10 years agobsg: fix sysfs link remove warning
Stanislaw Gruszka [Wed, 8 Feb 2012 19:02:03 +0000 (20:02 +0100)]
bsg: fix sysfs link remove warning

commit 37b40adf2d1b4a5e51323be73ccf8ddcf3f15dd3 upstream.

We create "bsg" link if q-> is not NULL, so remove it only
when the same condition is true.


WARNING: at fs/sysfs/inode.c:323 sysfs_hash_and_remove+0x2b/0x77()
sysfs: can not remove 'bsg', no directory
Call Trace:
  [<c0429683>] warn_slowpath_common+0x6a/0x7f
  [<c0537a68>] ? sysfs_hash_and_remove+0x2b/0x77
  [<c042970b>] warn_slowpath_fmt+0x2b/0x2f
  [<c0537a68>] sysfs_hash_and_remove+0x2b/0x77
  [<c053969a>] sysfs_remove_link+0x20/0x23
  [<c05d88f1>] bsg_unregister_queue+0x40/0x6d
  [<c0692263>] __scsi_remove_device+0x31/0x9d
  [<c069149f>] scsi_forget_host+0x41/0x52
  [<c0689fa9>] scsi_remove_host+0x71/0xe0
  [<f7de5945>] quiesce_and_remove_host+0x51/0x83 [usb_storage]
  [<f7de5a1e>] usb_stor_disconnect+0x18/0x22 [usb_storage]
  [<c06c29de>] usb_unbind_interface+0x4e/0x109
  [<c067a80f>] __device_release_driver+0x6b/0xa6
  [<c067a861>] device_release_driver+0x17/0x22
  [<c067a46a>] bus_remove_device+0xd6/0xe6
  [<c06785e2>] device_del+0xf2/0x137
  [<c06c101f>] usb_disable_device+0x94/0x1a0

Signed-off-by: Stanislaw Gruszka <>
Signed-off-by: Jens Axboe <>
Signed-off-by: Tim Gardner <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoASoC: i.MX SSI: Fix DSP_A format.
Javier Martin [Thu, 23 Feb 2012 14:43:18 +0000 (15:43 +0100)]
ASoC: i.MX SSI: Fix DSP_A format.

commit 5ed80a75b248bfaf840ea6b38f941edcf6ee7dc7 upstream.

According to i.MX27 Reference Manual (p 1593) TXBIT0 bit selects
whether the most significant or the less significant part of the
data word written to the FIFO is transmitted.

As DSP_A is the same as DSP_B with a data offset of 1 bit, it
doesn't make any sense to remove TXBIT0 bit here.

Signed-off-by: Javier Martin <>
Acked-by: Sascha Hauer <>
Signed-off-by: Mark Brown <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoASoC: dapm: Check for bias level when powering down
Mark Brown [Wed, 22 Feb 2012 15:52:56 +0000 (15:52 +0000)]
ASoC: dapm: Check for bias level when powering down

commit 7679e42ec833ed70aa34790a5f39dcb7e5bda4fe upstream.

Recent enhancements in the bias management means that we might not be
in standby when the CODEC is idle and can have active widgets without
being in full power mode but the shutdown functionality assumes these
things. Add checks for the bias level at each stage so that we don't
do transitions other than the ON->PREPARE->STANDBY->OFF ones that the
drivers are expecting.

Signed-off-by: Mark Brown <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoviafb: fix IGA1 modesetting on VX900
Florian Tobias Schandinat [Wed, 22 Feb 2012 18:53:08 +0000 (18:53 +0000)]
viafb: fix IGA1 modesetting on VX900

commit e29206381a1436e0f47c0f5b9a23159a03c57715 upstream.

Even if the documentation calls this bit "Reserved" it has to be set
to 0 for correct modesetting on IGA1.

Signed-off-by: Florian Tobias Schandinat <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoviafb: select HW scaling on VX900 for IGA2
Florian Tobias Schandinat [Wed, 22 Feb 2012 18:53:07 +0000 (18:53 +0000)]
viafb: select HW scaling on VX900 for IGA2

commit 050f0e02c8dc38b2b4f2df345ac760d22ca5c7ba upstream.

VX900 can do hardware scaling for both IGAs in contrast to previous
hardware which could do it only for IGA2. This patch ensures that
we set the parameter for IGA2 and not for IGA1. This fixes hardware
scaling on VX900 until we have the infrastructure to support it for
both IGAs.

Signed-off-by: Florian Tobias Schandinat <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoosd_uld: Bump MAX_OSD_DEVICES from 64 to 1,048,576
Boaz Harrosh [Wed, 25 Jan 2012 19:42:58 +0000 (21:42 +0200)]
osd_uld: Bump MAX_OSD_DEVICES from 64 to 1,048,576

commit 41f8ad76362e7aefe3a03949c43e23102dae6e0b upstream.

It used to be that minors where 8 bit. But now they
are actually 20 bit. So the fix is simplicity itself.

I've tested with 300 devices and all user-mode utils
work just fine. I have also mechanically added 10,000
to the ida (so devices are /dev/osd10000, /dev/osd10001 ...)
and was able to mkfs an exofs filesystem and access osds
from user-mode.

All the open-osd user-mode code uses the same library
to access devices through their symbolic names in
/dev/osdX so I'd say it's pretty safe. (Well tested)

This patch is very important because some of the systems
that will be deploying the 3.2 pnfs-objects code are larger
than 64 OSDs and will stop to work properly when reaching
that number.

Signed-off-by: Boaz Harrosh <>
Signed-off-by: James Bottomley <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agocrypto: mv_cesa - fix final callback not ignoring input data
Phil Sutter [Mon, 27 Feb 2012 11:17:04 +0000 (12:17 +0100)]
crypto: mv_cesa - fix final callback not ignoring input data

commit f8f54e190ddb4ed697036b60f5e2ae6dd45b801c upstream.

Broken by commit 6ef84509f3d439ed2d43ea40080643efec37f54f for users
passing a request with non-zero 'nbytes' field, like e.g. testmgr.

Signed-off-by: Phil Sutter <>
Signed-off-by: Herbert Xu <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoHID: usbhid: Add NOGET quirk for the AIREN Slim+ keyboard
Alan Stern [Mon, 27 Feb 2012 16:23:45 +0000 (11:23 -0500)]
HID: usbhid: Add NOGET quirk for the AIREN Slim+ keyboard

commit 37891abc8464637964a26ae4b61d307fef831f80 upstream.

This patch (as1531) adds a NOGET quirk for the Slim+ keyboard marketed
by AIREN.  This keyboard seems to have a lot of bugs; NOGET works
around only one of them.

Signed-off-by: Alan Stern <>
Reported-by: okias <>
Signed-off-by: Jiri Kosina <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agorapidio/tsi721: fix queue wrapping bug in inbound doorbell handler
Alexandre Bounine [Mon, 5 Mar 2012 22:59:21 +0000 (14:59 -0800)]
rapidio/tsi721: fix queue wrapping bug in inbound doorbell handler

commit b24823e61bfd93d0e72088e4f5245287582ed289 upstream.

Fix a bug that causes a kernel panic when the number of received doorbells
is larger than number of entries in the inbound doorbell queue (current
default value = 512).

Another possible indication for this bug is large number of spurious
doorbells reported by tsi721 driver after reaching the queue size maximum.

Signed-off-by: Alexandre Bounine <>
Cc: Chul Kim <>
Cc: Matt Porter <>
Signed-off-by: Andrew Morton <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoS390: qdio: fix handler function arguments for zfcp data router
Steffen Maier [Fri, 2 Mar 2012 16:32:58 +0000 (17:32 +0100)]
S390: qdio: fix handler function arguments for zfcp data router

commit 7b3cc67d4445995a025a4b55a7dc687b6829b4ca upstream.

Git commit 25f269f17316549e "[S390] qdio: EQBS retry after CCQ 96"
introduced a regression in regard to the zfcp data router.
Revoke the incorrect simplification of the function call arguments
for the qdio handler to make the zfcp hardware data router working

This is applicable to 3.2+ kernels.

Signed-off-by: Steffen Maier <>
Reviewed-by: Jan Glauber <>
Signed-off-by: Martin Schwidefsky <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agotty/powerpc: early udbg consoles can't be modules
Stephen Rothwell [Sun, 19 Feb 2012 20:22:38 +0000 (07:22 +1100)]
tty/powerpc: early udbg consoles can't be modules

commit f21c6d4a49179f91fd70a41382382f08c780d425 upstream.

Fixes these build errors:

ERROR: ".udbg_printf" [drivers/tty/ehv_bytechan.ko] undefined!
ERROR: ".register_early_udbg_console" [drivers/tty/ehv_bytechan.ko] undefined!
ERROR: "udbg_putc" [drivers/tty/ehv_bytechan.ko] undefined!

Cc: Timur Tabi <>
Signed-off-by: Stephen Rothwell <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoiwlwifi: fix key removal
Johannes Berg [Fri, 17 Feb 2012 17:47:14 +0000 (09:47 -0800)]
iwlwifi: fix key removal

commit 5dcbf480473f6c3f06ad2426b7517038a2a18911 upstream.

When trying to remove a key, we always send key
flags just setting the key type, not including
the multicast flag and the key ID. As a result,
whenever any key was removed, the unicast key 0
would be removed, causing a complete connection
loss after the second rekey (the first doesn't
cause a key removal). Fix the key removal code
to include the key ID and multicast flag, thus
removing the correct key.

Reported-by: Alexander Schnaidt <>
Tested-by: Alexander Schnaidt <>
Signed-off-by: Johannes Berg <>
Signed-off-by: Wey-Yi Guy <>
Signed-off-by: John W. Linville <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agomm: thp: fix BUG on mm->nr_ptes
Andrea Arcangeli [Mon, 5 Mar 2012 22:59:20 +0000 (14:59 -0800)]
mm: thp: fix BUG on mm->nr_ptes

commit 1c641e84719429bbfe62a95ed3545ee7fe24408f upstream.

Dave Jones reports a few Fedora users hitting the BUG_ON(mm->nr_ptes...)
in exit_mmap() recently.

Quoting Hugh's discovery and explanation of the SMP race condition:

  "mm->nr_ptes had unusual locking: down_read mmap_sem plus
   page_table_lock when incrementing, down_write mmap_sem (or mm_users
   0) when decrementing; whereas THP is careful to increment and
   decrement it under page_table_lock.

   Now most of those paths in THP also hold mmap_sem for read or write
   (with appropriate checks on mm_users), but two do not: when
   split_huge_page() is called by hwpoison_user_mappings(), and when
   called by add_to_swap().

   It's conceivable that the latter case is responsible for the
   exit_mmap() BUG_ON mm->nr_ptes that has been reported on Fedora."

The simplest way to fix it without having to alter the locking is to make
split_huge_page() a noop in nr_ptes terms, so by counting the preallocated
pagetables that exists for every mapped hugepage.  It was an arbitrary
choice not to count them and either way is not wrong or right, because
they are not used but they're still allocated.

Reported-by: Dave Jones <>
Reported-by: Hugh Dickins <>
Signed-off-by: Andrea Arcangeli <>
Acked-by: Hugh Dickins <>
Cc: David Rientjes <>
Cc: Josh Boyer <>
Signed-off-by: Andrew Morton <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agokprobes: return proper error code from register_kprobe()
Prashanth Nageshappa [Mon, 5 Mar 2012 22:59:12 +0000 (14:59 -0800)]
kprobes: return proper error code from register_kprobe()

commit f986a499ef6f317d906e6f6f281be966e1237a10 upstream.

register_kprobe() aborts if the address of the new request falls in a
prohibited area (such as ftrace pouch, __kprobes annotated functions,
non-kernel text addresses, jump label text).  We however don't return the
right error on this abort, resulting in a silent failure - incorrect
adding/reporting of kprobes ('perf probe do_fork+18' or 'perf probe
mcount' for instance).

In V2 we are incorporating Masami Hiramatsu's  feedback.

This patch fixes it by returning -EINVAL upon failure.

While we are here, rename the label used for exit to be more appropriate.

Signed-off-by: Ananth N Mavinakayanahalli <>
Signed-off-by: Prashanth K Nageshappa <>
Acked-by: Masami Hiramatsu <>
Cc: Jason Baron <>
Signed-off-by: Andrew Morton <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoath9k_hw: prevent writes to const data on AR9160
Felix Fietkau [Wed, 15 Feb 2012 18:31:20 +0000 (19:31 +0100)]
ath9k_hw: prevent writes to const data on AR9160

commit 9bbb8168ed3d8b946f9c1901a63a675012de88f2 upstream.

Duplicate the data for iniAddac early on, to avoid having to do redundant
memcpy calls later. While we're at it, make AR5416 < v2.2 use the same
codepath. Fixes a reported crash on x86.

Signed-off-by: Felix Fietkau <>
Reported-by: Magnus Määttä <>
Signed-off-by: John W. Linville <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agomac80211: zero initialize count field in ieee80211_tx_rate
Mohammed Shafi Shajakhan [Mon, 20 Feb 2012 04:35:31 +0000 (10:05 +0530)]
mac80211: zero initialize count field in ieee80211_tx_rate

commit 8617b093d0031837a7be9b32bc674580cfb5f6b5 upstream.

rate control algorithms concludes the rate as invalid
with rate[i].idx < -1 , while they do also check for rate[i].count is
non-zero. it would be safer to zero initialize the 'count' field.
recently we had a ath9k rate control crash where the ath9k rate control
in ath_tx_status assumed to check only for rate[i].count being non-zero
in one instance and ended up in using invalid rate index for
'connection monitoring NULL func frames' which eventually lead to the crash.
thanks to Pavel Roskin for fixing it and finding the root cause.

Cc: Pavel Roskin <>
Signed-off-by: Mohammed Shafi Shajakhan <>
Signed-off-by: John W. Linville <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agocifs: fix dentry refcount leak when opening a FIFO on lookup
Jeff Layton [Thu, 23 Feb 2012 14:37:45 +0000 (09:37 -0500)]
cifs: fix dentry refcount leak when opening a FIFO on lookup

commit 5bccda0ebc7c0331b81ac47d39e4b920b198b2cd upstream.

The cifs code will attempt to open files on lookup under certain
circumstances. What happens though if we find that the file we opened
was actually a FIFO or other special file?

Currently, the open filehandle just ends up being leaked leading to
a dentry refcount mismatch and oops on umount. Fix this by having the
code close the filehandle on the server if it turns out not to be a
regular file. While we're at it, change this spaghetti if statement
into a switch too.

Reported-by: CAI Qian <>
Tested-by: CAI Qian <>
Reviewed-by: Shirish Pargaonkar <>
Signed-off-by: Jeff Layton <>
Signed-off-by: Steve French <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoNOMMU: Don't need to clear vm_mm when deleting a VMA
David Howells [Thu, 23 Feb 2012 13:51:00 +0000 (13:51 +0000)]
NOMMU: Don't need to clear vm_mm when deleting a VMA

commit b94cfaf6685d691dc3fab023cf32f65e9b7be09c upstream.

Don't clear vm_mm in a deleted VMA as it's unnecessary and might
conceivably break the filesystem or driver VMA close routine.

Reported-by: Al Viro <>
Signed-off-by: David Howells <>
Acked-by: Al Viro <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agomm: memcg: Correct unregistring of events attached to the same eventfd
Anton Vorontsov [Fri, 24 Feb 2012 01:14:46 +0000 (05:14 +0400)]
mm: memcg: Correct unregistring of events attached to the same eventfd

commit 371528caec553785c37f73fa3926ea0de84f986f upstream.

There is an issue when memcg unregisters events that were attached to
the same eventfd:

- On the first call mem_cgroup_usage_unregister_event() removes all
  events attached to a given eventfd, and if there were no events left,
  thresholds->primary would become NULL;

- Since there were several events registered, cgroups core will call
  mem_cgroup_usage_unregister_event() again, but now kernel will oops,
  as the function doesn't expect that threshold->primary may be NULL.

That's a good question whether mem_cgroup_usage_unregister_event()
should actually remove all events in one go, but nowadays it can't
do any better as cftype->unregister_event callback doesn't pass
any private event-associated cookie. So, let's fix the issue by
simply checking for threshold->primary.

FWIW, w/o the patch the following oops may be observed:

 BUG: unable to handle kernel NULL pointer dereference at 0000000000000004
 IP: [<ffffffff810be32c>] mem_cgroup_usage_unregister_event+0x9c/0x1f0
 Pid: 574, comm: kworker/0:2 Not tainted 3.3.0-rc4+ #9 Bochs Bochs
 RIP: 0010:[<ffffffff810be32c>]  [<ffffffff810be32c>] mem_cgroup_usage_unregister_event+0x9c/0x1f0
 RSP: 0018:ffff88001d0b9d60  EFLAGS: 00010246
 Process kworker/0:2 (pid: 574, threadinfo ffff88001d0b8000, task ffff88001de91cc0)
 Call Trace:
  [<ffffffff8107092b>] cgroup_event_remove+0x2b/0x60
  [<ffffffff8103db94>] process_one_work+0x174/0x450
  [<ffffffff8103e413>] worker_thread+0x123/0x2d0

Signed-off-by: Anton Vorontsov <>
Acked-by: KAMEZAWA Hiroyuki <>
Cc: Kirill A. Shutemov <>
Cc: Michal Hocko <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoaio: wake up waiters when freeing unused kiocbs
Jeff Moyer [Mon, 5 Mar 2012 22:59:12 +0000 (14:59 -0800)]
aio: wake up waiters when freeing unused kiocbs

commit 880641bb9da2473e9ecf6c708d993b29928c1b3c upstream.

Bart Van Assche reported a hung fio process when either hot-removing
storage or when interrupting the fio process itself.  The (pruned) call
trace for the latter looks like so:

  fio             D 0000000000000001     0  6849   6848 0x00000004
   ffff880092541b88 0000000000000046 ffff880000000000 ffff88012fa11dc0
   ffff88012404be70 ffff880092541fd8 ffff880092541fd8 ffff880092541fd8
   ffff880128b894d0 ffff88012404be70 ffff880092541b88 000000018106f24d
  Call Trace:

The problem lies with the allocation batching code.  It will
opportunistically allocate kiocbs, and then trim back the list of iocbs
when there is not enough room in the completion ring to hold all of the

In the case above, what happens is that the pruning back of events ends
up freeing up the last active request and the context is marked as dead,
so it is thus responsible for waking up waiters.  Unfortunately, the
code does not check for this condition, so we end up with a hung task.

Signed-off-by: Jeff Moyer <>
Reported-by: Bart Van Assche <>
Tested-by: Bart Van Assche <>
Signed-off-by: Andrew Morton <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agommc: sdhci-esdhc-imx: fix for mmc cards on i.MX5
Sascha Hauer [Fri, 17 Feb 2012 10:51:49 +0000 (11:51 +0100)]
mmc: sdhci-esdhc-imx: fix for mmc cards on i.MX5

commit 5b6b0ad6e572b32a641116aaa5f897ffebe31e44 upstream.

On i.MX53 we have to write a special SDHCI_CMD_ABORTCMD to the
command. This works for SD cards. However, with MMC cards
the MMC_SET_BLOCK_COUNT command is used instead, but this
needs the same handling. Fix MMC cards by testing for the
MMC_SET_BLOCK_COUNT command aswell. Tested on a custom i.MX53
board with a Transcend MMC+ card and eMMC.

The kernel started used MMC_SET_BLOCK_COUNT in 3.0, so this
is a regression for these boards introduced in 3.0; it should
go to 3.0/3.1/3.2-stable.

Signed-off-by: Sascha Hauer <>
Acked-by: Shawn Guo <>
Signed-off-by: Chris Ball <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agommc: atmel-mci: don't use dma features when using DMA with no chan available
Ludovic Desroches [Thu, 9 Feb 2012 15:33:53 +0000 (16:33 +0100)]
mmc: atmel-mci: don't use dma features when using DMA with no chan available

commit ef8781989a1bcd05aa47e853917c37df44917194 upstream.

Some callbacks are set too early -- i.e. we can have dma capabilities but
we can't get a dma channel. So wait to get the dma channel before setting
callbacks and change logs consequently.

Signed-off-by: Ludovic Desroches <>
Signed-off-by: Nicolas Ferre <>
Signed-off-by: Chris Ball <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoalpha: fix 32/64-bit bug in futex support
Andrew Morton [Mon, 5 Mar 2012 22:59:19 +0000 (14:59 -0800)]
alpha: fix 32/64-bit bug in futex support

commit 62aca403657fe30e5235c5331e9871e676d9ea0a upstream.

Michael Cree said:

: : I have noticed some user space problems (pulseaudio crashes in pthread
: : code, glibc/nptl test suite failures, java compiler freezes on SMP alpha
: : systems) that arise when using a 2.6.39 or later kernel on Alpha.
: : Bisecting between 2.6.38 and 2.6.39 (using glibc/nptl test suite as
: : criterion for good/bad kernel) eventually leads to:
: :
: : 8d7718aa082aaf30a0b4989e1f04858952f941bc is the first bad commit
: : commit 8d7718aa082aaf30a0b4989e1f04858952f941bc
: : Author: Michel Lespinasse <>
: : Date:   Thu Mar 10 18:50:58 2011 -0800
: :
: :     futex: Sanitize futex ops argument types
: :
: :     Change futex_atomic_op_inuser and futex_atomic_cmpxchg_inatomic
: :     prototypes to use u32 types for the futex as this is the data type the
: :     futex core code uses all over the place.
: :
: : Looking at the commit I see there is a change of the uaddr argument in
: : the Alpha architecture specific code for futexes from int to u32, but I
: : don't see why this should cause a problem.

Richard Henderson said:

: futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
:                               u32 oldval, u32 newval)
: ...
:         :       "r"(uaddr), "r"((long)oldval), "r"(newval)
: There is no 32-bit compare instruction.  These are implemented by
: consistently extending the values to a 64-bit type.  Since the
: load instruction sign-extends, we want to sign-extend the other
: quantity as well (despite the fact it's logically unsigned).
: So:
: -        :       "r"(uaddr), "r"((long)oldval), "r"(newval)
: +        :       "r"(uaddr), "r"((long)(int)oldval), "r"(newval)
: should do the trick.

Michael said:

: This fixes the glibc test suite failures and the pulseaudio related
: crashes, but it does not fix the java compiiler lockups that I was (and
: are still) observing.  That is some other problem.

Reported-by: Michael Cree <>
Tested-by: Michael Cree <>
Acked-by: Phil Carmody <>
Cc: Richard Henderson <>
Cc: Michel Lespinasse <>
Cc: Ivan Kokshaysky <>
Reviewed-by: Matt Turner <>
Signed-off-by: Andrew Morton <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoMove Logitech Harmony 900 from cdc_ether to zaurus
Scott Talbert [Tue, 21 Feb 2012 13:06:00 +0000 (13:06 +0000)]
Move Logitech Harmony 900 from cdc_ether to zaurus

commit ee932bf9acb2e2c6a309e808000f24856330e3f9 upstream.

In the current kernel implementation, the Logitech Harmony 900 remote
control is matched to the cdc_ether driver through the generic
USB_CDC_SUBCLASS_MDLM entry.  However, this device appears to be of the
pseudo-MDLM (Belcarra) type, rather than the standard one.  This patch
blacklists the Harmony 900 from the cdc_ether driver and whitelists it for
the pseudo-MDLM driver in zaurus.

Signed-off-by: Scott Talbert <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoARM: S3C24XX: DMA resume regression fix
Gusakov Andrey [Fri, 2 Mar 2012 22:32:36 +0000 (07:32 +0900)]
ARM: S3C24XX: DMA resume regression fix

commit e39d40c65dfd8390b50c03482ae9e289b8a8f351 upstream.

s3c2410_dma_suspend suspends channels from 0 to dma_channels.
s3c2410_dma_resume resumes channels in reverse order. So
pointer should be decremented instead of being incremented.

Signed-off-by: Gusakov Andrey <>
Reviewed-by: Heiko Stuebner <>
Signed-off-by: Kukjin Kim <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agogenirq: Clear action->thread_mask if IRQ_ONESHOT is not set
Thomas Gleixner [Tue, 6 Mar 2012 22:18:54 +0000 (23:18 +0100)]
genirq: Clear action->thread_mask if IRQ_ONESHOT is not set

commit 52abb700e16a9aa4cbc03f3d7f80206cbbc80680 upstream.

Xommit ac5637611(genirq: Unmask oneshot irqs when thread was not woken)
fails to unmask when a !IRQ_ONESHOT threaded handler is handled by

This happens because thread_mask is or'ed unconditionally in
irq_wake_thread(), but for !IRQ_ONESHOT interrupts never cleared.  So
the check for !desc->thread_active fails and keeps the interrupt

Keep the thread_mask zero for !IRQ_ONESHOT interrupts.

Document the thread_mask magic while at it.

Reported-and-tested-by: Sven Joachim <>
Reported-and-tested-by: Stefan Lippers-Hollmann <>
Signed-off-by: Thomas Gleixner <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agomfd: Test for jack detection when deciding if wm8994 should suspend
Mark Brown [Mon, 20 Feb 2012 21:32:32 +0000 (21:32 +0000)]
mfd: Test for jack detection when deciding if wm8994 should suspend

commit e7c248a049c2aac21bded0b0722caee6f0e57256 upstream.

The jack detection on WM1811 is often required during system suspend, add
it as another check when deciding if we should suspend.

Signed-off-by: Mark Brown <>
Signed-off-by: Samuel Ortiz <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agomfd: Fix ACPI conflict check
Jean Delvare [Sat, 18 Feb 2012 16:54:23 +0000 (17:54 +0100)]
mfd: Fix ACPI conflict check

commit 81b5482c32769abb6dfb979560dab2f952ba86fa upstream.

The code is currently always checking the first resource of every
device only (several times.) This has been broken since the ACPI check
was added in February 2010 in commit

Fix the check to run on each resource individually, once.

Signed-off-by: Jean Delvare <>
Signed-off-by: Samuel Ortiz <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoregset: Return -EFAULT, not -EIO, on host-side memory fault
H. Peter Anvin [Fri, 2 Mar 2012 18:43:49 +0000 (10:43 -0800)]
regset: Return -EFAULT, not -EIO, on host-side memory fault

commit 5189fa19a4b2b4c3bec37c3a019d446148827717 upstream.

There is only one error code to return for a bad user-space buffer
pointer passed to a system call in the same address space as the
system call is executed, and that is EFAULT.  Furthermore, the
low-level access routines, which catch most of the faults, return
EFAULT already.

Signed-off-by: H. Peter Anvin <>
Reviewed-by: Oleg Nesterov <>
Acked-by: Roland McGrath <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoregset: Prevent null pointer reference on readonly regsets
H. Peter Anvin [Fri, 2 Mar 2012 18:43:48 +0000 (10:43 -0800)]
regset: Prevent null pointer reference on readonly regsets

commit c8e252586f8d5de906385d8cf6385fee289a825e upstream.

The regset common infrastructure assumed that regsets would always
have .get and .set methods, but not necessarily .active methods.
Unfortunately people have since written regsets without .set methods.

Rather than putting in stub functions everywhere, handle regsets with
null .get or .set methods explicitly.

Signed-off-by: H. Peter Anvin <>
Reviewed-by: Oleg Nesterov <>
Acked-by: Roland McGrath <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoALSA: hda - Always set HP pin in unsol handler for STAC/IDT codecs
Takashi Iwai [Wed, 29 Feb 2012 08:41:17 +0000 (09:41 +0100)]
ALSA: hda - Always set HP pin in unsol handler for STAC/IDT codecs

commit 7bff172a352a2fbe9856bba517d71a2072aab041 upstream.

A bug report with an old Sony laptop showed that we can't rely on BIOS
setting the pins of headphones but the driver should set always by

Signed-off-by: Takashi Iwai <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoALSA: hda - Add a fake mute feature
Takashi Iwai [Mon, 27 Feb 2012 14:00:58 +0000 (15:00 +0100)]
ALSA: hda - Add a fake mute feature

commit 3868137ea41866773e75d9ac4b9988dcc361ff1d upstream.

Some codecs don't supply the mute amp-capabilities although the lowest
volume gives the mute.  It'd be handy if the parser provides the mute
mixers in such a case.

This patch adds an extension amp-cap bit (which is used only in the
driver) to represent the min volume = mute state.  Also modified the
amp cache code to support the fake mute feature when this bit is set
but the real mute bit is unset.

In addition, conexant cx5051 parser uses this new feature to implement
the missing mute controls.


Signed-off-by: Takashi Iwai <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoALSA: hda/realtek - Fix resume of multiple input sources
Takashi Iwai [Sat, 25 Feb 2012 10:13:16 +0000 (11:13 +0100)]
ALSA: hda/realtek - Fix resume of multiple input sources

commit 068b939431486f524438330b0848a8222e33d421 upstream.

When there are multiple input sources, the driver wrongly overwrites with
the value of the last input source on other slots at resume.  Thus the
primary input source may be shown wrongly.

Reported-and-tested-by: Julian Sikorski <>
Signed-off-by: Takashi Iwai <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoperf/x86/kvm: Fix Host-Only/Guest-Only counting with SVM disabled
Joerg Roedel [Wed, 29 Feb 2012 13:57:32 +0000 (14:57 +0100)]
perf/x86/kvm: Fix Host-Only/Guest-Only counting with SVM disabled

commit 1018faa6cf23b256bf25919ef203cd7c129f06f2 upstream.

It turned out that a performance counter on AMD does not
count at all when the GO or HO bit is set in the control
register and SVM is disabled in EFER.

This patch works around this issue by masking out the HO bit
in the performance counter control register when SVM is not

The GO bit is not touched because it is only set when the
user wants to count in guest-mode only. So when SVM is
disabled the counter should not run at all and the
not-counting is the intended behaviour.

Signed-off-by: Joerg Roedel <>
Signed-off-by: Peter Zijlstra <>
Cc: Avi Kivity <>
Cc: Stephane Eranian <>
Cc: David Ahern <>
Cc: Gleb Natapov <>
Cc: Robert Richter <>
Signed-off-by: Ingo Molnar <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoS390: KEYS: Enable the compat keyctl wrapper on s390x
David Howells [Fri, 24 Feb 2012 17:01:27 +0000 (18:01 +0100)]
S390: KEYS: Enable the compat keyctl wrapper on s390x

commit 1d057720609ed052a6371fe1d53300e5e6328e94 upstream.

Enable the compat keyctl wrapper on s390x so that 32-bit s390 userspace can
call the keyctl() syscall.

There's an s390x assembly wrapper that truncates all the register values to
32-bits and this then calls compat_sys_keyctl() - but the latter only exists if
CONFIG_KEYS_COMPAT is enabled, and the s390 Kconfig doesn't enable it.

Without this patch, 32-bit calls to the keyctl() syscall are given an ENOSYS

[root@devel4 ~]# keyctl show
Session Keyring
-3: key inaccessible (Function not implemented)

Signed-off-by: David Howells <>
Cc: Carsten Otte <>
Reviewed-by: Christian Borntraeger <>
Signed-off-by: Heiko Carstens <>
Signed-off-by: Martin Schwidefsky <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoregulator: fix the ldo configure according to 88pm860x spec
Jett.Zhou [Thu, 23 Feb 2012 11:52:08 +0000 (19:52 +0800)]
regulator: fix the ldo configure according to 88pm860x spec

commit 3380643b0eaa7ecf99c4f095bdfcb6e5df471616 upstream.

Signed-off-by: Jett.Zhou <>
Signed-off-by: Mark Brown <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoi2c: mxs: only flag completion when queue is completely done
Wolfram Sang [Fri, 13 Jan 2012 11:14:26 +0000 (12:14 +0100)]
i2c: mxs: only flag completion when queue is completely done

commit 844990daa2e69a4258049ba9c2bae1180657dac3 upstream.

The hardware generates an interrupt for every completed command in the
queue while the code assumed that it will only generate one interrupt
when the queue is empty. So, explicitly check if the queue is really
empty. This patch fixed problems which occurred due to high traffic on
the bus. While we are here, move the completion-initialization after the
parameter error checking.

Signed-off-by: Wolfram Sang <>
Cc: Shawn Guo <>
Cc: Marek Vasut <>
Cc: Lothar Waßmann <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agowatchdog: hpwdt: clean up set_memory_x call for 32 bit
Maxim Uvarov [Mon, 16 Jan 2012 04:02:50 +0000 (20:02 -0800)]
watchdog: hpwdt: clean up set_memory_x call for 32 bit

commit 97d2a10d5804d585ab0b58efbd710948401b886a upstream.

1. address has to be page aligned.
2. set_memory_x uses page size argument, not size.
Bug causes with following commit:
commit da28179b4e90dda56912ee825c7eaa62fc103797
Author: Mingarelli, Thomas <>
Date:   Mon Nov 7 10:59:00 2011 +0100

     watchdog: hpwdt: Changes to handle NX secure bit in 32bit path

    commit e67d668e147c3b4fec638c9e0ace04319f5ceccd upstream.

    This patch makes use of the set_memory_x() kernel API in order
    to make necessary BIOS calls to source NMIs.

Signed-off-by: Maxim Uvarov <>
Signed-off-by: Wim Van Sebroeck <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoARM: LPC32xx: Fix irq on GPI_28
Roland Stigge [Mon, 27 Feb 2012 16:28:02 +0000 (17:28 +0100)]
ARM: LPC32xx: Fix irq on GPI_28

commit f6737055c1c432a9628a9a731f9881ad8e0a9eee upstream.

The GPI_28 IRQ was not registered properly. The registration of
IRQ_LPC32XX_GPI_28 was added and the (wrong) IRQ_LPC32XX_GPI_11 at
LPC32XX_SIC1_IRQ(4) was replaced by IRQ_LPC32XX_GPI_28 (see manual of
LPC32xx / interrupt controller).

Signed-off-by: Roland Stigge <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoARM: LPC32xx: Fix interrupt controller init
Roland Stigge [Mon, 27 Feb 2012 16:28:02 +0000 (17:28 +0100)]
ARM: LPC32xx: Fix interrupt controller init

commit 35dd0a75d4a382e7f769dd0277732e7aa5235718 upstream.

This patch fixes the initialization of the interrupt controller of the LPC32xx
by correctly setting up SIC1 and SIC2 instead of (wrongly) using the same value
as for the Main Interrupt Controller (MIC).

Signed-off-by: Roland Stigge <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoARM: LPC32xx: irq.c: Clear latched event
Roland Stigge [Mon, 27 Feb 2012 16:28:02 +0000 (17:28 +0100)]
ARM: LPC32xx: irq.c: Clear latched event

commit 94ed7830cba4dce57b18a2926b5d826bfd184bd6 upstream.

This patch fixes the wakeup disable function by clearing latched events.

Signed-off-by: Roland Stigge <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoARM: LPC32xx: serial.c: Fixed loop limit
Roland Stigge [Mon, 27 Feb 2012 16:28:03 +0000 (17:28 +0100)]
ARM: LPC32xx: serial.c: Fixed loop limit

commit ff424aa4c89d19082e8ae5a3351006bc8a4cd91b upstream.

This patch fixes a wrong loop limit on UART init.

Signed-off-by: Roland Stigge <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoARM: LPC32xx: serial.c: HW bug workaround
Roland Stigge [Mon, 27 Feb 2012 16:28:02 +0000 (17:28 +0100)]
ARM: LPC32xx: serial.c: HW bug workaround

commit 2707208ee8a80dbbd5426f5aa1a934f766825bb5 upstream.

This patch fixes a HW bug by flushing RX FIFOs of the UARTs on init. It was
ported from NXP's tree.

Signed-off-by: Roland Stigge <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agodrm/i915: Prevent a machine hang by checking crtc->active before loading lut
Alban Browaeys [Fri, 24 Feb 2012 17:12:45 +0000 (17:12 +0000)]
drm/i915: Prevent a machine hang by checking crtc->active before loading lut

commit aed3f09db39596e539f90b11a5016aea4d8442e1 upstream.

Before loading the lut (gamma), check the active state of intel_crtc,
otherwise at least on gen2 hang ensue.

This is reproducible in Xorg via:
  xset dpms force off
  xgamma -rgamma 2.0 # freeze.

Signed-off-by: Alban Browaeys <>
Signed-off-by: Chris Wilson <>
Reviewed-by: Jesse Barnes <>
Signed-off-by: Jesse Barnes <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agocompat: fix compile breakage on s390
Heiko Carstens [Mon, 27 Feb 2012 09:01:52 +0000 (10:01 +0100)]
compat: fix compile breakage on s390

commit 048cd4e51d24ebf7f3552226d03c769d6ad91658 upstream.

The new is_compat_task() define for the !COMPAT case in
include/linux/compat.h conflicts with a similar define in

This is the minimal patch which fixes the build issues.

Signed-off-by: Heiko Carstens <>
Signed-off-by: Linus Torvalds <>
Cc: Jonathan Nieder <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoFix autofs compile without CONFIG_COMPAT
Linus Torvalds [Sun, 26 Feb 2012 17:44:55 +0000 (09:44 -0800)]
Fix autofs compile without CONFIG_COMPAT

commit 3c761ea05a8900a907f32b628611873f6bef24b2 upstream.

The autofs compat handling fix caused a compile failure when
CONFIG_COMPAT isn't defined.

Instead of adding random #ifdef'fery in autofs, let's just make the
compat helpers earlier to use: without CONFIG_COMPAT, is_compat_task()
just hardcodes to zero.

We could probably do something similar for a number of other cases where
we have #ifdef's in code, but this is the low-hanging fruit.

Reported-and-tested-by: Andreas Schwab <>
Signed-off-by: Linus Torvalds <>
Cc: Jonathan Nieder <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoautofs: work around unhappy compat problem on x86-64
Ian Kent [Wed, 22 Feb 2012 12:45:44 +0000 (20:45 +0800)]
autofs: work around unhappy compat problem on x86-64

commit a32744d4abae24572eff7269bc17895c41bd0085 upstream.

When the autofs protocol version 5 packet type was added in commit
5c0a32fc2cd0 ("autofs4: add new packet type for v5 communications"), it
obvously tried quite hard to be word-size agnostic, and uses explicitly
sized fields that are all correctly aligned.

However, with the final "char name[NAME_MAX+1]" array at the end, the
actual size of the structure ends up being not very well defined:
because the struct isn't marked 'packed', doing a "sizeof()" on it will
align the size of the struct up to the biggest alignment of the members
it has.

And despite all the members being the same, the alignment of them is
different: a "__u64" has 4-byte alignment on x86-32, but native 8-byte
alignment on x86-64.  And while 'NAME_MAX+1' ends up being a nice round
number (256), the name[] array starts out a 4-byte aligned.

End result: the "packed" size of the structure is 300 bytes: 4-byte, but
not 8-byte aligned.

As a result, despite all the fields being in the same place on all
architectures, sizeof() will round up that size to 304 bytes on
architectures that have 8-byte alignment for u64.

Note that this is *not* a problem for 32-bit compat mode on POWER, since
there __u64 is 8-byte aligned even in 32-bit mode.  But on x86, 32-bit
and 64-bit alignment is different for 64-bit entities, and as a result
the structure that has exactly the same layout has different sizes.

So on x86-64, but no other architecture, we will just subtract 4 from
the size of the structure when running in a compat task.  That way we
will write the properly sized packet that user mode expects.

Not pretty.  Sadly, this very subtle, and unnecessary, size difference
has been encoded in user space that wants to read packets of *exactly*
the right size, and will refuse to touch anything else.

Reported-and-tested-by: Thomas Meyer <>
Signed-off-by: Ian Kent <>
Signed-off-by: Linus Torvalds <>
Cc: Jonathan Nieder <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoARM: OMAP: make iommu subsys_initcall to fix builtin omap3isp
Ohad Ben-Cohen [Sun, 26 Feb 2012 10:14:14 +0000 (12:14 +0200)]
ARM: OMAP: make iommu subsys_initcall to fix builtin omap3isp

commit 435792d93410f008120c4dbab148019a3cc31dbc upstream.

omap3isp depends on omap's iommu and will fail to probe if
initialized before it (which always happen if they are builtin).

Make omap's iommu subsys_initcall as an interim solution until
the probe deferral mechanism is merged.

Reported-by: James <>
Debugged-by: Laurent Pinchart <>
Signed-off-by: Ohad Ben-Cohen <>
Cc: Tony Lindgren <>
Cc: Hiroshi Doyu <>
Cc: Joerg Roedel <>
Signed-off-by: Joerg Roedel <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoLinux 3.2.9 v3.2.9
Greg Kroah-Hartman [Thu, 1 Mar 2012 00:32:49 +0000 (16:32 -0800)]
Linux 3.2.9

10 years agocdrom: use copy_to_user() without the underscores
Dan Carpenter [Mon, 6 Feb 2012 09:20:45 +0000 (10:20 +0100)]
cdrom: use copy_to_user() without the underscores

commit 822bfa51ce44f2c63c300fdb76dc99c4d5a5ca9f upstream.

"nframes" comes from the user and "nframes * CD_FRAMESIZE_RAW" can wrap
on 32 bit systems.  That would have been ok if we used the same wrapped
value for the copy, but we use a shifted value.  We should just use the
checked version of copy_to_user() because it's not going to make a
difference to the speed.

Signed-off-by: Dan Carpenter <>
Signed-off-by: Jens Axboe <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoepoll: limit paths
Jason Baron [Fri, 13 Jan 2012 01:17:43 +0000 (17:17 -0800)]
epoll: limit paths

commit 28d82dc1c4edbc352129f97f4ca22624d1fe61de upstream.

The current epoll code can be tickled to run basically indefinitely in
both loop detection path check (on ep_insert()), and in the wakeup paths.
The programs that tickle this behavior set up deeply linked networks of
epoll file descriptors that cause the epoll algorithms to traverse them
indefinitely.  A couple of these sample programs have been previously
posted in this thread:

To fix the loop detection path check algorithms, I simply keep track of
the epoll nodes that have been already visited.  Thus, the loop detection
becomes proportional to the number of epoll file descriptor and links.
This dramatically decreases the run-time of the loop check algorithm.  In
one diabolical case I tried it reduced the run-time from 15 mintues (all
in kernel time) to .3 seconds.

Fixing the wakeup paths could be done at wakeup time in a similar manner
by keeping track of nodes that have already been visited, but the
complexity is harder, since there can be multiple wakeups on different
cpus...Thus, I've opted to limit the number of possible wakeup paths when
the paths are created.

This is accomplished, by noting that the end file descriptor points that
are found during the loop detection pass (from the newly added link), are
actually the sources for wakeup events.  I keep a list of these file
descriptors and limit the number and length of these paths that emanate
from these 'source file descriptors'.  In the current implemetation I
allow 1000 paths of length 1, 500 of length 2, 100 of length 3, 50 of
length 4 and 10 of length 5.  Note that it is sufficient to check the
'source file descriptors' reachable from the newly added link, since no
other 'source file descriptors' will have newly added links.  This allows
us to check only the wakeup paths that may have gotten too long, and not
re-check all possible wakeup paths on the system.

In terms of the path limit selection, I think its first worth noting that
the most common case for epoll, is probably the model where you have 1
epoll file descriptor that is monitoring n number of 'source file
descriptors'.  In this case, each 'source file descriptor' has a 1 path of
length 1.  Thus, I believe that the limits I'm proposing are quite
reasonable and in fact may be too generous.  Thus, I'm hoping that the
proposed limits will not prevent any workloads that currently work to

In terms of locking, I have extended the use of the 'epmutex' to all
epoll_ctl add and remove operations.  Currently its only used in a subset
of the add paths.  I need to hold the epmutex, so that we can correctly
traverse a coherent graph, to check the number of paths.  I believe that
this additional locking is probably ok, since its in the setup/teardown
paths, and doesn't affect the running paths, but it certainly is going to
add some extra overhead.  Also, worth noting is that the epmuex was
recently added to the ep_ctl add operations in the initial path loop
detection code using the argument that it was not on a critical path.

Another thing to note here, is the length of epoll chains that is allowed.
Currently, eventpoll.c defines:

/* Maximum number of nesting allowed inside epoll sets */
#define EP_MAX_NESTS 4

This basically means that I am limited to a graph depth of 5 (EP_MAX_NESTS
+ 1).  However, this limit is currently only enforced during the loop
check detection code, and only when the epoll file descriptors are added
in a certain order.  Thus, this limit is currently easily bypassed.  The
newly added check for wakeup paths, stricly limits the wakeup paths to a
length of 5, regardless of the order in which ep's are linked together.
Thus, a side-effect of the new code is a more consistent enforcement of
the graph depth.

Thus far, I've tested this, using the sample programs previously
mentioned, which now either return quickly or return -EINVAL.  I've also
testing using the piptest.c epoll tester, which showed no difference in
performance.  I've also created a number of different epoll networks and
tested that they behave as expectded.

I believe this solves the original diabolical test cases, while still
preserving the sane epoll nesting.

Signed-off-by: Jason Baron <>
Cc: Nelson Elhage <>
Cc: Davide Libenzi <>
Signed-off-by: Andrew Morton <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoepoll: ep_unregister_pollwait() can use the freed pwq->whead
Oleg Nesterov [Fri, 24 Feb 2012 19:07:29 +0000 (20:07 +0100)]
epoll: ep_unregister_pollwait() can use the freed pwq->whead

commit 971316f0503a5c50633d07b83b6db2f15a3a5b00 upstream.

signalfd_cleanup() ensures that ->signalfd_wqh is not used, but
this is not enough. eppoll_entry->whead still points to the memory
we are going to free, ep_unregister_pollwait()->remove_wait_queue()
is obviously unsafe.

Change ep_poll_callback(POLLFREE) to set eppoll_entry->whead = NULL,
change ep_unregister_pollwait() to check pwq->whead != NULL under
rcu_read_lock() before remove_wait_queue(). We add the new helper,
ep_remove_wait_queue(), for this.

This works because sighand_cachep is SLAB_DESTROY_BY_RCU and because
->signalfd_wqh is initialized in sighand_ctor(), not in copy_sighand.
ep_unregister_pollwait()->remove_wait_queue() can play with already
freed and potentially reused ->sighand, but this is fine. This memory
must have the valid ->signalfd_wqh until rcu_read_unlock().

Reported-by: Maxime Bizon <>
Signed-off-by: Oleg Nesterov <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoepoll: introduce POLLFREE to flush ->signalfd_wqh before kfree()
Oleg Nesterov [Fri, 24 Feb 2012 19:07:11 +0000 (20:07 +0100)]
epoll: introduce POLLFREE to flush ->signalfd_wqh before kfree()

commit d80e731ecab420ddcb79ee9d0ac427acbc187b4b upstream.

This patch is intentionally incomplete to simplify the review.
It ignores ep_unregister_pollwait() which plays with the same wqh.
See the next change.

epoll assumes that the EPOLL_CTL_ADD'ed file controls everything
f_op->poll() needs. In particular it assumes that the wait queue
can't go away until eventpoll_release(). This is not true in case
of signalfd, the task which does EPOLL_CTL_ADD uses its ->sighand
which is not connected to the file.

This patch adds the special event, POLLFREE, currently only for
epoll. It expects that init_poll_funcptr()'ed hook should do the
necessary cleanup. Perhaps it should be defined as EPOLLFREE in

__cleanup_sighand() is changed to do wake_up_poll(POLLFREE) if
->signalfd_wqh is not empty, we add the new signalfd_cleanup()

ep_poll_callback(POLLFREE) simply does list_del_init(task_list).
This make this poll entry inconsistent, but we don't care. If you
share epoll fd which contains our sigfd with another process you
should blame yourself. signalfd is "really special". I simply do
not know how we can define the "right" semantics if it used with

The main problem is, epoll calls signalfd_poll() once to establish
the connection with the wait queue, after that signalfd_poll(NULL)
returns the different/inconsistent results depending on who does
EPOLL_CTL_MOD/signalfd_read/etc. IOW: apart from sigmask, signalfd
has nothing to do with the file, it works with the current thread.

In short: this patch is the hack which tries to fix the symptoms.
It also assumes that nobody can take tasklist_lock under epoll
locks, this seems to be true.


- we do not have wake_up_all_poll() but wake_up_poll()
  is fine, poll/epoll doesn't use WQ_FLAG_EXCLUSIVE.

- signalfd_cleanup() uses POLLHUP along with POLLFREE,
  we need a couple of simple changes in eventpoll.c to
  make sure it can't be "lost".

Reported-by: Maxime Bizon <>
Signed-off-by: Oleg Nesterov <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agohwmon: (f75375s) Fix register write order when setting fans to full speed
Nikolaus Schulz [Wed, 22 Feb 2012 22:18:44 +0000 (23:18 +0100)]
hwmon: (f75375s) Fix register write order when setting fans to full speed

commit c1c1a3d012fe5e82a9a025fb4b5a4f8ee67a53f6 upstream.

By hwmon sysfs interface convention, setting pwm_enable to zero sets a fan
to full speed.  In the f75375s driver, this need be done by enabling
manual fan control, plus duty mode for the F875387 chip, and then setting
the maximum duty cycle.  Fix a bug where the two necessary register writes
were swapped, effectively discarding the setting to full-speed.

Signed-off-by: Nikolaus Schulz <>
Cc: Riku Voipio <>
Signed-off-by: Guenter Roeck <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoimon: don't wedge hardware after early callbacks
Jarod Wilson [Thu, 26 Jan 2012 15:04:11 +0000 (12:04 -0300)]
imon: don't wedge hardware after early callbacks

commit 8791d63af0cf113725ae4cb8cba9492814c59a93 upstream.

This patch is just a minor update to one titled "imon: Input from ffdc
device type ignored" from Corinna Vinschen. An earlier patch to prevent
an oops when we got early callbacks also has the nasty side-effect of
wedging imon hardware, as we don't acknowledge the urb. Rework the check
slightly here to bypass processing the packet, as the driver isn't yet
fully initialized, but still acknowlege the urb and submit a new rx_urb.
Do this for both interfaces -- irrelevant for ffdc hardware, but
relevant for newer hardware, though newer hardware doesn't spew the
constant stream of data as soon as the hardware is initialized like the
older ffdc devices, so they'd be less likely to trigger this anyway...

Tested with both an ffdc device and an 0042 device.

Reported-by: Corinna Vinschen <>
Signed-off-by: Jarod Wilson <>
Signed-off-by: Mauro Carvalho Chehab <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agohdpvr: fix race conditon during start of streaming
Janne Grunau [Thu, 2 Feb 2012 16:35:21 +0000 (13:35 -0300)]
hdpvr: fix race conditon during start of streaming

commit afa159538af61f1a65d48927f4e949fe514fb4fc upstream.

status has to be set to STREAMING before the streaming worker is
queued. hdpvr_transmit_buffers() will exit immediately otherwise.

Reported-by: Joerg Desch <>
Signed-off-by: Mauro Carvalho Chehab <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agocan: sja1000: fix isr hang when hw is unplugged under load
Oliver Hartkopp [Wed, 15 Feb 2012 16:51:56 +0000 (17:51 +0100)]
can: sja1000: fix isr hang when hw is unplugged under load

commit a7762b10c12a70c5dbf2253142764b728ac88c3a upstream.

In the case of hotplug enabled devices (PCMCIA/PCIeC) the removal of the
hardware can cause an infinite loop in the common sja1000 isr.

Use the already retrieved status register to indicate a possible hardware
removal and double check by reading the mode register in sja1000_is_absent.

Signed-off-by: Oliver Hartkopp <>
Acked-by: Wolfgang Grandegger <>
Signed-off-by: Marc Kleine-Budde <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agobuilddeb: Don't create files in /tmp with predictable names
Ben Hutchings [Wed, 15 Feb 2012 14:17:29 +0000 (14:17 +0000)]
builddeb: Don't create files in /tmp with predictable names

commit 6c635224602d760c1208ada337562f40d8ae93a5 upstream.

The current use of /tmp for file lists is insecure.  Put them under
$objtree/debian instead.

Signed-off-by: Ben Hutchings <>
Acked-by: maximilian attems <>
Signed-off-by: Michal Marek <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agodavinci_emac: Do not free all rx dma descriptors during init
Christian Riesch [Thu, 23 Feb 2012 01:14:17 +0000 (01:14 +0000)]
davinci_emac: Do not free all rx dma descriptors during init

commit 5d69703263d588dbb03f4e57091afd8942d96e6d upstream.

This patch fixes a regression that was introduced by

commit 0a5f38467765ee15478db90d81e40c269c8dda20
davinci_emac: Add Carrier Link OK check in Davinci RX Handler

Said commit adds a check whether the carrier link is ok. If the link is
not ok, the skb is freed and no new dma descriptor added to the rx dma
channel. This causes trouble during initialization when the carrier
status has not yet been updated. If a lot of packets are received while
netif_carrier_ok returns false, all dma descriptors are freed and the
rx dma transfer is stopped.

The bug occurs when the board is connected to a network with lots of
traffic and the ifconfig down/up is done, e.g., when reconfiguring
the interface with DHCP.

The bug can be reproduced by flood pinging the davinci board while doing
ifconfig eth0 down
ifconfig eth0 up
on the board.

After that, the rx path stops working and the overrun value reported
by ifconfig is counting up.

This patch reverts commit 0a5f38467765ee15478db90d81e40c269c8dda20
and instead issues warnings only if cpdma_chan_submit returns -ENOMEM.

Signed-off-by: Christian Riesch <>
Cc: Cyril Chemparathy <>
Cc: Sascha Hauer <>
Tested-by: Rajashekhara, Sudhakar <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agojme: Fix FIFO flush issue
Guo-Fu Tseng [Wed, 22 Feb 2012 08:58:10 +0000 (08:58 +0000)]
jme: Fix FIFO flush issue

commit ba9adbe67e288823ac1deb7f11576ab5653f833e upstream.

Set the RX FIFO flush watermark lower.
According to Federico and JMicron's reply,
setting it to 16QW would be stable on most platforms.
Otherwise, user might experience packet drop issue.

Reported-by: Federico Quagliata <>
Fixed-by: Federico Quagliata <>
Signed-off-by: Guo-Fu Tseng <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoipvs: fix matching of fwmark templates during scheduling
Simon Horman [Fri, 27 Jan 2012 01:45:27 +0000 (10:45 +0900)]
ipvs: fix matching of fwmark templates during scheduling

commit e0aac52e17a3db68fe2ceae281780a70fc69957f upstream.

Commit f11017ec2d1859c661f4e2b12c4a8d250e1f47cf (2.6.37)
moved the fwmark variable in subcontext that is invalidated before
reaching the ip_vs_ct_in_get call. As vaddr is provided as pointer
in the param structure make sure the fwmark variable is in
same context. As the fwmark templates can not be matched,
more and more template connections are created and the
controlled connections can not go to single real server.

Signed-off-by: Julian Anastasov <>
Signed-off-by: Simon Horman <>
Signed-off-by: Pablo Neira Ayuso <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoscsi_pm: Fix bug in the SCSI power management handler
Alan Stern [Fri, 17 Feb 2012 21:25:08 +0000 (16:25 -0500)]
scsi_pm: Fix bug in the SCSI power management handler

commit fea6d607e154cf96ab22254ccb48addfd43d4cb5 upstream.

This patch (as1520) fixes a bug in the SCSI layer's power management

LUN scanning can be carried out asynchronously in do_scan_async(), and
sd uses an asynchronous thread for the time-consuming parts of disk
probing in sd_probe_async().  Currently nothing coordinates these
async threads with system sleep transitions; they can and do attempt
to continue scanning/probing SCSI devices even after the host adapter
has been suspended.  As one might expect, the outcome is not ideal.

This is what the "prepare" stage of system suspend was created for.
After the prepare callback has been called for a host, target, or
device, drivers are not allowed to register any children underneath
them.  Currently the SCSI prepare callback is not implemented; this
patch rectifies that omission.

For SCSI hosts, the prepare routine calls scsi_complete_async_scans()
to wait until async scanning is finished.  It might be slightly more
efficient to wait only until the host in question has been scanned,
but there's currently no way to do that.  Besides, during a sleep
transition we will ultimately have to wait until all the host scanning
has finished anyway.

For SCSI devices, the prepare routine calls async_synchronize_full()
to wait until sd probing is finished.  The routine does nothing for
SCSI targets, because asynchronous target scanning is done only as
part of host scanning.

Signed-off-by: Alan Stern <>
Signed-off-by: James Bottomley <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoscsi_scan: Fix 'Poison overwritten' warning caused by using freed 'shost'
Huajun Li [Sun, 12 Feb 2012 11:59:14 +0000 (19:59 +0800)]
scsi_scan: Fix 'Poison overwritten' warning caused by using freed 'shost'

commit 267a6ad4aefaafbde607804c60945bcf97f91c1b upstream.

In do_scan_async(), calling scsi_autopm_put_host(shost) may reference
freed shost, and cause Posison overwitten warning.
Yes, this case can happen, for example, an USB is disconnected just
when do_scan_async() thread starts to run, then scsi_host_put() called
in scsi_finish_async_scan() will lead to shost be freed(because the
refcount of shost->shost_gendev decreases to 1 after USB disconnects),
at this point, if references shost again, system will show following
warning msg.

To make scsi_autopm_put_host(shost) always reference a valid shost,
put it just before scsi_host_put() in function

[  299.281565] =============================================================================
[  299.281634] BUG kmalloc-4096 (Tainted: G          I ): Poison overwritten
[  299.281682] -----------------------------------------------------------------------------
[  299.281684]
[  299.281752] INFO: 0xffff880056c305d0-0xffff880056c305d0. First byte
0x6a instead of 0x6b
[  299.281816] INFO: Allocated in scsi_host_alloc+0x4a/0x490 age=1688
cpu=1 pid=2004
[  299.281870]  __slab_alloc+0x617/0x6c1
[  299.281901]  __kmalloc+0x28c/0x2e0
[  299.281931]  scsi_host_alloc+0x4a/0x490
[  299.281966]  usb_stor_probe1+0x5b/0xc40 [usb_storage]
[  299.282010]  storage_probe+0xa4/0xe0 [usb_storage]
[  299.282062]  usb_probe_interface+0x172/0x330 [usbcore]
[  299.282105]  driver_probe_device+0x257/0x3b0
[  299.282138]  __driver_attach+0x103/0x110
[  299.282171]  bus_for_each_dev+0x8e/0xe0
[  299.282201]  driver_attach+0x26/0x30
[  299.282230]  bus_add_driver+0x1c4/0x430
[  299.282260]  driver_register+0xb6/0x230
[  299.282298]  usb_register_driver+0xe5/0x270 [usbcore]
[  299.282337]  0xffffffffa04ab03d
[  299.282364]  do_one_initcall+0x47/0x230
[  299.282396]  sys_init_module+0xa0f/0x1fe0
[  299.282429] INFO: Freed in scsi_host_dev_release+0x18a/0x1d0 age=85
cpu=0 pid=2008
[  299.282482]  __slab_free+0x3c/0x2a1
[  299.282510]  kfree+0x296/0x310
[  299.282536]  scsi_host_dev_release+0x18a/0x1d0
[  299.282574]  device_release+0x74/0x100
[  299.282606]  kobject_release+0xc7/0x2a0
[  299.282637]  kobject_put+0x54/0xa0
[  299.282668]  put_device+0x27/0x40
[  299.282694]  scsi_host_put+0x1d/0x30
[  299.282723]  do_scan_async+0x1fc/0x2b0
[  299.282753]  kthread+0xdf/0xf0
[  299.282782]  kernel_thread_helper+0x4/0x10
[  299.282817] INFO: Slab 0xffffea00015b0c00 objects=7 used=7 fp=0x
      (null) flags=0x100000000004080
[  299.282882] INFO: Object 0xffff880056c30000 @offset=0 fp=0x          (null)
[  299.282884]

Signed-off-by: Huajun Li <>
Acked-by: Alan Stern <>
Signed-off-by: James Bottomley <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agogenirq: Handle pending irqs in irq_startup()
Thomas Gleixner [Wed, 8 Feb 2012 10:57:52 +0000 (11:57 +0100)]
genirq: Handle pending irqs in irq_startup()

commit b4bc724e82e80478cba5fe9825b62e71ddf78757 upstream.

An interrupt might be pending when irq_startup() is called, but the
startup code does not invoke the resend logic. In some cases this
prevents the device from issuing another interrupt which renders the
device non functional.

Call the resend function in irq_startup() to keep things going.

Reported-and-tested-by: Russell King <>
Signed-off-by: Thomas Gleixner <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agogenirq: Unmask oneshot irqs when thread was not woken
Thomas Gleixner [Tue, 7 Feb 2012 16:58:03 +0000 (17:58 +0100)]
genirq: Unmask oneshot irqs when thread was not woken

commit ac5637611150281f398bb7a47e3fcb69a09e7803 upstream.

When the primary handler of an interrupt which is marked IRQ_ONESHOT
returns IRQ_HANDLED or IRQ_NONE, then the interrupt thread is not
woken and the unmask logic of the interrupt line is never
invoked. This keeps the interrupt masked forever.

This was not noticed as most IRQ_ONESHOT users wake the thread
unconditionally (usually because they cannot access the underlying
device from hard interrupt context). Though this behaviour was nowhere
documented and not necessarily intentional. Some drivers can avoid the
thread wakeup in certain cases and run into the situation where the
interrupt line s kept masked.

Handle it gracefully.

Reported-and-tested-by: Lothar Wassmann <>
Signed-off-by: Thomas Gleixner <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoath9k: stop on rates with idx -1 in ath9k rate control's .tx_status
Pavel Roskin [Sat, 11 Feb 2012 15:01:53 +0000 (10:01 -0500)]
ath9k: stop on rates with idx -1 in ath9k rate control's .tx_status

commit 2504a6423b9ab4c36df78227055995644de19edb upstream.

Rate control algorithms are supposed to stop processing when they
encounter a rate with the index -1.  Checking for rate->count not being
zero is not enough.

Allowing a rate with negative index leads to memory corruption in

One consequence of the bug is discussed at

Signed-off-by: Pavel Roskin <>
Signed-off-by: John W. Linville <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agox86/amd: Fix L1i and L2 cache sharing information for AMD family 15h processors
Andreas Herrmann [Wed, 8 Feb 2012 19:52:29 +0000 (20:52 +0100)]
x86/amd: Fix L1i and L2 cache sharing information for AMD family 15h processors

commit 32c3233885eb10ac9cb9410f2f8cd64b8df2b2a1 upstream.

For L1 instruction cache and L2 cache the shared CPU information
is wrong. On current AMD family 15h CPUs those caches are shared
between both cores of a compute unit.

This fixes

Signed-off-by: Andreas Herrmann <>
Cc: Petkov Borislav <>
Cc: Dave Jones <>
Signed-off-by: Ingo Molnar <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoARM: omap: fix oops in arch/arm/mach-omap2/vp.c when pmic is not found
Russell King [Tue, 7 Feb 2012 09:42:11 +0000 (09:42 +0000)]
ARM: omap: fix oops in arch/arm/mach-omap2/vp.c when pmic is not found

commit d980e0f8d858c6963d676013e976ff00ab7acb2b upstream.

When the PMIC is not found, voltdm->pmic will be NULL.  vp.c's
initialization function tries to dereferences this, which causes an

Unable to handle kernel NULL pointer dereference at virtual address 00000000
pgd = c0004000
[00000000] *pgd=00000000
Internal error: Oops: 5 [#1] PREEMPT
Modules linked in:
CPU: 0    Not tainted  (3.3.0-rc2+ #204)
PC is at omap_vp_init+0x5c/0x15c
LR is at omap_vp_init+0x58/0x15c
pc : [<c03db880>]    lr : [<c03db87c>]    psr: 60000013
sp : c181ff30  ip : c181ff68  fp : c181ff64
r10: c0407808  r9 : c040786c  r8 : c0407814
r7 : c0026868  r6 : c00264fc  r5 : c040ad6c  r4 : 00000000
r3 : 00000040  r2 : 000032c8  r1 : 0000fa00  r0 : 000032c8
Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment kernel
Control: 10c5387d  Table: 80004019  DAC: 00000015
Process swapper (pid: 1, stack limit = 0xc181e2e8)
Stack: (0xc181ff30 to 0xc1820000)
ff20:                                     c0381d00 c02e9c6d c0383582 c040786c
ff40: c040ad6c c00264fc c0026868 c0407814 00000000 c03d9de4 c181ff8c c181ff68
ff60: c03db448 c03db830 c02e982c c03fdfb8 c03fe004 c0039988 00000013 00000000
ff80: c181ff9c c181ff90 c03d9df8 c03db390 c181ffdc c181ffa0 c0008798 c03d9df0
ffa0: c181ffc4 c181ffb0 c0055a44 c0187050 c0039988 c03fdfb8 c03fe004 c0039988
ffc0: 00000013 00000000 00000000 00000000 c181fff4 c181ffe0 c03d1284 c0008708
ffe0: 00000000 c03d1208 00000000 c181fff8 c0039988 c03d1214 1077ce40 01f7ee08
[<c03db824>] (omap_vp_init+0x0/0x15c) from [<c03db448>] (omap_voltage_late_init+0xc4/0xfc)
[<c03db384>] (omap_voltage_late_init+0x0/0xfc) from [<c03d9df8>] (omap2_common_pm_late_init+0x14/0x54)
 r8:00000000 r7:00000013 r6:c0039988 r5:c03fe004 r4:c03fdfb8
[<c03d9de4>] (omap2_common_pm_late_init+0x0/0x54) from [<c0008798>] (do_one_initcall+0x9c/0x164)
[<c00086fc>] (do_one_initcall+0x0/0x164) from [<c03d1284>] (kernel_init+0x7c/0x120)
[<c03d1208>] (kernel_init+0x0/0x120) from [<c0039988>] (do_exit+0x0/0x2cc)
 r5:c03d1208 r4:00000000
Code: e5ca300b e5900034 ebf69027 e5994024 (e5941000)
---[ end trace aed617dddaf32c3d ]---
Kernel panic - not syncing: Attempted to kill init!

Signed-off-by: Russell King <>
Cc: Igor Grinberg <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoARM: omap: fix oops in drivers/video/omap2/dss/dpi.c
Russell King [Tue, 7 Feb 2012 09:44:55 +0000 (09:44 +0000)]
ARM: omap: fix oops in drivers/video/omap2/dss/dpi.c

commit 40410715715178ec196314dd0c19150c06901f80 upstream.

When a PMIC is not found, this driver is unable to obtain its
'vdds_dsi_reg' regulator.  Even through its initialization function
fails, other code still calls its enable function, which fails to
check whether it has this regulator before asking for it to be enabled.

This fixes the oops, however a better fix would be to sort out the
upper layers to prevent them calling into a module which failed to

Unable to handle kernel NULL pointer dereference at virtual address 00000038
pgd = c0004000
[00000038] *pgd=00000000
Internal error: Oops: 5 [#1] PREEMPT
Modules linked in:
CPU: 0    Not tainted  (3.3.0-rc2+ #228)
PC is at regulator_enable+0x10/0x70
LR is at omapdss_dpi_display_enable+0x54/0x15c
pc : [<c01b9a08>]    lr : [<c01af994>]    psr: 60000013
sp : c181fd90  ip : c181fdb0  fp : c181fdac
r10: c042eff0  r9 : 00000060  r8 : c044a164
r7 : c042c0e4  r6 : c042bd60  r5 : 00000000  r4 : c042bd60
r3 : c084de48  r2 : c181e000  r1 : c042bd60  r0 : 00000000
Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment kernel
Control: 10c5387d  Table: 80004019  DAC: 00000015
Process swapper (pid: 1, stack limit = 0xc181e2e8)
Stack: (0xc181fd90 to 0xc1820000)
fd80:                                     c001754c c042bd60 00000000 c042bd60
fda0: c181fdcc c181fdb0 c01af994 c01b9a04 c0016104 c042bd60 c042bd60 c044a338
fdc0: c181fdec c181fdd0 c01b5ed0 c01af94c c042bd60 c042bd60 c1aa8000 c1aa8a0c
fde0: c181fe04 c181fdf0 c01b5f54 c01b5ea8 c02fc18c c042bd60 c181fe3c c181fe08
fe00: c01b2a18 c01b5f48 c01aed14 c02fc160 c01df8ec 00000002 c042bd60 00000003
fe20: c042bd60 c1aa8000 c1aa8a0c c042eff8 c181fe84 c181fe40 c01b3874 c01b29fc
fe40: c042eff8 00000000 c042f000 c0449db8 c044ed78 00000000 c181fe74 c042eff8
fe60: c042eff8 c0449db8 c0449db8 c044ed78 00000000 00000000 c181fe94 c181fe88
fe80: c01e452c c01b35e8 c181feb4 c181fe98 c01e2fdc c01e4518 c042eff8 c0449db8
fea0: c0449db8 c181fef0 c181fecc c181feb8 c01e3104 c01e2f48 c042eff8 c042f02c
fec0: c181feec c181fed0 c01e3190 c01e30c0 c01e311c 00000000 c01e311c c0449db8
fee0: c181ff14 c181fef0 c01e1998 c01e3128 c18330a8 c1892290 c04165e8 c0449db8
ff00: c0449db8 c1ab60c0 c181ff24 c181ff18 c01e2e28 c01e194c c181ff54 c181ff28
ff20: c01e2218 c01e2e14 c039afed c181ff38 c04165e8 c041660c c0449db8 00000013
ff40: 00000000 c03ffdb8 c181ff7c c181ff58 c01e384c c01e217c c181ff7c c04165e8
ff60: c041660c c003a37c 00000013 00000000 c181ff8c c181ff80 c01e488c c01e3790
ff80: c181ff9c c181ff90 c03ffdcc c01e484c c181ffdc c181ffa0 c0008798 c03ffdc4
ffa0: c181ffc4 c181ffb0 c0056440 c0187810 c003a37c c04165e8 c041660c c003a37c
ffc0: 00000013 00000000 00000000 00000000 c181fff4 c181ffe0 c03ea284 c0008708
ffe0: 00000000 c03ea208 00000000 c181fff8 c003a37c c03ea214 1073cec0 01f7ee08
[<c01b99f8>] (regulator_enable+0x0/0x70) from [<c01af994>] (omapdss_dpi_display_enable+0x54/0x15c)
 r6:c042bd60 r5:00000000 r4:c042bd60
[<c01af940>] (omapdss_dpi_display_enable+0x0/0x15c) from [<c01b5ed0>] (generic_dpi_panel_power_on+0x34/0x78)
 r6:c044a338 r5:c042bd60 r4:c042bd60
[<c01b5e9c>] (generic_dpi_panel_power_on+0x0/0x78) from [<c01b5f54>] (generic_dpi_panel_enable+0x18/0x28)
 r7:c1aa8a0c r6:c1aa8000 r5:c042bd60 r4:c042bd60
[<c01b5f3c>] (generic_dpi_panel_enable+0x0/0x28) from [<c01b2a18>] (omapfb_init_display+0x28/0x150)
[<c01b29f0>] (omapfb_init_display+0x0/0x150) from [<c01b3874>] (omapfb_probe+0x298/0x318)
 r8:c042eff8 r7:c1aa8a0c r6:c1aa8000 r5:c042bd60 r4:00000003
[<c01b35dc>] (omapfb_probe+0x0/0x318) from [<c01e452c>] (platform_drv_probe+0x20/0x24)
[<c01e450c>] (platform_drv_probe+0x0/0x24) from [<c01e2fdc>] (really_probe+0xa0/0x178)
[<c01e2f3c>] (really_probe+0x0/0x178) from [<c01e3104>] (driver_probe_device+0x50/0x68)
 r7:c181fef0 r6:c0449db8 r5:c0449db8 r4:c042eff8
[<c01e30b4>] (driver_probe_device+0x0/0x68) from [<c01e3190>] (__driver_attach+0x74/0x98)
 r5:c042f02c r4:c042eff8
[<c01e311c>] (__driver_attach+0x0/0x98) from [<c01e1998>] (bus_for_each_dev+0x58/0x98)
 r6:c0449db8 r5:c01e311c r4:00000000
[<c01e1940>] (bus_for_each_dev+0x0/0x98) from [<c01e2e28>] (driver_attach+0x20/0x28)
 r7:c1ab60c0 r6:c0449db8 r5:c0449db8 r4:c04165e8
[<c01e2e08>] (driver_attach+0x0/0x28) from [<c01e2218>] (bus_add_driver+0xa8/0x22c)
[<c01e2170>] (bus_add_driver+0x0/0x22c) from [<c01e384c>] (driver_register+0xc8/0x154)
[<c01e3784>] (driver_register+0x0/0x154) from [<c01e488c>] (platform_driver_register+0x4c/0x60)
 r8:00000000 r7:00000013 r6:c003a37c r5:c041660c r4:c04165e8
[<c01e4840>] (platform_driver_register+0x0/0x60) from [<c03ffdcc>] (omapfb_init+0x14/0x34)
[<c03ffdb8>] (omapfb_init+0x0/0x34) from [<c0008798>] (do_one_initcall+0x9c/0x164)
[<c00086fc>] (do_one_initcall+0x0/0x164) from [<c03ea284>] (kernel_init+0x7c/0x120)
[<c03ea208>] (kernel_init+0x0/0x120) from [<c003a37c>] (do_exit+0x0/0x2d8)
 r5:c03ea208 r4:00000000
Code: e1a0c00d e92dd870 e24cb004 e24dd004 (e5906038)
---[ end trace 9e2474c2e193b223 ]---

Acked-by: Tony Lindgren <>
Signed-off-by: Russell King <>
Cc: Igor Grinberg <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agohwmon: (ads1015) Fix file leak in probe function
Guenter Roeck [Wed, 22 Feb 2012 16:13:52 +0000 (08:13 -0800)]
hwmon: (ads1015) Fix file leak in probe function

commit 363434b5dc352464ac7601547891e5fc9105f124 upstream.

An error while creating sysfs attribute files in the driver's probe function
results in an error abort, but already created files are not removed. This patch
fixes the problem.

Signed-off-by: Guenter Roeck <>
Cc: Dirk Eibach <>
Acked-by: Jean Delvare <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agohwmon: (max6639) Fix PPR register initialization to set both channels
Chris D Schimp [Mon, 20 Feb 2012 22:44:59 +0000 (17:44 -0500)]
hwmon: (max6639) Fix PPR register initialization to set both channels

commit 2f2da1ac0ba5b6cc6e1957c4da5ff20e67d8442b upstream.

Initialize PPR register for both channels, and set correct PPR register bits.
Also remove unnecessary variable initializations.

Signed-off-by: Chris D Schimp <>
[ Merged two patches into one]
Signed-off-by: Guenter Roeck <>
Acked-by: Roland Stigge <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agohwmon: (max6639) Fix FAN_FROM_REG calculation
Chris D Schimp [Mon, 20 Feb 2012 21:59:24 +0000 (16:59 -0500)]
hwmon: (max6639) Fix FAN_FROM_REG calculation

commit b63d97a36edb1aecf8c13e5f5783feff4d64c24b upstream.

RPM calculation from tachometer value does not depend on PPR.
Also, do not report negative RPM values.

Signed-off-by: Chris D Schimp <>
[ do not report negative RPM values]
Signed-off-by: Guenter Roeck <>
Acked-by: Roland Stigge <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoNOMMU: Lock i_mmap_mutex for access to the VMA prio list
David Howells [Thu, 23 Feb 2012 13:50:35 +0000 (13:50 +0000)]
NOMMU: Lock i_mmap_mutex for access to the VMA prio list

commit 918e556ec214ed2f584e4cac56d7b29e4bb6bf27 upstream.

Lock i_mmap_mutex for access to the VMA prio list to prevent concurrent
access.  Currently, certain parts of the mmap handling are protected by
the region mutex, but not all.

Reported-by: Al Viro <>
Signed-off-by: David Howells <>
Acked-by: Al Viro <>
Signed-off-by: Linus Torvalds <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoALSA: hda/realtek - Fix surround output regression on Acer Aspire 5935
Takashi Iwai [Fri, 17 Feb 2012 09:12:38 +0000 (10:12 +0100)]
ALSA: hda/realtek - Fix surround output regression on Acer Aspire 5935

commit ef8d60fb79614a86a82720dc2402631dbcafb315 upstream.

The previous fix for the speaker on Acer Aspire 59135 introduced
another problem for surround outputs.  It changed the connections on
the line-in/mic pins for limiting the routes, but it left the modified
connections.  Thus wrong connection indices were written when set to
4ch or 6ch mode.

This patch fixes it by restoring the right connections just after
parsing the tree but before the initialization.


Signed-off-by: Takashi Iwai <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoALSA: hda/realtek - Fix overflow of vol/sw check bitmap
Takashi Iwai [Thu, 16 Feb 2012 15:38:07 +0000 (16:38 +0100)]
ALSA: hda/realtek - Fix overflow of vol/sw check bitmap

commit c14c95f62ecb8710af14ae0d48e01991b70bb6f4 upstream.

The bitmap introduced in the commit [527e4d73: ALSA: hda/realtek - Fix
missing volume controls with ALC260] is too narrow for some codecs,
which may have more NIDs than 0x20, thus it may overflow the bitmap
array on them.

Just double the number to cover all and also add a sanity-check code
to be safer.

Signed-off-by: Takashi Iwai <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoASoC: wm8962: Fix sidetone enumeration texts
Mark Brown [Tue, 14 Feb 2012 06:00:47 +0000 (22:00 -0800)]
ASoC: wm8962: Fix sidetone enumeration texts

commit 31794bc37bf2db84f085da52b72bfba65739b2d2 upstream.

The sidetone enumeration texts have left and right swapped.

Signed-off-by: Mark Brown <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agotarget: Allow control CDBs with data > 1 page
Andy Grover [Tue, 17 Jan 2012 00:57:08 +0000 (16:57 -0800)]
target: Allow control CDBs with data > 1 page

commit 4949314c7283ea4f9ade182ca599583b89f7edd6 upstream.

We need to handle >1 page control cdbs, so extend the code to do a vmap
if bigger than 1 page. It seems like kmap() is still preferable if just
a page, fewer TLB shootdowns(?), so keep using that when possible.

Rename function pair for their new scope.

Signed-off-by: Andy Grover <>
Signed-off-by: Nicholas Bellinger <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agousb-storage: fix freezing of the scanning thread
Alan Stern [Tue, 21 Feb 2012 18:16:32 +0000 (13:16 -0500)]
usb-storage: fix freezing of the scanning thread

commit bb94a406682770a35305daaa241ccdb7cab399de upstream.

This patch (as1521b) fixes the interaction between usb-storage's
scanning thread and the freezer.  The current implementation has a
race: If the device is unplugged shortly after being plugged in and
just as a system sleep begins, the scanning thread may get frozen
before the khubd task.  Khubd won't be able to freeze until the
disconnect processing is complete, and the disconnect processing can't
proceed until the scanning thread finishes, so the sleep transition
will fail.

The implementation in the 3.2 kernel suffers from an additional
problem.  There the scanning thread calls set_freezable_with_signal(),
and the signals sent by the freezer will mess up the thread's I/O
delays, which are all interruptible.

The solution to both problems is the same: Replace the kernel thread
used for scanning with a delayed-work routine on the system freezable
work queue.  Freezable work queues have the nice property that you can
cancel a work item even while the work queue is frozen, and no signals
are needed.

The 3.2 version of this patch solves the problem in Bugzilla #42730.

Signed-off-by: Alan Stern <>
Acked-by: Seth Forshee <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoUSB: Set hub depth after USB3 hub reset
Elric Fu [Sat, 18 Feb 2012 05:32:27 +0000 (13:32 +0800)]
USB: Set hub depth after USB3 hub reset

commit a45aa3b30583e7d54e7cf4fbcd0aa699348a6e5c upstream.

The superspeed device attached to a USB 3.0 hub(such as VIA's)
doesn't respond the address device command after resume. The
root cause is the superspeed hub will miss the Hub Depth value
that is used as an offset into the route string to locate the
bits it uses to determine the downstream port number after
reset, and all packets can't be routed to the device attached
to the superspeed hub.

Hub driver sends a Set Hub Depth request to the superspeed hub
except for USB 3.0 root hub when the hub is initialized and
doesn't send the request again after reset due to the resume
process. So moving the code that sends the Set Hub Depth request
to the superspeed hub from hub_configure() to hub_activate()
is to cover those situations include initialization and reset.

The patch should be backported to kernels as old as 2.6.39.

Signed-off-by: Elric Fu <>
Signed-off-by: Sarah Sharp <>
Acked-by: Alan Stern <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoUSB: Don't fail USB3 probe on missing legacy PCI IRQ.
Sarah Sharp [Tue, 14 Feb 2012 00:25:57 +0000 (16:25 -0800)]
USB: Don't fail USB3 probe on missing legacy PCI IRQ.

commit 68d07f64b8a11a852d48d1b05b724c3e20c0d94b upstream.

Intel has a PCI USB xhci host controller on a new platform. It doesn't
have a line IRQ definition in BIOS.  The Linux driver refuses to
initialize this controller, but Windows works well because it only depends
on MSI.

Actually, Linux also can work for MSI.  This patch avoids the line IRQ
checking for USB3 HCDs in usb core PCI probe.  It allows the xHCI driver
to try to enable MSI or MSI-X first.  It will fail the probe if MSI
enabling failed and there's no legacy PCI IRQ.

This patch should be backported to kernels as old as 2.6.32.

Signed-off-by: Alex Shi <>
Signed-off-by: Sarah Sharp <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoxhci: Fix encoding for HS bulk/control NAK rate.
Sarah Sharp [Mon, 13 Feb 2012 22:42:11 +0000 (14:42 -0800)]
xhci: Fix encoding for HS bulk/control NAK rate.

commit 340a3504fd39dad753ba908fb6f894ee81fc3ae2 upstream.

The xHCI 0.96 spec says that HS bulk and control endpoint NAK rate must
be encoded as an exponent of two number of microframes.  The endpoint
descriptor has the NAK rate encoded in number of microframes.  We were
just copying the value from the endpoint descriptor into the endpoint
context interval field, which was not correct.  This lead to the VIA
host rejecting the add of a bulk OUT endpoint from any USB 2.0 mass
storage device.

The fix is to use the correct encoding.  Refactor the code to convert
number of frames to an exponential number of microframes, and make sure
we convert the number of microframes in HS bulk and control endpoints to
an exponent.

This should be back ported to kernels as old as 2.6.31, that contain the
commit dfa49c4ad120a784ef1ff0717168aa79f55a483a "USB: xhci - fix math
in xhci_get_endpoint_interval"

Signed-off-by: Sarah Sharp <>
Tested-by: Felipe Contreras <>
Suggested-by: Andiry Xu <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoxhci: Fix oops caused by more USB2 ports than USB3 ports.
Sarah Sharp [Thu, 9 Feb 2012 22:43:44 +0000 (14:43 -0800)]
xhci: Fix oops caused by more USB2 ports than USB3 ports.

commit 3278a55a1aebe2bbd47fbb5196209e5326a88b56 upstream.

The code to set the device removable bits in the USB 2.0 roothub
descriptor was accidentally looking at the USB 3.0 port registers
instead of the USB 2.0 registers.  This can cause an oops if there are
more USB 2.0 registers than USB 3.0 registers.

This should be backported to kernels as old as 2.6.39, that contain the
commit 4bbb0ace9a3de8392527e3c87926309d541d3b00 "xhci: Return a USB 3.0
hub descriptor for USB3 roothub."

Signed-off-by: Sarah Sharp <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoUSB: Fix handoff when BIOS disables host PCI device.
Sarah Sharp [Tue, 7 Feb 2012 23:11:46 +0000 (15:11 -0800)]
USB: Fix handoff when BIOS disables host PCI device.

commit cab928ee1f221c9cc48d6615070fefe2e444384a upstream.

On some systems with an Intel Panther Point xHCI host controller, the
BIOS disables the xHCI PCI device during boot, and switches the xHCI
ports over to EHCI.  This allows the BIOS to access USB devices without
having xHCI support.

The downside is that the xHCI BIOS handoff mechanism will fail because
memory mapped I/O is not enabled for the disabled PCI device.
Jesse Barnes says this is expected behavior.  The PCI core will enable
BARs before quirks run, but it will leave it in an undefined state, and
it may not have memory mapped I/O enabled.

Make the generic USB quirk handler call pci_enable_device() to re-enable
MMIO, and call pci_disable_device() once the host-specific BIOS handoff
is finished.  This will balance the ref counts in the PCI core.  When
the PCI probe function is called, usb_hcd_pci_probe() will call
pci_enable_device() again.

This should be back ported to kernels as old as 2.6.31.  That was the
first kernel with xHCI support, and no one has complained about BIOS
handoffs failing due to memory mapped I/O being disabled on other hosts

Signed-off-by: Sarah Sharp <>
Acked-by: Oliver Neukum <>
Cc: Jesse Barnes <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoUSB: Remove duplicate USB 3.0 hub feature #defines.
Sarah Sharp [Fri, 6 Jan 2012 00:28:54 +0000 (16:28 -0800)]
USB: Remove duplicate USB 3.0 hub feature #defines.

commit d9f5343e35d9138432657202afa8e3ddb2ade360 upstream.

Somehow we ended up with duplicate hub feature #defines in ch11.h.
Tatyana Brokhman first created the USB 3.0 hub feature macros in 2.6.38
with commit 0eadcc09203349b11ca477ec367079b23d32ab91 "usb: USB3.0 ch11
definitions".  In 2.6.39, I modified a patch from John Youn that added
similar macros in a different place in the same file, and committed
dbe79bbe9dcb22cb3651c46f18943477141ca452 "USB 3.0 Hub Changes".

Some of the #defines used different names for the same values.  Others
used exactly the same names with the same values, like these gems:

 #define USB_PORT_FEAT_BH_PORT_RESET            28

According to my very geeky husband (who looked it up in the C99 spec),
it is allowed to have object-like macros with duplicate names as long as
the replacement list is exactly the same.  However, he recalled that
some compilers will give warnings when they find duplicate macros.  It's
probably best to remove the duplicates in the stable tree, so that the
code compiles for everyone.

The macros are now fixed to move the feature requests that are specific
to USB 3.0 hubs into a new section (out of the USB 2.0 hub feature
section), and use the most common macro name.

This patch should be backported to 2.6.39.

Signed-off-by: Sarah Sharp <>
Cc: Tatyana Brokhman <>
Cc: John Youn <>
Cc: Jamey Sharp <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoUSB: Serial: ti_usb_3410_5052: Add Abbot Diabetes Care cable id
Andrew Lunn [Mon, 20 Feb 2012 08:31:57 +0000 (09:31 +0100)]
USB: Serial: ti_usb_3410_5052: Add Abbot Diabetes Care cable id

commit 7fd25702ba616d9ba56e2a625472f29e5aff25ee upstream.

This USB-serial cable with mini stereo jack enumerates as:
Bus 001 Device 004: ID 1a61:3410 Abbott Diabetes Care

It is a TI3410 inside.

Signed-off-by: Andrew Lunn <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoUSB: option: cleanup zte 3g-dongle's pid in option.c
Rui li [Tue, 14 Feb 2012 02:35:01 +0000 (10:35 +0800)]
USB: option: cleanup zte 3g-dongle's pid in option.c

commit b9e44fe5ecda4158c22bc1ea4bffa378a4f83f65 upstream.

  1. Remove all old mass-storage ids's pid:
  2. As the pid from 0x1401 to 0x1510 which have not surely assigned to
     use for serial-port or mass-storage port,so i think it should be
     removed now, and will re-add after it have assigned in future;
  3. sort the pid to WCDMA and CDMA.

Signed-off-by: Rui li <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoUSB: Added Kamstrup VID/PIDs to cp210x serial driver.
Bruno Thomsen [Tue, 21 Feb 2012 22:41:37 +0000 (23:41 +0100)]
USB: Added Kamstrup VID/PIDs to cp210x serial driver.

commit c6c1e4491dc8d1ed2509fa6aacffa7f34614fc38 upstream.

Signed-off-by: Bruno Thomsen <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agotcp: fix tcp_shifted_skb() adjustment of lost_cnt_hint for FACK
Neal Cardwell [Mon, 13 Feb 2012 20:22:08 +0000 (20:22 +0000)]
tcp: fix tcp_shifted_skb() adjustment of lost_cnt_hint for FACK

[ Upstream commit 0af2a0d0576205dda778d25c6c344fc6508fc81d ]

This commit ensures that lost_cnt_hint is correctly updated in
tcp_shifted_skb() for FACK TCP senders. The lost_cnt_hint adjustment
in tcp_sacktag_one() only applies to non-FACK senders, so FACK senders
need their own adjustment.

This applies the spirit of 1e5289e121372a3494402b1b131b41bfe1cf9b7f -
except now that the sequence range passed into tcp_sacktag_one() is
correct we need only have a special case adjustment for FACK.

Signed-off-by: Neal Cardwell <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agotcp: fix range tcp_shifted_skb() passes to tcp_sacktag_one()
Neal Cardwell [Sun, 12 Feb 2012 18:37:10 +0000 (18:37 +0000)]
tcp: fix range tcp_shifted_skb() passes to tcp_sacktag_one()

[ Upstream commit daef52bab1fd26e24e8e9578f8fb33ba1d0cb412 ]

Fix the newly-SACKed range to be the range of newly-shifted bytes.

Previously - since 832d11c5cd076abc0aa1eaf7be96c81d1a59ce41 -
tcp_shifted_skb() incorrectly called tcp_sacktag_one() with the start
and end sequence numbers of the skb it passes in set to the range just
beyond the range that is newly-SACKed.

This commit also removes a special-case adjustment to lost_cnt_hint in
tcp_shifted_skb() since the pre-existing adjustment of lost_cnt_hint
in tcp_sacktag_one() now properly handles this things now that the
correct start sequence number is passed in.

Signed-off-by: Neal Cardwell <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agotcp: allow tcp_sacktag_one() to tag ranges not aligned with skbs
Neal Cardwell [Sun, 12 Feb 2012 18:37:09 +0000 (18:37 +0000)]
tcp: allow tcp_sacktag_one() to tag ranges not aligned with skbs

[ Upstream commit cc9a672ee522d4805495b98680f4a3db5d0a0af9 ]

This commit allows callers of tcp_sacktag_one() to pass in sequence
ranges that do not align with skb boundaries, as tcp_shifted_skb()
needs to do in an upcoming fix in this patch series.

In fact, now tcp_sacktag_one() does not need to depend on an input skb
at all, which makes its semantics and dependencies more clear.

Signed-off-by: Neal Cardwell <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agogro: more generic L2 header check
Eric Dumazet [Wed, 8 Feb 2012 08:51:50 +0000 (08:51 +0000)]
gro: more generic L2 header check

[ Upstream commit 5ca3b72c5da47d95b83857b768def6172fbc080a ]

Shlomo Pongratz reported GRO L2 header check was suited for Ethernet
only, and failed on IB/ipoib traffic.

He provided a patch faking a zeroed header to let GRO aggregates frames.

Roland Dreier, Herbert Xu, and others suggested we change GRO L2 header
check to be more generic, ie not assuming L2 header is 14 bytes, but
taking into account hard_header_len.

__napi_gro_receive() has special handling for the common case (Ethernet)
to avoid a memcmp() call and use an inline optimized function instead.

Signed-off-by: Eric Dumazet <>
Reported-by: Shlomo Pongratz <>
Cc: Roland Dreier <>
Cc: Or Gerlitz <>
Cc: Herbert Xu <>
Tested-by: Sean Hefty <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoIPoIB: Stop lying about hard_header_len and use skb->cb to stash LL addresses
Roland Dreier [Tue, 7 Feb 2012 14:51:21 +0000 (14:51 +0000)]
IPoIB: Stop lying about hard_header_len and use skb->cb to stash LL addresses

[ Upstream commit 936d7de3d736e0737542641269436f4b5968e9ef ]

Commit a0417fa3a18a ("net: Make qdisc_skb_cb upper size bound
explicit.") made it possible for a netdev driver to use skb->cb
between its header_ops.create method and its .ndo_start_xmit
method.  Use this in ipoib_hard_header() to stash away the LL address
(GID + QPN), instead of the "ipoib_pseudoheader" hack.  This allows
IPoIB to stop lying about its hard_header_len, which will let us fix
the L2 check for GRO.

Signed-off-by: Roland Dreier <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agonet: Make qdisc_skb_cb upper size bound explicit.
David S. Miller [Mon, 6 Feb 2012 20:14:37 +0000 (15:14 -0500)]
net: Make qdisc_skb_cb upper size bound explicit.

[ Upstream commit 16bda13d90c8d5da243e2cfa1677e62ecce26860 ]

Just like skb->cb[], so that qdisc_skb_cb can be encapsulated inside
of other data structures.

This is intended to be used by IPoIB so that it can remember
addressing information stored at hard_header_ops->create() time that
it can fetch when the packet gets to the transmit routine.

Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoipv4: Fix wrong order of ip_rt_get_source() and update iph->daddr.
Li Wei [Thu, 9 Feb 2012 21:15:25 +0000 (21:15 +0000)]
ipv4: Fix wrong order of ip_rt_get_source() and update iph->daddr.

[ Upstream commit 5dc7883f2a7c25f8df40d7479687153558cd531b ]

This patch fix a bug which introduced by commit ac8a4810 (ipv4: Save
nexthop address of LSRR/SSRR option to IPCB.).In that patch, we saved
the nexthop of SRR in ip_option->nexthop and update iph->daddr until
we get to ip_forward_options(), but we need to update it before
ip_rt_get_source(), otherwise we may get a wrong src.

Signed-off-by: Li Wei <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agotcp_v4_send_reset: binding oif to iif in no sock case
Shawn Lu [Sat, 4 Feb 2012 12:38:09 +0000 (12:38 +0000)]
tcp_v4_send_reset: binding oif to iif in no sock case

[ Upstream commit e2446eaab5585555a38ea0df4e01ff313dbb4ac9 ]

Binding RST packet outgoing interface to incoming interface
for tcp v4 when there is no socket associate with it.
when sk is not NULL, using sk->sk_bound_dev_if instead.
(suggested by Eric Dumazet).

This has few benefits:
1. tcp_v6_send_reset already did that.
2. This helps tcp connect with SO_BINDTODEVICE set. When
connection is lost, we still able to sending out RST using
same interface.
3. we are sending reply, it is most likely to be succeed
if iif is used

Signed-off-by: Shawn Lu <>
Acked-by: Eric Dumazet <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoipv4: reset flowi parameters on route connect
Julian Anastasov [Sat, 4 Feb 2012 13:04:46 +0000 (13:04 +0000)]
ipv4: reset flowi parameters on route connect

[ Upstream commit e6b45241c57a83197e5de9166b3b0d32ac562609 ]

Eric Dumazet found that commit 813b3b5db83
(ipv4: Use caller's on-stack flowi as-is in output
route lookups.) that comes in 3.0 added a regression.
The problem appears to be that resulting flowi4_oif is
used incorrectly as input parameter to some routing lookups.
The result is that when connecting to local port without
listener if the IP address that is used is not on a loopback
interface we incorrectly assign RTN_UNICAST to the output
route because no route is matched by oif=lo. The RST packet
can not be sent immediately by tcp_v4_send_reset because
it expects RTN_LOCAL.

So, change ip_route_connect and ip_route_newports to
update the flowi4 fields that are input parameters because
we do not want unnecessary binding to oif.

To make it clear what are the input parameters that
can be modified during lookup and to show which fields of
floiw4 are reused add a new function to update the flowi4
structure: flowi4_update_output.

Thanks to Yurij M. Plotnikov for providing a bug report including a
program to reproduce the problem.

Thanks to Eric Dumazet for tracking the problem down to
tcp_v4_send_reset and providing initial fix.

Reported-by: Yurij M. Plotnikov <>
Signed-off-by: Julian Anastasov <>
Acked-by: Eric Dumazet <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agovia-velocity: S3 resume fix.
David Lv [Sat, 4 Feb 2012 23:22:26 +0000 (23:22 +0000)]
via-velocity: S3 resume fix.

[ Upstream commit b530b1930bbd9d005345133f0ff0c556d2a52b19 ]

Initially diagnosed on Ubuntu 11.04 with kernel 2.6.38.

velocity_close is not called during a suspend / resume cycle in this
driver and it has no business playing directly with power states.

Signed-off-by: David Lv <>
Acked-by: Francois Romieu <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agoveth: Enforce minimum size of VETH_INFO_PEER
Hagen Paul Pfeifer [Wed, 15 Feb 2012 04:09:46 +0000 (04:09 +0000)]
veth: Enforce minimum size of VETH_INFO_PEER

[ Upstream commit 237114384ab22c174ec4641e809f8e6cbcfce774 ]

VETH_INFO_PEER carries struct ifinfomsg plus optional IFLA
attributes. A minimal size of sizeof(struct ifinfomsg) must be
enforced or we may risk accessing that struct beyond the limits
of the netlink message.

Signed-off-by: Thomas Graf <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agonet_sched: Bug in netem reordering
Hagen Paul Pfeifer [Wed, 4 Jan 2012 17:35:26 +0000 (17:35 +0000)]
net_sched: Bug in netem reordering

[ Upstream commit eb10192447370f19a215a8c2749332afa1199d46 ]

Not now, but it looks you are correct. q->qdisc is NULL until another
additional qdisc is attached (beside tfifo). See 50612537e9ab2969312.
The following patch should work.

From: Hagen Paul Pfeifer <>

netem: catch NULL pointer by updating the real qdisc statistic

Reported-by: Vijay Subramanian <>
Signed-off-by: Hagen Paul Pfeifer <>
Acked-by: Eric Dumazet <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>
10 years agonetpoll: netpoll_poll_dev() should access dev->flags
Eric Dumazet [Tue, 14 Feb 2012 10:11:59 +0000 (10:11 +0000)]
netpoll: netpoll_poll_dev() should access dev->flags

[ Upstream commit 58e05f357a039a94aa36475f8c110256f693a239 ]

commit 5a698af53f (bond: service netpoll arp queue on master device)
tested IFF_SLAVE flag against dev->priv_flags instead of dev->flags

Signed-off-by: Eric Dumazet <>
Cc: WANG Cong <>
Acked-by: Neil Horman <>
Signed-off-by: David S. Miller <>
Signed-off-by: Greg Kroah-Hartman <>