firefly-linux-kernel-4.4.55.git
14 years agoLinux 2.6.32.14
Greg Kroah-Hartman [Wed, 26 May 2010 21:29:57 +0000 (14:29 -0700)]
Linux 2.6.32.14

14 years agoRevert "parisc: Set PCI CLS early in boot."
Greg Kroah-Hartman [Mon, 24 May 2010 21:58:13 +0000 (14:58 -0700)]
Revert "parisc: Set PCI CLS early in boot."

This reverts the following patch, which shouldn't have been applied
to the .32 stable tree as it causes problems.

  commit 5fd4514bb351b5ecb0da3692fff70741e5ed200c upstream.

  Set the PCI CLS early in the boot process to prevent
  device failures. In pcibios_set_master use the new
  pci_cache_line_size instead of a hard-coded value.

Signed-off-by: Carlos O'Donell <carlos@codesourcery.com>
Reviewed-by: Grant Grundler <grundler@google.com>
Signed-off-by: Kyle McMartin <kyle@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agocrypto: authenc - Add EINPROGRESS check
Herbert Xu [Mon, 26 Apr 2010 01:14:05 +0000 (09:14 +0800)]
crypto: authenc - Add EINPROGRESS check

commit 180ce7e81030e1ef763d58f97f9ab840ff57d848 upstream.

When Steffen originally wrote the authenc async hash patch, he
correctly had EINPROGRESS checks in place so that we did not invoke
the original completion handler with it.

Unfortuantely I told him to remove it before the patch was applied.

As only MAY_BACKLOG request completion handlers are required to
handle EINPROGRESS completions, those checks are really needed.

This patch restores them.

Reported-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoRevert "ath9k: fix lockdep warning when unloading module" on stable kernels
Luis R. Rodriguez [Wed, 19 May 2010 21:33:49 +0000 (17:33 -0400)]
Revert "ath9k: fix lockdep warning when unloading module" on stable kernels

Johannes' patch 34e8950 titled:

mac80211: allow station add/remove to sleep

changed the way mac80211 adds and removes peers. The new
sta_add() / sta_remove() callbacks allowed the driver callbacks
to sleep. Johannes also ported ath9k to use sta_add() / sta_remove()
via the patch 4ca7786 titled:

ath9k: convert to new station add/remove callbacks

but this patch forgot to address a change in locking issue which
Ming Lei eventually found on his 2.6.33-wl #12 build. The 2.6.33-wl
build includes code for the 802.11 subsystem for 2.6.34 though so did
already have the above two patches (ath9k_sta_remove() on his trace),
the 2.6.33 kernel did not however have these two patches. Ming eventually
cured his lockdep warnign via the patch a9f042c titled:

ath9k: fix lockdep warning when unloading module

This went in to 2.6.34 and although it was not marked as a stable
fix it did get trickled down and applied on both 2.6.33 and 2.6.32.

In review, the culprits:

mac80211: allow station add/remove to sleep
git describe --contains 34e895075e21be3e21e71d6317440d1ee7969ad0
v2.6.34-rc1~233^2~49^2~107

ath9k: convert to new station add/remove callbacks
git describe --contains 4ca778605cfec53d8a689f0b57babb93b030c784
v2.6.34-rc1~233^2~49^2~10

ath9k: fix lockdep warning when unloading module

This last one trickled down to 2.6.33 (OK), 2.6.33 (invalid) and 2.6.32 (invalid).

git describe --contains a9f042cbe5284f34ccff15f3084477e11b39b17b
v2.6.34-rc2~48^2~77^2~7
git describe --contains 0524bcfa80f1fffb4e1fe18a0a28900869a58a7c
v2.6.33.2~125
git describe --contains 0dcc9985f34aef3c60bffab3dfc7f7ba3748f35a
v2.6.32.11~79

The patch titled "ath9k: fix lockdep warning when unloading module"
should be reverted on both 2.6.33 and 2.6.32 as it is invalid and
actually ended up causing the following warning:

ADDRCONF(NETDEV_CHANGE): wlan31: link becomes ready
phy0: WMM queue=2 aci=0 acm=0 aifs=3 cWmin=15 cWmax=1023 txop=0
phy0: WMM queue=3 aci=1 acm=0 aifs=7 cWmin=15 cWmax=1023 txop=0
phy0: WMM queue=1 aci=2 acm=0 aifs=2 cWmin=7 cWmax=15 txop=94
phy0: WMM queue=0 aci=3 acm=0 aifs=2 cWmin=3 cWmax=7 txop=47
phy0: device now idle
------------[ cut here ]------------
WARNING: at kernel/softirq.c:143 local_bh_enable_ip+0x7b/0xa0()
Hardware name: 7660A14
Modules linked in: ath9k(-) mac80211 ath cfg80211 <whatever-bleh-etc>
Pid: 2003, comm: rmmod Not tainted 2.6.32.11 #6
Call Trace:
 [<ffffffff8105d178>] warn_slowpath_common+0x78/0xb0
 [<ffffffff8105d1bf>] warn_slowpath_null+0xf/0x20
 [<ffffffff81063f8b>] local_bh_enable_ip+0x7b/0xa0
 [<ffffffff815121e4>] _spin_unlock_bh+0x14/0x20
 [<ffffffffa034aea5>] ath_tx_node_cleanup+0x185/0x1b0 [ath9k]
 [<ffffffffa0345597>] ath9k_sta_notify+0x57/0xb0 [ath9k]
 [<ffffffffa02ac51a>] __sta_info_unlink+0x15a/0x260 [mac80211]
 [<ffffffffa02ac658>] sta_info_unlink+0x38/0x60 [mac80211]
 [<ffffffffa02b3fbe>] ieee80211_set_disassoc+0x1ae/0x210 [mac80211]
 [<ffffffffa02b42d9>] ieee80211_mgd_deauth+0x109/0x110 [mac80211]
 [<ffffffffa02ba409>] ieee80211_deauth+0x19/0x20 [mac80211]
 [<ffffffffa028160e>] __cfg80211_mlme_deauth+0xee/0x130 [cfg80211]
 [<ffffffff81118540>] ? init_object+0x50/0x90
 [<ffffffffa0285429>] __cfg80211_disconnect+0x159/0x1d0 [cfg80211]
 [<ffffffffa027125f>] cfg80211_netdev_notifier_call+0x10f/0x450 [cfg80211]
 [<ffffffff81514ca7>] notifier_call_chain+0x47/0x90
 [<ffffffff8107f501>] raw_notifier_call_chain+0x11/0x20
 [<ffffffff81442d66>] call_netdevice_notifiers+0x16/0x20
 [<ffffffff8144352d>] dev_close+0x4d/0xa0
 [<ffffffff814439a8>] rollback_registered+0x48/0x120
 [<ffffffff81443a9d>] unregister_netdevice+0x1d/0x70
 [<ffffffffa02b6cc4>] ieee80211_remove_interfaces+0x84/0xc0 [mac80211]
 [<ffffffffa02aa072>] ieee80211_unregister_hw+0x42/0xf0 [mac80211]
 [<ffffffffa0347bde>] ath_detach+0x8e/0x180 [ath9k]
 [<ffffffffa0347ce1>] ath_cleanup+0x11/0x50 [ath9k]
 [<ffffffffa0351a2c>] ath_pci_remove+0x1c/0x20 [ath9k]
 [<ffffffff8129d712>] pci_device_remove+0x32/0x60
 [<ffffffff81332373>] __device_release_driver+0x53/0xb0
 [<ffffffff81332498>] driver_detach+0xc8/0xd0
 [<ffffffff81331405>] bus_remove_driver+0x85/0xe0
 [<ffffffff81332a5a>] driver_unregister+0x5a/0x90
 [<ffffffff8129da00>] pci_unregister_driver+0x40/0xb0
 [<ffffffffa03518d0>] ath_pci_exit+0x10/0x20 [ath9k]
 [<ffffffffa0353cd5>] ath9k_exit+0x9/0x2a [ath9k]
 [<ffffffff81092838>] sys_delete_module+0x1a8/0x270
 [<ffffffff8107ebe9>] ? up_read+0x9/0x10
 [<ffffffff81011f82>] system_call_fastpath+0x16/0x1b
---[ end trace fad957019ffdd40b ]---
phy0: Removed STA 00:22:6b:56:fd:e8
phy0: Destroyed STA 00:22:6b:56:fd:e8
wlan31: deauthenticating from 00:22:6b:56:fd:e8 by local choice (reason=3)
ath9k 0000:16:00.0: PCI INT A disabled

The original lockdep fixed an issue where due to the new changes
the driver was not disabling the bottom halves but it is incorrect
to do this on the older kernels since IRQs are already disabled.

Cc: Ming Lei <tom.leiming@gmail.com>
Cc: Johannes Berg <johannes@sipsolutions.net>
Cc: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agonilfs2: fix sync silent failure
Ryusuke Konishi [Mon, 3 May 2010 12:00:48 +0000 (21:00 +0900)]
nilfs2: fix sync silent failure

commit 973bec34bfc1bc2465646181653d67f767d418c8 upstream.

As of 32a88aa1, __sync_filesystem() will return 0 if s_bdi is not set.
And nilfs does not set s_bdi anywhere.  I noticed this problem by the
warning introduced by the recent commit 5129a469 ("Catch filesystem
lacking s_bdi").

 WARNING: at fs/super.c:959 vfs_kern_mount+0xc5/0x14e()
 Hardware name: PowerEdge 2850
 Modules linked in: nilfs2 loop tpm_tis tpm tpm_bios video shpchp pci_hotplug output dcdbas
 Pid: 3773, comm: mount.nilfs2 Not tainted 2.6.34-rc6-debug #38
 Call Trace:
  [<c1028422>] warn_slowpath_common+0x60/0x90
  [<c102845f>] warn_slowpath_null+0xd/0x10
  [<c1095936>] vfs_kern_mount+0xc5/0x14e
  [<c1095a03>] do_kern_mount+0x32/0xbd
  [<c10a811e>] do_mount+0x671/0x6d0
  [<c1073794>] ? __get_free_pages+0x1f/0x21
  [<c10a684f>] ? copy_mount_options+0x2b/0xe2
  [<c107b634>] ? strndup_user+0x48/0x67
  [<c10a81de>] sys_mount+0x61/0x8f
  [<c100280c>] sysenter_do_call+0x12/0x32

This ensures to set s_bdi for nilfs and fixes the sync silent failure.

Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Acked-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agommap_min_addr check CAP_SYS_RAWIO only for write
Kees Cook [Thu, 22 Apr 2010 19:19:17 +0000 (12:19 -0700)]
mmap_min_addr check CAP_SYS_RAWIO only for write

commit 4ae69e6b718589abe97c9625ccbb1e0bc95a8c0e upstream.

Redirecting directly to lsm, here's the patch discussed on lkml:
http://lkml.org/lkml/2010/4/22/219

The mmap_min_addr value is useful information for an admin to see without
being root ("is my system vulnerable to kernel NULL pointer attacks?") and
its setting is trivially easy for an attacker to determine by calling
mmap() in PAGE_SIZE increments starting at 0, so trying to keep it private
has no value.

Only require CAP_SYS_RAWIO if changing the value, not reading it.

Comment from Serge :

  Me, I like to write my passwords with light blue pen on dark blue
  paper, pasted on my window - if you're going to get my password, you're
  gonna get a headache.

Signed-off-by: Kees Cook <kees.cook@canonical.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: James Morris <jmorris@namei.org>
(cherry picked from commit 822cceec7248013821d655545ea45d1c6a9d15b3)
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agomegaraid_sas: fix for 32bit apps
Tomas Henzl [Thu, 11 Feb 2010 17:01:50 +0000 (18:01 +0100)]
megaraid_sas: fix for 32bit apps

commit b3dc1a212e5167984616445990c76056034f8eeb upstream.

It looks like this patch -

commit 7b2519afa1abd1b9f63aa1e90879307842422dae
Author: Yang, Bo <Bo.Yang@lsi.com>
Date:   Tue Oct 6 14:52:20 2009 -0600

    [SCSI] megaraid_sas: fix 64 bit sense pointer truncation

has caused a problem for 32bit programs with 64bit os -

http://bugzilla.kernel.org/show_bug.cgi?id=15001

fix by converting the user space 32bit pointer to a 64 bit one when
needed.

[jejb: fix up some 64 bit warnings]
Signed-off-by: Tomas Henzl <thenzl@redhat.com>
Cc: Bo Yang <Bo.Yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoCacheFiles: Fix error handling in cachefiles_determine_cache_security()
David Howells [Wed, 12 May 2010 14:34:03 +0000 (15:34 +0100)]
CacheFiles: Fix error handling in cachefiles_determine_cache_security()

commit 7ac512aa8237c43331ffaf77a4fd8b8d684819ba upstream.

cachefiles_determine_cache_security() is expected to return with a
security override in place.  However, if set_create_files_as() fails, we
fail to do this.  In this case, we should just reinstate the security
override that was set by the caller.

Furthermore, if set_create_files_as() fails, we should dispose of the
new credentials we were in the process of creating.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agop54: disable channels with incomplete calibration data sets
Christian Lamparter [Sat, 31 Oct 2009 21:59:27 +0000 (22:59 +0100)]
p54: disable channels with incomplete calibration data sets

commit 93a59d7527147e3656664aa3179f8d19de256081 upstream.

James Grossmann [1] reported that p54 spews out confusing
messages instead of preventing the mayhem from happening.

the reason is that "p54: generate channel list dynamically"
is not perfect. It didn't discard incomplete channel data
sets and therefore p54 advertised to support them as well.

[1]: http://marc.info/?l=linux-wireless&m=125699830215890

Cc: Larry Finger <Larry.Finger@lwfinger.net>
Reported-by: James Grossmann <cctsurf@gmail.com>
Signed-off-by: Christian Lamparter <chunkeey@web.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoiwlwifi: clear all the stop_queue flag after load firmware
Wey-Yi Guy [Tue, 9 Feb 2010 16:14:11 +0000 (08:14 -0800)]
iwlwifi: clear all the stop_queue flag after load firmware

commit a9e10fb9b1c6ad16e73cf2656951fce3a817611e upstream.

All the queues are awake and ready to use after loading firmware,
for firmware reload case, if any queues was stopped before
reload, mac80211 will wake those queues after restart hardware, so make
sure all the flag used to keep track of the queue status are
reset correctly.

Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agorevert "procfs: provide stack information for threads" and its fixup commits
Robin Holt [Tue, 11 May 2010 21:06:46 +0000 (14:06 -0700)]
revert "procfs: provide stack information for threads" and its fixup commits

commit 34441427aab4bdb3069a4ffcda69a99357abcb2e upstream.

Originally, commit d899bf7b ("procfs: provide stack information for
threads") attempted to introduce a new feature for showing where the
threadstack was located and how many pages are being utilized by the
stack.

Commit c44972f1 ("procfs: disable per-task stack usage on NOMMU") was
applied to fix the NO_MMU case.

Commit 89240ba0 ("x86, fs: Fix x86 procfs stack information for threads on
64-bit") was applied to fix a bug in ia32 executables being loaded.

Commit 9ebd4eba7 ("procfs: fix /proc/<pid>/stat stack pointer for kernel
threads") was applied to fix a bug which had kernel threads printing a
userland stack address.

Commit 1306d603f ('proc: partially revert "procfs: provide stack
information for threads"') was then applied to revert the stack pages
being used to solve a significant performance regression.

This patch nearly undoes the effect of all these patches.

The reason for reverting these is it provides an unusable value in
field 28.  For x86_64, a fork will result in the task->stack_start
value being updated to the current user top of stack and not the stack
start address.  This unpredictability of the stack_start value makes
it worthless.  That includes the intended use of showing how much stack
space a thread has.

Other architectures will get different values.  As an example, ia64
gets 0.  The do_fork() and copy_process() functions appear to treat the
stack_start and stack_size parameters as architecture specific.

I only partially reverted c44972f1 ("procfs: disable per-task stack usage
on NOMMU") .  If I had completely reverted it, I would have had to change
mm/Makefile only build pagewalk.o when CONFIG_PROC_PAGE_MONITOR is
configured.  Since I could not test the builds without significant effort,
I decided to not change mm/Makefile.

I only partially reverted 89240ba0 ("x86, fs: Fix x86 procfs stack
information for threads on 64-bit") .  I left the KSTK_ESP() change in
place as that seemed worthwhile.

Signed-off-by: Robin Holt <holt@sgi.com>
Cc: Stefani Seibold <stefani@seibold.net>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Michal Simek <monstr@monstr.eu>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoproc: partially revert "procfs: provide stack information for threads"
KOSAKI Motohiro [Fri, 8 Jan 2010 22:42:56 +0000 (14:42 -0800)]
proc: partially revert "procfs: provide stack information for threads"

commit 1306d603fcf1f6682f8575d1ff23631a24184b21 upstream.

Commit d899bf7b (procfs: provide stack information for threads) introduced
to show stack information in /proc/{pid}/status.  But it cause large
performance regression.  Unfortunately /proc/{pid}/status is used ps
command too and ps is one of most important component.  Because both to
take mmap_sem and page table walk are heavily operation.

If many process run, the ps performance is,

[before d899bf7b]

% perf stat ps >/dev/null

 Performance counter stats for 'ps':

     4090.435806  task-clock-msecs         #      0.032 CPUs
             229  context-switches         #      0.000 M/sec
               0  CPU-migrations           #      0.000 M/sec
             234  page-faults              #      0.000 M/sec
      8587565207  cycles                   #   2099.425 M/sec
      9866662403  instructions             #      1.149 IPC
      3789415411  cache-references         #    926.409 M/sec
        30419509  cache-misses             #      7.437 M/sec

   128.859521955  seconds time elapsed

[after d899bf7b]

% perf stat  ps  > /dev/null

 Performance counter stats for 'ps':

     4305.081146  task-clock-msecs         #      0.028 CPUs
             480  context-switches         #      0.000 M/sec
               2  CPU-migrations           #      0.000 M/sec
             237  page-faults              #      0.000 M/sec
      9021211334  cycles                   #   2095.480 M/sec
     10605887536  instructions             #      1.176 IPC
      3612650999  cache-references         #    839.160 M/sec
        23917502  cache-misses             #      5.556 M/sec

   152.277819582  seconds time elapsed

Thus, this patch revert it. Fortunately /proc/{pid}/task/{tid}/smaps
provide almost same information. we can use it.

Commit d899bf7b introduced two features:

 1) Add the annotattion of [thread stack: xxxx] mark to
    /proc/{pid}/task/{tid}/maps.
 2) Add StackUsage field to /proc/{pid}/status.

I only revert (2), because I haven't seen (1) cause regression.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Stefani Seibold <stefani@seibold.net>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: hda - New Intel HDA controller
Vitaliy Kulikov [Mon, 15 Mar 2010 08:01:26 +0000 (09:01 +0100)]
ALSA: hda - New Intel HDA controller

commit c602c8ad45d6ee6ad91fc544513cc96f70790983 upstream.

Added a PCI controller id on new Dell laptops.

Signed-off-by: Vitaliy Kulikov <Vitaliy.Kulikov@idt.com>
Cc: AmenophisIII <AmenophisIII@gmx.at>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoBtrfs: check for read permission on src file in the clone ioctl
Dan Rosenberg [Sat, 15 May 2010 15:27:37 +0000 (11:27 -0400)]
Btrfs: check for read permission on src file in the clone ioctl

commit 5dc6416414fb3ec6e2825fd4d20c8bf1d7fe0395 upstream.

The existing code would have allowed you to clone a file that was
only open for writing

Signed-off-by: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agox86, amd: Check X86_FEATURE_OSVW bit before accessing OSVW MSRs
Andreas Herrmann [Tue, 27 Apr 2010 10:13:48 +0000 (12:13 +0200)]
x86, amd: Check X86_FEATURE_OSVW bit before accessing OSVW MSRs

commit f01487119dda3d9f58c9729c7361ecc50a61c188 upstream.

If host CPU is exposed to a guest the OSVW MSRs are not guaranteed
to be present and a GP fault occurs. Thus checking the feature flag is
essential.

Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
LKML-Reference: <20100427101348.GC4489@alberich.amd.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agox86, cacheinfo: Turn off L3 cache index disable feature in virtualized environments
Frank Arnold [Thu, 22 Apr 2010 14:06:59 +0000 (16:06 +0200)]
x86, cacheinfo: Turn off L3 cache index disable feature in virtualized environments

commit 7f284d3cc96e02468a42e045f77af11e5ff8b095 upstream.

When running a quest kernel on xen we get:

BUG: unable to handle kernel NULL pointer dereference at 0000000000000038
IP: [<ffffffff8142f2fb>] cpuid4_cache_lookup_regs+0x2ca/0x3df
PGD 0
Oops: 0000 [#1] SMP
last sysfs file:
CPU 0
Modules linked in:

Pid: 0, comm: swapper Tainted: G        W  2.6.34-rc3 #1 /HVM domU
RIP: 0010:[<ffffffff8142f2fb>]  [<ffffffff8142f2fb>] cpuid4_cache_lookup_regs+0x
2ca/0x3df
RSP: 0018:ffff880002203e08  EFLAGS: 00010046
RAX: 0000000000000000 RBX: 0000000000000003 RCX: 0000000000000060
RDX: 0000000000000000 RSI: 0000000000000040 RDI: 0000000000000000
RBP: ffff880002203ed8 R08: 00000000000017c0 R09: ffff880002203e38
R10: ffff8800023d5d40 R11: ffffffff81a01e28 R12: ffff880187e6f5c0
R13: ffff880002203e34 R14: ffff880002203e58 R15: ffff880002203e68
FS:  0000000000000000(0000) GS:ffff880002200000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 0000000000000038 CR3: 0000000001a3c000 CR4: 00000000000006f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process swapper (pid: 0, threadinfo ffffffff81a00000, task ffffffff81a44020)
Stack:
 ffffffff810d7ecb ffff880002203e20 ffffffff81059140 ffff880002203e30
<0> ffffffff810d7ec9 0000000002203e40 000000000050d140 ffff880002203e70
<0> 0000000002008140 0000000000000086 ffff880040020140 ffffffff81068b8b
Call Trace:
 <IRQ>
 [<ffffffff810d7ecb>] ? sync_supers_timer_fn+0x0/0x1c
 [<ffffffff81059140>] ? mod_timer+0x23/0x25
 [<ffffffff810d7ec9>] ? arm_supers_timer+0x34/0x36
 [<ffffffff81068b8b>] ? hrtimer_get_next_event+0xa7/0xc3
 [<ffffffff81058e85>] ? get_next_timer_interrupt+0x19a/0x20d
 [<ffffffff8142fa23>] get_cpu_leaves+0x5c/0x232
 [<ffffffff8106a7b1>] ? sched_clock_local+0x1c/0x82
 [<ffffffff8106a9a0>] ? sched_clock_tick+0x75/0x7a
 [<ffffffff8107748c>] generic_smp_call_function_single_interrupt+0xae/0xd0
 [<ffffffff8101f6ef>] smp_call_function_single_interrupt+0x18/0x27
 [<ffffffff8100a773>] call_function_single_interrupt+0x13/0x20
 <EOI>
 [<ffffffff8143c468>] ? notifier_call_chain+0x14/0x63
 [<ffffffff810295c6>] ? native_safe_halt+0xc/0xd
 [<ffffffff810114eb>] ? default_idle+0x36/0x53
 [<ffffffff81008c22>] cpu_idle+0xaa/0xe4
 [<ffffffff81423a9a>] rest_init+0x7e/0x80
 [<ffffffff81b10dd2>] start_kernel+0x40e/0x419
 [<ffffffff81b102c8>] x86_64_start_reservations+0xb3/0xb7
 [<ffffffff81b103c4>] x86_64_start_kernel+0xf8/0x107
Code: 14 d5 40 ff ae 81 8b 14 02 31 c0 3b 15 47 1c 8b 00 7d 0e 48 8b 05 36 1c 8b
 00 48 63 d2 48 8b 04 d0 c7 85 5c ff ff ff 00 00 00 00 <8b> 70 38 48 8d 8d 5c ff
 ff ff 48 8b 78 10 ba c4 01 00 00 e8 eb
RIP  [<ffffffff8142f2fb>] cpuid4_cache_lookup_regs+0x2ca/0x3df
 RSP <ffff880002203e08>
CR2: 0000000000000038
---[ end trace a7919e7f17c0a726 ]---

The L3 cache index disable feature of AMD CPUs has to be disabled if the
kernel is running as guest on top of a hypervisor because northbridge
devices are not available to the guest. Currently, this fixes a boot
crash on top of Xen. In the future this will become an issue on KVM as
well.

Check if northbridge devices are present and do not enable the feature
if there are none.

[ hpa: backported to 2.6.34 ]

Signed-off-by: Frank Arnold <frank.arnold@amd.com>
LKML-Reference: <1271945222-5283-3-git-send-email-bp@amd64.org>
Acked-by: Borislav Petkov <borislav.petkov@amd.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agox86, k8: Fix build error when K8_NB is disabled
Borislav Petkov [Sat, 24 Apr 2010 07:56:53 +0000 (09:56 +0200)]
x86, k8: Fix build error when K8_NB is disabled

commit ade029e2aaacc8965a548b0b0f80c5bee97ffc68 upstream.

K8_NB depends on PCI and when the last is disabled (allnoconfig) we fail
at the final linking stage due to missing exported num_k8_northbridges.
Add a header stub for that.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
LKML-Reference: <20100503183036.GJ26107@aftab>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoprofile: fix stats and data leakage
Hugh Dickins [Sat, 15 May 2010 02:44:10 +0000 (19:44 -0700)]
profile: fix stats and data leakage

commit 16a2164bb03612efe79a76c73da6da44445b9287 upstream.

If the kernel is large or the profiling step small, /proc/profile
leaks data and readprofile shows silly stats, until readprofile -r
has reset the buffer: clear the prof_buffer when it is vmalloc()ed.

Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoinotify: don't leak user struct on inotify release
Pavel Emelyanov [Wed, 12 May 2010 22:34:07 +0000 (15:34 -0700)]
inotify: don't leak user struct on inotify release

commit b3b38d842fa367d862b83e7670af4e0fd6a80fc0 upstream.

inotify_new_group() receives a get_uid-ed user_struct and saves the
reference on group->inotify_data.user.  The problem is that free_uid() is
never called on it.

Issue seem to be introduced by 63c882a0 (inotify: reimplement inotify
using fsnotify) after 2.6.30.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Eric Paris <eparis@parisplace.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Eric Paris <eparis@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoinotify: race use after free/double free in inotify inode marks
Eric Paris [Tue, 11 May 2010 21:17:40 +0000 (17:17 -0400)]
inotify: race use after free/double free in inotify inode marks

commit e08733446e72b983fed850fc5d8bd21b386feb29 upstream.

There is a race in the inotify add/rm watch code.  A task can find and
remove a mark which doesn't have all of it's references.  This can
result in a use after free/double free situation.

Task A Task B
------------ -----------
inotify_new_watch()
 allocate a mark (refcnt == 1)
 add it to the idr
inotify_rm_watch()
 inotify_remove_from_idr()
  fsnotify_put_mark()
      refcnt hits 0, free
 take reference because we are on idr
 [at this point it is a use after free]
 [time goes on]
 refcnt may hit 0 again, double free

The fix is to take the reference BEFORE the object can be found in the
idr.

Signed-off-by: Eric Paris <eparis@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: hda: Fix 0 dB for Lenovo models using Conexant CX20549 (Venice)
Daniel T Chen [Mon, 10 May 2010 19:50:04 +0000 (21:50 +0200)]
ALSA: hda: Fix 0 dB for Lenovo models using Conexant CX20549 (Venice)

commit 0ebf9e3692d640917fb792a7494d05e1f5b1058f upstream.

Reference: http://mailman.alsa-project.org/pipermail/alsa-devel/2010-May/027525.html

As reported on the mailing list, we also need to cap to the 0 dB offset
for Lenovo models, else the sound will be distorted.

Reported-and-Tested-by: Tim Starling <tstarling@wikimedia.org>
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: ice1724 - Fix ESI Maya44 capture source control
Takashi Iwai [Wed, 12 May 2010 14:43:32 +0000 (16:43 +0200)]
ALSA: ice1724 - Fix ESI Maya44 capture source control

commit 8213466596bf10b75887754773ee13c10cf86f5c upstream.

The capture source control of maya44 was wrongly coded with the bit
shift instead of the bit mask.  Also, the slot for line-in was
wrongly assigned (slot 5 instead of 4).

Reported-by: Alex Chernyshoff <alexdsp@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoserial: imx.c: fix CTS trigger level lower to avoid lost chars
Valentin Longchamp [Wed, 5 May 2010 09:47:07 +0000 (11:47 +0200)]
serial: imx.c: fix CTS trigger level lower to avoid lost chars

commit 1c5250d6163dac28be3afabdfb6c723f107051b7 upstream.

The imx CTS trigger level is left at its reset value that is 32
chars. Since the RX FIFO has 32 entries, when CTS is raised, the
FIFO already is full. However, some serial port devices first empty
their TX FIFO before stopping when CTS is raised, resulting in lost
chars.

This patch sets the trigger level lower so that other chars arrive
after CTS is raised, there is still room for 16 of them.

Signed-off-by: Valentin Longchamp<valentin.longchamp@epfl.ch>
Tested-by: Philippe Rétornaz<philippe.retornaz@epfl.ch>
Acked-by: Wolfram Sang<w.sang@pengutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agocifs: guard against hardlinking directories
Jeff Layton [Tue, 11 May 2010 18:59:55 +0000 (14:59 -0400)]
cifs: guard against hardlinking directories

commit 3d69438031b00c601c991ab447cafb7d5c3c59a6 upstream.

When we made serverino the default, we trusted that the field sent by the
server in the "uniqueid" field was actually unique. It turns out that it
isn't reliably so.

Samba, in particular, will just put the st_ino in the uniqueid field when
unix extensions are enabled. When a share spans multiple filesystems, it's
quite possible that there will be collisions. This is a server bug, but
when the inodes in question are a directory (as is often the case) and
there is a collision with the root inode of the mount, the result is a
kernel panic on umount.

Fix this by checking explicitly for directory inodes with the same
uniqueid. If that is the case, then we can assume that using server inode
numbers will be a problem and that they should be disabled.

Fixes Samba bugzilla 7407

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Reviewed-and-Tested-by: Suresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: Steve French <sfrench@us.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agopowerpc/perf_event: Fix oops due to perf_event_do_pending call
Paul Mackerras [Tue, 13 Apr 2010 20:46:04 +0000 (20:46 +0000)]
powerpc/perf_event: Fix oops due to perf_event_do_pending call

commit 0fe1ac48bef018bed896307cd12f6ca9b5e704ab upstream.

Anton Blanchard found that large POWER systems would occasionally
crash in the exception exit path when profiling with perf_events.
The symptom was that an interrupt would occur late in the exit path
when the MSR[RI] (recoverable interrupt) bit was clear.  Interrupts
should be hard-disabled at this point but they were enabled.  Because
the interrupt was not recoverable the system panicked.

The reason is that the exception exit path was calling
perf_event_do_pending after hard-disabling interrupts, and
perf_event_do_pending will re-enable interrupts.

The simplest and cleanest fix for this is to use the same mechanism
that 32-bit powerpc does, namely to cause a self-IPI by setting the
decrementer to 1.  This means we can remove the tests in the exception
exit path and raw_local_irq_restore.

This also makes sure that the call to perf_event_do_pending from
timer_interrupt() happens within irq_enter/irq_exit.  (Note that
calling perf_event_do_pending from timer_interrupt does not mean that
there is a possible 1/HZ latency; setting the decrementer to 1 ensures
that the timer interrupt will happen immediately, i.e. within one
timebase tick, which is a few nanoseconds or 10s of nanoseconds.)

Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoptrace: fix return value of do_syscall_trace_enter()
Gerald Schaefer [Wed, 12 May 2010 07:32:12 +0000 (09:32 +0200)]
ptrace: fix return value of do_syscall_trace_enter()

commit 545c174d1f093a462b4bb9131b23d5ea72a600e1 upstream.

strace may change the system call number, so regs->gprs[2] must not
be read before tracehook_report_syscall_entry(). This fixes a bug
where "strace -f" will hang after a vfork().

Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agommc: atmel-mci: remove data error interrupt after xfer
Nicolas Ferre [Tue, 11 May 2010 21:06:50 +0000 (14:06 -0700)]
mmc: atmel-mci: remove data error interrupt after xfer

commit abc2c9fdf636c4335a8d72ac3c5ae152bca44b68 upstream.

Disable data error interrupts while we are actually recording that there
is not such errors.  This will prevent, in some cases, the warning message
printed at new request queuing (in atmci_start_request()).

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: <linux-mmc@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agommc: atmel-mci: prevent kernel oops while removing card
Nicolas Ferre [Tue, 11 May 2010 21:06:49 +0000 (14:06 -0700)]
mmc: atmel-mci: prevent kernel oops while removing card

commit 009a891b22395fc86e5f34057d79fffee4509ab5 upstream.

The removing of an SD card in certain circumstances can lead to a kernel
oops if we do not make sure that the "data" field of the host structure is
valid.  This patch adds a test in atmci_dma_cleanup() function and also
calls atmci_stop_dma() before throwing away the reference to data.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: <linux-mmc@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agommc: atmel-mci: fix two parameters swapped
Nicolas Ferre [Tue, 11 May 2010 21:06:48 +0000 (14:06 -0700)]
mmc: atmel-mci: fix two parameters swapped

commit ebb1fea9b3adf25d7e2f643c614163af4f93a17f upstream.

Two parameters were swapped in the calls to atmci_init_slot().

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Reported-by: Anders Grahn <anders.grahn@hd-wireless.se>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: <linux-mmc@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoACPI: sleep: eliminate duplicate entries in acpisleep_dmi_table[]
Alex Chiang [Tue, 11 May 2010 16:21:38 +0000 (10:21 -0600)]
ACPI: sleep: eliminate duplicate entries in acpisleep_dmi_table[]

commit 7d6fb7bd1919517937ec390f6ca2d7bcf4f89fb6 upstream.

Duplicate entries ended up acpisleep_dmi_table[] by accident.
They don't hurt functionality, but they are ugly, so let's get
rid of them.

Signed-off-by: Alex Chiang <achiang@canonical.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agodma-mapping: fix dma_sync_single_range_*
FUJITA Tomonori [Tue, 11 May 2010 21:06:43 +0000 (14:06 -0700)]
dma-mapping: fix dma_sync_single_range_*

commit f33d7e2d2d113a63772bbc993cdec3b5327f0ef1 upstream.

dma_sync_single_range_for_cpu() and dma_sync_single_range_for_device() use
a wrong address with a partial synchronization.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agohugetlbfs: kill applications that use MAP_NORESERVE with SIGBUS instead of OOM-killer
Mel Gorman [Tue, 11 May 2010 21:06:53 +0000 (14:06 -0700)]
hugetlbfs: kill applications that use MAP_NORESERVE with SIGBUS instead of OOM-killer

commit 4a6018f7f4f1075c1a5403b5ec0ee7262187b86c upstream.

Ordinarily, application using hugetlbfs will create mappings with
reserves.  For shared mappings, these pages are reserved before mmap()
returns success and for private mappings, the caller process is guaranteed
and a child process that cannot get the pages gets killed with sigbus.

An application that uses MAP_NORESERVE gets no reservations and mmap()
will always succeed at the risk the page will not be available at fault
time.  This might be used for example on very large sparse mappings where
the developer is confident the necessary huge pages exist to satisfy all
faults even though the whole mapping cannot be backed by huge pages.
Unfortunately, if an allocation does fail, VM_FAULT_OOM is returned to the
fault handler which proceeds to trigger the OOM-killer.  This is
unhelpful.

Even without hugetlbfs mounted, a user using mmap() can trivially trigger
the OOM-killer because VM_FAULT_OOM is returned (will provide example
program if desired - it's a whopping 24 lines long).  It could be
considered a DOS available to an unprivileged user.

This patch alters hugetlbfs to kill a process that uses MAP_NORESERVE
where huge pages were not available with SIGBUS instead of triggering the
OOM killer.

This change affects hugetlb_cow() as well.  I feel there is a failure case
in there, but I didn't create one.  It would need a fairly specific target
in terms of the faulting application and the hugepage pool size.  The
hugetlb_no_page() path is much easier to hit but both might as well be
closed.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agofbdev: bfin-t350mcqb-fb: fix fbmem allocation with blanking lines
Michael Hennerich [Tue, 11 May 2010 21:07:00 +0000 (14:07 -0700)]
fbdev: bfin-t350mcqb-fb: fix fbmem allocation with blanking lines

commit de145b44b95b9d3212a82d1c0f29b09778ef33c5 upstream.

The current allocation does not include the memory required for blanking
lines.  So avoid memory corruption when multiple devices are using the DMA
memory near each other.

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agohp_accel: fix race in device removal
Oliver Neukum [Tue, 11 May 2010 21:07:03 +0000 (14:07 -0700)]
hp_accel: fix race in device removal

commit 06efbeb4a47b6f865e1c9d175ab9d6e90b69ae9e upstream.

The work queue has to be flushed after the device has been made
inaccessible.  The patch closes a window during which a work queue might
remain active after the device is removed and would then lead to ACPI
calls with undefined behavior.

Signed-off-by: Oliver Neukum <oneukum@suse.de>
Acked-by: Eric Piel <eric.piel@tremplin-utc.net>
Acked-by: Pavel Machek <pavel@ucw.cz>
Cc: Pavel Herrmann <morpheus.ibis@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoipv4: udp: fix short packet and bad checksum logging
Bjørn Mork [Thu, 6 May 2010 03:44:34 +0000 (03:44 +0000)]
ipv4: udp: fix short packet and bad checksum logging

commit ccc2d97cb7c798e785c9f198de243e2b59f7073b upstream.

commit 2783ef23 moved the initialisation of saddr and daddr after
pskb_may_pull() to avoid a potential data corruption.  Unfortunately
also placing it after the short packet and bad checksum error paths,
where these variables are used for logging.  The result is bogus
output like

[92238.389505] UDP: short packet: From 2.0.0.0:65535 23715/178 to 0.0.0.0:65535

Moving the saddr and daddr initialisation above the error paths, while still
keeping it after the pskb_may_pull() to keep the fix from commit 2783ef23.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoRevert "module: fix __module_ref_addr()"
Greg Kroah-Hartman [Wed, 12 May 2010 22:11:42 +0000 (15:11 -0700)]
Revert "module: fix __module_ref_addr()"

This reverts commit d150a2b96558a7349cbf3a72a279c37bc67d50fb.

Thanks to Jiri Benc for finding the problem that this patch is
not correct for the 2.6.32-stable series.

Cc: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoLinux 2.6.32.13
Greg Kroah-Hartman [Wed, 12 May 2010 21:59:32 +0000 (14:59 -0700)]
Linux 2.6.32.13

14 years agoMIPS: Sibyte: Apply M3 workaround only on affected chip types and versions.
Ralf Baechle [Fri, 23 Apr 2010 01:56:38 +0000 (02:56 +0100)]
MIPS: Sibyte: Apply M3 workaround only on affected chip types and versions.

(cherry picked from commit e65c7f33d75e977350ca350573d93c517ec02776)

Previously it was unconditionally used on all Sibyte family SOCs.  The
M3 bug has to be handled in the TLB exception handler which is extremly
performance sensitive, so this modification is expected to deliver around
2-3% performance improvment.  This is important as required changes to the
M3 workaround will make it more costly.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoSCSI: Retry commands with UNIT_ATTENTION sense codes to fix ext3/ext4 I/O error
James Bottomley [Tue, 4 May 2010 20:51:40 +0000 (16:51 -0400)]
SCSI: Retry commands with UNIT_ATTENTION sense codes to fix ext3/ext4 I/O error

commit 77a4229719e511a0d38d9c355317ae1469adeb54 upstream.

There's nastyness in the way we currently handle barriers (and
discards): They're effectively filesystem commands, but they get
processed as BLOCK_PC commands.  Unfortunately BLOCK_PC commands are
taken by SCSI to be SG_IO commands and the issuer expects to see and
handle any returned errors, however trivial.  This leads to a huge
problem, because the block layer doesn't expect this to happen and any
trivially retryable error on a barrier causes an immediate I/O error
to the filesystem.

The only real way to hack around this is to take the usual class of
offending errors (unit attentions) and make them all retryable in the
case of a REQ_HARDBARRIER.  A correct fix would involve a rework of
the entire block and SCSI submit system, and so is out of scope for a
quick fix.

Cc: Hannes Reinecke <hare@suse.de>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoEnable retries for SYNCRONIZE_CACHE commands to fix I/O error
Hannes Reinecke [Tue, 4 May 2010 14:49:21 +0000 (16:49 +0200)]
Enable retries for SYNCRONIZE_CACHE commands to fix I/O error

commit c213e1407be6b04b144794399a91472e0ef92aec upstream.

Some arrays are giving I/O errors with ext3 filesystems when
SYNCHRONIZE_CACHE gets a UNIT_ATTENTION.  What is happening is that
these commands have no retries, so the UNIT_ATTENTION causes the
barrier to fail.  We should be enable retries here to clear any
transient error and allow the barrier to succeed.

Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoscsi_debug: virtual_gb ignores sector_size
Douglas Gilbert [Sun, 25 Apr 2010 10:30:23 +0000 (12:30 +0200)]
scsi_debug: virtual_gb ignores sector_size

commit 5447ed6c968e7270b656afa273c2b79d15d82edd upstream.

In the scsi_debug driver, the virtual_gb option ignores the
sector_size, implicitly assuming that is 512 bytes.  So if
'virtual_gb=1 sector_size=4096' the result is an 8 GB (virtual) disk.

Signed-off-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoSCSI: libiscsi: regression: fix header digest errors
Mike Christie [Sat, 24 Apr 2010 21:21:19 +0000 (16:21 -0500)]
SCSI: libiscsi: regression: fix header digest errors

commit 96b1f96dcab87756c0a1e7ba76bc5dc2add82b88 upstream.

This fixes a regression introduced with this commit:

commit d3305f3407fa3e9452079ec6cc8379067456e4aa
Author: Mike Christie <michaelc@cs.wisc.edu>
Date:   Thu Aug 20 15:10:58 2009 -0500

    [SCSI] libiscsi: don't increment cmdsn if cmd is not sent

in 2.6.32.

When I moved the hdr->cmdsn after init_task, I added
a bug when header digests are used. The problem is
that the LLD may calculate the header digest in init_task,
so if we then set the cmdsn after the init_task call we
change what the digest will be calculated by the target.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoSCSI: fix locking around blk_abort_request()
Tejun Heo [Thu, 15 Apr 2010 00:00:08 +0000 (09:00 +0900)]
SCSI: fix locking around blk_abort_request()

commit 70b25f890ce9f0520c64075ce9225a5b020a513e upstream.

blk_abort_request() expects queue lock to be held by the caller.
Grab it before calling the function.

Lack of this synchronization led to infinite loop on corrupt
q->timeout_list.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agopxa/colibri: fix missing #include <mach/mfp.h> in colibri.h
Jakob Viketoft [Wed, 5 May 2010 10:25:27 +0000 (18:25 +0800)]
pxa/colibri: fix missing #include <mach/mfp.h> in colibri.h

commit ccb8d8d070b8f25f0163da5c9ceacf63a5169540 upstream.

The use of mfp_cfg_t causes build errors without including <mach/mfp.h>.

CC: Daniel Mack <daniel@caiaq.de>
Signed-off-by: Jakob Viketoft <jakob.viketoft@bitsim.com>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agocpuidle: Fix incorrect optimization
Arjan van de Ven [Sat, 8 May 2010 22:47:37 +0000 (15:47 -0700)]
cpuidle: Fix incorrect optimization

commit 1c6fe0364fa7bf28248488753ee0afb6b759cd04 upstream.

commit 672917dcc78 ("cpuidle: menu governor: reduce latency on exit")
added an optimization, where the analysis on the past idle period moved
from the end of idle, to the beginning of the new idle.

Unfortunately, this optimization had a bug where it zeroed one key
variable for new use, that is needed for the analysis.  The fix is
simple, zero the variable after doing the work from the previous idle.

During the audit of the code that found this issue, another issue was
also found; the ->measured_us data structure member is never set, a
local variable is always used instead.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoACPI: sleep: init_set_sci_en_on_resume for Dell Studio 155x
Kamal Mostafa [Tue, 27 Apr 2010 21:02:40 +0000 (14:02 -0700)]
ACPI: sleep: init_set_sci_en_on_resume for Dell Studio 155x

commit ea5bc73f4f56449b2d450068d492bcd17a675d7a upstream.

Add Dell Studio models (1558, 1557, 1555) to the 'set_sci_en_on_resume'
list to fix hang on resume.

BugLink: http://bugs.launchpad.net/bugs/553498
Signed-off-by: Kamal Mostafa <kamal@canonical.com>
Acked-by: Alex Chiang <achiang@canonical.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agopower_meter: acpi_device_class "power_meter_resource" too long
Dan Carpenter [Tue, 27 Apr 2010 21:01:07 +0000 (14:01 -0700)]
power_meter: acpi_device_class "power_meter_resource" too long

commit 18262714ca0fb65c290b8ea1807b2b02bb52d0e3 upstream.

acpi_device_class can only be 19 characters and a NULL terminator.

The current code has a buffer overflow in acpi_power_meter_add():
       strcpy(acpi_device_class(device), ACPI_POWER_METER_CLASS);

Signed-off-by: Dan Carpenter <error27@gmail.com>
Cc: Len Brown <lenb@kernel.org>
Cc: "Darrick J. Wong" <djwong@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoACPI: DMI init_set_sci_en_on_resume for multiple Lenovo ThinkPads
Alex Chiang [Tue, 20 Apr 2010 14:03:14 +0000 (08:03 -0600)]
ACPI: DMI init_set_sci_en_on_resume for multiple Lenovo ThinkPads

commit 07bedca29b0973f36a6b6db36936deed367164ed upstream.

Multiple Lenovo ThinkPad models with Intel Core i5/i7 CPUs can
successfully suspend/resume once, and then hang on the second s/r
cycle.

We got confirmation that this was due to a BIOS defect. The BIOS
did not properly set SCI_EN coming out of S3. The BIOS guys
hinted that The Other Leading OS ignores the fact that hardware
owns the bit and sets it manually.

In any case, an existing DMI table exists for machines where this
defect is a known problem. Lenovo promise to fix their BIOS, but
for folks who either won't or can't upgrade their BIOS, allow
Linux to workaround the issue.

https://bugzilla.kernel.org/show_bug.cgi?id=15407
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/532374

Confirmed by numerous testers in the launchpad bug that using
acpi_sleep=sci_force_enable fixes the issue. We add the machines
to acpisleep_dmi_table[] to automatically enable this workaround.

Cc: Colin King <colin.king@canonical.com>
Signed-off-by: Alex Chiang <achiang@canonical.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoV4L/DVB: budget: Oops: "BUG: unable to handle kernel NULL pointer dereference"
Bjørn Mork [Wed, 24 Mar 2010 10:57:57 +0000 (07:57 -0300)]
V4L/DVB: budget: Oops: "BUG: unable to handle kernel NULL pointer dereference"

commit 6f550dc08369ee0bc6402963c377e65f0f2e3b71 upstream.

Never call dvb_frontend_detach if we failed to attach a frontend. This fixes
the following oops, which will be triggered by a missing stv090x module:

[    8.172997] DVB: registering new adapter (TT-Budget S2-1600 PCI)
[    8.209018] adapter has MAC addr = 00:d0:5c:cc:a7:29
[    8.328665] Intel ICH 0000:00:1f.5: PCI INT B -> GSI 17 (level, low) -> IRQ 17
[    8.328753] Intel ICH 0000:00:1f.5: setting latency timer to 64
[    8.562047] DVB: Unable to find symbol stv090x_attach()
[    8.562117] BUG: unable to handle kernel NULL pointer dereference at 000000ac
[    8.562239] IP: [<e08b04a3>] dvb_frontend_detach+0x4/0x67 [dvb_core]

Ref http://bugs.debian.org/575207

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agomd/raid6: Fix raid-6 read-error correction in degraded state
Gabriele A. Trombetti [Wed, 28 Apr 2010 01:51:17 +0000 (11:51 +1000)]
md/raid6: Fix raid-6 read-error correction in degraded state

commit 87aa63000c484bfb9909989316f615240dfee018 upstream.

Fix: Raid-6 was not trying to correct a read-error when in
singly-degraded state and was instead dropping one more device, going to
doubly-degraded state. This patch fixes this behaviour.

Tested-by: Janos Haar <janos.haar@netcenter.hu>
Signed-off-by: Gabriele A. Trombetti <g.trombetti.lkrnl1213@logicschema.com>
Reported-by: Janos Haar <janos.haar@netcenter.hu>
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agovirtio: initialize earlier
Stijn Tintel [Fri, 7 May 2010 04:58:34 +0000 (14:28 +0930)]
virtio: initialize earlier

commit e2dbe06c271f3bb2a495627980aad3d1d8ccef2a upstream.

Move initialization of the virtio framework before the initialization of
mtd, so that block2mtd can be used on virtio-based block devices.

Addresses https://bugzilla.kernel.org/show_bug.cgi?id=15644

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agomd: restore ability of spare drives to spin down.
NeilBrown [Fri, 7 May 2010 09:44:26 +0000 (19:44 +1000)]
md: restore ability of spare drives to spin down.

commit 1176568de7e066c0be9e46c37503b9fd4730edcf upstream.

Some time ago we stopped the clean/active metadata updates
from being written to a 'spare' device in most cases so that
it could spin down and say spun down.  Device failure/removal
etc are still recorded on spares.

However commit 51d5668cb2e3fd1827a55 broke this 50% of the time,
depending on whether the event count is even or odd.
The change log entry said:

   This means that the alignment between 'odd/even' and
    'clean/dirty' might take a little longer to attain,

how ever the code makes no attempt to create that alignment, so it
could take arbitrarily long.

So when we find that clean/dirty is not aligned with odd/even,
force a second metadata-update immediately.  There are already cases
where a second metadata-update is needed immediately (e.g. when a
device fails during the metadata update).  We just piggy-back on that.

Reported-by: Joe Bryant <tenminjoe@yahoo.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agosecurity: testing the wrong variable in create_by_name()
Dan Carpenter [Thu, 22 Apr 2010 10:05:35 +0000 (12:05 +0200)]
security: testing the wrong variable in create_by_name()

commit b338cc8207eae46640a8d534738fda7b5e48511d upstream.

There is a typo here.  We should be testing "*dentry" instead of
"dentry".  If "*dentry" is an ERR_PTR, it gets dereferenced in either
mkdir() or create() which would cause an OOPs.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: James Morris <jmorris@namei.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agotracing: Fix ftrace_event_call alignment for use with gcc 4.5
Jeff Mahoney [Wed, 24 Feb 2010 18:59:23 +0000 (13:59 -0500)]
tracing: Fix ftrace_event_call alignment for use with gcc 4.5

commit 86c38a31aa7f2dd6e74a262710bf8ebf7455acc5 upstream.

GCC 4.5 introduces behavior that forces the alignment of structures to
 use the largest possible value. The default value is 32 bytes, so if
 some structures are defined with a 4-byte alignment and others aren't
 declared with an alignment constraint at all - it will align at 32-bytes.

 For things like the ftrace events, this results in a non-standard array.
 When initializing the ftrace subsystem, we traverse the _ftrace_events
 section and call the initialization callback for each event. When the
 structures are misaligned, we could be treating another part of the
 structure (or the zeroed out space between them) as a function pointer.

 This patch forces the alignment for all the ftrace_event_call structures
 to 4 bytes.

 Without this patch, the kernel fails to boot very early when built with
 gcc 4.5.

 It's trivial to check the alignment of the members of the array, so it
 might be worthwhile to add something to the build system to do that
 automatically. Unfortunately, that only covers this case. I've asked one
 of the gcc developers about adding a warning when this condition is seen.

Cc: stable@kernel.org
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
LKML-Reference: <4B85770B.6010901@suse.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Andreas Radke <a.radke@arcor.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agobnx2: Fix lost MSI-X problem on 5709 NICs.
Michael Chan [Tue, 27 Apr 2010 11:28:09 +0000 (11:28 +0000)]
bnx2: Fix lost MSI-X problem on 5709 NICs.

commit c441b8d2cb2194b05550a558d6d95d8944e56a84 upstream.

It has been reported that under certain heavy traffic conditions in MSI-X
mode, the driver can lose an MSI-X vector causing all packets in the
associated rx/tx ring pair to be dropped.  The problem is caused by
the chip dropping the write to unmask the MSI-X vector by the kernel
(when migrating the IRQ for example).

This can be prevented by increasing the GRC timeout value for these
register read and write operations.

Thanks to Dell for helping us debug this problem.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoqla2xxx: Properly handle UNDERRUN completion statuses.
Lalit Chandivade [Tue, 13 Oct 2009 22:16:52 +0000 (15:16 -0700)]
qla2xxx: Properly handle UNDERRUN completion statuses.

commit 0f00a206ccb1dc644b6770ef25f185610fee6962 upstream.

Correct issues where the lower scsi-status would be improperly
cleared, instead, allow the midlayer to process the status after
the proper residual-count checks are performed.  Finally,
validate firmware status flags prior to assigning values from the
FCP_RSP frame.

Signed-off-by: Lalit Chandivade <lalit.chandivade@qlogic.com>
Signed-off-by: Michael Hernandez <michael.hernandez@qlogic.com>
Signed-off-by: Ravi Anand <ravi.anand@qlogic.com>
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoparisc: Set PCI CLS early in boot.
Carlos O'Donell [Mon, 22 Feb 2010 23:25:59 +0000 (23:25 +0000)]
parisc: Set PCI CLS early in boot.

commit 5fd4514bb351b5ecb0da3692fff70741e5ed200c upstream.

Set the PCI CLS early in the boot process to prevent
device failures. In pcibios_set_master use the new
pci_cache_line_size instead of a hard-coded value.

Signed-off-by: Carlos O'Donell <carlos@codesourcery.com>
Reviewed-by: Grant Grundler <grundler@google.com>
Signed-off-by: Kyle McMartin <kyle@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoxfs: add a shrinker to background inode reclaim
Dave Chinner [Tue, 4 May 2010 02:58:20 +0000 (12:58 +1000)]
xfs: add a shrinker to background inode reclaim

commit 9bf729c0af67897ea8498ce17c29b0683f7f2028 upstream

On low memory boxes or those with highmem, kernel can OOM before the
background reclaims inodes via xfssyncd. Add a shrinker to run inode
reclaim so that it inode reclaim is expedited when memory is low.

This is more complex than it needs to be because the VM folk don't
want a context added to the shrinker infrastructure. Hence we need
to add a global list of XFS mount structures so the shrinker can
traverse them.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Alex Elder <aelder@sgi.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agotg3: Fix INTx fallback when MSI fails
Andre Detsch [Mon, 26 Apr 2010 07:27:07 +0000 (07:27 +0000)]
tg3: Fix INTx fallback when MSI fails

commit dc8bf1b1a6edfc92465526de19772061302f0929 upstream.

tg3: Fix INTx fallback when MSI fails

MSI setup changes the value of irq_vec in struct tg3 *tp.
This attribute must be taken into account and restored before
we try to do a new request_irq for INTx fallback.

In powerpc, the original code was leading to an EINVAL return within
request_irq, because the driver was trying to use the disabled MSI
virtual irq number instead of tp->pdev->irq.

Signed-off-by: Andre Detsch <adetsch@br.ibm.com>
Acked-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Cc: Brandon Philips <bphilips@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoskip sense logging for some ATA PASS-THROUGH cdbs
Douglas Gilbert [Sun, 3 Jan 2010 18:51:15 +0000 (13:51 -0500)]
skip sense logging for some ATA PASS-THROUGH cdbs

commit e7efe5932b1d3916c79326a4221693ea90a900e2 upstream.

Further to the lsml thread titled:
"does scsi_io_completion need to dump sense data for ata pass through (ck_cond =
1) ?"

This is a patch to skip logging when the sense data is
associated with a SENSE_KEY of "RECOVERED_ERROR" and the
additional sense code is "ATA PASS-THROUGH INFORMATION
AVAILABLE". This only occurs with the SAT ATA PASS-THROUGH
commands when CK_COND=1 (in the cdb). It indicates that
the sense data contains ATA registers.

Smartmontools uses such commands on ATA disks connected via
SAT. Periodic checks such as those done by smartd cause
nuisance entries into logs that are:
    - neither errors nor warnings
    - pointless unless the cdb that caused them are also logged

Signed-off-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoPCI: Ensure we re-enable devices on resume
Matthew Garrett [Thu, 22 Apr 2010 13:30:51 +0000 (09:30 -0400)]
PCI: Ensure we re-enable devices on resume

commit cc2893b6af5265baa1d68b17b136cffca9e40cfa upstream.

If the firmware puts a device back into D0 state at resume time, we'll
update its state in resume_noirq and thus skip the platform resume code.
Calling that code twice should be safe and we ought to avoid getting to
that point anyway, so remove the check and also allow the platform pci
code to be called for D0.

Fixes USB not being powered after resume on recent Lenovo machines.

Acked-by: Alex Chiang <achiang@canonical.com>
Acked-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agor8169: more broken register writes workaround
françois romieu [Mon, 26 Apr 2010 11:42:58 +0000 (11:42 +0000)]
r8169: more broken register writes workaround

commit 908ba2bfd22253f26fa910cd855e4ccffb1467d0 upstream.

78f1cd02457252e1ffbc6caa44a17424a45286b8 ("fix broken register writes")
does not work for Al Viro's r8169 (XID 18000000).

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agor8169: fix broken register writes
Francois Romieu [Sun, 28 Mar 2010 02:35:46 +0000 (19:35 -0700)]
r8169: fix broken register writes

commit 78f1cd02457252e1ffbc6caa44a17424a45286b8 upstream.

This is quite similar to b39fe41f481d20c201012e4483e76c203802dda7
though said registers are not even documented as 64-bit registers
- as opposed to the initial TxDescStartAddress ones - but as single
bytes which must be combined into 32 bits at the MMIO read/write
level before being merged into a 64 bit logical entity.

Credits go to Ben Hutchings <ben@decadent.org.uk> for the MAR
registers (aka "multicast is broken for ages on ARM) and to
Timo Teräs <timo.teras@iki.fi> for the MAC registers.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agor8169: use correct barrier between cacheable and non-cacheable memory
David Dillow [Wed, 3 Mar 2010 16:33:10 +0000 (16:33 +0000)]
r8169: use correct barrier between cacheable and non-cacheable memory

commit 4c020a961a812ffae9846b917304cea504c3a733 upstream.

r8169 needs certain writes to be visible to other CPUs or the NIC before
touching the hardware, but was using smp_wmb() which is only required to
order cacheable memory access. Switch to wmb() which is required to
order both cacheable and non-cacheable memory.

Noticed by Catalin Marinas and Paul Mackerras.

Signed-off-by: David Dillow <dave@thedillows.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agokgdb: don't needlessly skip PAGE_USER test for Fsl booke
Wufei [Wed, 28 Apr 2010 21:42:32 +0000 (17:42 -0400)]
kgdb: don't needlessly skip PAGE_USER test for Fsl booke

commit 56151e753468e34aeb322af4b0309ab727c97d2e upstream.

The bypassing of this test is a leftover from 2.4 vintage
kernels, and is no longer appropriate, or even used by KGDB.
Currently KGDB uses probe_kernel_write() for all access to
memory via the KGDB core, so it can simply be deleted.

This fixes CVE-2010-1446.

CC: Benjamin Herrenschmidt <benh@kernel.crashing.org>
CC: Paul Mackerras <paulus@samba.org>
CC: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Wufei <fei.wu@windriver.com>
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoKVM: remove unused load_segment_descriptor_to_kvm_desct
Marcelo Tosatti [Tue, 27 Apr 2010 16:35:26 +0000 (13:35 -0300)]
KVM: remove unused load_segment_descriptor_to_kvm_desct

Commit 78ce64a384 in v2.6.32.12 introduced a warning due to unused
load_segment_descriptor_to_kvm_desct helper, which has been opencoded by
this commit.

On upstream, the helper was removed as part of a different commit.

Remove the now unused function.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agodccp_probe: Fix module load dependencies between dccp and dccp_probe
Neil Horman [Fri, 15 Jan 2010 09:40:55 +0000 (01:40 -0800)]
dccp_probe: Fix module load dependencies between dccp and dccp_probe

commit 38ff3e6bb987ec583268da8eb22628293095d43b upstream.

This was just recently reported to me.  When built as modules, the
dccp_probe module has a silent dependency on the dccp module.  This
stems from the fact that the module_init routine of dccp_probe
registers a jprobe on the dccp_sendmsg symbol.  Since the symbol is
only referenced as a text string (the .symbol_name field in the jprobe
struct) rather than the address of the symbol itself, depmod never
picks this dependency up, and so if you load the dccp_probe module
without the dccp module loaded, the register_jprobe call fails with an
-EINVAL, and the whole module load fails.

The fix is pretty easy, we can just wrap the register_jprobe call in a
try_then_request_module call, which forces the dependency to get
satisfied prior to the probe registration.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Cc: maximilian attems <max@stro.at>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agodrivers/net/wireless/p54/txrx.c Fix off by one error
Darren Jenkins [Wed, 17 Feb 2010 12:40:15 +0000 (23:40 +1100)]
drivers/net/wireless/p54/txrx.c Fix off by one error

commit 088ea189c4c75cdf211146faa4b341a0f7476be6 upstream.

fix off by one error in the queue size check of p54_tx_qos_accounting_alloc()

Coverity CID: 13314

Signed-off-by: Darren Jenkins <darrenrjenkins@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agop54pci: rx frame length check
Christian Lamparter [Sun, 17 Jan 2010 23:07:38 +0000 (00:07 +0100)]
p54pci: rx frame length check

commit f5300e04df78feae8107c1846dd3a9e27c071b2f upstream.

A long time ago, a user reported several crashes due to
data corruptions which are likely the result of a
not-100%-supported, or faulty? PCI bridge.
( http://patchwork.kernel.org/patch/53004/ )

This patch fixes entry #1.
"1.  p54p_check_rx_ring - skb_over_panic: Under a ping flood
or just left running for a bit would panic with a skb_over_panic."
As described in the mail: The invalid frame length causes
skb_put to bailout and trigger a crash.

Note:
Simply dropping the frame is problematic, because if its content
contains a tx feedback we would lose some portion of the device
memory space.... And the driver/mac80211 should handle all other
invalid data.

Reported-by: Quintin Pitts <geek4linux@gmail.com>
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoACPI: introduce kernel parameter acpi_sleep=sci_force_enable
Zhang Rui [Wed, 30 Dec 2009 07:36:42 +0000 (15:36 +0800)]
ACPI: introduce kernel parameter acpi_sleep=sci_force_enable

commit d7f0eea9e431e1b8b0742a74db1a9490730b2a25 upstream.

Introduce kernel parameter acpi_sleep=sci_force_enable

some laptop requires SCI_EN being set directly on resume,
or else they hung somewhere in the resume code path.

We already have a blacklist for these laptops but we still need
this option, especially when debugging some suspend/resume problems,
in case there are systems that need this workaround and are not yet
in the blacklist.

Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Acked-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agojfs: fix diAllocExt error in resizing filesystem
Bill Pemberton [Fri, 16 Apr 2010 13:01:20 +0000 (08:01 -0500)]
jfs: fix diAllocExt error in resizing filesystem

commit 2b0b39517d1af5294128dbc2fd7ed39c8effa540 upstream.

Resizing the filesystem would result in an diAllocExt error in some
instances because changes in bmp->db_agsize would not get noticed if
goto extendBmap was called.

Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Signed-off-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Cc: jfs-discussion@lists.sourceforge.net
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoCRED: Fix a race in creds_are_invalid() in credentials debugging
David Howells [Wed, 21 Apr 2010 09:28:25 +0000 (10:28 +0100)]
CRED: Fix a race in creds_are_invalid() in credentials debugging

commit e134d200d57d43b171dcb0b55c178a1a0c7db14a upstream.

creds_are_invalid() reads both cred->usage and cred->subscribers and then
compares them to make sure the number of processes subscribed to a cred struct
never exceeds the refcount of that cred struct.

The problem is that this can cause a race with both copy_creds() and
exit_creds() as the two counters, whilst they are of atomic_t type, are only
atomic with respect to themselves, and not atomic with respect to each other.

This means that if creds_are_invalid() can read the values on one CPU whilst
they're being modified on another CPU, and so can observe an evolving state in
which the subscribers count now is greater than the usage count a moment
before.

Switching the order in which the counts are read cannot help, so the thing to
do is to remove that particular check.

I had considered rechecking the values to see if they're in flux if the test
fails, but I can't guarantee they won't appear the same, even if they've
changed several times in the meantime.

Note that this can only happen if CONFIG_DEBUG_CREDENTIALS is enabled.

The problem is only likely to occur with multithreaded programs, and can be
tested by the tst-eintr1 program from glibc's "make check".  The symptoms look
like:

CRED: Invalid credentials
CRED: At include/linux/cred.h:240
CRED: Specified credentials: ffff88003dda5878 [real][eff]
CRED: ->magic=43736564, put_addr=(null)
CRED: ->usage=766, subscr=766
CRED: ->*uid = { 0,0,0,0 }
CRED: ->*gid = { 0,0,0,0 }
CRED: ->security is ffff88003d72f538
CRED: ->security {359, 359}
------------[ cut here ]------------
kernel BUG at kernel/cred.c:850!
...
RIP: 0010:[<ffffffff81049889>]  [<ffffffff81049889>] __invalid_creds+0x4e/0x52
...
Call Trace:
 [<ffffffff8104a37b>] copy_creds+0x6b/0x23f

Note the ->usage=766 and subscr=766.  The values appear the same because
they've been re-read since the check was made.

Reported-by: Roland McGrath <roland@redhat.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: James Morris <jmorris@namei.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agodrm/i915: fix tiling limits for i915 class hw v2
Daniel Vetter [Sat, 17 Apr 2010 13:12:03 +0000 (15:12 +0200)]
drm/i915: fix tiling limits for i915 class hw v2

commit c36a2a6de59e4a141a68b7575de837d3b0bd96b3 upstream.

Current code is definitely crap: Largest pitch allowed spills into
the TILING_Y bit of the fence registers ... :(

I've rewritten the limits check under the assumption that 3rd gen hw
has a 3d pitch limit of 8kb (like 2nd gen). This is supported by an
otherwise totally misleading XXX comment.

This bug mostly resulted in tiling-corrupted pixmaps because the kernel
allowed too wide buffers to be tiled. Bug brought to the light by the
xf86-video-intel 2.11 release because that unconditionally enabled
tiling for pixmaps, relying on the kernel to check things. Tiling for
the framebuffer was not affected because the ddx does some additional
checks there ensure the buffer is within hw-limits.

v2: Instead of computing the value that would be written into the
hw fence registers and then checking the limits simply check whether
the stride is above the 8kb limit. To better document the hw, add
some WARN_ONs in i915_write_fence_reg like I've done for the i830
case (using the right limits).

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=27449
Tested-by: Alexander Lam <lambchop468@gmail.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoinitramfs: handle unrecognised decompressor when unpacking
Phillip Lougher [Fri, 23 Apr 2010 17:18:11 +0000 (13:18 -0400)]
initramfs: handle unrecognised decompressor when unpacking

commit df37bd156dcb4f5441beaf5bde444adac974e9a0 upstream.

The unpack routine fails to handle the decompress_method() returning
unrecognised decompressor (compress_name == NULL).  This results in the
routine looping eventually oopsing on an out of bounds memory access.

Note this bug is usually hidden, only triggering on trailing junk after
one or more correct compressed blocks.  The case of the compressed archive
being complete junk is (by accident?) caught by the if (state != Reset)
check because state is initialised to Start, but not updated due to the
decompressor not having been called.  Obviously if the junk is trailing a
correctly decompressed buffer, state == Reset from the previous call to
the decompressor.

Signed-off-by: Phillip Lougher <phillip@lougher.demon.co.uk>
Reported-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoext4: correctly calculate number of blocks for fiemap
Leonard Michlmayr [Thu, 4 Mar 2010 22:07:28 +0000 (17:07 -0500)]
ext4: correctly calculate number of blocks for fiemap

commit aca92ff6f57c000d1b4523e383c8bd6b8269b8b1 upstream.

ext4_fiemap() rounds the length of the requested range down to
blocksize, which is is not the true number of blocks that cover the
requested region.  This problem is especially impressive if the user
requests only the first byte of a file: not a single extent will be
reported.

We fix this by calculating the last block of the region and then
subtract to find the number of blocks in the extents.

Signed-off-by: Leonard Michlmayr <leonard.michlmayr@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agolibata: Fix accesses at LBA28 boundary (old bug, but nasty) (v2)
Mark Lord [Wed, 7 Apr 2010 17:52:08 +0000 (13:52 -0400)]
libata: Fix accesses at LBA28 boundary (old bug, but nasty) (v2)

commit 45c4d015a92f72ec47acd0c7557abdc0c8a6499d upstream.

Most drives from Seagate, Hitachi, and possibly other brands,
do not allow LBA28 access to sector number 0x0fffffff (2^28 - 1).
So instead use LBA48 for such accesses.

This bug could bite a lot of systems, especially when the user has
taken care to align partitions to 4KB boundaries. On misaligned systems,
it is less likely to be encountered, since a 4KB read would end at
0x10000000 rather than at 0x0fffffff.

Signed-off-by: Mark Lord <mlord@pobox.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: hda: Use olpc-xo-1_5 quirk for Toshiba Satellite P500-PSPGSC-01800T
Daniel T Chen [Wed, 5 May 2010 02:07:58 +0000 (22:07 -0400)]
ALSA: hda: Use olpc-xo-1_5 quirk for Toshiba Satellite P500-PSPGSC-01800T

commit c53666813813a0ea3d0391e1911eefc05a5e6b4f upstream.

BugLink: https://launchpad.net/bugs/549267
The OR verified that using the olpc-xo-1_5 model quirk allows the
headphones to be audible when inserted into the jack. Capture was
also verified to work correctly.

Reported-by: Richard Gagne
Tested-by: Richard Gagne
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: hda: Use olpc-xo-1_5 quirk for Toshiba Satellite Pro T130-15F
Daniel T Chen [Tue, 4 May 2010 00:39:31 +0000 (20:39 -0400)]
ALSA: hda: Use olpc-xo-1_5 quirk for Toshiba Satellite Pro T130-15F

commit 4442dd4613fe3795b4c8a5f42fc96b7ffb90d01a upstream.

BugLink: https://launchpad.net/bugs/573284
The OR verified that using the olpc-xo-1_5 model quirk allows the
headphones to be audible when inserted into the jack. Capture was
also verified to work correctly.

Reported-by: Andy Couldrake <acouldrake@googlemail.com>
Tested-by: Andy Couldrake <acouldrake@googlemail.com>
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: hda: Fix 0 dB for Packard Bell models using Conexant CX20549 (Venice)
Daniel T Chen [Wed, 28 Apr 2010 22:00:11 +0000 (18:00 -0400)]
ALSA: hda: Fix 0 dB for Packard Bell models using Conexant CX20549 (Venice)

commit 8f0f5ff6777104084b4b2e1ae079541c2a6ed6d9 upstream.

BugLink: https://launchpad.net/bugs/541802
The OR's hardware distorts at PCM 100% because it does not correspond to
0 dB. Fix this in patch_cxt5045() for all Packard Bell models.

Reported-by: Valombre
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: hda: Fix max PCM level to 0 dB for Fujitsu-Siemens laptops using CX20549 (Venice)
Daniel T Chen [Wed, 25 Nov 2009 23:27:20 +0000 (18:27 -0500)]
ALSA: hda: Fix max PCM level to 0 dB for Fujitsu-Siemens laptops using CX20549 (Venice)

commit 0b587fc4d35afb1bc0fc3d890084bb14c78372dc upstream.

BugLink: https://bugtrack.alsa-project.org/alsa-bug/view.php?id=4792
Cristian reported that these models have really bad sound above 6 dB
and proposed the original patch. I've updated the comment to reflect
this change.

Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
Reported-by: Cristian Klein
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: snd-meastro3: Ignore spurious HV interrupts during suspend / resume
Hans de Goede [Wed, 21 Apr 2010 15:04:08 +0000 (11:04 -0400)]
ALSA: snd-meastro3: Ignore spurious HV interrupts during suspend / resume

commit 715aa675338ce6e1a3b4f77cf87ea611f93058a8 upstream.

Ignore spurious HV interrupts during suspend / resume, this avoids
mistaking them for a mute button press. This is not very pretty but
it seems the only way to fix the master volume control gets muted
after suspend issue I'm seeing. Note that the es1968 driver is doing
exactly the same.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: snd-meastro3: Add amp_gpio quirk for Compaq EVO N600C
Hans de Goede [Wed, 21 Apr 2010 15:04:06 +0000 (11:04 -0400)]
ALSA: snd-meastro3: Add amp_gpio quirk for Compaq EVO N600C

commit 7efbfd1ae98ef9efe06352e2a1ad83e8c14ceeb1 upstream.

Without this quirk sound stops working after suspend resume. With this quirk,
one still needs to manually unmute the master volume control after a suspend /
/ resume cycle. That is fixed in another patch in this set.

Note that this patch was submitted to the alsa bug tracker a long time ago:
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=4319

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: hda: Use ALC880_F1734 quirk for Fujitsu Siemens AMILO Xi 1526
Daniel T Chen [Thu, 22 Apr 2010 11:15:26 +0000 (07:15 -0400)]
ALSA: hda: Use ALC880_F1734 quirk for Fujitsu Siemens AMILO Xi 1526

commit 3353541fe533350a22a03e2fb7dc085b35912575 upstream.

BugLink: https://launchpad.net/bugs/567494
The OR has verified that the existing model quirk, ALC880_UNIWILL,
is insufficient for audible playback and capture by default. Instead,
the ALC880_F1734 model quirk needs to be used.

This change is necessary for both 2.6.32.11 and 2.6.33.2.

Reported-by: Arnaud Malpeyre <amalpeyre@gmail.com>
Tested-by: Arnaud Malpeyre <amalpeyre@gmail.com>
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: hda: Use STAC_DELL_M6_BOTH quirk for Dell Studio 1558
Daniel T Chen [Thu, 22 Apr 2010 21:54:45 +0000 (17:54 -0400)]
ALSA: hda: Use STAC_DELL_M6_BOTH quirk for Dell Studio 1558

commit 5c1bccf645d4ab65e4c7502acb42e8b9afdb5bdc upstream.

BugLink: https://launchpad.net/bugs/568600
The OR has verified that the dell-m6 model quirk is necessary for audio
to be audible by default on the Dell Studio XPS 1645.

This change is necessary for 2.6.32.11 and 2.6.33.2 alike.

Reported-by: Andy Ross <andy@plausible.org>
Tested-by: Andy Ross <andy@plausible.org>
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: hda: Use STAC_DELL_M6_BOTH quirk for Dell Studio XPS 1645
Daniel T Chen [Thu, 22 Apr 2010 00:41:52 +0000 (20:41 -0400)]
ALSA: hda: Use STAC_DELL_M6_BOTH quirk for Dell Studio XPS 1645

commit aac78daf8f37256283f56820ae858add7139c56c upstream.

BugLink: https://launchpad.net/bugs/553002
The OR has verified that the dell-m6 model quirk is necessary for audio
to be audible by default on the Dell Studio XPS 1645.

This change is necessary for 2.6.32.11 and 2.6.33.2 alike.

Reported-by: Robert Chambers
Tested-by: Robert Chambers
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: hda - Add PCI quirk for HP dv6-1110ax.
Kunal Gangakhedkar [Sat, 20 Mar 2010 17:38:01 +0000 (23:08 +0530)]
ALSA: hda - Add PCI quirk for HP dv6-1110ax.

commit e3d2530a6cea80987f77b75d8784a00f3aaf22ff upstream.

Adding this PCI quirk fixes the board config detection.
This also fixes jack sensing by using "hp_detect=1" via properly detected
board config.

Signed-off-by: Kunal Gangakhedkar <kunal.gangakhedkar@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoALSA: hda: Use LPIB quirk for DG965OT board version AAD63733-203
Daniel T Chen [Wed, 21 Apr 2010 23:55:43 +0000 (19:55 -0400)]
ALSA: hda: Use LPIB quirk for DG965OT board version AAD63733-203

commit 0e0280dc2b0c7395a880d25544b47f3e3e3f79db upstream.

BugLink: https://launchpad.net/bugs/459083
The OR has verified with 2.6.32.11 and the latest alsa-driver stable
daily snapshot that position_fix=1 is necessary for the external mic
to work and for PulseAudio not to crash constantly.

This patch is necessary also for 2.6.32.11 and 2.6.33.2.

Reported-by: <imwithid@yahoo.com>
Tested-by: <imwithid@yahoo.com>
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agox86, AMD: Fix stale cpuid4_info shared_map data in shared_cpu_map cpumasks
Prarit Bhargava [Wed, 9 Dec 2009 18:36:45 +0000 (13:36 -0500)]
x86, AMD: Fix stale cpuid4_info shared_map data in shared_cpu_map cpumasks

commit ebb682f522411abbe358059a256a8672ec0bd55b upstream.

The per_cpu cpuid4_info shared_map can contain stale data when CPUs are added
and removed.

The stale data can lead to a NULL pointer derefernce panic on a remove of a
CPU that has had siblings previously removed.

This patch resolves the panic by verifying a cpu is actually online before
adding it to the shared_cpu_map, only examining cpus that are part of
the same lower level cache, and by updating other siblings lowest level cache
maps when a cpu is added.

Signed-off-by: Prarit Bhargava <prarit@redhat.com>
LKML-Reference: <20091209183336.17855.98708.sendpatchset@prarit.bos.redhat.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agox86, k8 nb: Fix boot crash: enable k8_northbridges unconditionally on AMD systems
Borislav Petkov [Fri, 12 Mar 2010 14:43:03 +0000 (15:43 +0100)]
x86, k8 nb: Fix boot crash: enable k8_northbridges unconditionally on AMD systems

commit 0e152cd7c16832bd5cadee0c2e41d9959bc9b6f9 upstream.

de957628ce7c84764ff41331111036b3ae5bad0f changed setting of the
x86_init.iommu.iommu_init function ptr only when GART IOMMU is
found.

One side effect of it is that num_k8_northbridges
is not initialized anymore if not explicitly
called. This resulted in uninitialized pointers in
<arch/x86/kernel/cpu/intel_cacheinfo.c:amd_calc_l3_indices()>,
for example, which uses the num_k8_northbridges thing through
node_to_k8_nb_misc().

Fix that through an initcall that runs right after the PCI
subsystem and does all the scanning. Then, remove initialization
in gart_iommu_init() which is a rootfs_initcall and we're
running before that.

What is more, since num_k8_northbridges is being used in other
places beside GART IOMMU, include it whenever we add AMD CPU
support. The previous dependency chain in kconfig contained

K8_NB depends on AGP_AMD64|GART_IOMMU

which was clearly incorrect. The more natural way in terms of
hardware dependency should be

AGP_AMD64|GART_IOMMU depends on K8_NB depends on CPU_SUP_AMD &&
PCI. Make it so Number One!

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
Cc: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Joerg Roedel <joerg.roedel@amd.com>
LKML-Reference: <20100312144303.GA29262@aftab>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Tested-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agox86: Disable large pages on CPUs with Atom erratum AAE44
H. Peter Anvin [Tue, 13 Apr 2010 21:40:54 +0000 (14:40 -0700)]
x86: Disable large pages on CPUs with Atom erratum AAE44

commit 7a0fc404ae663776e96db43879a0fa24fec1fa3a upstream.

Atom erratum AAE44/AAF40/AAG38/AAH41:

"If software clears the PS (page size) bit in a present PDE (page
directory entry), that will cause linear addresses mapped through this
PDE to use 4-KByte pages instead of using a large page after old TLB
entries are invalidated. Due to this erratum, if a code fetch uses
this PDE before the TLB entry for the large page is invalidated then
it may fetch from a different physical address than specified by
either the old large page translation or the new 4-KByte page
translation. This erratum may also cause speculative code fetches from
incorrect addresses."

[http://download.intel.com/design/processor/specupdt/319536.pdf]

Where as commit 211b3d03c7400f48a781977a50104c9d12f4e229 seems to
workaround errata AAH41 (mixed 4K TLBs) it reduces the window of
opportunity for the bug to occur and does not totally remove it.  This
patch disables mixed 4K/4MB page tables totally avoiding the page
splitting and not tripping this processor issue.

This is based on an original patch by Colin King.

Originally-by: Colin Ian King <colin.king@canonical.com>
Cc: Colin Ian King <colin.king@canonical.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
LKML-Reference: <1269271251-19775-1-git-send-email-colin.king@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agox86-64: Clear a 64-bit FS/GS base on fork if selector is nonzero
H. Peter Anvin [Fri, 23 Apr 2010 23:17:40 +0000 (16:17 -0700)]
x86-64: Clear a 64-bit FS/GS base on fork if selector is nonzero

commit 7ce5a2b9bb2e92902230e3121d8c3047fab9cb47 upstream.

When we do a thread switch, we clear the outgoing FS/GS base if the
corresponding selector is nonzero.  This is taken by __switch_to() as
an entry invariant; it does not verify that it is true on entry.
However, copy_thread() doesn't enforce this constraint, which can
result in inconsistent results after fork().

Make copy_thread() match the behavior of __switch_to().

Reported-and-tested-by: Samuel Thibault <samuel.thibault@inria.fr>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
LKML-Reference: <4BD1E061.8030605@zytor.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoedac, mce: Fix wrong mask and macro usage
Borislav Petkov [Fri, 30 Apr 2010 13:19:02 +0000 (15:19 +0200)]
edac, mce: Fix wrong mask and macro usage

commit 35d824b28fc5544d1eb7c1e3db15a1740df8ec4b upstream.

Correct two mishaps which prevented reporting error type (CECC vs UECC)
and extended error description.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agop54pci: fix bugs in p54p_check_tx_ring
Hans de Goede [Thu, 22 Apr 2010 17:52:16 +0000 (19:52 +0200)]
p54pci: fix bugs in p54p_check_tx_ring

commit 0250ececdf6813457c98719e2d33b3684881fde0 upstream.

Hans de Goede identified a bug in p54p_check_tx_ring:

there are two ring indices. 1 => tx data and 3 => tx management.
But the old code had a constant "1" and this resulted in spurious
dma unmapping failures.

Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=583623
Bug-Identified-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agodm9601: fix phy/eeprom write routine
Peter Korsgaard [Mon, 3 May 2010 10:01:26 +0000 (10:01 +0000)]
dm9601: fix phy/eeprom write routine

commit e9162ab1610531d6ea6c1833daeb2613e44275e8 upstream.

Use correct bit positions in DM_SHARED_CTRL register for writes.

Michael Planes recently encountered a 'KY-RS9600 USB-LAN converter', which
came with a driver CD containing a Linux driver. This driver turns out to
be a copy of dm9601.c with symbols renamed and my copyright stripped.
That aside, it did contain 1 functional change in dm_write_shared_word(),
and after checking the datasheet the original value was indeed wrong
(read versus write bits).

On Michaels HW, this change bumps receive speed from ~30KB/s to ~900KB/s.
On other devices the difference is less spectacular, but still significant
(~30%).

Reported-by: Michael Planes <michael.planes@free.fr>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoblock: ensure jiffies wrap is handled correctly in blk_rq_timed_out_timer
Richard Kennedy [Wed, 14 Apr 2010 18:54:03 +0000 (20:54 +0200)]
block: ensure jiffies wrap is handled correctly in blk_rq_timed_out_timer

commit a534dbe96e9929c7245924d8252d89048c23d569 upstream.

blk_rq_timed_out_timer() relied on blk_add_timer() never returning a
timer value of zero, but commit 7838c15b8dd18e78a523513749e5b54bda07b0cb
removed the code that bumped this value when it was zero.
Therefore when jiffies is near wrap we could get unlucky & not set the
timeout value correctly.

This patch uses a flag to indicate that the timeout value was set and so
handles jiffies wrap correctly, and it keeps all the logic in one
function so should be easier to maintain in the future.

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoserial: 8250_pnp - add Fujitsu Wacom device
Ping Cheng [Mon, 22 Mar 2010 20:40:29 +0000 (13:40 -0700)]
serial: 8250_pnp - add Fujitsu Wacom device

commit d9901660b53b92f0f3551c06588b8be38224b245 upstream.

Add Fujitsu Wacom 1FGT Tablet PC device

Signed-off-by: Ping Cheng <pingc@wacom.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoraid6: fix recovery performance regression
Dan Williams [Wed, 5 May 2010 03:41:56 +0000 (20:41 -0700)]
raid6: fix recovery performance regression

commit 5157b4aa5b7de8787b6318e61bcc285031bb9088 upstream.

The raid6 recovery code should immediately drop back to the optimized
synchronous path when a p+q dma resource is not available.  Otherwise we
run the non-optimized/multi-pass async code in sync mode.

Verified with raid6test (NDISKS=255)

Applies to kernels >= 2.6.32.

Acked-by: NeilBrown <neilb@suse.de>
Reported-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoperf: Fix resource leak in failure path of perf_event_open()
Tejun Heo [Sat, 1 May 2010 08:11:35 +0000 (10:11 +0200)]
perf: Fix resource leak in failure path of perf_event_open()

commit 048c852051d2bd5da54a4488bc1f16b0fc74c695 upstream.

perf_event_open() kfrees event after init failure which doesn't
release all resources allocated by perf_event_alloc().  Use
free_event() instead.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Paul Mackerras <paulus@au1.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <4BDBE237.1040809@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoi2c: Fix probing of FSC hardware monitoring chips
Jean Delvare [Tue, 4 May 2010 09:09:28 +0000 (11:09 +0200)]
i2c: Fix probing of FSC hardware monitoring chips

commit b1d4b390ea4bb480e65974ce522a04022608a8df upstream.

Some FSC hardware monitoring chips (Syleus at least) doesn't like
quick writes we typically use to probe for I2C chips. Use a regular
byte read instead for the address they live at (0x73). These are the
only known chips living at this address on PC systems.

For clarity, this fix should not be needed for kernels 2.6.30 and
later, as we started instantiating the hwmon devices explicitly based
on DMI data. Still, this fix is valuable in the following two cases:
* Support for recent FSC chips on older kernels. The DMI-based device
  instantiation is more difficult to backport than the device support
  itself.
* Case where the DMI-based device instantiation fails, whatever the
  reason. We fall back to probing in that case, so it should work.

This fixes kernel bug #15634:
https://bugzilla.kernel.org/show_bug.cgi?id=15634

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
14 years agoStaging: hv: name network device ethX rather than sethX
Stephen Hemminger [Thu, 11 Mar 2010 17:11:37 +0000 (09:11 -0800)]
Staging: hv: name network device ethX rather than sethX

commit 546d9e101e7a71e6202f47a13ddcd9b8fb05a52e upstream.

This patch makes the HyperV network device use the same naming scheme as
other virtual drivers (Xen, KVM). In an ideal world, userspace tools
would not care what the name is, but some users and applications do
care. Vyatta CLI is one of the tools that does depend on what the name
is.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Cc: Hank Janssen <hjanssen@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>