Jon Nelson has found a test case which causes postgresql to fail with
the error:
psql:t.sql:4: ERROR: invalid page header in block 38269 of relation base/16384/16581
Under memory pressure, it looks like part of a file can end up getting
replaced by zero's. Until we can figure out the cause, we'll roll
back the change and use block_write_full_page() instead of
ext4_bio_write_page(). The new, more efficient writing function can
be used via the mount option mblk_io_submit, so we can test and fix
the new page I/O code.
To reproduce the problem, install postgres 8.4 or 9.0, and pin enough
memory such that the system just at the end of triggering writeback
before running the following sql script:
begin;
create temporary table foo as select x as a, ARRAY[x] as b FROM
generate_series(1, 10000000 ) AS x;
create index foo_a_idx on foo (a);
create index foo_b_idx on foo USING GIN (b);
rollback;
If the temporary table is created on a hard drive partition which is
encrypted using dm_crypt, then under memory pressure, approximately
30-40% of the time, pgsql will issue the above failure.
This patch should fix this problem, and the problem will come back if
the file system is mounted with the mblk_io_submit mount option.
Reported-by: Jon Nelson <jnelson@jamponi.net>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
* 'for-2.6.37' of git://linux-nfs.org/~bfields/linux:
nfsd: Fix possible BUG_ON firing in set_change_info
sunrpc: prevent use-after-free on clearing XPT_BUSY
* git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
Btrfs: prevent RAID level downgrades when space is low
Btrfs: account for missing devices in RAID allocation profiles
Btrfs: EIO when we fail to read tree roots
Btrfs: fix compiler warnings
Btrfs: Make async snapshot ioctl more generic
Btrfs: pwrite blocked when writing from the mmaped buffer of the same page
Btrfs: Fix a crash when mounting a subvolume
Btrfs: fix sync subvol/snapshot creation
Btrfs: Fix page leak in compressed writeback path
Btrfs: do not BUG if we fail to remove the orphan item for dead snapshots
Btrfs: fixup return code for btrfs_del_orphan_item
Btrfs: do not do fast caching if we are allocating blocks for tree_root
Btrfs: deal with space cache errors better
Btrfs: fix use after free in O_DIRECT
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
ceph: fix ioctl magic
ceph: Behave better when handling file lock replies.
ceph: pass lock information by struct file_lock instead of as individual params.
ceph: Handle file locks in replies from the MDS.
ceph: avoid possible null deref in readdir after dir llseek
* 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6:
NFS: Fix panic after nfs_umount()
nfs: remove extraneous and problematic calls to nfs_clear_request
nfs: kernel should return EPROTONOSUPPORT when not support NFSv4
NFS: Fix fcntl F_GETLK not reporting some conflicts
nfs: Discard ACL cache on mode update
NFS: Readdir cleanups
NFS: nfs_readdir_search_for_cookie() don't mark as eof if cookie not found
NFS: Fix a memory leak in nfs_readdir
Call the filesystem back whenever a page is removed from the page cache
NFS: Ensure we use the correct cookie in nfs_readdir_xdr_filler
Running the annotate branch profiler on three boxes, including my
main box that runs firefox, evolution, xchat, and is part of the distcc farm,
showed this with the likelys in the workqueue code:
correct incorrect % Function File Line
------- --------- - -------- ---- ----
96 996253 99 wq_worker_sleeping workqueue.c 703
96 996247 99 wq_worker_waking_up workqueue.c 677
The likely()s in this case were assuming that WORKER_NOT_RUNNING will
most likely be false. But this is not the case. The reason is
(and shown by adding trace_printks and testing it) that most of the time
WORKER_PREP is set.
In worker_thread() we have:
worker_clr_flags(worker, WORKER_PREP);
[ do work stuff ]
worker_set_flags(worker, WORKER_PREP, false);
(that 'false' means not to wake up an idle worker)
The wq_worker_sleeping() is called from schedule when a worker thread
is putting itself to sleep. Which happens most of the time outside
of that [ do work stuff ].
The wq_worker_waking_up is called by the wakeup worker code, which
is also callod outside that [ do work stuff ].
Thus, the likely and unlikely used by those two functions are actually
backwards.
Remove the annotation and let gcc figure it out.
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
Refresh rate nearness is not calculated or reset when nearest resolution
changes.
This patch resets the refresh rate differential measurement whenever a
new nearest resolution is discovered. This fixes two error cases;
first, wherein the first mode's refresh rate differential is never
calculated and second, when the closest refresh rate from a previous
nearest resolution is erroneously preserved.
Signed-off-by: Andrew Kephart <andrew.kephart@alereon.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Buffer transfer size is the number of transfers to be performed in
relation with the width of the _source_ interface.
So in the DMA_FROM_DEVICE case, it should be the register width that
should be taken into account.
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The commit 111c182340 (kgdboc: reset
input devices (keyboards) when exiting debugger) introduced a
regression in early debugging such that you get a kernel oops on
continue (with the go command) if you boot a kernel with:
earlyprintk=vga ekgdboc=kbd kgdbwait
The restore kgdboc_restore_input() routine schedules work for the
purpose of sending key release events for any keys that were in the
depressed state prior to entering the kernel debugger. A simple fix
to the crash is to not invoke the schedule_work() if the kernel
system_state is anything other than SYSTEM_RUNNING.
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Acked-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Reviewed-by: Sergei Shtylyov <sshtylyov@mvista.com>
The saving of the NVS memory area during suspend and restoring it
during resume causes problems to appear on Sony Vaio VGN-NW130D, so
blacklist that machine to avoid those problems.
Addresses https://bugzilla.kernel.org/show_bug.cgi?id=23002
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Reported-and-tested-by: Adriano <adriano.vilela@yahoo.com>
Signed-off-by: Len Brown <len.brown@intel.com>
The extent allocator has code that allows us to fill
allocations from any available block group, even if it doesn't
match the raid level we've requested.
This was put in because adding a new drive to a filesystem
made with the default mkfs options actually upgrades the metadata from
single spindle dup to full RAID1.
But, the code also allows us to allocate from a raid0 chunk when we
really want a raid1 or raid10 chunk. This can cause big trouble because
mkfs creates a small (4MB) raid0 chunk for data and metadata which then
goes unused for raid1/raid10 installs.
The allocator will happily wander in and allocate from that chunk when
things get tight, which is not correct.
The fix here is to make sure that we provide duplication when the
caller has asked for it. It does all the dups to be any raid level,
which preserves the dup->raid1 upgrade abilities.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
When we mount in RAID degraded mode without adding a new device to
replace the failed one, we can end up using the wrong RAID flags for
allocations.
This results in strange combinations of block groups (raid1 in a raid10
filesystem) and corruptions when we try to allocate blocks from single
spindle chunks on drives that are actually missing.
The first device has two small 4MB chunks in it that mkfs creates and
these are usually unused in a raid1 or raid10 setup. But, in -o degraded,
the allocator will fall back to these because the mask of desired raid groups
isn't correct.
The fix here is to count the missing devices as we build up the list
of devices in the system. This count is used when picking the
raid level to make sure we continue using the same levels that were
in place before we lost a drive.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Fault handling is getting enabled after enabling the interrupt-remapping (as
the success of interrupt-remapping can affect the apic mode and hence the
fault handling mode).
Hence there can potentially be some faults between the window of enabling
interrupt-remapping in the vt-d and the fault-handling of the vt-d units.
Handle any previous faults after enabling the vt-d fault handling.
For v2.6.38 cleanup, need to check if we can remove the dmar_fault() in the
enable_intr_remapping() and see if we can enable fault handling along with
enabling intr-remapping.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20101201062244.630417138@intel.com>
Cc: stable@kernel.org [v2.6.32+]
Acked-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Interrupt-remapping gets enabled very early in the boot, as it determines the
apic mode that the processor can use. And the current code enables the vt-d
fault handling before the setup_local_APIC(). And hence the APIC LDR registers
and data structure in the memory may not be initialized. So the vt-d fault
handling in logical xapic/x2apic modes were broken.
Fix this by enabling the vt-d fault handling in the end_local_APIC_setup()
A cleaner fix of enabling fault handling while enabling intr-remapping
will be addressed for v2.6.38. [ Enabling intr-remapping determines the
usage of x2apic mode and the apic mode determines the fault-handling
configuration. ]
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
LKML-Reference: <20101201062244.541996375@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: stable@kernel.org [v2.6.32+]
Acked-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
In x2apic mode, we need to set the upper address register of the fault
handling interrupt register of the vt-d hardware. Without this
irq migration of the vt-d fault handling interrupt is broken.
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
LKML-Reference: <1291225233.2648.39.camel@sbsiddha-MOBL3>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: stable@kernel.org [v2.6.32+]
Acked-by: Chris Wright <chrisw@sous-sol.org>
Tested-by: Takao Indoh <indou.takao@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
On platforms with Intel 7500 chipset, there were some reports of system
hang/NMI's during kexec/kdump in the presence of interrupt-remapping enabled.
During kdump, there is a window where the devices might be still using old
kernel's interrupt information, while the kdump kernel is coming up. This can
cause vt-d faults as the interrupt configuration from the old kernel map to
null IRTE entries in the new kernel etc. (with out interrupt-remapping enabled,
we still have the same issue but in this case we will see benign spurious
interrupt hit the new kernel).
Based on platform config settings, these platforms seem to generate NMI/SMI
when a vt-d fault happens and there were reports that the resulting SMI causes
the system to hang.
Fix it by masking vt-d spec defined errors to platform error reporting logic.
VT-d spec related errors are already handled by the VT-d OS code, so need to
report the same error through other channels.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1291667190.2675.8.camel@sbsiddha-MOBL3.sc.intel.com>
Cc: stable@kernel.org [v2.6.32+]
Reported-by: Max Asbock <masbock@linux.vnet.ibm.com>
Reported-and-tested-by: Takao Indoh <indou.takao@jp.fujitsu.com>
Acked-by: Chris Wright <chrisw@sous-sol.org>
Acked-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Alignment of alloc_bootmem() depends on the value of
L1_CACHE_SHIFT. What we need here, however, is 64 byte alignment. Use
alloc_bootmem_align() and explicitly specify the alignment instead.
This fixes a kernel boot crash reported by Jody when the cpu in .config
is set to MPENTIUMII but the kernel is booted on a xsave-capable CPU.
Reported-by: Jody Bruchon <jody@nctritech.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20101116212442.059967454@sbsiddha-MOBL3.sc.intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: <stable@kernel.org>
Add an alloc_bootmem_align() interface to allocate bootmem with
specified alignment. This is necessary to be able to allocate the
xsave area in a subsequent patch.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20101116212441.977574826@sbsiddha-MOBL3.sc.intel.com>
Acked-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: <stable@kernel.org>
The vdso Makefile passes linker-style -m options not to the linker but
to gcc. This happens to work with earlier gcc, but fails with gcc
4.6. Pass gcc-style -m options, instead.
Note: all currently supported versions of gcc supports -m32, so there
is no reason to conditionalize it any more.
Reported-by: H. J. Lu <hjl.tools@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
LKML-Reference: <tip-*@git.kernel.org>
Cc: <stable@kernel.org>
Fixed fsl dma slow issue by initializing dma mode register with
bandwidth control. It boosts dma performance and should works
with 85xx board.
Signed-off-by: Forrest Shi <b29237@freescale.com>
Signed-off-by: Li Yang <leoli@freescale.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
If we just get a plain IO error when we read tree roots, the code
wasn't properly sending that error up the chain. This allowed mounts to
continue when they should failed, and allowed operations
on partially setup root structs. The end result was usually oopsen
on spinlocks that hadn't been spun up correctly.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The ltc4215 driver used the chip's "power good" status bit to provide
the power1_alarm file. This is wrong: the chip is really reporting the
status of one of the monitored voltages.
Change the sysfs file from power1_alarm to in2_min_alarm instead. This
matches the voltage that the chip is raising an alarm for.
Signed-off-by: Ira W. Snyder <iws@ovro.caltech.edu>
Signed-off-by: Guenter Roeck <guenter.roeck@ericsson.com>
When adjusting the code to handle removing the old nmi watchdog,
I forgot to consider the compile case when the local apic is not
enabled.
This change fixes the following build error:
arch/x86/kernel/apic/hw_nmi.c:28:6: error: redefinition of ‘touch_nmi_watchdog’
Signed-off-by: Don Zickus <dzickus@redhat.com>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Rakib Mullick <rakib.mullick@gmail.com>
LKML-Reference: <20101213153719.GD18577@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Currently bias of non-DAPM codec will be powered down (standby/off) whenever
there is a stream stop. This is wrong in simultaneous playback/capture since
the bias is put down immediately after stopping the first stream.
Fix this by using the codec->active count when figuring out the needed bias
level after stream stop.
Signed-off-by: Jarkko Nikula <jhnikula@gmail.com>
Acked-by: Liam Girdwood <lrg@slimlogic.co.uk>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
commit 995bd3bb5 (x86: Hpet: Avoid the comparator readback penalty)
chose 8 HPET cycles as a safe value for the ETIME check, as we had the
confirmation that the posted write to the comparator register is
delayed by two HPET clock cycles on Intel chipsets which showed
readback problems.
After that patch hit mainline we got reports from machines with newer
AMD chipsets which seem to have an even longer delay. See
http://thread.gmane.org/gmane.linux.kernel/1054283 and
http://thread.gmane.org/gmane.linux.kernel/1069458 for further
information.
Boris tried to come up with an ACPI based selection of the minimum
HPET cycles, but this failed on a couple of test machines. And of
course we did not get any useful information from the hardware folks.
For now our only option is to chose a paranoid high and safe value for
the minimum HPET cycles used by the ETIME check. Adjust the minimum ns
value for the HPET clockevent accordingly.
Reported-Bistected-and-Tested-by: Markus Trippelsdorf <markus@trippelsdorf.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <alpine.LFD.2.00.1012131222420.2653@localhost6.localdomain6>
Cc: Simon Kirby <sim@hostway.ca>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Andreas Herrmann <Andreas.Herrmann3@amd.com>
Cc: John Stultz <johnstul@us.ibm.com>
There are still quite a number of MFD and GPIO expander drivers that are
using the old irq_chip APIs that haven't had a chance to update during
the .37 cycle, resulting in allyes/modconfig errors on some
configurations.
Mark Brown has done most of the legwork to get these fixed up in .38,
so this should just be a .37 stop-gap that we can drop at the end of the
.38 merge window.
Reported-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
The current implementation of the v7_coherent_*_range function assumes
that the D and I cache lines have the same size, which is incorrect
architecturally. This patch adds the icache_line_size macro which reads
the CTR register. The main loop in v7_coherent_*_range is split in two
independent loops or the D and I caches. This also has the performance
advantage that the DSB is moved outside the main loop.
Reported-by: Kevin Sapp <ksapp@quicinc.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The current implementation of the dcache_line_size macro reads the L1
cache size from the CCSIDR register. This, however, is not guaranteed to
be the smallest cache line in the cache hierarchy. The patch changes to
the macro to use the more architecturally correct CTR register.
Reported-by: Kevin Sapp <ksapp@quicinc.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
__pppoe_xmit function return value was invalid resulting in
additional call to kfree_skb on already freed skb. This resulted in
memory corruption and consequent kernel panic after PPPoE peer
terminated the link.
This fixes commit 55c95e738d.
Reported-by: Gorik Van Steenberge <gvs@zemos.net>
Reported-by: Daniel Kenzelmann <kernel.bugzilla@kenzelmann.dyndns.info>
Reported-by: Denys Fedoryshchenko <nuclearcat@nuclearcat.com>
Reported-by: Pawel Staszewski <pstaszewski@artcom.pl>
Diagnosed-by: Andrej Ota <andrej@ota.si>
Diagnosed-by: Eric Dumazet <eric.dumazet@gmail.com>
Tested-by: Denys Fedoryshchenko <nuclearcat@nuclearcat.com>
Tested-by: Pawel Staszewski <pstaszewski@artcom.pl>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: Andrej Ota <andrej@ota.si>
Signed-off-by: David S. Miller <davem@davemloft.net>
We must not wake the TX queue without free TX descriptors.
sca_xmit() expects at least one free descriptor and BUGs otherwise.
Problem reported and fix tested by Bernie Innocenti and Ward Vandewege.
Signed-off-by: Krzysztof Hałasa <khc@pm.waw.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Vladislav Zolotarov <vladz@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
bnx2x_src_init_t2() is used only when BCM_CNIC is defined.
So, to avoid a compilation warning, we won't define it unless
BCM_CNIC is defined.
Signed-off-by: Vladislav Zolotarov <vladz@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make the LSO code work on BE platforms: parsing_data field of
a parsing BD (PBD) for 57712 was improperly composed which made FW read wrong
values for TCP header's length and offset and, as a result, the corresponding
PCI device was performing bad DMA reads triggering EEH.
Signed-off-by: Vladislav Zolotarov <vladz@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Removing usage of rtnl_lock() to protect firmware interface registers.
These registers are accessed in some worker threads and can create a
deadlock if rtnl_lock is taken by upper layers while the worker is still
pending.
We remove rtnl_lock and use a driver mutex just while mailboxes are
accessed.
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Agere FW643 rev 06, listed as "11c1:5901 (rev 06) (prog-if 10 [OHCI])",
produced SBP-2 I/O errors since kernel 2.6.36. Disabling MSI fixes it.
Since MSI work on Agere FW643-E (same vendor and device ID, but rev 07),
introduce a device revision field into firewire-ohci's quirks list so
that different quirks can be defined for older and newer revisions.
Reported-by: Jonathan Isom <jeisom@gmail.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Cc: <stable@kernel.org> # 2.6.36.y
"VIA Technologies, Inc. VT6315 Series Firewire Controller [1106:3403]"
does not generate any interrupts if Message Signaled Interrupts were
enabled. This is a regression since kernel 2.6.36 in which MSI support
was added to firewire-ohci. Hence blacklist MSI on all VIA controllers.
Reported-by: Robin Cook <rcook@wyrms.net>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Cc: <stable@kernel.org> # 2.6.36.y
Add the features struct and device table entry to enable yet another
version of Wacom Bamboo Pen (CTL460, Product ID 0xD4).
Signed-off-by: Kevin Granade <kevin.granade@gmail.com>
Acked-by: Ping Cheng <pingc@wacom.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Properly const-, __init-, and __read_mostly-annotate this code.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Theodore Ts'o