Crouse <jordan.crouse@amd.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Found these two bugs while browsing through the code. The first one is
a cut-n-paste bug, instead of disabling the clock when request_irq()
fails, it enabled it once more. The second one fixes a debug printout,
AT91_SSC_IER is write only, AT91_SSC_IMR is readable (the printed string
actually says imr).
Frank Mandarino was busy so he asked me to send these to this list.
/Patrik
Signed-off-by: Patrik Sevallius <patrik.sevallius@enea.com> Acked-by: Frank Mandarino <fmandarino@endrelia.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Alexey Dobriyan [Thu, 8 May 2008 09:58:39 +0000 (10:58 +0100)]
[ARM] lubbock: fix compilation
arch/arm/mach-pxa/lubbock.c:399: error: expected '}' before ';' token
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Oliver Hartkopp [Thu, 8 May 2008 09:49:55 +0000 (02:49 -0700)]
can: Fix can_send() handling on dev_queue_xmit() failures
The tx packet counting and the local loopback of CAN frames should
only happen in the case that the CAN frame has been enqueued to the
netdevice tx queue successfully.
Thanks to Andre Naujoks <nautsch@gmail.com> for reporting this issue.
Signed-off-by: Oliver Hartkopp <oliver@hartkopp.net> Signed-off-by: Urs Thuermann <urs@isnogud.escape.de> Signed-off-by: David S. Miller <davem@davemloft.net>
Pavel Emelyanov [Thu, 8 May 2008 08:24:25 +0000 (01:24 -0700)]
netns: Fix arbitrary net_device-s corruptions on net_ns stop.
When a net namespace is destroyed, some devices (those, not killed
on ns stop explicitly) are moved back to init_net.
The problem, is that this net_ns change has one point of failure -
the __dev_alloc_name() may be called if a name collision occurs (and
this is easy to trigger). This allocator performs a likely-to-fail
GFP_ATOMIC allocation to find a suitable number. Other possible
conditions that may cause error (for device being ns local or not
registered) are always false in this case.
So, when this call fails, the device is unregistered. But this is
*not* the right thing to do, since after this the device may be
released (and kfree-ed) improperly. E. g. bridges require more
actions (sysfs update, timer disarming, etc.), some other devices
want to remove their private areas from lists, etc.
I. e. arbitrary use-after-free cases may occur.
The proposed fix is the following: since the only reason for the
dev_change_net_namespace to fail is the name generation, we may
give it a unique fall-back name w/o %d-s in it - the dev<ifindex>
one, since ifindexes are still unique.
So make this change, raise the failure-case printk loglevel to
EMERG and replace the unregister_netdevice call with BUG().
[ Use snprintf() -DaveM ]
Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Thu, 8 May 2008 08:15:21 +0000 (01:15 -0700)]
netfilter: nf_conntrack_sip: restrict RTP expect flushing on error to last request
Some Inovaphone PBXs exhibit very stange behaviour: when dialing for
example "123", the device sends INVITE requests for "1", "12" and
"123" back to back. The first requests will elicit error responses
from the receiver, causing the SIP helper to flush the RTP
expectations even though we might still see a positive response.
Note the sequence number of the last INVITE request that contained a
media description and only flush the expectations when receiving a
negative response for that sequence number.
Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Thu, 8 May 2008 08:13:31 +0000 (01:13 -0700)]
macvlan: Fix memleak on device removal/crash on module removal
As noticed by Ben Greear, macvlan crashes the kernel when unloading the
module. The reason is that it tries to clean up the macvlan_port pointer
on the macvlan device itself instead of the underlying device. A non-NULL
pointer is taken as indication that the macvlan_handle_frame_hook is
valid, when receiving the next packet on the underlying device it tries
to call the NULL hook and crashes.
Clean up the macvlan_port on the correct device to fix this.
Signed-off-by; Patrick McHardy <kaber@trash.net> Tested-by: Ben Greear <greearb@candelatech.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: J.H.M. Dassen (Ray) <jdassen@debian.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Ilpo Järvinen [Thu, 8 May 2008 08:09:11 +0000 (01:09 -0700)]
tcp FRTO: SACK variant is errorneously used with NewReno
Note: there's actually another bug in FRTO's SACK variant, which
is the causing failure in NewReno case because of the error
that's fixed here. I'll fix the SACK case separately (it's
a separate bug really, though related, but in order to fix that
I need to audit tp->snd_nxt usage a bit).
There were two places where SACK variant of FRTO is getting
incorrectly used even if SACK wasn't negotiated by the TCP flow.
This leads to incorrect setting of frto_highmark with NewReno
if a previous recovery was interrupted by another RTO.
An eventual fallback to conventional recovery then incorrectly
considers one or couple of segments as forward transmissions
though they weren't, which then are not LOST marked during
fallback making them "non-retransmittable" until the next RTO.
In a bad case, those segments are really lost and are the only
one left in the window. Thus TCP needs another RTO to continue.
The next FRTO, however, could again repeat the same events
making the progress of the TCP flow extremely slow.
In order for these events to occur at all, FRTO must occur
again in FRTOs step 3 while the key segments must be lost as
well, which is not too likely in practice. It seems to most
frequently with some small devices such as network printers
that *seem* to accept TCP segments only in-order. In cases
were key segments weren't lost, things get automatically
resolved because those wrongly marked segments don't need to be
retransmitted in order to continue.
I found a reproducer after digging up relevant reports (few
reports in total, none at netdev or lkml I know of), some
cases seemed to indicate middlebox issues which seems now
to be a false assumption some people had made. Bugzilla
#10063 _might_ be related. Damon L. Chesser <damon@damtek.com>
had a reproducable case and was kind enough to tcpdump it
for me. With the tcpdump log it was quite trivial to figure
out.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> Signed-off-by: David S. Miller <davem@davemloft.net>
[POWERPC] spufs: don't requeue victim contex in find_victim if it's not in spu_run
We should not requeue the victim context in find_victim if the owner is
not in spu_run. It's first not needed because leaving the context on
the spu is an optimization and second is harmful because it means the
owner could re-enter spu_run when the context is on the runqueue and
trip the BUG_ON in __spu_update_sched_info.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
Magnus Damm [Thu, 24 Apr 2008 12:47:15 +0000 (21:47 +0900)]
sh: no high level trigger on some sh3 cpus
The processor models sh7706, sh7707 and sh7709 don't support high
level trigger sense configuration. And the intc code looks like
crap these days so what's the difference.
Signed-off-by: Magnus Damm <damm@igel.co.jp> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Magnus Damm [Thu, 24 Apr 2008 12:41:12 +0000 (21:41 +0900)]
sh: clean up sh7710 and sh7720 intc tables
Clean up the intc tables by removing unneeded #ifdefs. The vector
list is what selects which interrupt sources that should be added,
having unsupported bitfields listed is ok as long as the vector
is excluded from the list.
Signed-off-by: Magnus Damm <damm@igel.co.jp> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Magnus Damm [Thu, 24 Apr 2008 12:36:34 +0000 (21:36 +0900)]
sh: add interrupt ack code to sh3
This patch adds interrupt acknowledge code for external interrupt
sources on sh3 processors. Only really required for edge triggered
interrupts, but we ack regardless of sense configuration.
Signed-off-by: Magnus Damm <damm@igel.co.jp> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Magnus Damm [Thu, 24 Apr 2008 12:30:09 +0000 (21:30 +0900)]
sh: unify external irq pin code for sh3
This patch unifies the sh3 external irq pin code. It buys us some
savings with reduced code redundancy, but the main feature with
this change is irq sense selection support for all sh3 processors.
Signed-off-by: Magnus Damm <damm@igel.co.jp> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
David S. Miller [Thu, 8 May 2008 01:54:05 +0000 (18:54 -0700)]
sparc: Fix SA_ONSTACK signal handling.
We need to be more liberal about the alignment of the buffer given to
us by sigaltstack(). The user should not need to be mindful of all of
the alignment constraints we have for the stack frame.
This mirrors how we handle this situation in clone() as well.
Also, we align the stack even in non-SA_ONSTACK cases so that signals
due to bad stack alignment can be delivered properly. This makes such
errors easier to debug and recover from.
Finally, add the sanity check x86 has to make sure we won't overflow
the signal stack.
This fixes glibc testcases nptl/tst-cancel20.c and
nptl/tst-cancelx20.c
Signed-off-by: David S. Miller <davem@davemloft.net>
Robert Jarzmik [Wed, 7 May 2008 19:39:06 +0000 (20:39 +0100)]
[ARM] 5032/1: Added cpufreq support for pxa27x CPU
PXA cpus maximum frequency depends on the cpu (624 for
pxa270, 520 for pxa272, 416 for pxa271). It should be
provided on kernel or module start (cpu-pxa
pxa27x_maxfreq parameter).
Make use of cpufreq_frequency_table_cpuinfo (patch by Bill
Reese provided by Philipp Zabel).
Some additionnal fixes from Philipp Zabel include :
* rename PXA cpufreq driver to reflect added PXA27x support
* remove unused variable ramstart from PXA cpufreq driver
Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com> Signed-off-by: Robert Jarzmik <rjarzmik@free.fr> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Commit 84f43c308b73a6a12128288721a1007ba4f1a8da "pxafb: introduce register
independent LCD connection type for pxafb" implements compatibility mode
for old style pxafb_mach_info initialization data wrongly, causing the
system to Oops repeatedly - first during probe, then when drawing. Fix it
and make pxafb_decode_mach_info void.
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de> Acked-by: Eric Miao <eric.miao@marvell.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
John Gregor [Wed, 7 May 2008 18:01:10 +0000 (11:01 -0700)]
IB/ipath: Fix SDMA error recovery in absence of link status change
What's fixed:
in ipath_cancel_sends()
We need to unconditionally set ABORTING. So, swap the tests
so the set_bit() isn't shadowed by the &&.
If we've disarmed the piobufs, then we need to unconditionally
set DISARMED. So, move it out from the overly protective if
at the bottom.
in sdma_abort_task()
Abort_task was written knowing that the SDMA engine would always
be reset (and restarted) on error. A recent change broke that
fundamental assumption by taking the restart portion and making
it conditional on a link status change. But, SDMA can go boom
without a link status change in some conditions.
Signed-off-by: John Gregor <john.gregor@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
Dave Olson [Wed, 7 May 2008 18:00:15 +0000 (11:00 -0700)]
IB/ipath: Need to always request and handle PIO avail interrupts
Now that we always use PIO for vl15 on 7220, we could get stuck forever
if we happened to run out of PIO buffers from the verbs code, because
the setup code wouldn't run; the interrupt was also ignored if SDMA was
supported. We also have to reduce the pio update threshold if we have
fewer kernel buffers than the existing threshold.
Clean up the initialization a bit to get ordering safer and more
sensible, and use the existing ipath_chg_kernavail call to do init,
rather than doing it separately.
Drop unnecessary clearing of pio buffer on pio parity error.
Drop incorrect updating of pioavailshadow when exitting freeze mode
(software state may not match chip state if buffer has been allocated
and not yet written).
If we couldn't get a kernel buffer for a while, make sure we are
in sync with hardware, mainly to handle the exitting freeze case.
Signed-off-by: Dave Olson <dave.olson@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
Michael Albaugh [Wed, 7 May 2008 17:59:23 +0000 (10:59 -0700)]
IB/ipath: Fix count of packets received by kernel
The loop in ipath_kreceive() that processes packets increments the
loop-index 'i' once too often, because the exit condition does not
depend on it, and is checked after the increment. By adding a check for
!last to the iterator in the for loop, we correct that in a way that is
not so likely to be re-broken by changes in the loop body.
Signed-off-by: Michael Albaugh <micheal.albaugh@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
Dave Olson [Wed, 7 May 2008 17:57:48 +0000 (10:57 -0700)]
IB/ipath: Fix bug that can leave sends disabled after freeze recovery
The semantics of cancel_sends changed, but the code using it was missed.
Don't leave sends and pioavail updates disabled, and add a comment as to
why the force update is needed.
Signed-off-by: Dave Olson <dave.olson@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
Michael Albaugh [Wed, 7 May 2008 17:56:47 +0000 (10:56 -0700)]
IB/ipath: Only warn about prototype chip during init
We warn about prototype chips, but the function that checks for
support is also called as a result of a get_portinfo request, which
can clutter the logs.
Restrict warning to only appear during initialization.
Signed-off-by: Michael Albaugh <michael.albaugh@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
Herbert Xu [Tue, 6 May 2008 12:46:49 +0000 (20:46 +0800)]
[CRYPTO] hmac: Avoid calling virt_to_page on key
When HMAC gets a key longer than the block size of the hash, it needs
to feed it as input to the hash to reduce it to a fixed length. As
it is HMAC converts the key to a scatter and gather list. However,
this doesn't work on certain platforms if the key is not allocated
via kmalloc. For example, the keys from tcrypt are stored in the
rodata section and this causes it to fail with HMAC on x86-64.
This patch fixes this by copying the key to memory obtained via
kmalloc before hashing it.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Jens Axboe [Wed, 7 May 2008 08:15:46 +0000 (10:15 +0200)]
block: avoid duplicate calls to get_part() in disk stat code
get_part() is fairly expensive, as it O(N) loops over partitions
to find the right one. In lots of normal IO paths we end up looking
up the partition twice, to make matters even worse. Change the
stat add code to accept a passed in partition instead.
Jens Axboe [Wed, 7 May 2008 07:51:23 +0000 (09:51 +0200)]
cfq-iosched: make io priorities inherit CPU scheduling class as well as nice
We currently set all processes to the best-effort scheduling class,
regardless of what CPU scheduling class they belong to. Improve that
so that we correctly track idle and rt scheduling classes as well.
Jan Kara [Tue, 6 May 2008 16:26:17 +0000 (18:26 +0200)]
udf: Fix memory corruption when fs mounted with noadinicb option
When UDF filesystem is mounted with noadinicb mount option, it
happens that we extend an empty directory with a block. A code in
udf_add_entry() didn't count with this possibility and used
uninitialized data leading to memory and filesystem corruption.
Add a check whether file already has some extents before operating
on them.
Jens Axboe [Wed, 7 May 2008 07:48:17 +0000 (09:48 +0200)]
block: optimize generic_unplug_device()
Original patch from Mikulas Patocka <mpatocka@redhat.com>
Mike Anderson was doing an OLTP benchmark on a computer with 48 physical
disks mapped to one logical device via device mapper.
He found that there was a slowdown on request_queue->lock in function
generic_unplug_device. The slowdown is caused by the fact that when some
code calls unplug on the device mapper, device mapper calls unplug on all
physical disks. These unplug calls take the lock, find that the queue is
already unplugged, release the lock and exit.
With the below patch, performance of the benchmark was increased by 18%
(the whole OLTP application, not just block layer microbenchmarks).
So I'm submitting this patch for upstream. I think the patch is correct,
because when more threads call simultaneously plug and unplug, it is
unspecified, if the queue is or isn't plugged (so the patch can't make
this worse). And the caller that plugged the queue should unplug it
anyway. (if it doesn't, there's 3ms timeout).
Miklos Szeredi [Wed, 7 May 2008 07:22:39 +0000 (09:22 +0200)]
vfs: splice remove_suid() cleanup
generic_file_splice_write() duplicates remove_suid() just because it
doesn't hold i_mutex. But it grabs i_mutex inside splice_from_pipe()
anyway, so this is rather pointless.
Move locking to generic_file_splice_write() and call remove_suid() and
__splice_from_pipe() instead.
Jens Axboe [Wed, 7 May 2008 07:17:12 +0000 (09:17 +0200)]
cfq-iosched: fix RCU race in the cfq io_context destructor handling
put_io_context() drops the RCU read lock before calling into cfq_dtor(),
however we need to hold off freeing there before grabbing and
dereferencing the first object on the list.
So extend the rcu_read_lock() scope to cover the calling of cfq_dtor(),
and optimize cfq_free_io_context() to use a new variant for
call_for_each_cic() that assumes the RCU read lock is already held.
Hit in the wild by Alexey Dobriyan <adobriyan@gmail.com>