Without this udev doesn't have a way to key the ne device to the platform
device.
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If we do not do device_lock in pci_walk_bus, we can avoid deadlock.
Device_lock in pci_walk_bus is introduced in commit: d71374dafbba7ec3f67371d3b7e9f6310a588808, corresponding email thread
is: https://lkml.org/lkml/2006/5/26/38. The patch author Zhang Yanmin
said device_lock is added to pci_walk_bus because:
Some error handling functions call pci_walk_bus. For example, PCIe
aer. Here we lock the device, so the driver wouldn't detach from the
device, as the cb might call driver's callback function.
So I fixed the deadlock as follows:
- remove device_lock from pci_walk_bus
- add device_lock into callback if callback will call driver's callback
I checked pci_walk_bus users one by one, and found only PCIe aer needs
device lock.
Commit 29c00b4a1d9e27 (rcu: Add event-tracing for RCU callback
invocation) added a regression in rcu_do_batch()
Under stress, RCU is supposed to allow to process all items in queue,
instead of a batch of 10 items (blimit), but an integer overflow makes
the effective limit being 1. So, unless there is frequent idle periods
(during which RCU ignores batch limits), RCU can be forced into a
state where it cannot keep up with the callback-generation rate,
eventually resulting in OOM.
This commit therefore converts a few variables in rcu_do_batch() from
int to long to fix this problem, along with the module parameters
controlling the batch limits.
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch (as1632b) fixes a bug in ehci-hcd. The USB core uses
urb->hcpriv to determine whether or not an URB is active; host
controller drivers are supposed to set this pointer to a non-NULL
value when an URB is queued. However ehci-hcd sets it to NULL for
isochronous URBs, which defeats the check in usbcore.
In itself this isn't a big deal. But people have recently found that
certain sequences of actions will cause the snd-usb-audio driver to
reuse URBs without waiting for them to complete. In the absence of
proper checking by usbcore, the URBs get added to their endpoint list
twice. This leads to list corruption and a system freeze.
The patch makes ehci-hcd assign a meaningful value to urb->hcpriv for
isochronous URBs. Improving robustness always helps.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Reported-by: Artem S. Tashkinov <t.artem@lycos.com> Reported-by: Christof Meerwald <cmeerw@cmeerw.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Recently I build perf and get a build error on builtin-test.c. The error is as
following:
$ make
CC perf.o
CC builtin-test.o
cc1: warnings being treated as errors
builtin-test.c: In function ‘sched__get_first_possible_cpu’:
builtin-test.c:977: warning: implicit declaration of function ‘CPU_ALLOC’
builtin-test.c:977: warning: nested extern declaration of ‘CPU_ALLOC’
builtin-test.c:977: warning: assignment makes pointer from integer without a cast
builtin-test.c:978: warning: implicit declaration of function ‘CPU_ALLOC_SIZE’
builtin-test.c:978: warning: nested extern declaration of ‘CPU_ALLOC_SIZE’
builtin-test.c:979: warning: implicit declaration of function ‘CPU_ZERO_S’
builtin-test.c:979: warning: nested extern declaration of ‘CPU_ZERO_S’
builtin-test.c:982: warning: implicit declaration of function ‘CPU_FREE’
builtin-test.c:982: warning: nested extern declaration of ‘CPU_FREE’
builtin-test.c:992: warning: implicit declaration of function ‘CPU_ISSET_S’
builtin-test.c:992: warning: nested extern declaration of ‘CPU_ISSET_S’
builtin-test.c:998: warning: implicit declaration of function ‘CPU_CLR_S’
builtin-test.c:998: warning: nested extern declaration of ‘CPU_CLR_S’
make: *** [builtin-test.o] Error 1
This problem is introduced in 3e7c439a. CPU_ALLOC and related macros are
missing in sched__get_first_possible_cpu function. In 54489c18, commiter
mentioned that CPU_ALLOC has been removed. So CPU_ALLOC calls in this
function are removed to let perf to be built.
Signed-off-by: Vinson Lee <vlee@twitter.com> Signed-off-by: Zheng Liu <wenqing.lz@taobao.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Vinson Lee <vlee@twitter.com> Cc: Zheng Liu <wenqing.lz@taobao.com> Link: http://lkml.kernel.org/r/1352422726-31114-1-git-send-email-vlee@twitter.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some devices (ex Nokia C7) simply don't respond at all when data is sent
to some of their USB interfaces. The data gets stuck in the TTYs queue
and sits there until close(2), which them blocks because closing_wait
defaults to 30 seconds (even though the fd is O_NONBLOCK). This is
rarely desired. Implement the standard mechanism to adjust closing_wait
and let applications handle it how they want to.
Signed-off-by: Dan Williams <dcbw@redhat.com> Acked-by: Oliver Neukum <oneukum@suse.de> Tested-by: Aleksander Morgado <aleksander@gnu.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The function rb_check_pages() was added to make sure the ring buffer's
pages were sane. This check is done when the ring buffer size is modified
as well as when the iterator is released (closing the "trace" file),
as that was considered a non fast path and a good place to do a sanity
check.
The problem is that the check does not have any locks around it.
If one process were to read the trace file, and another were to read
the raw binary file, the check could happen while the reader is reading
the file.
The issues with this is that the check requires to clear the HEAD page
before doing the full check and it restores it afterward. But readers
require the HEAD page to exist before it can read the buffer, otherwise
it gives a nasty warning and disables the buffer.
By adding the reader lock around the check, this keeps the race from
happening.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The function rb_set_head_page() searches the list of ring buffer
pages for a the page that has the HEAD page flag set. If it does
not find it, it will do a WARN_ON(), disable the ring buffer and
return NULL, as this should never happen.
But if this bug happens to happen, not all callers of this function
can handle a NULL pointer being returned from it. That needs to be
fixed.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Ali reports that plugging a device into the Fresco Logic xHCI host with
PCI device ID 1400 produces an IRQ error:
do_IRQ: 3.176 No irq handler for vector (irq -1)
Other early Fresco Logic host revisions don't support MSI, even though
their PCI config space claims they do. Extend the quirk to disabling
MSI to this chipset revision. Also enable the short transfer quirk,
since it's likely this revision also has that quirk, and it should be
harmless to enable.
This patch should be backported to stable kernels as old as 2.6.36, that
contain the commit f5182b4155b9d686c5540a6822486400e34ddd98 "xhci:
Disable MSI for some Fresco Logic hosts."
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Reported-by: A Sh <smr.ash1991@gmail.com> Tested-by: A Sh <smr.ash1991@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch (as1636) is a partial workaround for a hardware bug
affecting OHCI controllers by NVIDIA at least, maybe others too. When
the controller retires a Transfer Descriptor, it is supposed to add
the TD onto the Done Queue. But sometimes this doesn't happen, with
the result that ohci-hcd never realizes the corresponding transfer has
finished. Symptoms can vary; a typical result is that USB audio stops
working after a while.
The patch works around the problem by recognizing that TDs are always
processed in order. Therefore, if a later TD is found on the Done
Queue than all the earlier TDs for the same endpoint must be finished
as well.
Unfortunately this won't solve the problem in cases where the missing
TD is the last one in the endpoint's queue. A complete fix would
require a signficant amount of change to the driver.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Tested-by: Oliver Neukum <oneukum@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The ACPI video driver can't control backlight correctly on
Asus UL30VT. Vendor driver (asus-laptop) can work. This patch is to
add "Asus UL30VT" to ACPI video detect blacklist in order to use
asus-laptop for video control on the "Asus UL30VT" rather than ACPI
video driver.
References: https://bugzilla.kernel.org/show_bug.cgi?id=32592 Reported-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Lan Tianyu <tianyu.lan@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Or else the laptop will boot with a dimmed screen.
References: https://bugzilla.kernel.org/show_bug.cgi?id=51141 Tested-by: Stefan Nagy <public@stefan-nagy.at> Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
During resume from system suspend the 'data' field of
struct pnp_dev in pnpacpi_set_resources() may be a stale pointer,
due to removal of the associated ACPI device node object in the
previous suspend-resume cycle. This happens, for example, if a
dockable machine is booted in the docking station and then suspended
and resumed and suspended again. If that happens,
pnpacpi_build_resource_template() called from pnpacpi_set_resources()
attempts to use that pointer and crashes.
However, pnpacpi_set_resources() actually checks the device's ACPI
handle, attempts to find the ACPI device node object attached to it
and returns an error code if that fails, so in fact it knows what the
correct value of dev->data should be. Use this observation to update
dev->data with the correct value if necessary and dump a call trace
if that's the case (once).
We still need to fix the root cause of this issue, but preventing
systems from crashing because of it is an improvement too.
Reported-and-tested-by: Zdenek Kabelac <zdenek.kabelac@gmail.com>
References: https://bugzilla.kernel.org/show_bug.cgi?id=51071 Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add a quirk to correctly report battery capacity on 2010 and 2011
Lenovo Thinkpad models.
The affected models that I tested (x201, t410, t410s, and x220)
exhibit a problem where, when battery capacity reporting unit is mAh,
the values being reported are wrong. Pre-2010 and 2012 models appear
to always report in mWh and are thus unaffected. Also, in mid-2012
Lenovo issued a BIOS update for the 2011 models that fixes the issue
(tested on x220 with a post-1.29 BIOS). No such update is available
for the 2010 models, so those still need this patch.
Problem description: for some reason, the affected Thinkpads switch
the reporting unit between mAh and mWh; generally, mAh is used when a
laptop is plugged in and mWh when it's unplugged, although a
suspend/resume or rmmod/modprobe is needed for the switch to take
effect. The values reported in mAh are *always* wrong. This does
not appear to be a kernel regression; I believe that the values were
never reported correctly. I tested back to kernel 2.6.34, with
multiple machines and BIOS versions.
Simply plugging a laptop into mains before turning it on is enough to
reproduce the problem. Here's a sample /proc/acpi/battery/BAT0/info
from Thinkpad x220 (before a BIOS update) with a 4-cell battery:
present: yes
design capacity: 2886 mAh
last full capacity: 2909 mAh
battery technology: rechargeable
design voltage: 14800 mV
design capacity warning: 145 mAh
design capacity low: 13 mAh
cycle count: 0
capacity granularity 1: 1 mAh
capacity granularity 2: 1 mAh
model number: 42T4899
serial number: 21064
battery type: LION
OEM info: SANYO
Once the laptop switches the unit to mWh (unplug from mains, suspend,
resume), the output changes to:
Can you see how the values for "design capacity", etc., differ by a
factor of 10 instead of 14.8 (the design voltage of this battery)?
On the battery itself it says: 14.8V, 1.95Ah, 29Wh, so clearly the
values reported in mWh are correct and the ones in mAh are not.
My guess is that this problem has been around ever since those
machines were released, but because the most common Thinkpad
batteries are rated at 10.8V, the error (8%) is small enough that it
simply hasn't been noticed or at least nobody could be bothered to
look into it.
My patch works around the problem by adjusting the incorrectly
reported mAh values by "10000 / design_voltage". The patch also has
code to figure out if it should be activated or not. It only
activates on Lenovo Thinkpads, only when the unit is mAh, and, as an
extra precaution, only when the battery capacity reported through
ACPI does not match what is reported through DMI (I've never
encountered a machine where the first two conditions would be true
but the last would not, but better safe than sorry).
I've been using this patch for close to a year on several systems
without any problems.
References: https://bugzilla.kernel.org/show_bug.cgi?id=41062 Acked-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
As reported https://bugzilla.kernel.org/show_bug.cgi?id=51031, the UAS
driver causes problems and has been asked to be not built into any of
the major distributions. To prevent users from running into problems
with it, and for distros that were not notified, just mark the whole
thing as broken.
Acked-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
BeagleBone A5+ devices ended up getting shipped with the
'BeagleBone/XDS100V2' product string, and not XDS100 like it
was agreed, so adjust the quirk to match.
The Newport AGILIS model AG-UC8 compact piezo motor controller
(http://search.newport.com/?q=*&x2=sku&q2=AG-UC8)
is yet another device using an FTDI USB-to-serial chip. It works
fine with the ftdi_sio driver when adding
options ftdi-sio product=0x3000 vendor=0x104d
to modprobe.d. udevadm reports "Newport" as the manufacturer,
and "Agilis" as the product name.
Signed-off-by: Martin Teichmann <lkb.teichmann@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Huawei E173 will normally appear as 12d1:1436 in Linux. But
the modem has another mode with different device ID and a slightly
different set of descriptors. This is the mode used by Windows like
this:
3Modem: USB\VID_12D1&PID_140C&MI_00\6&3A1D2012&0&0000
Networkcard: USB\VID_12D1&PID_140C&MI_01\6&3A1D2012&0&0001
Appli.Inter: USB\VID_12D1&PID_140C&MI_02\6&3A1D2012&0&0002
PC UI Inter: USB\VID_12D1&PID_140C&MI_03\6&3A1D2012&0&0003
All interfaces have the same ff/ff/ff class codes in this mode.
Blacklisting the network interface to allow it to be picked up by
the network driver.
HPET_TN_FSB is not a proper mask bit; it merely toggles between MSI and
legacy interrupt delivery. The proper mask bit is HPET_TN_ENABLE, so
use both bits when (un)masking the interrupt.
We need to first destroy the floppy_wq workqueue before cleaning up
the queue. Otherwise we might race with still pending work with the
workqueue, but all the block queue already gone. This might lead to
various oopses, such as
dmapool always calls dma_alloc_coherent() with GFP_ATOMIC flag,
regardless the flags provided by the caller. This causes excessive
pruning of emergency memory pools without any good reason. Additionaly,
on ARM architecture any driver which is using dmapools will sooner or
later trigger the following error:
"ERROR: 256 KiB atomic DMA coherent pool is too small!
Please increase it with coherent_pool= kernel parameter!".
Increasing the coherent pool size usually doesn't help much and only
delays such error, because all GFP_ATOMIC DMA allocations are always
served from the special, very limited memory pool.
This patch changes the dmapool code to correctly use gfp flags provided
by the dmapool caller.
Reported-by: Soeren Moch <smoch@web.de> Reported-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Tested-by: Andrew Lunn <andrew@lunn.ch> Tested-by: Soeren Moch <smoch@web.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drm/i915: parse eDP panel color depth from VBT block
originally introduced parsing bpp for eDP from VBT, with a default of 18
bpp if the eDP BIOS data block is not present. Turns out that default seems
to break the Macbook Pro with retina display, as noted in
Since we can't ignore bpc settings from VBT completely after all, get rid
of the default. Do not clamp eDP to 18 bpp by default if the eDP BDB is
missing from VBT.
Signed-off-by: Jani Nikula <jani.nikula@intel.com> Tested-by: Henrik Rydberg <rydberg@euromail.se>
[danvet: paste in the updated commit message from irc.] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
but putting the VBT check after the EDID check to see them both in dmesg if
this clamps more than the EDID. We have enough history with bpc clamping to
warrant the extra debug info.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=47641
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=56401 Signed-off-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
8852aac25e ("workqueue: mod_delayed_work_on() shouldn't queue timer on
0 delay") unexpectedly uncovered a very nasty abuse of delayed_work in
megaraid - it allocated work_struct, casted it to delayed_work and
then pass that into queue_delayed_work().
Previously, this was okay because 0 @delay short-circuited to
queue_work() before doing anything with delayed_work. 8852aac25e
moved 0 @delay test into __queue_delayed_work() after sanity check on
delayed_work making megaraid trigger BUG_ON().
Although megaraid is already fixed by c1d390d8e6 ("megaraid: fix
BUG_ON() from incorrect use of delayed work"), this patch converts
BUG_ON()s in __queue_delayed_work() to WARN_ON_ONCE()s so that such
abusers, if there are more, trigger warning but don't crash the
machine.
The driver is currently filling data in a wrong way, on drivers
for csrows-based memory controller, when the first layer is a
csrow.
This is not easily to notice, as, in general, memories are
filed in dual, interleaved, symetric mode, as very few memory
controllers support asymetric modes.
While digging into a bug for i82795_edac driver, the asymetric
mode there is now working, allowing us to fill the machine with
4x1GB ranks at channel 0, and 2x512GB at channel 1:
Channel 0 ranks:
EDAC DEBUG: i82975x_init_csrows: DIMM A0: from page 0x00000000 to 0x0003ffff (size: 0x00040000 pages)
EDAC DEBUG: i82975x_init_csrows: DIMM A1: from page 0x00040000 to 0x0007ffff (size: 0x00040000 pages)
EDAC DEBUG: i82975x_init_csrows: DIMM A2: from page 0x00080000 to 0x000bffff (size: 0x00040000 pages)
EDAC DEBUG: i82975x_init_csrows: DIMM A3: from page 0x000c0000 to 0x000fffff (size: 0x00040000 pages)
Channel 1 ranks:
EDAC DEBUG: i82975x_init_csrows: DIMM B0: from page 0x00100000 to 0x0011ffff (size: 0x00020000 pages)
EDAC DEBUG: i82975x_init_csrows: DIMM B1: from page 0x00120000 to 0x0013ffff (size: 0x00020000 pages)
Instead of properly showing the memories as such, before this patch, it
shows the memory layout as:
The matrix-keymap module is currently lacking a proper module license,
add one so we don't have this module tainting the entire kernel. This
issue has been present since commit 1932811f426f ("Input: matrix-keymap
- uninline and prepare for device tree support")
soc-dmaengine-pcm library need to be part of the snd-soc-core in order to
be able to compile ASoC as modules when dmaengine is enabled on the platform.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Cc: Florian Fainelli <florian@openwrt.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
After commit 846a136881b8f73c1f74250bf6acfaa309cab1f2 ("ARM: vfp: fix
saving d16-d31 vfp registers on v6+ kernels"), the OMAP 2430SDP board
started crashing during boot with omap2plus_defconfig:
[ 3.875122] mmcblk0: mmc0:e624 SD04G 3.69 GiB
[ 3.915954] mmcblk0: p1
[ 4.086639] Internal error: Oops - undefined instruction: 0 [#1] SMP ARM
[ 4.093719] Modules linked in:
[ 4.096954] CPU: 0 Not tainted (3.6.0-02232-g759e00b #570)
[ 4.103149] PC is at vfp_reload_hw+0x1c/0x44
[ 4.107666] LR is at __und_usr_fault_32+0x0/0x8
It turns out that the context save/restore fix unmasked a latent bug
in commit 5aaf254409f8d58229107b59507a8235b715a960 ("ARM: 6203/1: Make
VFPv3 usable on ARMv6"). When CONFIG_VFPv3 is set, but the kernel is
booted on a pre-VFPv3 core, the code attempts to save and restore the
d16-d31 VFP registers. These are only present on non-D16 VFPv3+, so
this results in an undefined instruction exception. The code didn't
crash before commit 846a136 because the save and restore code was
only touching d0-d15, present on all VFP.
Fix by implementing a request from Russell King to add a new HWCAP
flag that affirmatively indicates the presence of the d16-d31
registers:
and some feedback from Måns to clarify the name of the HWCAP flag.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Cc: Tony Lindgren <tony@atomide.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Dave Martin <dave.martin@linaro.org> Cc: Måns Rullgård <mans.rullgard@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Ben Hutchings <ben@decadent.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On some systems, e.g., kzm9g, MMCIF interfaces can produce spurious
interrupts without any active request. To prevent the Oops, that results
in such cases, don't dereference the mmc request pointer until we make
sure, that we are indeed processing such a request.
This fixes a regression in 3.7-rc, which has since gone into stable.
Commit 00442ad04a5e ("mempolicy: fix a memory corruption by refcount
imbalance in alloc_pages_vma()") changed get_vma_policy() to raise the
refcount on a shmem shared mempolicy; whereas shmem_alloc_page() went
on expecting alloc_page_vma() to drop the refcount it had acquired.
This deserves a rework: but for now fix the leak in shmem_alloc_page().
Hugh: shmem_swapin() did not need a fix, but surely it's clearer to use
the same refcounting there as in shmem_alloc_page(), delete its onstack
mempolicy, and the strange mpol_cond_copy() and __mpol_cond_copy() -
those were invented to let swapin_readahead() make an unknown number of
calls to alloc_pages_vma() with one mempolicy; but since 00442ad04a5e,
alloc_pages_vma() has kept refcount in balance, so now no problem.
Signed-off-by: Francois Romieu <romieu@fr.zoreil.com> Acked-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: CAI Qian <caiqian@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Way Access Filter in recent AMD CPUs may hurt the performance of
some workloads, caused by aliasing issues in the L1 cache.
This patch disables it on the affected CPUs.
The issue is similar to that one of last year:
http://lkml.indiana.edu/hypermail/linux/kernel/1107.3/00041.html
This new patch does not replace the old one, we just need another
quirk for newer CPUs.
The performance penalty without the patch depends on the
circumstances, but is a bit less than the last year's 3%.
The workloads affected would be those that access code from the same
physical page under different virtual addresses, so different
processes using the same libraries with ASLR or multiple instances of
PIE-binaries. The code needs to be accessed simultaneously from both
cores of the same compute unit.
More details can be found here:
http://developer.amd.com/Assets/SharedL1InstructionCacheonAMD15hCPU.pdf
CPUs affected are anything with the core known as Piledriver.
That includes the new parts of the AMD A-Series (aka Trinity) and the
just released new CPUs of the FX-Series (aka Vishera).
The model numbering is a bit odd here: FX CPUs have model 2,
A-Series has model 10h, with possible extensions to 1Fh. Hence the
range of model ids.
We handle NOTIFY_THROTTLING so don't then fall through to unsupported event.
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Peter Huewe <peterhuewe@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Huawei E173 is a QMI/wwan device which normally appear
as 12d1:1436 in Linux. The descriptors displayed in that
mode will be picked up by cdc_ether. But the modem has
another mode with a different device ID and a slightly
different set of descriptors. This is the mode used by
Windows like this:
3Modem: USB\VID_12D1&PID_140C&MI_00\6&3A1D2012&0&0000
Networkcard: USB\VID_12D1&PID_140C&MI_01\6&3A1D2012&0&0001
Appli.Inter: USB\VID_12D1&PID_140C&MI_02\6&3A1D2012&0&0002
PC UI Inter: USB\VID_12D1&PID_140C&MI_03\6&3A1D2012&0&0003
Reported-by: Thomas Schäfer <tschaefer@t-online.de> Signed-off-by: Bjørn Mork <bjorn@mork.no> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Peter Huewe <peterhuewe@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Bjørn Mork <bjorn@mork.no> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Peter Huewe <peterhuewe@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The i7core_edac addrmatch_dev and chancounts_dev have sysfs files
associated with them. The sysfs files, however, are coded so that the
parent device is is the mci device. This is incorrect and the mci struct
should be obtained through the addrmatch_dev and chancounts_dev device's
private data field which is populated in i7core_create_sysfs_devices().
fix bug where a register which was only meant to be read in 578xx/57712
devices causes a bogus error message to be logged when read from other
devices.
Signed-off-by: Ariel Elior <ariele@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: CAI Qian <caiqian@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Between commits 8323f26ce342 ("sched: Fix race in task_group()") and 800d4d30c8f2 ("sched, autogroup: Stop going ahead if autogroup is
disabled"), autogroup is a wreck.
With both applied, all you have to do to crash a box is disable
autogroup during boot up, then reboot.. boom, NULL pointer dereference
due to commit 800d4d30c8f2 not allowing autogroup to move things, and
commit 8323f26ce342 making that the only way to switch runqueues:
* Right-shift the values in GET_FBD_FAT_IDX and GET_FBD_NF_IDX, so
that the callers get the result they expect.
* Fix definition of FERR_FAT_FBD_ERR_MASK.
* Call GET_FBD_NF_IDX, not GET_FBD_FAT_IDX, when operating on
register FERR_NF_FBD. We were lucky they have the same definition.
This fixes kernel bug #44131:
https://bugzilla.kernel.org/show_bug.cgi?id=44131
If the raid1 or raid10 unplug function gets called
from a make_request function (which is very possible) when
there are bios on the current->bio_list list, then it will not
be able to successfully call bitmap_unplug() and it could
need to submit more bios and wait for them to complete.
But they won't complete while current->bio_list is non-empty.
So detect that case and handle the unplugging off to another thread
just like we already do when called from within the scheduler.
RAID1 version of bug was introduced in 3.6, so that part of fix is
suitable for 3.6.y. RAID10 part won't apply.
Reported-by: Torsten Kaiser <just.for.lkml@googlemail.com> Reported-by: Peter Maloney <peter.maloney@brockmann-consult.de> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Felix Liao reported that when an interface is set DOWN
while another interface is executing a ROC, the warning
in ieee80211_start_next_roc() (about the first item on
the list having started already) triggers.
This is because ieee80211_roc_purge() calls it even if
it never actually changed the list of ROC items. To fix
this, simply remove the function call. If it is needed
then it will be done by the ieee80211_sw_roc_work()
function when the ROC item that is being removed while
active is cleaned up.
Reported-by: Felix Liao <Felix.Liao@watchguard.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A rescue thread exiting TASK_INTERRUPTIBLE can lead to a task scheduling
off, never to be seen again. In the case where this occurred, an exiting
thread hit reiserfs homebrew conditional resched while holding a mutex,
bringing the box to its knees.
When a cpu enters S3 state, the FPU state is lost.
After resuming for S3, if we try to lazy restore the FPU for a process running
on the same CPU, this will result in a corrupted FPU context.
Ensure that "fpu_owner_task" is properly invalided when (re-)initializing a CPU,
so nobody will try to lazy restore a state which doesn't exist in the hardware.
Tested with a 64-bit kernel on a 4-core Ivybridge CPU with eagerfpu=off,
by doing thousands of suspend/resume cycles with 4 processes doing FPU
operations running. Without the patch, a process is killed after a
few hundreds cycles by a SIGFPE.
Signed-off-by: Vincent Palatin <vpalatin@chromium.org> Cc: Duncan Laurie <dlaurie@chromium.org> Cc: Olof Johansson <olofj@chromium.org> Link: http://lkml.kernel.org/r/1354306532-1014-1-git-send-email-vpalatin@chromium.org Signed-off-by: H. Peter Anvin <hpa@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
- The code relies on rc_pci_fixup being called, which only happens
when CONFIG_PCI_QUIRKS is enabled, so add that to Kconfig. Omitting
this causes a booting failure with a non-obvious cause.
- Update rc_pci_fixup to set the class properly, copying the
more modern style from other places
- Correct the rc_pci_fixup comment
Signed-off-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com> Signed-off-by: Jason Cooper <jason@lakedaemon.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
- If we stop processing an already-aborted command in
target_execute_cmd(), then we need to complete t_transport_stop_comp
to wake up the the TMR handling thread, or else it will end up
waiting forever.
- If we've a already sent an "aborted" status for a command in
transport_check_aborted_status() then we should bail out of
transport_send_task_abort() to avoid freeing the command twice.
Signed-off-by: Roland Dreier <roland@purestorage.com> Signed-off-by: Nicholas Bellinger <nab@risingtidesystems.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Kswapd does not in all places have the same criteria for a balanced
zone. Zones are only being reclaimed when their high watermark is
breached, but compaction checks loop over the zonelist again when the
zone does not meet the low watermark plus two times the size of the
allocation. This gets kswapd stuck in an endless loop over a small
zone, like the DMA zone, where the high watermark is smaller than the
compaction requirement.
Add a function, zone_balanced(), that checks the watermark, and, for
higher order allocations, if compaction has enough free memory. Then
use it uniformly to check for balanced zones.
This makes sure that when the compaction watermark is not met, at least
reclaim happens and progress is made - or the zone is declared
unreclaimable at some point and skipped entirely.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reported-by: George Spelvin <linux@horizon.com> Reported-by: Johannes Hirte <johannes.hirte@fem.tu-ilmenau.de> Reported-by: Tomas Racek <tracek@redhat.com> Tested-by: Johannes Hirte <johannes.hirte@fem.tu-ilmenau.de> Reviewed-by: Rik van Riel <riel@redhat.com> Cc: Mel Gorman <mel@csn.ul.ie> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
I enable CONFIG_DEBUG_VIRTUAL and CONFIG_SPARSEMEM_VMEMMAP, when doing
memory hotremove, there is a kernel BUG at arch/x86/mm/physaddr.c:20.
It is caused by free_section_usemap()->virt_to_page(), virt_to_page() is
only used for kernel direct mapping address, but sparse-vmemmap uses
vmemmap address, so it is going wrong here.
PMU interrupts start at IRQ_DOVE_PMU_START, not IRQ_DOVE_PMU_START + 1.
Fix the condition. (It may have been less likely to occur had the code
been written "if (irq >= IRQ_DOVE_PMU_START" which imho is the easier
to understand notation, and matches the normal way of thinking about
these things.)
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Jason Cooper <jason@lakedaemon.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix the acknowledgement of PMU interrupts on Dove: some Dove hardware
has not been sensibly designed so that interrupts can be handled in a
race free manner. The PMU is one such instance.
The pending (aka 'cause') register is a bunch of RW bits, meaning that
these bits can be both cleared and set by software (confirmed on the
Armada-510 on the cubox.)
Hardware sets the appropriate bit when an interrupt is asserted, and
software is required to clear the bits which are to be processed. If
we write ~(1 << bit), then we end up asserting every other interrupt
except the one we're processing. So, we need to do a read-modify-write
cycle to clear the asserted bit.
However, any interrupts which occur in the middle of this cycle will
also be written back as zero, which will also clear the new interrupts.
The upshot of this is: there is _no_ way to safely clear down interrupts
in this register (and other similarly behaving interrupt pending
registers on this device.) The patch below at least stops us creating
new interrupts.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Jason Cooper <jason@lakedaemon.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When a replacement operation completes there is a small window
when the original device is marked 'faulty' and the replacement
still looks like a replacement. The faulty should be removed and
the replacement moved in place very quickly, bit it isn't instant.
So the code write out to the array must handle the possibility that
the only working device for some slot in the replacement - but it
doesn't. If the primary device is faulty it just gives up. This
can lead to corruption.
So make the code more robust: if either the primary or the
replacement is present and working, write to them. Only when
neither are present do we give up.
This bug has been present since replacement was introduced in
3.3, so it is suitable for any -stable kernel since then.
The EEH core is talking with the PCI device driver to determine the
action (purely reset, or PCI device removal). During the period, the
driver might be unloaded and in turn causes kernel crash as follows:
The patch increases the reference of the corresponding driver modules
while EEH core does the negotiation with PCI device driver so that the
corresponding driver modules can't be unloaded during the period and
we're safe to refer the callbacks.
Reported-by: Alexey Kardashevskiy <aik@ozlabs.ru> Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
[ herton: backported for 3.5, adjusted driver assignments, return 0
instead of NULL, assume dev is not NULL ] Signed-off-by: Herton Ronaldo Krzesinski <herton.krzesinski@canonical.com> Cc: Ben Hutchings <ben@decadent.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The update_backups() function is used to backup all the metadata
blocks, so we should not take it for granted that 'data' is pointed to
a super block and use ext4_superblock_csum_set to calculate the
checksum there. In case where the data is a group descriptor block,
it will corrupt the last group descriptor, and then e2fsck will
complain about it it.
As all the metadata checksums should already be OK when we do the
backup, remove the wrong ext4_superblock_csum_set and it should be
just fine.
Reported-by: "Theodore Ts'o" <tytso@mit.edu> Signed-off-by: Tao Ma <boyu.mt@taobao.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: Ben Hutchings <ben@decadent.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since 4.4 GCC on MIPS no longer recognizes the "h" constraint,
leading to this build failure:
CC lib/mpi/generic_mpih-mul1.o
lib/mpi/generic_mpih-mul1.c: In function 'mpihelp_mul_1':
lib/mpi/generic_mpih-mul1.c:50:3: error: impossible constraint in 'asm'
This patch updates MPI with the latest umul_ppm implementations for MIPS.
Commit 5515061d22f0 ("mm: throttle direct reclaimers if PF_MEMALLOC
reserves are low and swap is backed by network storage") introduced a
check for fatal signals after a process gets throttled for network
storage. The intention was that if a process was throttled and got
killed that it should not trigger the OOM killer. As pointed out by
Minchan Kim and David Rientjes, this check is in the wrong place and too
broad. If a system is in am OOM situation and a process is exiting, it
can loop in __alloc_pages_slowpath() and calling direct reclaim in a
loop. As the fatal signal is pending it returns 1 as if it is making
forward progress and can effectively deadlock.
This patch moves the fatal_signal_pending() check after throttling to
throttle_direct_reclaim() where it belongs. If the process is killed
while throttled, it will return immediately without direct reclaim
except now it will have TIF_MEMDIE set and will use the PFMEMALLOC
reserves.
Minchan pointed out that it may be better to direct reclaim before
returning to avoid using the reserves because there may be pages that
can easily reclaim that would avoid using the reserves. However, we do
no such targetted reclaim and there is no guarantee that suitable pages
are available. As it is expected that this throttling happens when
swap-over-NFS is used there is a possibility that the process will
instead swap which may allocate network buffers from the PFMEMALLOC
reserves. Hence, in the swap-over-nfs case where a process can be
throtted and be killed it can use the reserves to exit or it can
potentially use reserves to swap a few pages and then exit. This patch
takes the option of using the reserves if necessary to allow the process
exit quickly.
If this patch passes review it should be considered a -stable candidate
for 3.6.
Signed-off-by: Mel Gorman <mgorman@suse.de> Cc: David Rientjes <rientjes@google.com> Cc: Luigi Semenzato <semenzato@google.com> Cc: Dan Magenheimer <dan.magenheimer@oracle.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Sonny Rao <sonnyrao@google.com> Cc: Minchan Kim <minchan@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: CAI Qian <caiqian@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
cmd was freed in pn533_dep_link_up regardless of
pn533_send_cmd_frame_async return code. Cmd is passed as argument to
pn533_in_dep_link_up_complete callback and should be freed there.
Signed-off-by: Szymon Janc <szymon.janc@tieto.com> Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: Peter Huewe <peterhuewe@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On hosts without the XSAVE support unprivileged local user can trigger
oops similar to the one below by setting X86_CR4_OSXSAVE bit in guest
cr4 register using KVM_SET_SREGS ioctl and later issuing KVM_RUN
ioctl.
QEMU first retrieves the supported features via KVM_GET_SUPPORTED_CPUID
and then sets them later. So guest's X86_FEATURE_XSAVE should be masked
out on hosts without X86_FEATURE_XSAVE, making kvm_set_cr4 with
X86_CR4_OSXSAVE fail. Userspaces that allow specifying guest cpuid with
X86_FEATURE_XSAVE even on hosts that do not support it, might be
susceptible to this attack from inside the guest as well.
Allow setting X86_CR4_OSXSAVE bit only if host has XSAVE support.
The delayed work function int_in_work() may call usb_reset_device()
and thus, indirectly, the driver's pre_reset method. Trying to
cancel the work synchronously in that situation would deadlock.
Fix by avoiding cancel_work_sync() in the pre_reset method.
If the reset was NOT initiated by int_in_work() this might cause
int_in_work() to run after the post_reset method, with urb_int_in
already resubmitted, so handle that case gracefully.
Signed-off-by: Tilman Schmidt <tilman@imap.cc> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 169ebd90131b ("writeback: Avoid iput() from flusher thread")
removed iget-iput pair from inode writeback. As a side effect, inodes
that are dirty during iput_final() call won't be ever added to inode LRU
(iput_final() doesn't add dirty inodes to LRU and later when the inode
is cleaned there's noone to add the inode there). Thus inodes are
effectively unreclaimable until someone looks them up again.
The practical effect of this bug is limited by the fact that inodes are
pinned by a dentry for long enough that the inode gets cleaned. But
still the bug can have nasty consequences leading up to OOM conditions
under certain circumstances. Following can easily reproduce the
problem:
for (( i = 0; i < 1000; i++ )); do
mkdir $i
for (( j = 0; j < 1000; j++ )); do
touch $i/$j
echo 2 > /proc/sys/vm/drop_caches
done
done
then one needs to run 'sync; ls -lR' to make inodes reclaimable again.
We fix the issue by inserting unused clean inodes into the LRU after
writeback finishes in inode_sync_complete().
Signed-off-by: Jan Kara <jack@suse.cz> Reported-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Dave Chinner <david@fromorbit.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
dev_pm_qos_add_request() can return 0, 1, or a negative error code,
therefore the correct error test is "if (error < 0)." Checking just for
non-zero return code leads to erroneous setting of the req->dev pointer
to NULL, which then leads to a repeated call to
dev_pm_qos_add_ancestor_request() in st1232_ts_irq_handler(). This in turn
leads to an Oops, when the I2C host adapter is unloaded and reloaded again
because of the inconsistent state of its QoS request list.
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If a signal handler is executed on altstack and another signal comes,
we will end up with rt_sigreturn() on return from the second handler
getting -EPERM from do_sigaltstack(). It's perfectly OK, since we
are not asking to change the settings; in fact, they couldn't have been
changed during the second handler execution exactly because we'd been
on altstack all along. 64bit sigreturn on sparc treats any error from
do_sigaltstack() as "SIGSEGV now"; we need to switch to the same semantics
we are using on other architectures.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 09e05d48 introduced a wait for transaction commit into
journal_unmap_buffer() in the case we are truncating a buffer undergoing commit
in the page stradding i_size on a filesystem with blocksize < pagesize. Sadly
we forgot to drop buffer lock before waiting for transaction commit and thus
deadlock is possible when kjournald wants to lock the buffer.
Fix the problem by dropping the buffer lock before waiting for transaction
commit. Since we are still holding page lock (and that is OK), buffer cannot
disappear under us.
Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Set in the rx_ifindex to pass the correct interface index in the case of a
message timeout detection. Usually the rx_ifindex value is set at receive
time. But when no CAN frame has been received the RX_TIMEOUT notification
did not contain a valid value.
Reported-by: Andre Naujoks <nautsch2@googlemail.com> Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net> Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The skb->tstamp is set to the hardware timestamp when available in the USB
urb message. This leads to user visible timestamps which contain the 'uptime'
of the USB adapter - and not the usual system generated timestamp.
Fix this wrong assignment by applying the available hardware timestamp to the
skb_shared_hwtstamps data structure - which is intended for this purpose.
Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net> Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Intel 82855PM host bridge / Mobility FireGL 9000 RV250 combination
in an (outdated) ThinkPad T41 needs AGPMode 1 for suspend/resume (under
KMS, that is). So add a quirk for it.
(Change R250 to RV250 in comment for preceding quirk too.)
Signed-off-by: Paul Bolle <pebolle@tiscali.nl> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The check whether the IBSS is active and can be removed should be
performed before deinitializing the fields used for the check/search.
Otherwise, the configured BSS will not be found and removed properly.
To make it more clear for the future, rename sdata->u.ibss to the
local pointer ifibss which is used within the checks.
Dave Jones reported a bug with futex_lock_pi() that his trinity test
exposed. Sometime between queue_me() and taking the q.lock_ptr, the
lock_ptr became NULL, resulting in a crash.
While futex_wake() is careful to not call wake_futex() on futex_q's with
a pi_state or an rt_waiter (which are either waiting for a
futex_unlock_pi() or a PI futex_requeue()), futex_wake_op() and
futex_requeue() do not perform the same test.
Update futex_wake_op() and futex_requeue() to test for q.pi_state and
q.rt_waiter and abort with -EINVAL if detected. To ensure any future
breakage is caught, add a WARN() to wake_futex() if the same condition
is true.
This fix has seen 3 hours of testing with "trinity -c futex" on an
x86_64 VM with 4 CPUS.
[akpm@linux-foundation.org: tidy up the WARN()] Signed-off-by: Darren Hart <dvhart@linux.intel.com> Reported-by: Dave Jones <davej@redat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: John Kacur <jkacur@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Request based dm attempts to re-run the request queue off the
request completion path. If used with a driver that potentially does
end_io from its request_fn, we could deadlock trying to recurse
back into request dispatch. Fix this by punting the request queue
run to kblockd.
Tested to fix a quickly reproducible deadlock in such a scenario.
After we've done __elv_add_request() and __blk_run_queue() in
blk_execute_rq_nowait(), the request might finish and be freed
immediately. Therefore checking if the type is REQ_TYPE_PM_RESUME
isn't safe afterwards, because if it isn't, rq might be gone.
Instead, check beforehand and stash the result in a temporary.
This fixes crashes in blk_execute_rq_nowait() I get occasionally when
running with lots of memory debugging options enabled -- I think this
race is usually harmless because the window for rq to be reallocated
is so small.
When a write to a replacement device completes, we carefully
and correctly found the rdev that the write actually went to
and the blithely called rdev_dec_pending on the primary rdev,
even if this write was to the replacement.
This means that any writes to an array while a replacement
was ongoing would cause the nr_pending count for the primary
device to go negative, so it could never be removed.
This bug has been present since replacement was introduced in
3.3, so it is suitable for any -stable kernel since then.
projects / thirdparty / kernel / stable.git / log
