platform/adaptation/renesas_rcar/renesas_kernel.git
13 years agodtc/powerpc: remove obsolete .gitignore entries
Wolfram Sang [Sat, 12 Mar 2011 16:44:10 +0000 (17:44 +0100)]
dtc/powerpc: remove obsolete .gitignore entries

dtc was moved and .gitignores have been added to the new location. So, we can
delete the old, forgotten ones.

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
13 years agoMerge branch 'pm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/suspe...
Linus Torvalds [Wed, 8 Jun 2011 22:58:51 +0000 (15:58 -0700)]
Merge branch 'pm-fixes' of git://git./linux/kernel/git/rafael/suspend-2.6

* 'pm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/suspend-2.6:
  PM / Runtime: Fix loops in pm_runtime_clk_notify()
  PM / Intel IOMMU: Fix init_iommu_pm_ops() for CONFIG_PM unset

13 years agovfs: reorganize 'struct inode' layout a bit
Linus Torvalds [Wed, 8 Jun 2011 22:18:19 +0000 (15:18 -0700)]
vfs: reorganize 'struct inode' layout a bit

This tries to make the 'struct inode' accesses denser in the data cache
by moving a commonly accessed field (i_security) closer to other fields
that are accessed often.

It also makes 'i_state' just an 'unsigned int' rather than 'unsigned
long', since we only use a few bits of that field, and moves it next to
the existing 'i_flags' so that we potentially get better structure
layout (although depending on config options, i_flags may already have
packed in the same word as i_lock, so this improves packing only for the
case of spinlock debugging)

Out 'struct inode' is still way too big, and we should probably move
some other fields around too (the acl fields in particular) for better
data cache access density.  Other fields (like the inode hash) are
likely to be entirely irrelevant under most loads.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoselinux: simplify and clean up inode_has_perm()
Linus Torvalds [Wed, 8 Jun 2011 22:11:56 +0000 (15:11 -0700)]
selinux: simplify and clean up inode_has_perm()

This is a rather hot function that is called with a potentially NULL
"struct common_audit_data" pointer argument.  And in that case it has to
provide and initialize its own dummy common_audit_data structure.

However, all the _common_ cases already pass it a real audit-data
structure, so that uncommon NULL case not only creates a silly run-time
test, more importantly it causes that function to have a big stack frame
for the dummy variable that isn't even used in the common case!

So get rid of that stupid run-time behavior, and make the (few)
functions that currently call with a NULL pointer just call a new helper
function instead (naturally called inode_has_perm_noapd(), since it has
no adp argument).

This makes the run-time test be a static code generation issue instead,
and allows for a much denser stack since none of the common callers need
the dummy structure.  And a denser stack not only means less stack space
usage, it means better cache behavior.  So we have a win-win-win from
this simplification: less code executed, smaller stack footprint, and
better cache behavior.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoMerge branch 'usb-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Wed, 8 Jun 2011 21:27:48 +0000 (14:27 -0700)]
Merge branch 'usb-linus' of git://git./linux/kernel/git/gregkh/usb-2.6

* 'usb-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6: (28 commits)
  MAINTAINERS: add a maintainer to Gadget Framework
  USB: serial: add another 4N-GALAXY.DE PID to ftdi_sio driver
  Revert "USB: option: add ID for ZTE MF 330"
  drivers/usb/host/ohci-pxa27x.c: add missing clk_put
  USB: CONFIG_USB_GADGET_DUALSPEED is not user-configurable
  USB: dummy-hcd needs the has_tt flag
  usb-storage: redo incorrect reads
  usb/renesas_usbhs: free uep on removal
  usb/s3c-hsudc: fix error path
  usb/pxa25x_udc: cleanup the LUBBOCK err path
  usb/mv_udc_core: fix compile
  usb: gadget: include <linux/prefetch.h> to fix compiling error
  USB: s3c-hsotg: Tone down debugging
  usb: remove bad dput after dentry_unhash
  USB: core: Tolerate protocol stall during hub and port status read
  musb: fix prefetch build failure
  USB: cdc-acm: Adding second ACM channel support for Nokia E7 and C7
  usb-gadget: unlock data->lock mutex on error path in ep_write()
  USB: option Add blacklist for ZTE K3765-Z (19d2:2002)
  option: add Prolink PH300 modem IDs
  ...

13 years agoMAINTAINERS: add a maintainer to Gadget Framework
Felipe Balbi [Wed, 8 Jun 2011 16:16:28 +0000 (19:16 +0300)]
MAINTAINERS: add a maintainer to Gadget Framework

I'll be continuing the amazing work Dave has
done with the Gadget Framework.

Signed-off-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoMerge branch 'spi/merge' of git://git.secretlab.ca/git/linux-2.6
Linus Torvalds [Wed, 8 Jun 2011 20:54:46 +0000 (13:54 -0700)]
Merge branch 'spi/merge' of git://git.secretlab.ca/git/linux-2.6

* 'spi/merge' of git://git.secretlab.ca/git/linux-2.6:
  spi/rtc-m41t93: Use spi_get_drvdata() for SPI devices
  spi/omap2: fix uninitialized variable

13 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6
Linus Torvalds [Wed, 8 Jun 2011 20:54:29 +0000 (13:54 -0700)]
Merge git://git./linux/kernel/git/sfrench/cifs-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
  cifs: trivial: add space in fsc error message
  cifs: silence printk when establishing first session on socket
  CIFS ACL support needs CONFIG_KEYS, so depend on it
  possible memory corruption in cifs_parse_mount_options()
  cifs: make CIFS depend on CRYPTO_ECB
  cifs: fix the kernel release version in the default security warning message

13 years agoMerge 3.0-rc2 + Linus's latest into usb-linus
Greg Kroah-Hartman [Wed, 8 Jun 2011 20:50:00 +0000 (13:50 -0700)]
Merge 3.0-rc2 + Linus's latest into usb-linus

This is needed to get the following MAINTAINERS patch to apply properly.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: serial: add another 4N-GALAXY.DE PID to ftdi_sio driver
Steffen Sledz [Tue, 7 Jun 2011 12:01:56 +0000 (14:01 +0200)]
USB: serial: add another 4N-GALAXY.DE PID to ftdi_sio driver

E.g. newer CAN 2.0 A/B <=> USB 2.0 converters report idProduct=f3c2.

Signed-off-by: Steffen Sledz <sledz@dresearch-fe.de>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoBtrfs: fix duplicate checking logic
Josef Bacik [Mon, 6 Jun 2011 14:50:35 +0000 (10:50 -0400)]
Btrfs: fix duplicate checking logic

When merging my code into the integration test the second check for duplicate
entries got screwed up.  This patch fixes it by dropping ret2 and just using ret
for the return value, and checking if we got an error before adding the bitmap
to the local list.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
13 years agoBtrfs: fix the allocator loop logic
Josef Bacik [Fri, 27 May 2011 20:11:38 +0000 (16:11 -0400)]
Btrfs: fix the allocator loop logic

I was testing with empty_cluster = 0 to try and reproduce a problem and kept
hitting early enospc panics.  This was because our loop logic was a little
confused.  So this is what I did

1) Make the loop variable the ultimate decider on wether we should loop again
isntead of checking to see if we had an uncached bg, empty size or empty
cluster.

2) Increment loop before checking to see what we are on to make the loop
definitions make more sense.

3) If we are on the chunk alloc loop don't set empty_size/empty_cluster to 0
unless we didn't actually allocate a chunk.  If we did allocate a chunk we
should be able to easily setup a new cluster so clearing
empty_size/empty_cluster makes us less efficient.

This kept me from hitting panics while trying to reproduce the other problem.
Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
13 years agoBtrfs: fix bitmap regression
Josef Bacik [Fri, 27 May 2011 18:07:49 +0000 (14:07 -0400)]
Btrfs: fix bitmap regression

In cleaning up the clustering code I accidently introduced a regression by
adding bitmap entries to the cluster rb tree.  The problem is if we've maxed out
the number of bitmaps we can have for the block group we can only add free space
to the bitmaps, but since the bitmap is on the cluster we can't find it and we
try to create another one.  This would result in a panic because the total
bitmaps was bigger than the max bitmaps that were allowed.  This patch fixes
this by checking to see if we have a cluster, and then looking at the cluster rb
tree to see if it has a bitmap entry and if it does and that space belongs to
that bitmap, go ahead and add it to that bitmap.

I could hit this panic every time with an fs_mark test within a couple of
minutes.  With this patch I no longer hit the panic and fs_mark goes to
completion.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
13 years agospi/rtc-m41t93: Use spi_get_drvdata() for SPI devices
Geert Uytterhoeven [Wed, 8 Jun 2011 19:48:03 +0000 (21:48 +0200)]
spi/rtc-m41t93: Use spi_get_drvdata() for SPI devices

One new offender detected by the recently increased type checking in
platform_get_drvdata():

drivers/rtc/rtc-m41t93.c: In function ‘m41t93_remove’:
drivers/rtc/rtc-m41t93.c:192: warning: passing argument 1 of ‘platform_get_drvdata’ from incompatible pointer type

Use spi_get_drvdata() instead of platform_get_drvdata(), cfr. commit
42fea15d6dc410e62dac6a764142045280624a5b ("spi/rtc-{ds1390,ds3234,m41t94}:
Use spi_get_drvdata() for SPI devices")

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
13 years agoBtrfs: don't commit the transaction if we dont have enough pinned bytes
Josef Bacik [Wed, 25 May 2011 17:10:16 +0000 (13:10 -0400)]
Btrfs: don't commit the transaction if we dont have enough pinned bytes

I noticed when running an enospc test that we would get stuck committing the
transaction in check_data_space even though we truly didn't have enough space.
So check to see if bytes_pinned is bigger than num_bytes, if it's not don't
commit the transaction.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
13 years agoBtrfs: noinline the cluster searching functions
Josef Bacik [Wed, 25 May 2011 17:07:37 +0000 (13:07 -0400)]
Btrfs: noinline the cluster searching functions

When profiling the find cluster code it's hard to tell where we are spending our
time because the bitmap and non-bitmap functions get inlined by the compiler, so
make that not happen.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
13 years agoBtrfs: cache bitmaps when searching for a cluster
Josef Bacik [Wed, 25 May 2011 17:03:16 +0000 (13:03 -0400)]
Btrfs: cache bitmaps when searching for a cluster

If we are looking for a cluster in a particularly sparse or fragmented block
group, we will do a lot of looping through the free space tree looking for
various things, and if we need to look at bitmaps we will endup doing the whole
dance twice.  So instead add the bitmap entries to a temporary list so if we
have to do the bitmap search we can just look through the list of entries we've
found quickly instead of having to loop through the entire tree again.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
13 years agoMerge branch 'stable/bug.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Wed, 8 Jun 2011 19:03:37 +0000 (12:03 -0700)]
Merge branch 'stable/bug.fixes' of git://git./linux/kernel/git/konrad/xen

* 'stable/bug.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
  xen: off by one errors in multicalls.c
  xen: use the trigger info we already have to choose the irq handler

13 years agoiwlegacy: fix channel switch locking
Stanislaw Gruszka [Wed, 8 Jun 2011 13:26:31 +0000 (15:26 +0200)]
iwlegacy: fix channel switch locking

We use priv->mutex to avoid race conditions between chswitch_done()
and mac_channel_switch(), when marking channel switch in
progress. But chswitch_done() can be called in atomic context
from rx_csa() or with mutex already taken from commit_rxon().

To fix remove mutex from chswitch_done() and use atomic bitops
for marking channel switch pending.

Cc: stable@kernel.org # 2.6.39+
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agomac80211: fix IBSS teardown race
Johannes Berg [Wed, 8 Jun 2011 11:27:29 +0000 (13:27 +0200)]
mac80211: fix IBSS teardown race

Ignacy reports that sometimes after leaving an IBSS
joining a new one didn't work because there still
were stations on the list. He fixed it by flushing
stations when attempting to join a new IBSS, but
this shouldn't be happening in the first case. When
I looked into it I saw a race condition in teardown
that could cause stations to be added after flush,
and thus cause this situation. Ignacy confirms that
after applying my patch he hasn't seen this happen
again.

Reported-by: Ignacy Gawedzki <i@lri.fr>
Debugged-by: Ignacy Gawedzki <i@lri.fr>
Tested-by: Ignacy Gawedzki <i@lri.fr>
Cc: stable@kernel.org
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoiwlagn: send tx power command if defer cause by RXON not match
Wey-Yi Guy [Fri, 27 May 2011 15:40:24 +0000 (08:40 -0700)]
iwlagn: send tx power command if defer cause by RXON not match

During channge channel, tx power will not send to uCode, the tx power command
should send after scan complete. but should also can send after RXON command.

Stable fix identified by Stanislaw Gruszka <sgruszka@redhat.com>.

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Cc: stable@kernel.org [2.6.38+]
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoStaging: cx23885: fix include of altera.h
Greg Kroah-Hartman [Wed, 8 Jun 2011 16:18:32 +0000 (09:18 -0700)]
Staging: cx23885: fix include of altera.h

The cx23885 driver was including staging/altera.h, but that file has
moved back into the driver directory.

Why a non-staging driver was including a staging driver is beyond me,
but this fixes the build so everything is happy for now.

For the record, it's not ok for a non-staging driver to depend on a
staging one, as that implies that the non-staging one should also be in
the staging tree if that's needed.

Cc: Igor M. Liplianin <liplianin@netup.ru>
Cc: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agocifs: trivial: add space in fsc error message
Jeff Layton [Wed, 8 Jun 2011 11:35:24 +0000 (07:35 -0400)]
cifs: trivial: add space in fsc error message

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
13 years agoMerge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Wed, 8 Jun 2011 15:36:15 +0000 (08:36 -0700)]
Merge branch 'perf-urgent-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  perf: Fix comments in include/linux/perf_event.h
  perf: Comment /proc/sys/kernel/perf_event_paranoid to be part of user ABI
  perf python: Fix argument name list of read_on_cpu()
  perf evlist: Don't die if sample_{id_all|type} is invalid
  perf python: Use exception to propagate errors
  perf evlist: Remove dependency on debug routines
  perf, cgroups: Fix up for new API

13 years agoMerge branch 'gpio/merge' of git://git.secretlab.ca/git/linux-2.6
Linus Torvalds [Wed, 8 Jun 2011 15:22:29 +0000 (08:22 -0700)]
Merge branch 'gpio/merge' of git://git.secretlab.ca/git/linux-2.6

* 'gpio/merge' of git://git.secretlab.ca/git/linux-2.6:
  gpio/samsung: make Kconfig options def_bool
  gpio/exynos4: Fix incorrect mapping of gpio pull-up macro to register setting
  GPIO: OMAP: add locking around calls to _set_gpio_triggering
  GPIO: OMAP: fix setting IRQWAKEN bits for OMAP4
  GPIO: OMAP: fix section mismatch warnings
  gpio: Fix gpio-exynos4 build fails in mainline

13 years agoMerge branch 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab...
Linus Torvalds [Wed, 8 Jun 2011 15:21:48 +0000 (08:21 -0700)]
Merge branch 'v4l_for_linus' of git://git./linux/kernel/git/mchehab/linux-2.6

* 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6:
  [media] soc_camera: preserve const attribute
  [media] uvc_entity: initialize return value
  [media] media: Fix media device minor registration
  [media] Make nchg variable signed because the code compares this variable against negative values
  [media] omap3isp: fix compiler warning
  [media] v4l: Fix media_entity_to_video_device macro argument name
  [media] ivtv: Internally separate encoder & decoder standard setting
  [media] ivtvfb: Add sanity check to ivtvfb_pan_display()
  [media] ivtvfb: use display information in info not in var for panning
  [media] ivtv: Make two ivtv_msleep_timeout calls uninterruptable
  [media] anysee: return EOPNOTSUPP for unsupported I2C messages
  [media] gspca - ov519: Set the default frame rate to 15 fps
  [media] gspca - stv06xx: Set a lower default value of gain for hdcs sensors
  [media] gspca: Remove coarse_expo_autogain.h
  [media] gspca - ov519: Change the ovfx2 bulk transfer size
  [media] gspca - ov519: Fix a regression for ovfx2 webcams

13 years agoMerge branch 'drm-radeon-urgent' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Wed, 8 Jun 2011 15:19:14 +0000 (08:19 -0700)]
Merge branch 'drm-radeon-urgent' of git://git./linux/kernel/git/airlied/drm-2.6

* 'drm-radeon-urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6:
  drm/radeon/kms: disable hdmi audio by default
  drm/radeon/kms: fix for radeon on systems >4GB without hardware iommu
  drm/radeon/kms: set family for use in parser.

13 years agoMerge branch 'for_3.0/gpio-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git...
Grant Likely [Wed, 8 Jun 2011 15:12:16 +0000 (09:12 -0600)]
Merge branch 'for_3.0/gpio-fixes' of git://git./linux/kernel/git/khilman/linux-omap-pm into gpio/merge

13 years agotty_buffer: get rid of 'seen_tail' logic in flush_to_ldisc
Linus Torvalds [Wed, 8 Jun 2011 14:46:30 +0000 (07:46 -0700)]
tty_buffer: get rid of 'seen_tail' logic in flush_to_ldisc

The flush_to_ldisc() work entry has special logic to notice when it has
seen the original tail of the data queue, and it avoids continuing the
flush if it sees that _original_ tail rather than the current tail.

This logic can trigger in case somebody is constantly adding new data to
the tty while the flushing is active - and the intent is to avoid
excessive CPU usage while flushing the tty, especially as we used to do
this from a softirq context which made it non-preemptible.

However, since we no longer re-arm the work-queue from within itself
(because that causes other trouble: see commit a5660b41af6a "tty: fix
endless work loop when the buffer fills up"), this just leads to
possible hung tty's (most easily seen in SMP and with a test-program
that floods a pty with data - nobody seems to have reported this for any
real-life situation yet).

And since the workqueue isn't done from timers and softirq's any more,
it's doubtful whether the CPU useage issue is really relevant any more.
So just remove the logic entirely, and see if anybody ever notices.

Alternatively, we might want to re-introduce the "re-arm the work" for
just this case, but then we'd have to re-introduce the delayed work
model or some explicit timer, which really doesn't seem worth it for
this.

Reported-and-tested-by: Guillaume Chazarain <guichaz@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoASoC: Fix WM8962 headphone volume update for use of advanced caches
Mark Brown [Tue, 7 Jun 2011 22:42:04 +0000 (23:42 +0100)]
ASoC: Fix WM8962 headphone volume update for use of advanced caches

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Liam Girdwood <lrg@ti.com>
Cc: stable@kernel.org
13 years agoMerge branch 'tip/perf/urgent-2' of git://git.kernel.org/pub/scm/linux/kernel/git...
Ingo Molnar [Wed, 8 Jun 2011 14:09:55 +0000 (16:09 +0200)]
Merge branch 'tip/perf/urgent-2' of git://git./linux/kernel/git/rostedt/linux-2.6-trace into perf/urgent

13 years agoMerge branch 'urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/rric/oprofile...
Ingo Molnar [Wed, 8 Jun 2011 13:49:03 +0000 (15:49 +0200)]
Merge branch 'urgent' of git://git./linux/kernel/git/rric/oprofile into perf/urgent

13 years agodrm/radeon/kms: disable hdmi audio by default
Alex Deucher [Mon, 6 Jun 2011 21:39:16 +0000 (17:39 -0400)]
drm/radeon/kms: disable hdmi audio by default

The current RE'd code causes blank screens and
display problems on a lot of systems.  So disable
it by default for now.  It can still be enabled
by setting the audio parameter to 1.  E.g.:
radeon.audio=1

Fixes:
https://bugs.freedesktop.org/show_bug.cgi?id=38010
https://bugs.freedesktop.org/show_bug.cgi?id=27731
https://bugs.freedesktop.org/show_bug.cgi?id=35970
https://bugs.freedesktop.org/show_bug.cgi?id=26195
and many other reported problems.

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
13 years agodrm/radeon/kms: fix for radeon on systems >4GB without hardware iommu
Daniel Haid [Wed, 8 Jun 2011 10:04:45 +0000 (20:04 +1000)]
drm/radeon/kms: fix for radeon on systems >4GB without hardware iommu

On my x86_64 system with >4GB of ram and swiotlb instead of
a hardware iommu (because I have a VIA chipset), the call
to pci_set_dma_mask (see below) with 40bits returns an error.

But it seems that the radeon driver is designed to have
need_dma32 = true exactly if pci_set_dma_mask is called
with 32 bits and false if it is called with 40 bits.

I have read somewhere that the default are 32 bits. So if the
call fails I suppose that need_dma32 should be set to true.

And indeed the patch fixes the problem I have had before
and which I had described here:
http://choon.net/forum/read.php?21,106131,115940

Acked-by: Alex Deucher <alexdeucher@gmail.com>
cc: stable@kernel.org
Signed-off-by: Dave Airlie <airlied@redhat.com>
13 years agodrm/radeon/kms: set family for use in parser.
Dave Airlie [Wed, 8 Jun 2011 09:58:29 +0000 (19:58 +1000)]
drm/radeon/kms: set family for use in parser.

Wierdly the kms parser never initialised the family, it wasn't really used
much, but the fmt checker patch started using it and it fell over.

Reported-by: Markus Trippelsdorf <markus@trippelsdorf.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
13 years agox86: cpu-hotplug: Prevent softirq wakeup on wrong CPU
Thomas Gleixner [Tue, 20 Jul 2010 12:34:50 +0000 (14:34 +0200)]
x86: cpu-hotplug: Prevent softirq wakeup on wrong CPU

After a newly plugged CPU sets the cpu_online bit it enables
interrupts and goes idle. The cpu which brought up the new cpu waits
for the cpu_online bit and when it observes it, it sets the cpu_active
bit for this cpu. The cpu_active bit is the relevant one for the
scheduler to consider the cpu as a viable target.

With forced threaded interrupt handlers which imply forced threaded
softirqs we observed the following race:

cpu 0                         cpu 1

bringup(cpu1);
                              set_cpu_online(smp_processor_id(), true);
              local_irq_enable();
while (!cpu_online(cpu1));
                              timer_interrupt()
                                -> wake_up(softirq_thread_cpu1);
                                     -> enqueue_on(softirq_thread_cpu1, cpu0);

                                                                        ^^^^

cpu_notify(CPU_ONLINE, cpu1);
  -> sched_cpu_active(cpu1)
     -> set_cpu_active((cpu1, true);

When an interrupt happens before the cpu_active bit is set by the cpu
which brought up the newly onlined cpu, then the scheduler refuses to
enqueue the woken thread which is bound to that newly onlined cpu on
that newly onlined cpu due to the not yet set cpu_active bit and
selects a fallback runqueue. Not really an expected and desirable
behaviour.

So far this has only been observed with forced hard/softirq threading,
but in theory this could happen without forced threaded hard/softirqs
as well. It's probably unobservable as it would take a massive
interrupt storm on the newly onlined cpu which causes the softirq loop
to wake up the softirq thread and an even longer delay of the cpu
which waits for the cpu_online bit.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Peter Zijlstra <peterz@infradead.org>
Cc: stable@kernel.org # 2.6.39
13 years agoceph: unwind canceled flock state
Sage Weil [Wed, 25 May 2011 21:56:12 +0000 (14:56 -0700)]
ceph: unwind canceled flock state

If we request a lock and then abort (e.g., ^C), we need to send a matching
unlock request to the MDS to unwind our lock attempt to avoid indefinitely
blocking other clients.

Reported-by: Brian Chrisman <brchrisman@gmail.com>
Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoceph: fix ENOENT logic in striped_read
Sage Weil [Wed, 8 Jun 2011 03:40:35 +0000 (20:40 -0700)]
ceph: fix ENOENT logic in striped_read

Getting ENOENT is equivalent to reading 0 bytes.  Make that correction
before setting up the hit_stripe and was_short flags.

Fixes the following case:
 dd if=/dev/zero of=/mnt/fs_depot/dd3 bs=1 seek=1048576 count=0
 dd if=/mnt/fs_depot/dd3 of=/root/ddout1 skip=8 bs=500 count=2 iflag=direct

Reported-by: Henry C Chang <henry.cy.chang@gmail.com>
Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoceph: fix short sync reads from the OSD
Sage Weil [Wed, 1 Jun 2011 23:08:44 +0000 (16:08 -0700)]
ceph: fix short sync reads from the OSD

If we get a short read from the OSD because the object is small, we need to
zero the remainder of the buffer.  For O_DIRECT reads, the attempted range
is not trimmed to i_size by the VFS, so we were actually looping
indefinitely.

Fix by trimming by i_size, and the unconditionally zeroing the trailing
range.

Reported-by: Jeff Wu <cpwu@tnsoft.com.cn>
Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoceph: fix sync vs canceled write
Sage Weil [Fri, 3 Jun 2011 16:37:09 +0000 (09:37 -0700)]
ceph: fix sync vs canceled write

If we cancel a write, trigger the safe completions to prevent a sync from
blocking indefinitely in ceph_osdc_sync().

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoceph: use ihold when we already have an inode ref
Sage Weil [Fri, 27 May 2011 16:24:26 +0000 (09:24 -0700)]
ceph: use ihold when we already have an inode ref

We should use ihold whenever we already have a stable inode ref, even
when we aren't holding i_lock.  This avoids adding new and unnecessary
locking dependencies.

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoxfrm: Fix off by one in the replay advance functions
Steffen Klassert [Sun, 5 Jun 2011 20:46:03 +0000 (20:46 +0000)]
xfrm: Fix off by one in the replay advance functions

We may write 4 byte too much when we reinitialize the anti replay
window in the replay advance functions. This patch fixes this by
adjusting the last index of the initialization loop.

Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoMerge branch 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Wed, 8 Jun 2011 02:21:11 +0000 (19:21 -0700)]
Merge branch 'irq-urgent-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  genirq: Ensure we locate the passed IRQ in irq_alloc_descs()
  genirq: Fix descriptor init on non-sparse IRQs
  irq: Handle spurios irq detection for threaded irqs
  genirq: Print threaded handler in spurious debug output

13 years agoMerge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Wed, 8 Jun 2011 02:20:53 +0000 (19:20 -0700)]
Merge branch 'x86-urgent-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86/amd-iommu: Fix boot crash with hidden PCI devices
  x86/amd-iommu: Use only per-device dma_ops
  x86/amd-iommu: Fix 3 possible endless loops

13 years agoMerge branch 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Wed, 8 Jun 2011 02:20:28 +0000 (19:20 -0700)]
Merge branch 'sched-urgent-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  sched: Fix/clarify set_task_cpu() locking rules
  lockdep: Fix lock_is_held() on recursion
  sched: Fix schedstat.nr_wakeups_migrate
  sched: Fix cross-cpu clock sync on remote wakeups

13 years agoMerge branch 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied...
Linus Torvalds [Wed, 8 Jun 2011 02:09:26 +0000 (19:09 -0700)]
Merge branch 'drm-fixes' of git://git./linux/kernel/git/airlied/drm-2.6

* 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6:
  drm/nv40: fall back to paged dma object for the moment
  drm/nouveau: fix leak of gart mm node
  drm/nouveau: fix vram page mapping when crossing page table boundaries
  drm/nv17-nv40: Fix modesetting failure when pitch == 4096px (fdo bug 35901).
  drm/nouveau: don't create accel engine objects when noaccel=1
  drm/nvc0: recognise 0xdX chipsets as NV_C0
  drm/i915: Add a no lvds quirk for the Asus EeeBox PC EB1007
  drm/i915: Share the common force-audio property between connectors
  drm/i915: Remove unused enum "chip_family"
  drm/915: fix relaxed tiling on gen2: tile height
  drm/i915/crt: Explicitly return false if connected to a digital monitor
  drm/i915: Replace ironlake_compute_wm0 with g4x_compute_wm0
  drm/i915: Only print out the actual number of fences for i915_error_state
  drm/i915: s/addr & ~PAGE_MASK/offset_in_page(addr)/
  drm: i915: correct return status in intel_hdmi_mode_valid()
  drm/i915: fix regression after clock gating init split
  drm/i915: fix if statement in ivybridge irq handler

13 years agoMerge branch 'drm-radeon-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Wed, 8 Jun 2011 02:09:17 +0000 (19:09 -0700)]
Merge branch 'drm-radeon-fixes' of git://git./linux/kernel/git/airlied/drm-2.6

* 'drm-radeon-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6:
  drm/radeon/kms/atom: fix PHY init
  drm/radeon/kms: add missing Evergreen texture formats to the CS parser
  drm/radeon/kms: viewport height has to be even
  drm/radeon/kms: remove duplicate reg from r600 safe regs
  drm/radeon/kms: add support for Llano Fusion APUs
  drm/radeon/kms: add llano pci ids
  drm/radeon/kms: fill in asic struct for llano
  drm/radeon/kms: add family ids for llano APUs
  drm/radeon: fix oops in ttm reserve when pageflipping (v2)
  drm/radeon/kms: clean up the radeon kms Kconfig
  drm/radeon/kms: fix thermal sensor reading on juniper
  drm/radeon/kms: add missing case for cayman thermal sensor
  drm/radeon/kms: add blit support for cayman (v2)
  drm/radeon/kms/blit: workaround some hw issues on evergreen+

13 years agoMerge branch 'timers-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Wed, 8 Jun 2011 02:07:22 +0000 (19:07 -0700)]
Merge branch 'timers-fixes-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'timers-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  timers: Consider slack value in mod_timer()
  clockevents: Handle empty cpumask gracefully

13 years agoMerge branch 'kvm-updates/3.0' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Linus Torvalds [Wed, 8 Jun 2011 02:06:28 +0000 (19:06 -0700)]
Merge branch 'kvm-updates/3.0' of git://git./virt/kvm/kvm

* 'kvm-updates/3.0' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: Initialize kvm before registering the mmu notifier
  KVM: x86: use proper port value when checking io instruction permission
  KVM: add missing void __user * cast to access_ok() call

13 years agoMAINTAINERS: Saying goodbye to David Brownell
Grant Likely [Mon, 6 Jun 2011 06:40:48 +0000 (00:40 -0600)]
MAINTAINERS: Saying goodbye to David Brownell

We had to say goodbye when David passed away recently.  David had a
huge impact on our community, both personally in the lives of the
people he worked with, and technically in the design and maintenance
of several subsystems.  He is greatly missed.

He also leaves behind a number of much loved subsystems now orphaned.
This patch updates the MAINTAINERS file for the areas that David was
responsible for and adds an entry for him to the CREDITS file.

Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
Acked-by: Harry Wei <harryxiyou@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/hirofumi/fatfs-2.6
Linus Torvalds [Wed, 8 Jun 2011 02:04:21 +0000 (19:04 -0700)]
Merge git://git./linux/kernel/git/hirofumi/fatfs-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/hirofumi/fatfs-2.6:
  fat: Fix corrupt inode flags when remove ATTR_SYS flag

13 years agoMN10300: Add missing _sdata declaration
David Howells [Mon, 6 Jun 2011 14:47:23 +0000 (15:47 +0100)]
MN10300: Add missing _sdata declaration

_sdata needs to be declared in the linker script now as of commit
a2d063ac216c ("extable, core_kernel_data(): Make sure all archs define
_sdata")

Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoMN10300: die_if_no_fixup() shouldn't use get_user() as it doesn't call set_fs()
David Howells [Mon, 6 Jun 2011 14:47:14 +0000 (15:47 +0100)]
MN10300: die_if_no_fixup() shouldn't use get_user() as it doesn't call set_fs()

die_if_no_fixup() shouldn't use get_user() as it doesn't call set_fs() to
indicate that it wants to probe a kernel address.  Instead it should use
probe_kernel_read().

This fixes the problem of gdb seeing SIGILL rather than SIGTRAP when hitting
the KGDB special breakpoint upon SysRq+g being seen.  The problem was that
die_if_no_fixup() was failing to read the opcode of the instruction that caused
the exception, and thus not fixing up the exception.

This caused gdb to get a S04 response to the $? request in its remote protocol
rather than S05 - which would then cause it to continue with $C04 rather than
$c in an attempt to pass the signal onto the inferior process.  The kernel,
however, does not support $Cnn, and so objects by returning an E22 response,
indicating an error.  gdb does not expect this and prints:

warning: Remote failure reply: E22

and then returns to the gdb command prompt unable to continue.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoMN10300: Fix one of the kernel debugger cacheflush variants
David Howells [Mon, 6 Jun 2011 14:47:05 +0000 (15:47 +0100)]
MN10300: Fix one of the kernel debugger cacheflush variants

One of the kernel debugger cacheflush variants escaped proper testing.  Two of
the labels are wrong, being derived from the code that was copied to construct
the variant.

The first label results in the following assembler message:

    AS      arch/mn10300/mm/cache-dbg-flush-by-reg.o
  arch/mn10300/mm/cache-dbg-flush-by-reg.S: Assembler messages:
  arch/mn10300/mm/cache-dbg-flush-by-reg.S:123: Error: symbol `debugger_local_cache_flushinv_no_dcache' is already defined

And the second label results in the following linker message:

  arch/mn10300/mm/built-in.o:(.text+0x1d39): undefined reference to `mn10300_local_icache_inv_range_reg_end'
  arch/mn10300/mm/built-in.o:(.text+0x1d39): relocation truncated to fit: R_MN10300_PCREL16 against undefined symbol `mn10300_local_icache_inv_range_reg_end'

To test this file the following configuration pieces must be set:

CONFIG_AM34=y
CONFIG_MN10300_CACHE_WBACK=y
CONFIG_MN10300_DEBUGGER_CACHE_FLUSH_BY_REG=y
CONFIG_MN10300_CACHE_MANAGE_BY_REG=y
CONFIG_AM34_HAS_CACHE_SNOOP=n

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoMerge branch 'for-linus' of git://git390.marist.edu/pub/scm/linux-2.6
Linus Torvalds [Wed, 8 Jun 2011 01:47:53 +0000 (18:47 -0700)]
Merge branch 'for-linus' of git://git390.marist.edu/linux-2.6

* 'for-linus' of git://git390.marist.edu/pub/scm/linux-2.6:
  [S390] fix kvm defines for 31 bit compile
  [S390] use generic RCU page-table freeing code
  [S390] qdio: Split SBAL entry flags
  [S390] kvm-s390: fix stfle facilities numbers >=64
  [S390] kvm-s390: Fix host crash on misbehaving guests

13 years agoMerge branch 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux...
Linus Torvalds [Wed, 8 Jun 2011 01:46:37 +0000 (18:46 -0700)]
Merge branch 'fixes' of git://git./linux/kernel/git/tmlind/linux-omap-2.6

* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6: (21 commits)
  ARM: OMAP4: MMC: increase delay for pbias
  arm: omap2plus: move NAND_BLOCK_SIZE out of boards
  omap4: hwmod: Enable the keypad
  omap3: Free Beagle rev gpios when they are read, so others can read them later
  arm: omap3: beagle: Ensure msecure is mux'd to be able to set the RTC
  omap: rx51: Don't power up speaker amplifier at bootup
  omap: rx51: Set regulator V28_A always on
  ARM: OMAP4: MMC: no regulator off during probe for eMMC
  arm: omap2plus: fix ads7846 pendown gpio request
  ARM: OMAP2: Add missing iounmap in omap4430_phy_init
  ARM: omap4: Pass core and wakeup mux tables to omap4_mux_init
  ARM: omap2+: mux: Allow board mux settings to be NULL
  OMAP4: fix return value of omap4_l3_init
  OMAP: iovmm: fix SW flags passed by user
  arch/arm/mach-omap1/dma.c: Invert calls to platform_device_put and platform_device_del
  OMAP2+: mux: fix compilation warnings
  OMAP: SRAM: Fix warning: format '%08lx' expects type 'long unsigned int'
  arm: omap3: cm-t3517: fix section mismatch warning
  OMAP2+: Fix 9 section mismatch(es) warnings from mach-omap2/built-in.o
  ARM: OMAP2: Add missing include of linux/gpio.h
  ...

13 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-fixes
Linus Torvalds [Wed, 8 Jun 2011 01:44:10 +0000 (18:44 -0700)]
Merge git://git./linux/kernel/git/steve/gfs2-2.6-fixes

* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-fixes:
  GFS2: Processes waiting on inode glock that no processes are holding

13 years agoMerge branch 'fbdev-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Wed, 8 Jun 2011 01:41:26 +0000 (18:41 -0700)]
Merge branch 'fbdev-fixes-for-linus' of git://git./linux/kernel/git/lethal/fbdev-3.x

* 'fbdev-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/lethal/fbdev-3.x:
  video: Fix use-after-free by vga16fb on rmmod
  video: Convert vmalloc/memset to vzalloc
  efifb: Disallow manual bind and unbind
  efifb: Fix mismatched request/release_mem_region
  efifb: Enable write-combining
  drivers/video/pxa168fb.c: add missing clk_put
  drivers/video/imxfb.c: add missing clk_put
  fbdev: bf537-lq035: add missing blacklight properties type
  savagefb: Use panel CVT mode as default
  fbdev: sh_mobile_lcdcfb: Fix up fallout from MERAM changes.

13 years agoMerge branch 'for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/viro...
Linus Torvalds [Wed, 8 Jun 2011 01:36:59 +0000 (18:36 -0700)]
Merge branch 'for-linus-2' of git://git./linux/kernel/git/viro/vfs-2.6

* 'for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:
  vfs: make unlink() and rmdir() return ENOENT in preference to EROFS
  lmLogOpen() broken failure exit
  usb: remove bad dput after dentry_unhash
  more conservative S_NOSEC handling

13 years agosparc: Remove unnecessary semicolons
Joe Perches [Fri, 3 Jun 2011 14:45:23 +0000 (14:45 +0000)]
sparc: Remove unnecessary semicolons

Semicolons are not necessary after switch/while/for/if braces
so remove them.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoAdd support for allocating irqs for bootbus devices
oftedal [Wed, 1 Jun 2011 11:11:41 +0000 (11:11 +0000)]
Add support for allocating irqs for bootbus devices

Some devices that can generate interrupts are connected directly to the
CPU through the bootbus on sun4d. This patch allows IRQs to be allocated
for such devices. The information used for allocating interrupts for
sbus devices are present at the corresponding SBI node. For bootbus
devices this information is present in the bootbus node.

Signed-off-by: Kjetil Oftedal <oftedal@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoDo not skip interrupt sources in sun4d interrupt handler and acknowledge interrupts...
oftedal [Wed, 1 Jun 2011 11:04:20 +0000 (11:04 +0000)]
Do not skip interrupt sources in sun4d interrupt handler and acknowledge interrupts correctly

During the introduction of genirq on sparc32 bugs were introduced in
the interrupt handler for sun4d. The interrupts handler checks the status
of the various sbus interfaces in the system and generates a virtual
interrupt, based upon the location of the interrupt source. This lookup
was broken by restructuring the code in such a way that index and shift
operations were performed prior to comparing this against the values
read from the interrupt controllers.

This could cause the handler to loop eternally as the interrupt source
could be skipped before any check was performed. Additionally
sun4d_encode_irq performs shifting internally, so it should not be performed
twice.

In sun4d_unmask interrupts were not correctly acknowledged, as the
corresponding bit it the interrupt mask was not actually cleared.

Signed-off-by: Kjetil Oftedal <oftedal@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoRestructure sun4d_build_device_irq so that timer interrupts can be allocated
oftedal [Wed, 1 Jun 2011 10:43:50 +0000 (10:43 +0000)]
Restructure sun4d_build_device_irq so that timer interrupts can be allocated

sun4d_build_device_irq was called without a valid platform_device when
the system timer was initialized on sun4d systems. This caused a NULL
pointer crash.

Josip Rodin suggested that the current sun4d_build_device_irq should be
split into two functions. So that the timer initialization could skip
the slot and sbus interface detection code in sun4d_build_device_irq, as
this does not make sence due to the timer interrupts not being generated
from a device located on sbus.

Signed-off-by: Kjetil Oftedal <oftedal@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agoRevert "USB: option: add ID for ZTE MF 330"
Greg Kroah-Hartman [Tue, 7 Jun 2011 22:03:37 +0000 (15:03 -0700)]
Revert "USB: option: add ID for ZTE MF 330"

This reverts commit a559d2c8c1bf652ea2d0ecd6ab4a250fcdb37db8.

Turns out that device id 0x1d6b:0x0002 is a USB hub, which causes havoc
when the option driver tries to bind to it.

So revert this as it doesn't seem to be needed at all.

Thanks to Michael Tokarev and Paweł Drobek for working on resolving this
issue.

Cc: Paweł Drobek <pawel.drobek@gmail.com>
Cc: Michael Tokarev <mjt@tls.msk.ru>
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoPM / Runtime: Fix loops in pm_runtime_clk_notify()
Rafael J. Wysocki [Tue, 7 Jun 2011 21:34:58 +0000 (23:34 +0200)]
PM / Runtime: Fix loops in pm_runtime_clk_notify()

The loops over connection ID strings in pm_runtime_clk_notify()
should actually iterate over the strings and not over the elements
of the first of them, so make them behave as appropriate.

This fixes a regression introduced by commit 600b776eb39a13a28b090
(OMAP1 / PM: Use generic clock manipulation routines for runtime PM).

Reported-and-tested-by: Janusz Krzysztofik <jkrzyszt@tis.icnet.pl>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
13 years agoigb: fix i350 SR-IOV failture
Williams, Mitch A [Tue, 7 Jun 2011 21:22:57 +0000 (14:22 -0700)]
igb: fix i350 SR-IOV failture

When SR-IOV is enabled, i350 devices fail to pass traffic. This is due to
the driver attempting to enable RSS on the PF device, which is not
supported by the i350.

When max_vfs is specified on an i350 adapter, set the number of RSS queues
to 1.

This issue affects 2.6.39 as well.

CC: stable@kernel.org
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agogpio/samsung: make Kconfig options def_bool
H Hartley Sweeten [Tue, 7 Jun 2011 20:37:27 +0000 (14:37 -0600)]
gpio/samsung: make Kconfig options def_bool

The Samsung GPIO drivers are always built-in when the relevant
platform is selected.  Change the Kconfig symbol to def_bool y
dependant on the platform.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
13 years agolibata: fix unexpectedly frozen port after ata_eh_reset()
Tejun Heo [Wed, 25 May 2011 11:19:39 +0000 (13:19 +0200)]
libata: fix unexpectedly frozen port after ata_eh_reset()

To work around controllers which can't properly plug events while
reset, ata_eh_reset() clears error states and ATA_PFLAG_EH_PENDING
after reset but before RESET is marked done.  As reset is the final
recovery action and full verification of devices including onlineness
and classfication match is done afterwards, this shouldn't lead to
lost devices or missed hotplug events.

Unfortunately, it forgot to thaw the port when clearing EH_PENDING, so
if the condition happens after resetting an empty port, the port could
be left frozen and EH will end without thawing it, making the port
unresponsive to further hotplug events.

Thaw if the port is frozen after clearing EH_PENDING.  This problem is
reported by Bruce Stenning in the following thread.

 http://thread.gmane.org/gmane.linux.kernel/1123265

stable: I think we should weather this patch a bit longer in -rcX
before sending it to -stable.  Please wait at least a month
after this patch makes upstream.  Thanks.

-v2: Fixed spelling in the comment per Dave Howorth.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Bruce Stenning <b.stenning@indigovision.com>
Cc: stable@kernel.org
Cc: Dave Howorth <dhoworth@mrc-lmb.cam.ac.uk>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
13 years agoPM / Intel IOMMU: Fix init_iommu_pm_ops() for CONFIG_PM unset
Rafael J. Wysocki [Tue, 7 Jun 2011 19:32:31 +0000 (21:32 +0200)]
PM / Intel IOMMU: Fix init_iommu_pm_ops() for CONFIG_PM unset

If CONFIG_PM is not set, init_iommu_pm_ops() introduced by commit
134fac3f457f3dd753ecdb25e6da3e5f6629f696 (PCI / Intel IOMMU: Use
syscore_ops instead of sysdev class and sysdev) is not defined
appropriately.  Fix this issue.

Reported-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
13 years agostaging: iio: error case memory leak fix
Andre Bartke [Wed, 1 Jun 2011 22:21:45 +0000 (00:21 +0200)]
staging: iio: error case memory leak fix

The data pointer should be freed in the error
cases of adis16400_trigger_handler().

Signed-off-by: Andre Bartke <andre.bartke@gmail.com>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: ath6kl: Fix a kernel panic during suspend/resume
Vasanthakumar Thiagarajan [Wed, 1 Jun 2011 08:14:11 +0000 (13:44 +0530)]
staging: ath6kl: Fix a kernel panic during suspend/resume

The kernel panic happens when we try to complete a pending
scan request while going to suspend state. The cause for this
kernel panic is accessing a freed memory (ar->arWmin). This
is freed before ar6k_cfg80211_scanComplete_event() getting
called where it is dereferenced.

RIP: 0010:[<ffffffffa042e726>]  [<ffffffffa042e726>] wlan_iterate_nodes+0x16/0xc0 [ath6kl]
RSP: 0018:ffff8800719fbce8  EFLAGS: 00010296
RAX: ffff880071bbcc00 RBX: ffff880037b22520 RCX: ffff880077413c80
RDX: ffff880037b221c0 RSI: ffffffffa041ef10 RDI: 0000000000000020
RBP: ffff8800719fbd18 R08: 0000000000000001 R09: 0000000000000001
R10: 0000000000000400 R11: 0000000000000000 R12: 0000000000000010
R13: ffff8800719fbdd8 R14: 00007fff83a84b60 R15: 0000000000000001
FS:  00007fdccb8a7700(0000) GS:ffff880077400000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 0000000000000148 CR3: 0000000070604000 CR4: 00000000000006f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process rmmod (pid: 1998, threadinfo ffff8800719fa000, task ffff880066712d80)
Stack:
0000000000000000 ffff880037b22520 0000000000000010 ffff8800719fbdd8
00007fff83a84b60 0000000000000001 ffff8800719fbd28 ffffffffa0429fe2
ffff8800719fbd58 ffffffffa041ee5f ffff8800719fbd58 ffff880037b22520
Call Trace:
[<ffffffffa0429fe2>] wmi_iterate_nodes+0x12/0x20 [ath6kl]
[<ffffffffa041ee5f>] ar6k_cfg80211_scanComplete_event+0x3f/0xf0 [ath6kl]
[<ffffffffa04245f1>] ar6000_close+0x61/0x100 [ath6kl]
[<ffffffff814d6736>] __dev_close_many+0x96/0x100
[<ffffffff814d688d>] dev_close_many+0x9d/0x120
[<ffffffff814d6a48>] rollback_registered_many+0xe8/0x290
[<ffffffff814d6d16>] unregister_netdevice_queue+0x96/0x100
[<ffffffff814d6ea0>] unregister_netdev+0x20/0x30
[<ffffffffa0420259>] ar6000_destroy+0x119/0x180 [ath6kl]
[<ffffffffa043182a>] ar6k_cleanup_module+0x2a/0x33 [ath6kl]
[<ffffffff81098fde>] sys_delete_module+0x19e/0x270
[<ffffffff815d7542>] system_call_fastpath+0x16/0x1b
Code: c3 0f 1f 40 00 48 89 df e8 68 ff ff ff eb df 66 0f 1f 44 00 00 55 48 89 e5 41 57 41 56 41 55 41 54 53 48 83 ec 08 0f 1f 44 00 00
8b af 28 01 00 00 4c 8d 7f 08 49 89 fc 48 89 f3 49 89 d6 41
RIP  [<ffffffffa042e726>] wlan_iterate_nodes+0x16/0xc0 [ath6kl]
RSP <ffff8800719fbce8>

Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: gma500: get control from firmware framebuffer if conflicts
Michael Chang [Mon, 30 May 2011 06:28:25 +0000 (14:28 +0800)]
staging: gma500: get control from firmware framebuffer if conflicts

Many Linux distributions would enable vesafb in order to display
early stage boot splash. In this case, we will get garbled X
Window screen if running X fbdev on psbfb.

This is because fb0 is occupied by vesafb while psbfb is on fb1.
They tried to drive the same pieces of hardware at the same
time. With unmodified X start-up, it would try to use default
fb0 framebuffer device and unfortunately it is now broken
becaues fb1 supersedes it.

We should let psbfb takeover framebuffer control from vesafb
to get around this problem.

See also commit : 4410f3910947dcea8672280b3adecd53cec4e85e

Signed-off-by: Michael Chang <mchang@novell.com>
Cc: Alan Cox <alan@linux.intel.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: gma500: Skip bogus LVDS VBT mode and check for LVDS before adding backlight
Patrik Jakobsson [Tue, 24 May 2011 12:04:56 +0000 (13:04 +0100)]
staging: gma500: Skip bogus LVDS VBT mode and check for LVDS before adding backlight

On the Fit-PC2 the VBT reports an invalid fixed panel mode for LVDS, this gets
in the way for SDVO. This patch makes VBT parsing skip the invalid mode. When
there is no LVDS output the backlight support crashes so the patch also checks
for this before enabling it.

Signed-off-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: usbip: bugfix prevent driver unbind
Arjan Mels [Fri, 20 May 2011 21:25:46 +0000 (23:25 +0200)]
staging: usbip: bugfix prevent driver unbind

Implemented pre_reset and post_reset methods of the driver to prevent the
driver from being unbound upon a device reset. Because of this also the
asynchronous reset introduced to prevent a race condition is no longer necessary
(and sometimes causes problems, because it comes later then expected).

Signed-off-by: Arjan Mels <arjan.mels@gmx.net>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Cc: Takahiro Hirofuchi <hirofuchi@users.sourceforge.net>
Cc: Max Vozeler <max@vozeler.com>
Cc: usbip-devel <usbip-devel@lists.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: iio: industrialio-trigger: set iio_poll_func private_data
Michael Hennerich [Wed, 25 May 2011 12:42:07 +0000 (14:42 +0200)]
staging: iio: industrialio-trigger: set iio_poll_func private_data

Failure to set iio_poll_func private_data, causes zero pointer access
violations in all consumer trigger handlers.

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Acked-by: Jonathan Cameron <jic23@cam.ac.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: rts_pstor: use bitwise operator instead of logical one
Nicolas Kaiser [Wed, 25 May 2011 00:03:02 +0000 (02:03 +0200)]
staging: rts_pstor: use bitwise operator instead of logical one

Looks like a typo.

Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
Reviewed-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: fix ath6kl build when CFG80211 is not enabled
Randy Dunlap [Tue, 17 May 2011 22:15:19 +0000 (15:15 -0700)]
staging: fix ath6kl build when CFG80211 is not enabled

Fix build errors when CONFIG_CFG80211 is not enabled:

drivers/built-in.o: In function `ar6k_cfg80211_deinit':
(.text+0x189b71): undefined reference to `cfg80211_scan_done'
drivers/built-in.o: In function `ar6k_cfg80211_deinit':
(.text+0x189b86): undefined reference to `wiphy_unregister'
drivers/built-in.o: In function `ar6k_cfg80211_deinit':
(.text+0x189b8d): undefined reference to `wiphy_free'
drivers/built-in.o: In function `ar6k_cfg80211_init':
(.text+0x18add7): undefined reference to `wiphy_new'
drivers/built-in.o: In function `ar6k_cfg80211_init':
(.text+0x18ae48): undefined reference to `wiphy_register'
drivers/built-in.o: In function `ar6k_cfg80211_tkip_micerr_event':
(.text+0x18ae95): undefined reference to `cfg80211_michael_mic_failure'
drivers/built-in.o: In function `ar6k_cfg80211_scan_node':
(.text+0x18afb5): undefined reference to `__ieee80211_get_channel'
drivers/built-in.o: In function `ar6k_cfg80211_scan_node':
(.text+0x18afd2): undefined reference to `cfg80211_inform_bss_frame'
drivers/built-in.o: In function `ar6k_cfg80211_disconnect_event':
(.text+0x18b046): undefined reference to `cfg80211_ibss_joined'
drivers/built-in.o: In function `ar6k_cfg80211_disconnect_event':
(.text+0x18b176): undefined reference to `cfg80211_connect_result'
drivers/built-in.o: In function `ar6k_cfg80211_disconnect_event':
(.text+0x18b190): undefined reference to `cfg80211_disconnected'
drivers/built-in.o: In function `ar6k_cfg80211_connect_event':
(.text+0x18b291): undefined reference to `cfg80211_get_bss'
drivers/built-in.o: In function `ar6k_cfg80211_connect_event':
(.text+0x18b457): undefined reference to `cfg80211_put_bss'
drivers/built-in.o: In function `ar6k_cfg80211_connect_event':
(.text+0x18b4fa): undefined reference to `cfg80211_roamed'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Luis R. Rodriguez <mcgrof@gmail.com>
Cc: Joe Perches <joe@perches.com>
Cc: Naveen Singh <nsingh@atheros.com>
Signed-off-by: Peter Foley <pefoley2@verizon.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: brcm80211: fix for 'multiple definition of wl_msg_level' build err
Roland Vossen [Tue, 24 May 2011 11:35:21 +0000 (13:35 +0200)]
staging: brcm80211: fix for 'multiple definition of wl_msg_level' build err

Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Roland Vossen <rvossen@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: fix olpc_dcon build, needs BACKLIGHT_CLASS_DEVICE
Randy Dunlap [Thu, 26 May 2011 02:17:06 +0000 (19:17 -0700)]
staging: fix olpc_dcon build, needs BACKLIGHT_CLASS_DEVICE

Fix olpc_dcon.c build by selecting the needed kconfig symbol
BACKLIGHT_CLASS_DEVICE.

olpc_dcon.c:(.text+0x11588b): undefined reference to `backlight_device_register'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Andres Salomon <dilinger@queued.net>
Cc: Chris Ball <cjb@laptop.org>
Cc: Jon Nettleton <jon.nettleton@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoStaging: remove STAGING_EXCLUDE_BUILD option
Greg Kroah-Hartman [Tue, 7 Jun 2011 19:23:57 +0000 (12:23 -0700)]
Staging: remove STAGING_EXCLUDE_BUILD option

Part of the requirement to be in the staging tree is that the code must
build, so let's make it easier for people to build the code to
test/prove this out.

Based on a recommendation from Linus to implement this.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoStaging: altera: move .h file to proper place
Greg Kroah-Hartman [Tue, 7 Jun 2011 19:19:14 +0000 (12:19 -0700)]
Staging: altera: move .h file to proper place

Staging drivers should be self-contained, without files in the include/
directories.  So move the altera.h file back to the driver directory for
now, until it moves out of the staging tree.

Cc: Igor M. Liplianin <liplianin@netup.ru>
Cc: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoftrace: Revert 8ab2b7efd ftrace: Remove unnecessary disabling of irqs
Steven Rostedt [Tue, 7 Jun 2011 13:26:46 +0000 (09:26 -0400)]
ftrace: Revert 8ab2b7efd ftrace: Remove unnecessary disabling of irqs

Revert the commit that removed the disabling of interrupts around
the initial modifying of mcount callers to nops, and update the comment.

The original comment was outdated and stated that the interrupts were
being disabled to prevent kstop machine, which was required with the
old ftrace daemon, but was no longer the case.

What the comment failed to mention was that interrupts needed to be
disabled to keep interrupts from preempting the modifying of the code
and then executing the code that was partially modified.

Revert the commit and update the comment.

Reported-by: Richard W.M. Jones <rjones@redhat.com>
Tested-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
13 years agokprobes/trace: Fix kprobe selftest for gcc 4.6
Steven Rostedt [Tue, 7 Jun 2011 02:35:13 +0000 (22:35 -0400)]
kprobes/trace: Fix kprobe selftest for gcc 4.6

With gcc 4.6, the self test kprobe function:

 kprobe_trace_selftest_target()

is optimized such that kallsyms does not list it. The kprobes
test uses this function to insert a probe and test it. But
it will fail the test if the function is not listed in kallsyms.

Adding a __used annotation keeps the symbol in the kallsyms table.

Suggested-by: David Daney <ddaney@caviumnetworks.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
13 years agonl80211: fix overflow in ssid_len
Luciano Coelho [Tue, 7 Jun 2011 17:42:26 +0000 (20:42 +0300)]
nl80211: fix overflow in ssid_len

When one of the SSID's length passed in a scan or sched_scan request
is larger than 255, there will be an overflow in the u8 that is used
to store the length before checking.  This causes the check to fail
and we overrun the buffer when copying the SSID.

Fix this by checking the nl80211 attribute length before copying it to
the struct.

This is a follow up for the previous commit
208c72f4fe44fe09577e7975ba0e7fa0278f3d03, which didn't fix the problem
entirely.

Reported-by: Ido Yariv <ido@wizery.com>
Signed-off-by: Luciano Coelho <coelho@ti.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agortlwifi: Avoid modifying skbs that are resubmitted
Mike McCormack [Mon, 6 Jun 2011 23:58:31 +0000 (08:58 +0900)]
rtlwifi: Avoid modifying skbs that are resubmitted

In the case we fail to allocate a new skb, the old skb should
be resubmitted unmodified.

Fixes bug introduced in a9e12869758430424804.

Signed-off-by: Mike McCormack <mikem@ring3k.org>
Acked-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agortlwifi: Fix logic in rx_interrupt
Mike McCormack [Mon, 30 May 2011 23:50:24 +0000 (08:50 +0900)]
rtlwifi: Fix logic in rx_interrupt

Should pass along packet if there's no CRC and no hardware error.

Signed-off-by: Mike McCormack <mikem@ring3k.org>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agoRevert "mac80211: stop queues before rate control updation"
John W. Linville [Tue, 7 Jun 2011 18:03:08 +0000 (14:03 -0400)]
Revert "mac80211: stop queues before rate control updation"

This reverts commit 1d38c16ce4156f63b45abbd09dd28ca2ef5172b4.

The mac80211 maintainer raised complaints about abuse of the CSA stop
reason, and about whether this patch actually serves its intended
purpose at all.

Signed-off-by: John W. Linville <linville@tuxdriver.com>
13 years agogpio/exynos4: Fix incorrect mapping of gpio pull-up macro to register setting
Thomas Abraham [Tue, 7 Jun 2011 08:34:49 +0000 (14:04 +0530)]
gpio/exynos4: Fix incorrect mapping of gpio pull-up macro to register setting

The S3C_GPIO_PULL_UP macro value incorrectly maps to a reserved setting of GPIO
pull up/down registers on Exynos4 platform. Fix this incorrect mapping by adding
wrappers to the s3c_gpio_setpull_updown and s3c_gpio_getpull_updown functions.

Signed-off-by: Thomas Abraham <thomas.ab@samsung.com>
Acked-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
13 years agodrivers/usb/host/ohci-pxa27x.c: add missing clk_put
Julia Lawall [Wed, 1 Jun 2011 17:10:06 +0000 (19:10 +0200)]
drivers/usb/host/ohci-pxa27x.c: add missing clk_put

Add a label before the call to clk_put and jump to that in the error
handling code that occurs after the call to clk_get has succeeded.

A simplified version of the semantic match that finds this problem is as
follows: (http://coccinelle.lip6.fr/)

// <smpl>
@r exists@
expression e1,e2;
statement S;
@@

e1 = clk_get@p1(...);
... when != e1 = e2
    when != clk_put(e1)
    when any
if (...) { ... when != clk_put(e1)
               when != if (...) { ... clk_put(e1) ... }
* return@p3 ...;
 } else S
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Acked-by: Eric Miao <eric.y.miao@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: CONFIG_USB_GADGET_DUALSPEED is not user-configurable
Alan Stern [Tue, 7 Jun 2011 15:31:05 +0000 (11:31 -0400)]
USB: CONFIG_USB_GADGET_DUALSPEED is not user-configurable

This patch (as1468) changes the Kconfig definition for
USB_GADGET_DUALSPEED.  This option is determined entirely by which
device controller drivers are to be built, through Select statements;
it does not need to be (and should not be) configurable by the user.

Also, the "default n" line is superfluous -- everything defaults to N.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Acked-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoUSB: dummy-hcd needs the has_tt flag
Alan Stern [Tue, 7 Jun 2011 15:33:01 +0000 (11:33 -0400)]
USB: dummy-hcd needs the has_tt flag

Like with other host controllers capable of operating at both high
speed and full speed, we need to indicate that the emulated controller
presented by dummy-hcd has this ability.  Otherwise usbcore will not
accept full-speed gadgets under dummy-hcd.  This patch (as1469) sets
the appropriate has_tt flag.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agousb-storage: redo incorrect reads
Alan Stern [Tue, 7 Jun 2011 15:35:52 +0000 (11:35 -0400)]
usb-storage: redo incorrect reads

Some USB mass-storage devices have bugs that cause them not to handle
the first READ(10) command they receive correctly.  The Corsair
Padlock v2 returns completely bogus data for its first read (possibly
it returns the data in encrypted form even though the device is
supposed to be unlocked).  The Feiya SD/SDHC card reader fails to
complete the first READ(10) command after it is plugged in or after a
new card is inserted, returning a status code that indicates it thinks
the command was invalid, which prevents the kernel from retrying the
read.

Since the first read of a new device or a new medium is for the
partition sector, the kernel is unable to retrieve the device's
partition table.  Users have to manually issue an "hdparm -z" or
"blockdev --rereadpt" command before they can access the device.

This patch (as1470) works around the problem.  It adds a new quirk
flag, US_FL_INVALID_READ10, indicating that the first READ(10) should
always be retried immediately, as should any failing READ(10) commands
(provided the preceding READ(10) command succeeded, to avoid getting
stuck in a loop).  The patch also adds appropriate unusual_devs
entries containing the new flag.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Tested-by: Sven Geggus <sven-usbst@geggus.net>
Tested-by: Paul Hartman <paul.hartman+linux@gmail.com>
CC: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
CC: <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agovfs: make unlink() and rmdir() return ENOENT in preference to EROFS
Theodore Ts'o [Mon, 6 Jun 2011 23:19:40 +0000 (19:19 -0400)]
vfs: make unlink() and rmdir() return ENOENT in preference to EROFS

If user space attempts to remove a non-existent file or directory, and
the file system is mounted read-only, return ENOENT instead of EROFS.
Either error code is arguably valid/correct, but ENOENT is a more
specific error message.

Reported-by: Michael Tokarev <mjt@tls.msk.ru>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
13 years agolmLogOpen() broken failure exit
Al Viro [Sun, 5 Jun 2011 18:22:56 +0000 (19:22 +0100)]
lmLogOpen() broken failure exit

Callers of lmLogOpen() expect it to return -E... on failure exits, which
is what it returns, except for the case of blkdev_get_by_dev() failure.
It that case lmLogOpen() return the error with the wrong sign...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
13 years agosched: Fix/clarify set_task_cpu() locking rules
Peter Zijlstra [Fri, 3 Jun 2011 15:37:07 +0000 (17:37 +0200)]
sched: Fix/clarify set_task_cpu() locking rules

Sergey reported a CONFIG_PROVE_RCU warning in push_rt_task where
set_task_cpu() was called with both relevant rq->locks held, which
should be sufficient for running tasks since holding its rq->lock
will serialize against sched_move_task().

Update the comments and fix the task_group() lockdep test.

Reported-and-tested-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1307115427.2353.3456.camel@twins
Signed-off-by: Ingo Molnar <mingo@elte.hu>
13 years agolockdep: Fix lock_is_held() on recursion
Peter Zijlstra [Mon, 6 Jun 2011 10:32:43 +0000 (12:32 +0200)]
lockdep: Fix lock_is_held() on recursion

The main lock_is_held() user is lockdep_assert_held(), avoid false
assertions in lockdep_off() sections by unconditionally reporting the
lock is taken.

[ the reason this is important is a lockdep_assert_held() in ttwu()
  which triggers a warning under lockdep_off() as in printk() which
  can trigger another wakeup and lock up due to spinlock
  recursion, as reported and heroically debugged by Arne Jansen ]

Reported-and-tested-by: Arne Jansen <lists@die-jansens.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: <stable@kernel.org>
Link: http://lkml.kernel.org/r/1307398759.2497.966.camel@laptop
Signed-off-by: Ingo Molnar <mingo@elte.hu>
13 years agoASoC: Blackfin: bf5xx-ad1836: Fix codec device name
Lars-Peter Clausen [Tue, 7 Jun 2011 08:24:46 +0000 (10:24 +0200)]
ASoC: Blackfin: bf5xx-ad1836: Fix codec device name

Fix the codec_name field of the dai_link to match the actual device name
of the codec. Otherwise the card won't be instantiated.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Liam Girdwood <lrg@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@kernel.org
13 years agox86/amd-iommu: Fix boot crash with hidden PCI devices
Joerg Roedel [Mon, 6 Jun 2011 14:50:14 +0000 (16:50 +0200)]
x86/amd-iommu: Fix boot crash with hidden PCI devices

Some PCIe cards ship with a PCI-PCIe bridge which is not
visible as a PCI device in Linux. But the device-id of the
bridge is present in the IOMMU tables which causes a boot
crash in the IOMMU driver.
This patch fixes by removing these cards from the IOMMU
handling. This is a pure -stable fix, a real fix to handle
this situation appriatly will follow for the next merge
window.

Cc: stable@kernel.org # > 2.6.32
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
13 years agonet: cpu offline cause napi stall
Heiko Carstens [Mon, 6 Jun 2011 20:50:03 +0000 (20:50 +0000)]
net: cpu offline cause napi stall

Frank Blaschka reported :
<quote>
  During heavy network load we turn off/on cpus.
  Sometimes this causes a stall on the network device.
  Digging into the dump I found out following:

  napi is scheduled but does not run. From the I/O buffers
  and the napi state I see napi/rx_softirq processing has stopped
  because the budget was reached. napi stays in the
  softnet_data poll_list and the rx_softirq was raised again.

  I assume at this time the cpu offline comes in,
  the rx softirq is raised/moved to another cpu but napi stays in the
  poll_list of the softnet_data of the now offline cpu.

  Reviewing dev_cpu_callback (net/core/dev.c) I did not find the
  poll_list is transfered to the new cpu.
</quote>

This patch is a straightforward implementation of Frank suggestion :

Transfert poll_list and trigger NET_RX_SOFTIRQ on new cpu.

Reported-by: Frank Blaschka <blaschka@linux.vnet.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Tested-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 years agobridge: provide a cow_metrics method for fake_ops
Alexander Holler [Tue, 7 Jun 2011 07:51:35 +0000 (00:51 -0700)]
bridge: provide a cow_metrics method for fake_ops

Like in commit 0972ddb237 (provide cow_metrics() methods to blackhole
dst_ops), we must provide a cow_metrics for bridges fake_dst_ops as
well.

This fixes a regression coming from commits 62fa8a846d7d (net: Implement
read-only protection and COW'ing of metrics.) and 33eb9873a28 (bridge:
initialize fake_rtable metrics)

ip link set mybridge mtu 1234
-->
[  136.546243] Pid: 8415, comm: ip Tainted: P
2.6.39.1-00006-g40545b7 #103 ASUSTeK Computer Inc.         V1Sn
        /V1Sn
[  136.546256] EIP: 0060:[<00000000>] EFLAGS: 00010202 CPU: 0
[  136.546268] EIP is at 0x0
[  136.546273] EAX: f14a389c EBX: 000005d4 ECX: f80d32c0 EDX: f80d1da1
[  136.546279] ESI: f14a3000 EDI: f255bf10 EBP: f15c3b54 ESP: f15c3b48
[  136.546285]  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
[  136.546293] Process ip (pid: 8415, ti=f15c2000 task=f4741f80
task.ti=f15c2000)
[  136.546297] Stack:
[  136.546301]  f80c658f f14a3000 ffffffed f15c3b64 c12cb9c8 f80d1b80
ffffffa1 f15c3bbc
[  136.546315]  c12da347 c12d9c7d 00000000 f7670b00 00000000 f80d1b80
ffffffa6 f15c3be4
[  136.546329]  00000004 f14a3000 f255bf20 00000008 f15c3bbc c11d6cae
00000000 00000000
[  136.546343] Call Trace:
[  136.546359]  [<f80c658f>] ? br_change_mtu+0x5f/0x80 [bridge]
[  136.546372]  [<c12cb9c8>] dev_set_mtu+0x38/0x80
[  136.546381]  [<c12da347>] do_setlink+0x1a7/0x860
[  136.546390]  [<c12d9c7d>] ? rtnl_fill_ifinfo+0x9bd/0xc70
[  136.546400]  [<c11d6cae>] ? nla_parse+0x6e/0xb0
[  136.546409]  [<c12db931>] rtnl_newlink+0x361/0x510
[  136.546420]  [<c1023240>] ? vmalloc_sync_all+0x100/0x100
[  136.546429]  [<c1362762>] ? error_code+0x5a/0x60
[  136.546438]  [<c12db5d0>] ? rtnl_configure_link+0x80/0x80
[  136.546446]  [<c12db27a>] rtnetlink_rcv_msg+0xfa/0x210
[  136.546454]  [<c12db180>] ? __rtnl_unlock+0x20/0x20
[  136.546463]  [<c12ee0fe>] netlink_rcv_skb+0x8e/0xb0
[  136.546471]  [<c12daf1c>] rtnetlink_rcv+0x1c/0x30
[  136.546479]  [<c12edafa>] netlink_unicast+0x23a/0x280
[  136.546487]  [<c12ede6b>] netlink_sendmsg+0x26b/0x2f0
[  136.546497]  [<c12bb828>] sock_sendmsg+0xc8/0x100
[  136.546508]  [<c10adf61>] ? __alloc_pages_nodemask+0xe1/0x750
[  136.546517]  [<c11d0602>] ? _copy_from_user+0x42/0x60
[  136.546525]  [<c12c5e4c>] ? verify_iovec+0x4c/0xc0
[  136.546534]  [<c12bd805>] sys_sendmsg+0x1c5/0x200
[  136.546542]  [<c10c2150>] ? __do_fault+0x310/0x410
[  136.546549]  [<c10c2c46>] ? do_wp_page+0x1d6/0x6b0
[  136.546557]  [<c10c47d1>] ? handle_pte_fault+0xe1/0x720
[  136.546565]  [<c12bd1af>] ? sys_getsockname+0x7f/0x90
[  136.546574]  [<c10c4ec1>] ? handle_mm_fault+0xb1/0x180
[  136.546582]  [<c1023240>] ? vmalloc_sync_all+0x100/0x100
[  136.546589]  [<c10233b3>] ? do_page_fault+0x173/0x3d0
[  136.546596]  [<c12bd87b>] ? sys_recvmsg+0x3b/0x60
[  136.546605]  [<c12bdd83>] sys_socketcall+0x293/0x2d0
[  136.546614]  [<c13629d0>] sysenter_do_call+0x12/0x26
[  136.546619] Code:  Bad EIP value.
[  136.546627] EIP: [<00000000>] 0x0 SS:ESP 0068:f15c3b48
[  136.546645] CR2: 0000000000000000
[  136.546652] ---[ end trace 6909b560e78934fa ]---

Signed-off-by: Alexander Holler <holler@ahsoftware.de>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>