9 years agoUSB: g_serial: don't set low_latency flag
Jon Povey [Mon, 14 Jun 2010 10:41:04 +0000]
USB: g_serial: don't set low_latency flag

No longer set low_latency flag as it causes this warning backtrace:

  WARNING: at kernel/mutex.c:207 __mutex_lock_slowpath+0x6c/0x288()

Fix associated locking and wakeups.

Signed-off-by: Jon Povey <jon.povey@racelogic.co.uk>
Cc: Maulik Mankad <x0082077@ti.com>
Cc: stable <stable@kernel.org>
Acked-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

9 years agoMerge branch 'for-linus' of git://neil.brown.name/md
Linus Torvalds [Mon, 28 Jun 2010 05:56:32 +0000]
Merge branch 'for-linus' of git://neil.brown.name/md

* 'for-linus' of git://neil.brown.name/md:
  md/raid5: don't include 'spare' drives when reshaping to fewer devices.
  md/raid5: add a missing 'continue' in a loop.
  md/raid5: Allow recovered part of partially recovered devices to be in-sync
  md/raid5: More careful check for "has array failed".
  md: Don't update ->recovery_offset when reshaping an array to fewer devices.
  md/raid5: avoid oops when number of devices is reduced then increased.
  md: enable raid4->raid0 takeover
  md: clear layout after ->raid0 takeover
  md: fix raid10 takeover: use new_layout for setup_conf
  md: fix handling of array level takeover that re-arranges devices.
  md: raid10: Fix null pointer dereference in fix_read_error()
  Restore partition detection of newly created md arrays.

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu
Linus Torvalds [Sun, 27 Jun 2010 18:33:44 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/tj/percpu

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu:
  percpu: fix first chunk match in per_cpu_ptr_to_phys()
  percpu: fix trivial bugs in pcpu_build_alloc_info()

9 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Sun, 27 Jun 2010 18:28:02 +0000]
Merge git://git./linux/kernel/git/davem/net-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (52 commits)
  phylib: Add autoload support for the LXT973 phy.
  ISDN: hysdn, fix potential NULL dereference
  vxge: fix memory leak in vxge_alloc_msix() error path
  isdn/gigaset: correct CAPI connection state storage
  isdn/gigaset: encode HLC and BC together
  isdn/gigaset: correct CAPI DATA_B3 Delivery Confirmation
  isdn/gigaset: correct CAPI voice connection encoding
  isdn/gigaset: honor CAPI application's buffer size request
  cpmac: do not leak struct net_device on phy_connect errors
  smc91c92_cs: fix the problem that lan & modem does not work simultaneously
  ipv6: fix NULL reference in proxy neighbor discovery
  Bluetooth: Bring back var 'i' increment
  xfrm: check bundle policy existance before dereferencing it
  sky2: enable rx/tx in sky2_phy_reinit()
  cnic: Disable statistics initialization for eth clients that do not support statistics
  net: add dependency on fw class module to qlcnic and netxen_nic
  snmp: fix SNMP_ADD_STATS()
  hso: remove setting of low_latency flag
  udp: Fix bogus UFO packet generation
  lasi82596: fix netdev_mc_count conversion
  ...

9 years agoMerge branch 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6
Linus Torvalds [Sun, 27 Jun 2010 16:04:02 +0000]
Merge branch 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6

* 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6:
  NFSv4: Fix an embarassing typo in encode_attrs()
  NFSv4: Ensure that /proc/self/mountinfo displays the minor version number
  NFSv4.1: Ensure that we initialise the session when following a referral
  SUNRPC: Fix a re-entrancy bug in xs_tcp_read_calldir()
  nfs4 use mandatory attribute file type in nfs4_get_root

9 years agoMerge branch 'sh/for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/lethal...
Linus Torvalds [Sun, 27 Jun 2010 15:18:20 +0000]
Merge branch 'sh/for-linus' of git://git./linux/kernel/git/lethal/sh-2.6

* 'sh/for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6:
  clocksource: sh_cmt: Fix up bogus shift value.
  arch/sh/mm: Eliminate a double lock
  sh: Fix up IORESOURCE_PCI_FIXED usage in pcibios_fixup_device_resources().
  sh: remove duplicated #include

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Linus Torvalds [Sun, 27 Jun 2010 15:03:00 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/dtor/input

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
  MAINTAINERS - Add an entry for the input MT protocol
  Input: wacom - fix serial number handling on Cintiq 21UX2
  Input: fixup X86_MRST selects
  Input: sysrq - fix "stuck" SysRq mode
  Input: ad7877 - fix spi word size to 16 bit
  Input: pcf8574_keypad - fix off by one in pcf8574_kp_irq_handler()

9 years agoMerge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux...
Linus Torvalds [Sun, 27 Jun 2010 14:50:47 +0000]
Merge branch 'for_linus' of git://git./linux/kernel/git/jack/linux-fs-2.6

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs-2.6:
  ext3: update ctime when changing the file's permission by setfacl
  ext2: update ctime when changing the file's permission by setfacl

9 years agoMerge branch 'virtio' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6
Linus Torvalds [Sun, 27 Jun 2010 14:49:45 +0000]
Merge branch 'virtio' of git://git./linux/kernel/git/rusty/linux-2.6

* 'virtio' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6:
  virtio-pci: disable msi at startup
  virtio: return ENOMEM on out of memory

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes...
Linus Torvalds [Sun, 27 Jun 2010 14:41:04 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/jbarnes/pci-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6:
  PCI/PM: Do not use native PCIe PME by default

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6
Linus Torvalds [Sun, 27 Jun 2010 14:39:57 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/tiwai/sound-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6:
  ALSA: usb/endpoint, fix dangling pointer use
  ALSA: asihpi - Get rid of incorrect "long" types and casts.
  ASoC: DaVinci: Fix McASP hardware FIFO configuration
  ALSA: hda - Fix line-in for mb5 model MacBook (Pro) 5,1 / 5,2
  ALSA: usb-audio: fix UAC2 control value queries
  ALSA: usb-audio: parse UAC2 sample rate ranges correctly
  ALSA: usb-audio: fix control messages for USB_RECIP_INTERFACE
  ALSA: usb-audio: add check for faulty clock in parse_audio_format_rates_v2()
  ALSA: hda - Don't check capture source mixer if no ADC is available

9 years agoMerge branch 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jdelv...
Linus Torvalds [Sun, 27 Jun 2010 14:39:38 +0000]
Merge branch 'hwmon-for-linus' of git://git./linux/kernel/git/jdelvare/staging

* 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jdelvare/staging:
  hwmon: (k8temp) Bypass core swapping on single-core processors
  hwmon: (i5k_amb) Fix sysfs attribute for lockdep
  hwmon: (k10temp) Do not blacklist known working CPU models

9 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/wim/linux-2.6-watchdog
Linus Torvalds [Sun, 27 Jun 2010 14:38:20 +0000]
Merge git://git./linux/kernel/git/wim/linux-2.6-watchdog

* git://git.kernel.org/pub/scm/linux/kernel/git/wim/linux-2.6-watchdog:
  watchdog: at32ap700x_wdt: register misc device last in probe() function
  watchdog: [PATCH 3/3] imx2_wdt: fix section mismatch

9 years agoMerge git://git.infradead.org/iommu-2.6
Linus Torvalds [Sun, 27 Jun 2010 14:37:51 +0000]
Merge git://git.infradead.org/iommu-2.6

* git://git.infradead.org/iommu-2.6:
  intel-iommu: Force-disable IOMMU for iGFX on broken Cantiga revisions.
  intel-iommu: Fix double lock in get_domain_for_dev()
  intel-iommu: Fix reference by physical address in intel_iommu_attach_device()

9 years agoMerge branch 'for-linus' of git://gitorious.org/linux-omap-dss2/linux
Linus Torvalds [Sun, 27 Jun 2010 14:36:16 +0000]
Merge branch 'for-linus' of git://gitorious.org/linux-omap-dss2/linux

* 'for-linus' of git://gitorious.org/linux-omap-dss2/linux:
  OMAP: OMAPFB: fix rfbi.c compile error
  OMAPFB: LCDC: change update_mode to DISABLED when going suspend

9 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6
Linus Torvalds [Sun, 27 Jun 2010 14:34:02 +0000]
Merge git://git./linux/kernel/git/sfrench/cifs-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
  MAINTAINERS: change mailing list address for CIFS
  cifs: remove bogus first_time check in NTLMv2 session setup code
  cifs: don't call cifs_new_fileinfo unless cifs_open succeeds
  cifs: don't ignore cifs_posix_open_inode_helper return value
  cifs: clean up arguments to cifs_open_inode_helper
  cifs: pass instantiated filp back after open call
  cifs: move cifs_new_fileinfo call out of cifs_posix_open
  cifs: implement drop_inode superblock op
  cifs: don't attempt busy-file rename unless it's in same directory

9 years agoMerge branch 'merge' of git://git.secretlab.ca/git/linux-2.6
Linus Torvalds [Sun, 27 Jun 2010 14:30:25 +0000]
Merge branch 'merge' of git://git.secretlab.ca/git/linux-2.6

* 'merge' of git://git.secretlab.ca/git/linux-2.6:
  powerpc/5200: fix lite5200 ethernet phy address
  powerpc/5200: Fix build error in sound code.
  powerpc/5200: fix oops during going to standby
  powerpc/5200: add lite5200 onboard I2C eeprom and flash
  maintainers: Add git trees for SPI and device tree
  of: Drop properties with "/" in their name

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mattst88...
Linus Torvalds [Sun, 27 Jun 2010 14:29:19 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/mattst88/alpha-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mattst88/alpha-2.6:
  alpha: Fix de2104x driver failing to readout MAC address correctly
  alpha: Detect Super IO chip, no IDE on Avanti, enable EPP19
  alpha: fix pci_mmap_resource API breakage
  alpha: fix __arch_hweight32 typo

9 years agoMerge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc
Linus Torvalds [Sun, 27 Jun 2010 14:15:53 +0000]
Merge branch 'merge' of git://git./linux/kernel/git/benh/powerpc

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
  powerpc: Fix mpic_resume on early G5 macs
  powerpc: rtas_flash needs to use rtas_data_buf
  powerpc: Unconditionally enabled irq stacks
  powerpc/kexec: Wait for online/possible CPUs only.
  powerpc: Disable CONFIG_SYSFS_DEPRECATED
  powerpc/boot: Remove addRamdisk.c since it is now unused
  powerpc: Move kdump default base address to 64MB on 64bit
  powerpc: Remove dead CONFIG_HIGHPTE
  powerpc/fsl-booke: Move loadcam_entry back to asm code to fix SMP ftrace
  powerpc/fsl-booke: Fix InstructionTLBError execute permission check

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris...
Linus Torvalds [Sun, 27 Jun 2010 14:13:13 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/jmorris/security-testing-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6:
  TPM: ReadPubEK output struct fix

9 years agoMerge branch 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuil...
Linus Torvalds [Sun, 27 Jun 2010 14:05:02 +0000]
Merge branch 'rc-fixes' of git://git./linux/kernel/git/mmarek/kbuild-2.6

* 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6:
  kbuild: fix LOCALVERSION handling to match description
  kbuild: Fix modpost segfault

9 years agoMerge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux...
Linus Torvalds [Sun, 27 Jun 2010 14:03:12 +0000]
Merge branch 'release' of git://git./linux/kernel/git/lenb/linux-acpi-2.6

* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6:
  ACPI / PM: Do not enable GPEs for system wakeup in advance
  ACPICA: Truncate I/O addresses to 16 bits for Windows compatibility
  ACPICA: Limit maximum time for Sleep() operator
  ACPICA: Fix namestring associated with AE_NO_HANDLER exception
  ACPI / ACPICA: Fix sysfs GPE interface
  ACPI / ACPICA: Fix GPE initialization
  ACPI / ACPICA: Avoid writing full enable masks to GPE registers
  ACPI / ACPICA: Fix low-level GPE manipulation code
  ACPI / ACPICA: Use helper function for computing GPE masks
  ACPI / ACPICA: Do not attempt to disable GPE when installing handler
  ACPI: Disable Vista compatibility for Sony VGN-NS50B_L
  ACPI: fan: fix unbalanced code block
  ACPI: Store NVS state even when entering suspend to RAM
  suspend: Move NVS save/restore code to generic suspend functionality
  ACPI: Do not try to set up acpi processor stuff on cores exceeding maxcpus=
  ACPI: acpi_pad: Don't needlessly mark LAPIC unstable

9 years agoKEYS: Propagate error code instead of returning -EINVAL
Dan Carpenter [Fri, 11 Jun 2010 16:30:05 +0000]
KEYS: Propagate error code instead of returning -EINVAL

This is from a Smatch check I'm writing.

strncpy_from_user() returns -EFAULT on error so the first change just
silences a warning but doesn't change how the code works.

The other change is a bug fix because install_thread_keyring_to_cred()
can return a variety of errors such as -EINVAL, -EEXIST, -ENOMEM or
-EKEYREVOKED.

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

9 years agophylib: Add autoload support for the LXT973 phy.
David Woodhouse [Tue, 22 Jun 2010 02:38:13 +0000]
phylib: Add autoload support for the LXT973 phy.

Commit e13647c1 (phylib: Add support for the LXT973 phy.) added a new ID
but neglected to also add it to the MODULE_DEVICE_TABLE.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoISDN: hysdn, fix potential NULL dereference
Jiri Slaby [Tue, 22 Jun 2010 01:41:36 +0000]
ISDN: hysdn, fix potential NULL dereference

Stanse found that lp is dereferenced earlier than checked for being
NULL in hysdn_rx_netpkt. Move the initialization below the test.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Karsten Keil <isdn@linux-pingi.de>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Patrick McHardy <kaber@trash.net>
Cc: netdev@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agovxge: fix memory leak in vxge_alloc_msix() error path
Michal Schmidt [Thu, 24 Jun 2010 04:13:44 +0000]
vxge: fix memory leak in vxge_alloc_msix() error path

When pci_enable_msix() returned ret<0, entries and vxge_entries were leaked.
While at it, use the centralized exit idiom in the function.

Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Acked-by: Ram Vepa <ram.vepa@exar.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoisdn/gigaset: correct CAPI connection state storage
Tilman Schmidt [Mon, 21 Jun 2010 13:55:20 +0000]
isdn/gigaset: correct CAPI connection state storage

CAPI applications can handle several connections in parallel,
so one connection state per application isn't sufficient.
Store the connection state in the channel structure instead.

Impact: bugfix
Signed-off-by: Tilman Schmidt <tilman@imap.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoisdn/gigaset: encode HLC and BC together
Tilman Schmidt [Mon, 21 Jun 2010 13:55:05 +0000]
isdn/gigaset: encode HLC and BC together

Adapt to buggy device firmware which accepts setting HLC only in the
same command line as BC, by encoding HLC and BC in a single command
if both are specified, and rejecting HLC without BC.

Impact: bugfix
Signed-off-by: Tilman Schmidt <tilman@imap.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoisdn/gigaset: correct CAPI DATA_B3 Delivery Confirmation
Tilman Schmidt [Mon, 21 Jun 2010 13:54:50 +0000]
isdn/gigaset: correct CAPI DATA_B3 Delivery Confirmation

The Gigaset CAPI driver handled all DATA_B3_REQ messages as if the
Delivery Confirmation flag bit was set, delaying the emission of the
DATA_B3_CONF reply until the data was actually transmitted. Some
CAPI applications (notably Asterisk) aren't happy with that
behaviour. Change it to actually evaluate the Delivery Confirmation
flag as described the CAPI specification.

Impact: bugfix
Signed-off-by: Tilman Schmidt <tilman@imap.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoisdn/gigaset: correct CAPI voice connection encoding
Tilman Schmidt [Mon, 21 Jun 2010 13:54:35 +0000]
isdn/gigaset: correct CAPI voice connection encoding

Make the Gigaset CAPI driver select L2_VOICE (AT^SBPR=2) as the
layer 2 encoding for transparent connections, like the ISDN4Linux
variant.  L2_BITSYNC (AT^SBPR=0) mutes internal connections and
distorts external ones.

Impact: bugfix
Signed-off-by: Tilman Schmidt <tilman@imap.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoisdn/gigaset: honor CAPI application's buffer size request
Tilman Schmidt [Mon, 21 Jun 2010 13:54:19 +0000]
isdn/gigaset: honor CAPI application's buffer size request

Fix the Gigaset CAPI driver to limit the length of a connection's
payload data receive buffers to the corresponding CAPI application's
data buffer size, as some real-life CAPI applications tend to be
rather unhappy if they receive bigger data blocks than requested.

Impact: bugfix
Signed-off-by: Tilman Schmidt <tilman@imap.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agocpmac: do not leak struct net_device on phy_connect errors
Florian Fainelli [Sun, 20 Jun 2010 22:07:48 +0000]
cpmac: do not leak struct net_device on phy_connect errors

If the call to phy_connect fails, we will return directly instead of freeing
the previously allocated struct net_device.

Signed-off-by: Florian Fainelli <florian@openwrt.org>
CC: stable@kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agosmc91c92_cs: fix the problem that lan & modem does not work simultaneously
Ken Kawasaki [Sat, 19 Jun 2010 15:24:27 +0000]
smc91c92_cs: fix the problem that lan & modem does not work simultaneously

smc91c92_cs:
  Fix the problem that lan & modem does not work simultaneously
  in the Megahertz multi-function card.
  We need to write MEGAHERTZ_ISR to retrigger interrupt.

Signed-off-by: Ken Kawasaki <ken_kawasaki@spring.nifty.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoipv6: fix NULL reference in proxy neighbor discovery
stephen hemminger [Mon, 21 Jun 2010 11:00:13 +0000]
ipv6: fix NULL reference in proxy neighbor discovery

The addition of TLLAO option created a kernel OOPS regression
for the case where neighbor advertisement is being sent via
proxy path.  When using proxy, ipv6_get_ifaddr() returns NULL
causing the NULL dereference.

Change causing the bug was:
commit f7734fdf61ec6bb848e0bafc1fb8bad2c124bb50
Author: Octavian Purdila <opurdila@ixiacom.com>
Date:   Fri Oct 2 11:39:15 2009 +0000

    make TLLAO option for NA packets configurable

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoBluetooth: Bring back var 'i' increment
Gustavo F. Padovan [Fri, 18 Jun 2010 14:24:00 +0000]
Bluetooth: Bring back var 'i' increment

commit ff6e2163f28a1094fb5ca5950fe2b43c3cf6bc7a accidentally added a
regression on the bnep code. Fixing it.

Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoMAINTAINERS - Add an entry for the input MT protocol
Henrik Rydberg [Fri, 25 Jun 2010 02:10:40 +0000]
MAINTAINERS - Add an entry for the input MT protocol

This patch adds a maintainer for the input multitouch (MT) protocol,
such that get_maintainer.pl selects it whenever an MT event is present
in the patch.

Signed-off-by: Henrik Rydberg <rydberg@euromail.se>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

9 years agoext3: update ctime when changing the file's permission by setfacl
Miao Xie [Thu, 27 May 2010 08:28:40 +0000]
ext3: update ctime when changing the file's permission by setfacl

ext3 didn't update the ctime of the file when its permission was changed.

Steps to reproduce:
 # touch aaa
 # stat -c %Z aaa
 1275289822
 # setfacl -m  'u::x,g::x,o::x' aaa
 # stat -c %Z aaa
 1275289822 <- unchanged

But, according to the spec of the ctime, ext3 must update it.

Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Jan Kara <jack@suse.cz>

9 years agoext2: update ctime when changing the file's permission by setfacl
Jan Kara [Wed, 2 Jun 2010 14:26:51 +0000]
ext2: update ctime when changing the file's permission by setfacl

ext2 didn't update the ctime of the file when its permission was changed.

Steps to reproduce:
 # touch aaa
 # stat -c %Z aaa
 1275289822
 # setfacl -m  'u::x,g::x,o::x' aaa
 # stat -c %Z aaa
 1275289822                         <- unchanged

But, according to the spec of the ctime, ext2 must update it.

Port of ext3 patch by Miao Xie <miaox@cn.fujitsu.com>.

Signed-off-by: Jan Kara <jack@suse.cz>

9 years agoxfrm: check bundle policy existance before dereferencing it
Timo Teräs [Thu, 24 Jun 2010 21:35:00 +0000]
xfrm: check bundle policy existance before dereferencing it

Fix the bundle validation code to not assume having a valid policy.
When we have multiple transformations for a xfrm policy, the bundle
instance will be a chain of bundles with only the first one having
the policy reference. When policy_genid is bumped it will expire the
first bundle in the chain which is equivalent of expiring the whole
chain.

Reported-bisected-and-tested-by: Justin P. Mattock <justinmattock@gmail.com>
Signed-off-by: Timo Teräs <timo.teras@iki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agomd/raid5: don't include 'spare' drives when reshaping to fewer devices.
NeilBrown [Thu, 17 Jun 2010 07:48:26 +0000]
md/raid5: don't include 'spare' drives when reshaping to fewer devices.

There are few situations where it would make any sense to add a spare
when reducing the number of devices in an array, but it is
conceivable:  A 6 drive RAID6 with two missing devices could be
reshaped to a 5 drive RAID6, and a spare could become available
just in time for the reshape, but not early enough to have been
recovered first.  'freezing' recovery can make this easy to
do without any races.

However doing such a thing is a bad idea.  md will not record the
partially-recovered state of the 'spare' and when the reshape
finished it will think that the spare is still spare.
Easiest way to avoid this confusion is to simply disallow it.

Signed-off-by: NeilBrown <neilb@suse.de>

9 years agomd/raid5: add a missing 'continue' in a loop.
NeilBrown [Thu, 17 Jun 2010 07:41:03 +0000]
md/raid5: add a missing 'continue' in a loop.

As the comment says, the tail of this loop only applies to devices
that are not fully in sync, so if In_sync was set, we should avoid
the rest of the loop.

This bug will hardly ever cause an actual problem.  The worst it
can do is allow an array to be assembled that is dirty and degraded,
which is not generally a good idea (without warning the sysadmin
first).

This will only happen if the array is RAID4 or a RAID5/6 in an
intermediate state during a reshape and so has one drive that is
all 'parity' - no data - while some other device has failed.

This is certainly possible, but not at all common.

Signed-off-by: NeilBrown <neilb@suse.de>

9 years agomd/raid5: Allow recovered part of partially recovered devices to be in-sync
NeilBrown [Thu, 17 Jun 2010 07:25:21 +0000]
md/raid5: Allow recovered part of partially recovered devices to be in-sync

During a recovery of reshape the early part of some devices might be
in-sync while the later parts are not.
We we know we are looking at an early part it is good to treat that
part as in-sync for stripe calculations.

This is particularly important for a reshape which suffers device
failure.  Treating the data as in-sync can mean the difference between
data-safety and data-loss.

Signed-off-by: NeilBrown <neilb@suse.de>

9 years agomd/raid5: More careful check for "has array failed".
NeilBrown [Wed, 16 Jun 2010 07:17:53 +0000]
md/raid5: More careful check for "has array failed".

When we are reshaping an array, the device failure combinations
that cause us to decide that the array as failed are more subtle.

In particular, any 'spare' will be fully in-sync in the section
of the array that has already been reshaped, thus failures that
affect only that section are less critical.

So encode this subtlety in a new function and call it as appropriate.

The case that showed this problem was a 4 drive RAID5 to 8 drive RAID6
conversion where the last two devices failed.
This resulted in:

  good good good good incomplete good good failed failed

while converting a 5-drive RAID6 to 8 drive RAID5
The incomplete device causes the whole array to look bad,
bad as it was actually good for the section that had been
converted to 8-drives, all the data was actually safe.

Reported-by: Terry Morris <tbmorris@tbmorris.com>
Signed-off-by: NeilBrown <neilb@suse.de>

9 years agomd: Don't update ->recovery_offset when reshaping an array to fewer devices.
NeilBrown [Wed, 16 Jun 2010 07:01:25 +0000]
md: Don't update ->recovery_offset when reshaping an array to fewer devices.

When an array is reshaped to have fewer devices, the reshape proceeds
from the end of the devices to the beginning.

If a device happens to be non-In_sync (which is possible but rare)
we would normally update the ->recovery_offset as the reshape
progresses. However that would be wrong as the recover_offset records
that the early part of the device is in_sync, while in fact it would
only be the later part that is in_sync, and in any case the offset
number would be measured from the wrong end of the device.

Relatedly, if after a reshape a spare is discovered to not be
recoverred all the way to the end, not allow spare_active
to incorporate it in the array.

This becomes relevant in the following sample scenario:

A 4 drive RAID5 is converted to a 6 drive RAID6 in a combined
operation.
The RAID5->RAID6 conversion will cause a 5 drive to be included as a
spare, then the 5drive -> 6drive reshape will effectively rebuild that
spare as it progresses.  The 6th drive is treated as in_sync the whole
time as there is never any case that we might consider reading from
it, but must not because there is no valid data.

If we interrupt this reshape part-way through and reverse it to return
to a 5-drive RAID6 (or event a 4-drive RAID5), we don't want to update
the recovery_offset - as that would be wrong - and we don't want to
include that spare as active in the 5-drive RAID6 when the reversed
reshape completed and it will be mostly out-of-sync still.

Signed-off-by: NeilBrown <neilb@suse.de>

9 years agomd/raid5: avoid oops when number of devices is reduced then increased.
NeilBrown [Wed, 16 Jun 2010 06:45:16 +0000]
md/raid5: avoid oops when number of devices is reduced then increased.

The entries in the stripe_cache maintained by raid5 are enlarged
when we increased the number of devices in the array, but not
shrunk when we reduce the number of devices.
So if entries are added after reducing the number of devices, we
much ensure to initialise the whole entry, not just the part that
is currently relevant.  Otherwise if we enlarge the array again,
we will reference uninitialised values.

As grow_buffers/shrink_buffer now want to use a count that is stored
explicity in the raid_conf, they should get it from there rather than
being passed it as a parameter.

Signed-off-by: NeilBrown <neilb@suse.de>

9 years agomd: enable raid4->raid0 takeover
Maciej Trela [Wed, 16 Jun 2010 10:56:12 +0000]
md: enable raid4->raid0 takeover

Only level 5 with layout=PARITY_N can be taken over to raid0 now.
Lets allow level 4 either.

Signed-off-by: Maciej Trela <maciej.trela@intel.com>
Signed-off-by: NeilBrown <neilb@suse.de>

9 years agomd: clear layout after ->raid0 takeover
Maciej Trela [Wed, 16 Jun 2010 10:55:14 +0000]
md: clear layout after ->raid0 takeover

After takeover from raid5/10 -> raid0 mddev->layout is not cleared.

Signed-off-by: Maciej Trela <maciej.trela@intel.com>
Signed-off-by: NeilBrown <neilb@suse.de>

9 years agomd: fix raid10 takeover: use new_layout for setup_conf
Maciej Trela [Wed, 16 Jun 2010 10:46:29 +0000]
md: fix raid10 takeover: use new_layout for setup_conf

Use mddev->new_layout in setup_conf.
Also use new_chunk, and don't set ->degraded in takeover().  That
gets set in run()

Signed-off-by: Maciej Trela <maciej.trela@intel.com>
Signed-off-by: NeilBrown <neilb@suse.de>

9 years agomd: fix handling of array level takeover that re-arranges devices.
NeilBrown [Tue, 15 Jun 2010 08:36:03 +0000]
md: fix handling of array level takeover that re-arranges devices.

Most array level changes leave the list of devices largely unchanged,
possibly causing one at the end to become redundant.
However conversions between RAID0 and RAID10 need to renumber
all devices (except 0).

This renumbering is currently being done in the ->run method when the
new personality takes over.  However this is too late as the common
code in md.c might already have invalidated some of the devices if
they had a ->raid_disk number that appeared to high.

Moving it into the ->takeover method is too early as the array is
still active at that time and wrong ->raid_disk numbers could cause
confusion.

So add a ->new_raid_disk field to mdk_rdev_s and use it to communicate
the new raid_disk number.
Now the common code knows exactly which devices need to be renumbered,
and which can be invalidated, and can do it all at a convenient time
when the array is suspend.
It can also update some symlinks in sysfs which previously were not be
updated correctly.

Reported-by: Maciej Trela <maciej.trela@intel.com>
Signed-off-by: NeilBrown <neilb@suse.de>

9 years agomd: raid10: Fix null pointer dereference in fix_read_error()
Prasanna S. Panchamukhi [Thu, 24 Jun 2010 03:31:03 +0000]
md: raid10: Fix null pointer dereference in fix_read_error()

Such NULL pointer dereference can occur when the driver was fixing the
read errors/bad blocks and the disk was physically removed
causing a system crash. This patch check if the
rcu_dereference() returns valid rdev before accessing it in fix_read_error().

Cc: stable@kernel.org
Signed-off-by: Prasanna S. Panchamukhi <prasanna.panchamukhi@riverbed.com>
Signed-off-by: Rob Becker <rbecker@riverbed.com>
Signed-off-by: NeilBrown <neilb@suse.de>

9 years agoRestore partition detection of newly created md arrays.
NeilBrown [Thu, 24 Jun 2010 03:31:03 +0000]
Restore partition detection of newly created md arrays.

Commit  b821eaa572fd737faaf6928ba046e571526c36c6 broke partition
detection for md arrays.

The logic was almost right.  However if revalidate_disk is called
when the device is not yet open, bdev->bd_disk won't be set, so the
flush_disk() Call will not set bd_invalidated.

So when md_open is called we still need to ensure that
->bd_invalidated gets set.  This is easily done with a call to
check_disk_size_change in the place where the offending commit removed
check_disk_change.  At the important times, the size will have changed
from 0 to non-zero, so check_disk_size_change will set bd_invalidated.

Tested-by: Duncan <1i5t5.duncan@cox.net>
Reported-by: Duncan <1i5t5.duncan@cox.net>
Signed-off-by: NeilBrown <neilb@suse.de>

9 years agosky2: enable rx/tx in sky2_phy_reinit()
Brandon Philips [Wed, 16 Jun 2010 16:21:58 +0000]
sky2: enable rx/tx in sky2_phy_reinit()

sky2_phy_reinit is called by the ethtool helpers sky2_set_settings,
sky2_nway_reset and sky2_set_pauseparam when netif_running.

However, at the end of sky2_phy_init GM_GP_CTRL has GM_GPCR_RX_ENA and
GM_GPCR_TX_ENA cleared. So, doing these commands causes the device to
stop working:

$ ethtool -r eth0
$ ethtool -A eth0 autoneg off

Fix this issue by enabling Rx/Tx after running sky2_phy_init in
sky2_phy_reinit.

Signed-off-by: Brandon Philips <bphilips@suse.de>
Tested-by: Brandon Philips <bphilips@suse.de>
Cc: stable@kernel.org
Tested-by: Mike McCormack <mikem@ring3k.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoMAINTAINERS: change mailing list address for CIFS
Jeff Layton [Sun, 6 Jun 2010 12:05:58 +0000]
MAINTAINERS: change mailing list address for CIFS

We're moving the mailing list to linux-cifs@vger.kernel.org.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>

9 years agocnic: Disable statistics initialization for eth clients that do not support statistics
Dmitry Kravkov [Wed, 23 Jun 2010 18:57:09 +0000]
cnic: Disable statistics initialization for eth clients that do not support statistics

Disable statistics initialization for eth clients that do not support
 statistics. This prevents memory corruption on bnx2x hw.

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>

9 years agonet: add dependency on fw class module to qlcnic and netxen_nic
Anirban Chakraborty [Wed, 23 Jun 2010 18:49:42 +0000]
net: add dependency on fw class module to qlcnic and netxen_nic

netxen_nic and qlcnic driver depends on firmware_class module.

Signed-off-by: Anirban Chakraborty <anirban.chakraborty@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agosnmp: fix SNMP_ADD_STATS()
Eric Dumazet [Wed, 23 Jun 2010 00:32:03 +0000]
snmp: fix SNMP_ADD_STATS()

commit aa2ea0586d9d (tcp: fix outsegs stat for TSO segments) incorrectly
assumed SNMP_ADD_STATS() was used from BH context.

Fix this using mib[!in_softirq()] instead of mib[0]

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agovirtio-pci: disable msi at startup
Michael S. Tsirkin [Thu, 24 Jun 2010 04:49:06 +0000]
virtio-pci: disable msi at startup

virtio-pci resets the device at startup by writing to the status
register, but this does not clear the pci config space,
specifically msi enable status which affects register
layout.

This breaks things like kdump when they try to use e.g. virtio-blk.

Fix by forcing msi off at startup. Since pci.c already has
a routine to do this, we export and use it instead of duplicating code.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Vivek Goyal <vgoyal@redhat.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: linux-pci@vger.kernel.org
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: stable@kernel.org

9 years agovirtio: return ENOMEM on out of memory
Michael S. Tsirkin [Thu, 10 Jun 2010 15:16:11 +0000]
virtio: return ENOMEM on out of memory

add_buf returns ring size on out of memory,
this is not what devices expect.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: stable@kernel.org # .34.x

9 years agoInput: wacom - fix serial number handling on Cintiq 21UX2
Ping Cheng [Tue, 22 Jun 2010 18:21:34 +0000]
Input: wacom - fix serial number handling on Cintiq 21UX2

Cintiq 21UX2 added 8 more bits for the tool serial number and more
buttons for the expresskey. We did not enable them properly in the
last patch.

Signed-off-by: Ping Cheng <pingc@wacom.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

9 years agoInput: fixup X86_MRST selects
Randy Dunlap [Fri, 18 Jun 2010 05:31:17 +0000]
Input: fixup X86_MRST selects

Some of the recent X86_MRST additions make some "select"s
conditional on X86_MRST but missed some related kconfig symbols,
causing:

drivers/built-in.o: In function `ps2_end_command':
(.text+0x257ab2): undefined reference to `i8042_check_port_owner'
drivers/built-in.o: In function `ps2_end_command':
(.text+0x257ae1): undefined reference to `i8042_unlock_chip'
drivers/built-in.o: In function `ps2_begin_command':
(.text+0x257b40): undefined reference to `i8042_check_port_owner'
drivers/built-in.o: In function `ps2_begin_command':
(.text+0x257b6f): undefined reference to `i8042_lock_chip'

when SERIO_I8042=m, SERIO_LIBPS2=y, KEYBOARD_ATKBD=y.

We need to make i8042 dependant upon !X86_MRST and allow deselecting
atkbd on Moorestown even when !CONFIG_EMBEDDED.

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Jacob Pan <jacob.jun.pan@intel.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>

9 years agoMerge commit 'v2.6.35-rc3' into for-linus
Dmitry Torokhov [Tue, 22 Jun 2010 18:32:04 +0000]
Merge commit 'v2.6.35-rc3' into for-linus

9 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wirel...
David S. Miller [Tue, 22 Jun 2010 17:54:12 +0000]
Merge branch 'master' of git://git./linux/kernel/git/linville/wireless-2.6

9 years agoNFSv4: Fix an embarassing typo in encode_attrs()
Trond Myklebust [Tue, 22 Jun 2010 12:52:39 +0000]
NFSv4: Fix an embarassing typo in encode_attrs()

Apparently, we have never been able to set the atime correctly from the
NFSv4 client.

Reported-by: 小倉一夫 <ka-ogura@bd6.so-net.ne.jp>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: stable@kernel.org

9 years agoNFSv4: Ensure that /proc/self/mountinfo displays the minor version number
Trond Myklebust [Fri, 18 Jun 2010 16:23:58 +0000]
NFSv4: Ensure that /proc/self/mountinfo displays the minor version number

Currently, we do not display the minor version mount parameter in the
/proc mount info.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: stable@kernel.org

9 years agoNFSv4.1: Ensure that we initialise the session when following a referral
Trond Myklebust [Thu, 17 Jun 2010 15:45:12 +0000]
NFSv4.1: Ensure that we initialise the session when following a referral

Put the code that is common to both the referral and ordinary mount cases
into a common helper routine.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>

9 years agoSUNRPC: Fix a re-entrancy bug in xs_tcp_read_calldir()
Trond Myklebust [Wed, 16 Jun 2010 17:57:32 +0000]
SUNRPC: Fix a re-entrancy bug in xs_tcp_read_calldir()

If the attempt to read the calldir fails, then instead of storing the read
bytes, we currently discard them. This leads to a garbage final result when
upon re-entry to the same routine, we read the remaining bytes.

Fixes the regression in bugzilla number 16213. Please see
    https://bugzilla.kernel.org/show_bug.cgi?id=16213

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: stable@kernel.org

9 years agonfs4 use mandatory attribute file type in nfs4_get_root
Andy Adamson [Wed, 16 Jun 2010 13:51:02 +0000]
nfs4 use mandatory attribute file type in nfs4_get_root

S_ISDIR(fsinfo.fattr->mode) checks the file type rather than the mode bits,
so we should be checking for the NFS_ATTR_FATTR_TYPE fattr property.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: stable@kernel.org

9 years agohso: remove setting of low_latency flag
Filip Aben [Tue, 22 Jun 2010 17:10:35 +0000]
hso: remove setting of low_latency flag

This patch removes the setting of the low_latency flag.
tty_flip_buffer_push() is occasionally being called in irq context, which
causes a hang if the low_latency flag is set.
Removing the low_latency flag only seems to impact the flush to ldisc,
which will now be put on a workqueue.

Signed-off-by: Filip Aben <f.aben@option.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoclocksource: sh_cmt: Fix up bogus shift value.
Paul Mundt [Tue, 22 Jun 2010 07:41:31 +0000]
clocksource: sh_cmt: Fix up bogus shift value.

The previous CMT fixup accidentally copied in the TMU shift value, reset
this back to its original value while preserving the TMU fix.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

9 years agoudp: Fix bogus UFO packet generation
Herbert Xu [Tue, 15 Jun 2010 01:52:25 +0000]
udp: Fix bogus UFO packet generation

It has been reported that the new UFO software fallback path
fails under certain conditions with NFS.  I tracked the problem
down to the generation of UFO packets that are smaller than the
MTU.  The software fallback path simply discards these packets.

This patch fixes the problem by not generating such packets on
the UFO path.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agolasi82596: fix netdev_mc_count conversion
Helge Deller [Mon, 21 Jun 2010 20:50:18 +0000]
lasi82596: fix netdev_mc_count conversion

Fix commit 4cd24eaf0 (net: use netdev_mc_count and netdev_mc_empty when
appropriate)

Signed-off-by: Helge Deller <deller@gmx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoNET: MIPSsim: Fix modpost warning.
Ralf Baechle [Mon, 21 Jun 2010 03:44:50 +0000]
NET: MIPSsim: Fix modpost warning.

$ make CONFIG_DEBUG_SECTION_MISMATCH=y
[...]
WARNING: drivers/net/built-in.o(.data+0x0): Section mismatch in reference from the variable mipsnet_driver to the function .init.text:mipsnet_probe()
The variable mipsnet_driver references
the function __init mipsnet_probe()
If the reference is valid then annotate the
variable with __init* or __refdata (see linux/init.h) or name the variable:
*_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console,
[...]

Fixed by making mipsnet_probe __devinit.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

 drivers/net/mipsnet.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoMerge branch 'fix/misc' into for-linus
Takashi Iwai [Mon, 21 Jun 2010 15:08:41 +0000]
Merge branch 'fix/misc' into for-linus

9 years agoALSA: usb/endpoint, fix dangling pointer use
Jiri Slaby [Mon, 21 Jun 2010 15:03:21 +0000]
ALSA: usb/endpoint, fix dangling pointer use

Stanse found that in snd_usb_parse_audio_endpoints, there is a
dangling pointer dereference. When snd_usb_parse_audio_format fails,
fp is freed, and continue invoked. On the next loop, there is
"fp && fp->altsetting == 1 && fp->channels == 1" test, but fp is set
from the last iteration (but is bogus) and thus ilegally dereferenced.

Set fp to NULL before "continue".

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Acked-by: Daniel Mack <daniel@caiaq.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

9 years agoarch/sh/mm: Eliminate a double lock
Julia Lawall [Sun, 20 Jun 2010 11:24:54 +0000]
arch/sh/mm: Eliminate a double lock

The function begins and ends with a read_lock.  The latter is changed to a
read_unlock.

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

// <smpl>
@locked@
expression E1;
position p;
@@

read_lock(E1@p,...);

@r exists@
expression x <= locked.E1;
expression locked.E1;
expression E2;
identifier lock;
position locked.p,p1,p2;
@@

*lock@p1 (E1@p,...);
... when != E1
    when != \(x = E2\|&x\)
*lock@p2 (E1,...);
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Acked-by: Matt Fleming <matt@console-pimps.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

9 years agoMerge branch 'fix/misc' into for-linus
Takashi Iwai [Sun, 20 Jun 2010 08:38:19 +0000]
Merge branch 'fix/misc' into for-linus

9 years agoMerge branch 'fix/asoc' into for-linus
Takashi Iwai [Sun, 20 Jun 2010 08:38:14 +0000]
Merge branch 'fix/asoc' into for-linus

9 years agohwmon: (k8temp) Bypass core swapping on single-core processors
Jean Delvare [Sun, 20 Jun 2010 07:22:32 +0000]
hwmon: (k8temp) Bypass core swapping on single-core processors

Commit a2e066bba2aad6583e3ff648bf28339d6c9f0898 introduced core
swapping for CPU models 64 and later. I recently had a report about
a Sempron 3200+, model 95, for which this patch broke temperature
reading. It happens that this is a single-core processor, so the
effect of the swapping was to read a temperature value for a core
that didn't exist, leading to an incorrect value (-49 degrees C.)

Disabling core swapping on singe-core processors should fix this.

Additional comment from Andreas:

The BKDG says

  Thermal Sensor Core Select (ThermSenseCoreSel)-Bit 2. This bit
  selects the CPU whose temperature is reported in the CurTemp
  field. This bit only applies to dual core processors. For
  single core processors CPU0 Thermal Sensor is always selected.

k8temp_probe() correctly detected that SEL_CORE can't be used on single
core CPU. Thus k8temp did never update the temperature values stored
in temp[1][x] and -49 degrees was reported. For single core CPUs we
must use the values read into temp[0][x].

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Tested-by: Rick Moritz <rhavin@gmx.net>
Acked-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: stable@kernel.org

9 years agohwmon: (i5k_amb) Fix sysfs attribute for lockdep
KAMEZAWA Hiroyuki [Sun, 20 Jun 2010 07:22:31 +0000]
hwmon: (i5k_amb) Fix sysfs attribute for lockdep

i5k_amb.ko uses dynamically allocated memory (by kmalloc) for
attributes passed to sysfs. So, sysfs_attr_init() should be called
for working happy with lockdep.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: stable@kernel.org [2.6.34 only]

9 years agohwmon: (k10temp) Do not blacklist known working CPU models
Jean Delvare [Sun, 20 Jun 2010 07:22:31 +0000]
hwmon: (k10temp) Do not blacklist known working CPU models

When detecting AM2+ or AM3 socket with DDR2, only blacklist cores
which are known to exist in AM2+ format.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Clemens Ladisch <clemens@ladisch.de>
Cc: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: stable@kernel.org

9 years agoath5k: initialize ah->ah_current_channel
Bob Copeland [Fri, 18 Jun 2010 17:15:23 +0000]
ath5k: initialize ah->ah_current_channel

ath5k assumes ah_current_channel is always a valid pointer in
several places, but a newly created interface may not have a
channel.  To avoid null pointer dereferences, set it up to point
to the first available channel until later reconfigured.

This fixes the following oops:
$ rmmod ath5k
$ insmod ath5k
$ iw phy0 set distance 11000

BUG: unable to handle kernel NULL pointer dereference at 00000006
IP: [<d0a1ff24>] ath5k_hw_set_coverage_class+0x74/0x1b0 [ath5k]
*pde = 00000000
Oops: 0000 [#1]
last sysfs file: /sys/devices/pci0000:00/0000:00:0e.0/ieee80211/phy0/index
Modules linked in: usbhid option usb_storage usbserial usblp evdev lm90
scx200_acb i2c_algo_bit i2c_dev i2c_core via_rhine ohci_hcd ne2k_pci
8390 leds_alix2 xt_IMQ imq nf_nat_tftp nf_conntrack_tftp nf_nat_irc nf_cc

Pid: 1597, comm: iw Not tainted (2.6.32.14 #8)
EIP: 0060:[<d0a1ff24>] EFLAGS: 00010296 CPU: 0
EIP is at ath5k_hw_set_coverage_class+0x74/0x1b0 [ath5k]
EAX: 000000c2 EBX: 00000000 ECX: ffffffff EDX: c12d2080
ESI: 00000019 EDI: cf8c0000 EBP: d0a30edc ESP: cfa09bf4
  DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 0068
Process iw (pid: 1597, ti=cfa09000 task=cf88a000 task.ti=cfa09000)
Stack:
  d0a34f35 d0a353f8 d0a30edc 000000fe cf8c0000 00000000 1900063d cfa8c9e0
<0> cfa8c9e8 cfa8c0c0 cfa8c000 d0a27f0c 199d84b4 cfa8c200 00000010 d09bfdc7
<0> 00000000 00000000 ffffffff d08e0d28 cf9263c0 00000001 cfa09cc4 00000000
Call Trace:
  [<d0a27f0c>] ? ath5k_hw_attach+0xc8c/0x3c10 [ath5k]
  [<d09bfdc7>] ? __ieee80211_request_smps+0x1347/0x1580 [mac80211]
  [<d08e0d28>] ? nl80211_send_scan_start+0x7b8/0x4520 [cfg80211]
  [<c10f5db9>] ? nla_parse+0x59/0xc0
  [<c11ca8d9>] ? genl_rcv_msg+0x169/0x1a0
  [<c11ca770>] ? genl_rcv_msg+0x0/0x1a0
  [<c11c7e68>] ? netlink_rcv_skb+0x38/0x90
  [<c11c9649>] ? genl_rcv+0x19/0x30
  [<c11c7c03>] ? netlink_unicast+0x1b3/0x220
  [<c11c893e>] ? netlink_sendmsg+0x26e/0x290
  [<c11a409e>] ? sock_sendmsg+0xbe/0xf0
  [<c1032780>] ? autoremove_wake_function+0x0/0x50
  [<c104d846>] ? __alloc_pages_nodemask+0x106/0x530
  [<c1074933>] ? do_lookup+0x53/0x1b0
  [<c10766f9>] ? __link_path_walk+0x9b9/0x9e0
  [<c11acab0>] ? verify_iovec+0x50/0x90
  [<c11a42b1>] ? sys_sendmsg+0x1e1/0x270
  [<c1048e50>] ? find_get_page+0x10/0x50
  [<c104a96f>] ? filemap_fault+0x5f/0x370
  [<c1059159>] ? __do_fault+0x319/0x370
  [<c11a55b4>] ? sys_socketcall+0x244/0x290
  [<c101962c>] ? do_page_fault+0x1ec/0x270
  [<c1019440>] ? do_page_fault+0x0/0x270
  [<c1002ae5>] ? syscall_call+0x7/0xb
Code: 00 b8 fe 00 00 00 b9 f8 53 a3 d0 89 5c 24 14 89 7c 24 10 89 44 24
0c 89 6c 24 08 89 4c 24 04 c7 04 24 35 4f a3 d0 e8 7c 30 60 f0 <0f> b7
43 06 ba 06 00 00 00 a8 10 75 0e 83 e0 20 83 f8 01 19 d2
EIP: [<d0a1ff24>] ath5k_hw_set_coverage_class+0x74/0x1b0 [ath5k] SS:ESP
0068:cfa09bf4
CR2: 0000000000000006
---[ end trace 54f73d6b10ceb87b ]---

Cc: stable@kernel.org
Reported-by: Steve Brown <sbrown@cortland.com>
Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

9 years agoPCI/PM: Do not use native PCIe PME by default
Rafael J. Wysocki [Fri, 18 Jun 2010 15:04:22 +0000]
PCI/PM: Do not use native PCIe PME by default

Commit c7f486567c1d0acd2e4166c47069835b9f75e77b
(PCI PM: PCIe PME root port service driver) causes the native PCIe
PME signaling to be used by default, if the BIOS allows the kernel to
control the standard configuration registers of PCIe root ports.
However, the native PCIe PME is coupled to the native PCIe hotplug
and calling pcie_pme_acpi_setup() makes some BIOSes expect that
the native PCIe hotplug will be used as well.  That, in turn, causes
problems to appear on systems where the PCIe hotplug driver is not
loaded.  The usual symptom, as reported by Jaroslav Kameník and
others, is that the ACPI GPE associated with PCIe hotplug keeps
firing continuously causing kacpid to take substantial percentage
of CPU time.

To work around this issue, change the default so that the native
PCIe PME signaling is only used if directly requested with the help
of the pcie_pme= command line switch.

Fixes https://bugzilla.kernel.org/show_bug.cgi?id=15924 , which is
a listed regression from 2.6.33.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Reported-by: Jaroslav Kameník <jaroslav@kamenik.cz>
Tested-by: Antoni Grzymala <antekgrzymala@gmail.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

9 years agopercpu: fix first chunk match in per_cpu_ptr_to_phys()
Tejun Heo [Fri, 18 Jun 2010 09:44:31 +0000]
percpu: fix first chunk match in per_cpu_ptr_to_phys()

per_cpu_ptr_to_phys() determines whether the passed in @addr belongs
to the first_chunk or not by just matching the address against the
address range of the base unit (unit0, used by cpu0).  When an adress
from another cpu was passed in, it will always determine that the
address doesn't belong to the first chunk even when it does.  This
makes the function return a bogus physical address which may lead to
crash.

This problem was discovered by Cliff Wickman while investigating a
crash during kdump on a SGI UV system.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Cliff Wickman <cpw@sgi.com>
Tested-by: Cliff Wickman <cpw@sgi.com>
Cc: stable@kernel.org

9 years agobridge: fdb cleanup runs too often
stephen hemminger [Tue, 15 Jun 2010 06:14:12 +0000]
bridge: fdb cleanup runs too often

It is common in end-node, non STP bridges to set forwarding
delay to zero; which causes the forwarding database cleanup
to run every clock tick. Change to run only as soon as needed
or at next ageing timer interval which ever is sooner.

Use round_jiffies_up macro rather than attempting round up
by changing value.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoMerge branch 'bugzilla-15951' into release
Len Brown [Thu, 17 Jun 2010 16:18:30 +0000]
Merge branch 'bugzilla-15951' into release

9 years agoACPI / PM: Do not enable GPEs for system wakeup in advance
Rafael J. Wysocki [Thu, 17 Jun 2010 15:40:57 +0000]
ACPI / PM: Do not enable GPEs for system wakeup in advance

After commit 9630bdd9b15d2f489c646d8bc04b60e53eb5ec78
(ACPI: Use GPE reference counting to support shared GPEs) the wakeup
enable mask bits of GPEs are set as soon as the GPEs are enabled to
wake up the system.  Unfortunately, this leads to a regression
reported by Michal Hocko, where a system is woken up from ACPI S5 by
a device that is not supposed to do that, because the wakeup enable
mask bit of this device's GPE is always set when
acpi_enter_sleep_state() calls acpi_hw_enable_all_wakeup_gpes(),
although it should only be set if the device is supposed to wake up
the system from the target state.

To work around this issue, rework the ACPI power management code so
that GPEs are not enabled to wake up the system upfront, but only
during a system state transition when the target state of the system
is known.  [Of course, this means that the reference counting of
"wakeup" GPEs doesn't really make sense and it is sufficient to
set/unset the wakeup mask bits for them during system sleep
transitions.  This will allow us to simplify the GPE handling code
quite a bit, but that change is too intrusive for 2.6.35.]

Fixes https://bugzilla.kernel.org/show_bug.cgi?id=15951

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Reported-and-tested-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Len Brown <len.brown@intel.com>

9 years agobnx2: fix dma_get_ops compilation breakage
FUJITA Tomonori [Thu, 17 Jun 2010 15:56:05 +0000]
bnx2: fix dma_get_ops compilation breakage

This removes dma_get_ops() prefetch optimization in bnx2.

bnx2 uses dma_get_ops() to see if dma_sync_single_for_cpu() is
noop. bnx2 does prefetch if it's noop.

But dma_get_ops() isn't available on all the architectures (only the
architectures that uses dma_map_ops struct have it). Using
dma_get_ops() in drivers leads to compilation breakage on many
architectures.

This patch removes dma_get_ops() and changes bnx2 to do prefetch on
all the architectures. This adds useless prefetch on non-coherent
architectures but this is harmless. It is also unlikely to cause the
performance drop.

[ Remove now unused local variable 'pdev' -DaveM ]

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agowatchdog: at32ap700x_wdt: register misc device last in probe() function
Hans-Christian Egtvedt [Tue, 8 Jun 2010 06:44:32 +0000]
watchdog: at32ap700x_wdt: register misc device last in probe() function

This patch reworks the probe() function in the at32ap700x_wdt driver, this to
make sure the miscdev is properly initialized and the driver is ready to be
accessed.

Reported-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Hans-Christian Egtvedt <hans-christian.egtvedt@atmel.com>
Signed-off-by: Wim Van sebroeck <wim@iguana.be>

9 years agopercpu: fix trivial bugs in pcpu_build_alloc_info()
Pavel V. Panteleev [Thu, 17 Jun 2010 08:07:25 +0000]
percpu: fix trivial bugs in pcpu_build_alloc_info()

Fix the following two trivial bugs in pcpu_build_alloc_info()

* we should memset group_cnt to 0 by size of group_cnt, not size of
  group_map (both are of the same size, so the bug isn't dangerous)

* we can delete useless variable group_cnt_max.

Signed-off-by: Pavel V. Panteleev <pp_84@mail.ru>
Signed-off-by: Tejun Heo <tj@kernel.org>

9 years agoALSA: asihpi - Get rid of incorrect "long" types and casts.
Eliot Blennerhassett [Thu, 17 Jun 2010 05:08:36 +0000]
ALSA: asihpi - Get rid of incorrect "long" types and casts.

These give incorrect results for index wrap on 64 bit.

Signed-off-by: Eliot Blennerhassett <eblennerhassett@audioscience.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

9 years agoucc_geth: fix for RX skb buffers recycling
Sergey Matyukevich [Mon, 14 Jun 2010 06:35:20 +0000]
ucc_geth: fix for RX skb buffers recycling

This patch implements a proper modification of RX skb buffers before
recycling. Adjusting only skb->data is not enough because after that
skb->tail and skb->len become incorrect.

Signed-off-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agopcnet_cs: add new id (TOSHIBA Modem/LAN Card)
Ken Kawasaki [Sat, 12 Jun 2010 00:17:28 +0000]
pcnet_cs: add new id (TOSHIBA Modem/LAN Card)

pcnet_cs:
serial_cs:
    add new id (TOSHIBA Modem/LAN Card)

Signed-off-by: Ken Kawasaki <ken_kawasaki@spring.nifty.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agogianfar: Fix oversized packets handling
Anton Vorontsov [Fri, 11 Jun 2010 10:51:03 +0000]
gianfar: Fix oversized packets handling

Issuing the following command on host:

$ ifconfig eth2 mtu 1600 ; ping 10.0.0.27 -s 1485 -c 1

Makes some boards (tested with MPC8315 rev 1.1 and MPC8313 rev 1.0)
oops like this:

  skb_over_panic: text:c0195914 len:1537 put:1537 head:c79e4800 data:c79e4880 tail:0xc79e4e81 end:0xc79e4e80 dev:eth1
  ------------[ cut here ]------------
  kernel BUG at net/core/skbuff.c:127!
  Oops: Exception in kernel mode, sig: 5 [#1]
  MPC831x RDB
  last sysfs file: /sys/kernel/uevent_seqnum
  Modules linked in:
  NIP: c01c1840 LR: c01c1840 CTR: c016d918
  [...]
  NIP [c01c1840] skb_over_panic+0x48/0x5c
  LR [c01c1840] skb_over_panic+0x48/0x5c
  Call Trace:
  [c0339d50] [c01c1840] skb_over_panic+0x48/0x5c (unreliable)
  [c0339d60] [c01c3020] skb_put+0x5c/0x60
  [c0339d70] [c0195914] gfar_clean_rx_ring+0x25c/0x3d0
  [c0339dc0] [c01976e8] gfar_poll+0x170/0x1bc

Dumped buffer descriptors showed that eTSEC's length/truncation
logic sometimes passes oversized packets, i.e. for the above ICMP
packet the following two buffer descriptors may become ready:

  status=1400 length=1536
  status=1800 length=1541

So, it seems that gianfar actually receives the whole big frame,
and it tries to place the packet into two BDs. This situation
confuses the driver, and so the skb_put() sanity check fails.

This patch fixes the issue by adding an appropriate check, i.e.
the driver should not try to process frames with buffer
descriptor's length over rx_buffer_size (i.e. maxfrm and mrblr).

Note that sometimes eTSEC works correctly, i.e. in the second
(last) buffer descriptor bits 'truncated' and 'crcerr' are set,
and so there's no oops. Though I couldn't find any logic when
it works correctly and when not.

Signed-off-by: Anton Vorontsov <avorontsov@mvista.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoehea: Fix kernel deadlock in DLPAR-mem processing
Jan-Bernd Themann [Tue, 15 Jun 2010 05:35:42 +0000]
ehea: Fix kernel deadlock in DLPAR-mem processing

Port reset operations and memory add/remove operations need to
be serialized to avoid a kernel deadlock. The deadlock is caused
by calling the napi_disable() function twice.
Therefore we have to employ the dlpar_mem_lock in the ehea_reset_port
function as well

Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoehea: fix delayed packet processing
Jan-Bernd Themann [Tue, 15 Jun 2010 05:35:16 +0000]
ehea: fix delayed packet processing

In the eHEA poll function an rmb() is required. Without that some packets
on the receive queue are not seen and thus delayed until the next interrupt
is handled for the same receive queue.

Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoixgbe: add comment on SFP+ ID for Active DA
Don Skidmore [Tue, 15 Jun 2010 09:23:17 +0000]
ixgbe: add comment on SFP+ ID for Active DA

These comments were forgotten in the initial patch to add this
functionality.  This patch corrects that.

Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoClear IFF_XMIT_DST_RELEASE for teql interfaces
Tom Hughes [Tue, 15 Jun 2010 22:24:28 +0000]
Clear IFF_XMIT_DST_RELEASE for teql interfaces

https://bugzilla.kernel.org/show_bug.cgi?id=16183

The sch_teql module, which can be used to load balance over a set of
underlying interfaces, stopped working after 2.6.30 and has been
broken in all kernels since then for any underlying interface which
requires the addition of link level headers.

The problem is that the transmit routine relies on being able to
access the destination address in the skb in order to do address
resolution once it has decided which underlying interface it is going
to transmit through.

In 2.6.31 the IFF_XMIT_DST_RELEASE flag was introduced, and set by
default for all interfaces, which causes the destination address to be
released before the transmit routine for the interface is called.

The solution is to clear that flag for teql interfaces.

Signed-off-by: Tom Hughes <tom@compton.nu>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agogianfar: Fix setup of RX time stamping
Manfred Rudigier [Fri, 11 Jun 2010 01:49:05 +0000]
gianfar: Fix setup of RX time stamping

Previously the RCTRL_TS_ENABLE bit was set unconditionally. However, if
the RCTRL_TS_ENABLE is set without TMR_CTRL[TE], the driver does not work
properly on some boards (Anton had problems with the MPC8313ERDB and
MPC8568EMDS).

With this patch the bit will only be set if requested from user space
with the SIOCSHWTSTAMP ioctl command, meaning that time stamping is
disabled during normal operation. Users who are not interested in time
stamps will not experience problems with buggy CPU revisions or
performance drops any more.

The setting of TMR_CTRL[TE] is still up to the user. This is considered
safe because users wanting HW timestamps must initialize the eTSEC clock
first anyway, e.g. with the recently submitted PTP clock driver.

Signed-off-by: Manfred Rudigier <manfred.rudigier@omicron.at>
Reviewed-by: Anton Vorontsov <cbouatmailru@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wirel...
David S. Miller [Wed, 16 Jun 2010 20:41:55 +0000]
Merge branch 'master' of git://git./linux/kernel/git/linville/wireless-2.6

9 years agomac80211: fix warn, enum may be used uninitialized
Christoph Fritz [Wed, 16 Jun 2010 14:37:34 +0000]
mac80211: fix warn, enum may be used uninitialized

regression introduced by b8d92c9c141ee3dc9b3537b1f0ffb4a54ea8d9b2

In function ‘ieee80211_work_rx_queued_mgmt’:
warning: ‘rma’ may be used uninitialized in this function

this re-adds default value WORK_ACT_NONE back to rma

Signed-off-by: Christoph Fritz <chf.fritz@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>