deliverable/linux.git
10 years agoASoC: fsl_ssi: Set default slot number for common cases
Nicolin Chen [Thu, 9 Jan 2014 10:42:48 +0000 (18:42 +0800)] 
ASoC: fsl_ssi: Set default slot number for common cases

For those platforms using DAI master mode like I2S, it's better to pre-set
a default slot number so that there's no need for these common cases to set
the slot number from its machine driver any more.

Signed-off-by: Nicolin Chen <Guangyu.Chen@freescale.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl-ssi: Add missing clk_disable_unprepare() on error in fsl_ssi_probe()
Wei Yongjun [Thu, 9 Jan 2014 14:27:31 +0000 (22:27 +0800)] 
ASoC: fsl-ssi: Add missing clk_disable_unprepare() on error in fsl_ssi_probe()

Add the missing clk_disable_unprepare() before return from
fsl_ssi_probe() in the request irq error handling case.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wirel...
John W. Linville [Thu, 9 Jan 2014 15:19:01 +0000 (10:19 -0500)] 
Merge branch 'master' of git://git./linux/kernel/git/linville/wireless into for-davem

10 years agoASoC: Allow PCMs to restrict the supported formats
Lars-Peter Clausen [Mon, 6 Jan 2014 13:19:16 +0000 (14:19 +0100)] 
ASoC: Allow PCMs to restrict the supported formats

Some DMA cores might add additional restrictions on which in memory audio
formats can be supported by the compound sound card. If the PCM component
specifies a set of formats it supports (by setting the formats field to non 0)
take these into account when calculating the format set for the compound sound
card.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Tested-by: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: ux500: Don't set unused struct snd_pcm_hardware fields
Lars-Peter Clausen [Mon, 6 Jan 2014 13:19:15 +0000 (14:19 +0100)] 
ASoC: ux500: Don't set unused struct snd_pcm_hardware fields

The ASoC core assumes that the PCM component of the ASoC card transparently
moves data around and does not impose any restrictions on the memory layout or
the transfer speed. It ignores all fields from the snd_pcm_hardware struct for
the PCM driver that are related to this. Setting these fields in the PCM driver
might suggest otherwise though, so rather not set them.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: sh: Don't set unused struct snd_pcm_hardware fields
Lars-Peter Clausen [Mon, 6 Jan 2014 13:19:14 +0000 (14:19 +0100)] 
ASoC: sh: Don't set unused struct snd_pcm_hardware fields

The ASoC core assumes that the PCM component of the ASoC card transparently
moves data around and does not impose any restrictions on the memory layout or
the transfer speed. It ignores all fields from the snd_pcm_hardware struct for
the PCM driver that are related to this. Setting these fields in the PCM driver
might suggest otherwise though, so rather not set them.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: nuc900: Don't set unused struct snd_pcm_hardware fields
Lars-Peter Clausen [Mon, 6 Jan 2014 13:19:13 +0000 (14:19 +0100)] 
ASoC: nuc900: Don't set unused struct snd_pcm_hardware fields

The ASoC core assumes that the PCM component of the ASoC card transparently
moves data around and does not impose any restrictions on the memory layout or
the transfer speed. It ignores all fields from the snd_pcm_hardware struct for
the PCM driver that are related to this. Setting these fields in the PCM driver
might suggest otherwise though, so rather not set them.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: kirkwood: Don't set unused struct snd_pcm_hardware fields
Lars-Peter Clausen [Mon, 6 Jan 2014 13:19:11 +0000 (14:19 +0100)] 
ASoC: kirkwood: Don't set unused struct snd_pcm_hardware fields

The ASoC core assumes that the PCM component of the ASoC card transparently
moves data around and does not impose any restrictions on the memory layout or
the transfer speed. It ignores all fields from the snd_pcm_hardware struct for
the PCM driver that are related to this. Setting these fields in the PCM driver
might suggest otherwise though, so rather not set them.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: intel: Don't set unused struct snd_pcm_hardware fields
Lars-Peter Clausen [Mon, 6 Jan 2014 13:19:10 +0000 (14:19 +0100)] 
ASoC: intel: Don't set unused struct snd_pcm_hardware fields

The ASoC core assumes that the PCM component of the ASoC card transparently
moves data around and does not impose any restrictions on the memory layout or
the transfer speed. It ignores all fields from the snd_pcm_hardware struct for
the PCM driver that are related to this. Setting these fields in the PCM driver
might suggest otherwise though, so rather not set them.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: adav80x: Use params_width() rather than memory format
Mark Brown [Wed, 8 Jan 2014 18:50:40 +0000 (18:50 +0000)] 
ASoC: adav80x: Use params_width() rather than memory format

Signed-off-by: Mark Brown <broonie@linaro.org>
Acked-by: Lars-Peter Clausen <lars@metafoo.de>
10 years agoASoC: adau1701: Use params_width() rather than memory format
Mark Brown [Wed, 8 Jan 2014 18:50:25 +0000 (18:50 +0000)] 
ASoC: adau1701: Use params_width() rather than memory format

Signed-off-by: Mark Brown <broonie@linaro.org>
Acked-by: Lars-Peter Clausen <lars@metafoo.de>
10 years agoASoC: adau1373: Use params_width() rather than memory format
Mark Brown [Wed, 8 Jan 2014 18:49:58 +0000 (18:49 +0000)] 
ASoC: adau1373: Use params_width() rather than memory format

Signed-off-by: Mark Brown <broonie@linaro.org>
Acked-by: Lars-Peter Clausen <lars@metafoo.de>
10 years agoASoC: ad193x: Use params_width() rather than memory format
Mark Brown [Wed, 8 Jan 2014 18:38:20 +0000 (18:38 +0000)] 
ASoC: ad193x: Use params_width() rather than memory format

Signed-off-by: Mark Brown <broonie@linaro.org>
Acked-by: Lars-Peter Clausen <lars@metafoo.de>
10 years agoASoC: ad1836: Use params_width() rather than explicit memory format
Mark Brown [Wed, 8 Jan 2014 18:37:22 +0000 (18:37 +0000)] 
ASoC: ad1836: Use params_width() rather than explicit memory format

Signed-off-by: Mark Brown <broonie@linaro.org>
Acked-by: Lars-Peter Clausen <lars@metafoo.de>
10 years agoASoC: fsl-ssi: Fix stats compile warning
Markus Pargmann [Thu, 9 Jan 2014 10:16:11 +0000 (11:16 +0100)] 
ASoC: fsl-ssi: Fix stats compile warning

single_open requires a function with signature
'int (*)(struct seq_file *, void *)'. This patch fixes the warning by
fixing the wrong return type of fsl_ssi_stats_show.

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl-ssi doc: Add list of supported compatibles
Markus Pargmann [Thu, 9 Jan 2014 10:16:10 +0000 (11:16 +0100)] 
ASoC: fsl-ssi doc: Add list of supported compatibles

There is no list of compatibles that are supported. This patch adds a
list of compatibles to the documentation.

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: wm_adsp: Mark wm_adsp2_boot_work as static
Charles Keepax [Thu, 9 Jan 2014 09:06:54 +0000 (09:06 +0000)] 
ASoC: wm_adsp: Mark wm_adsp2_boot_work as static

Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoRevert "intel_idle: mark states tables with __initdata tag"
Jiang Liu [Thu, 9 Jan 2014 07:30:26 +0000 (15:30 +0800)] 
Revert "intel_idle: mark states tables with __initdata tag"

This reverts commit 9d046ccb98085f1d437585f84748c783a04ba240.

Commit 9d046ccb98085 marks all state tables with __initdata, but
the state table may be accessed when doing CPU online, which then
causing system crash as below:

[  204.188841] BUG: unable to handle kernel paging request at ffffffff8227cce8
[  204.196844] IP: [<ffffffff814aa1c0>] intel_idle_cpu_init+0x40/0x130
[  204.203996] PGD 1e11067 PUD 1e12063 PMD 455859063 PTE 800000000227c062
[  204.211638] Oops: 0000 [#1] SMP DEBUG_PAGEALLOC
[  204.216975] Modules linked in: x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd gpio_ich microcode joydev sb_edac edac_core ipmi_si lpc_ich ipmi_msghandler lp tpm_tis parport wmi mac_hid acpi_pad hid_generic ixgbe isci usbhid dca hid libsas ptp ahci libahci scsi_transport_sas megaraid_sas pps_core mdio
[  204.262815] CPU: 11 PID: 1489 Comm: bash Not tainted 3.13.0-rc7+ #48
[  204.269993] Hardware name: Intel Corporation BRICKLAND/BRICKLAND, BIOS BRIVTIN1.86B.0047.L09.1312061514 12/06/2013
[  204.281646] task: ffff8804303a24a0 ti: ffff880440fac000 task.ti: ffff880440fac000
[  204.290311] RIP: 0010:[<ffffffff814aa1c0>]  [<ffffffff814aa1c0>] intel_idle_cpu_init+0x40/0x130
[  204.300184] RSP: 0018:ffff880440fadd28  EFLAGS: 00010286
[  204.306192] RAX: ffffffff8227cca0 RBX: ffffe8fff1a03400 RCX: 0000000000000007
[  204.314244] RDX: ffff88045f400000 RSI: 0000000000000009 RDI: 0000000000001120
[  204.322296] RBP: ffff880440fadd38 R08: 0000000000000000 R09: 0000000000000001
[  204.330411] R10: 0000000000000001 R11: 0000000000000000 R12: 000000000000001e
[  204.338482] R13: 00000000ffffffdb R14: 0000000000000001 R15: 0000000000000000
[  204.346743] FS:  00007f64f7b0c740(0000) GS:ffff88045ce00000(0000) knlGS:0000000000000000
[  204.355919] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  204.362449] CR2: ffffffff8227cce8 CR3: 0000000444ab0000 CR4: 00000000001407e0
[  204.370520] Stack:
[  204.372853]  000000000000001e ffffffff81f10240 ffff880440fadd50 ffffffff814aa307
[  204.381519]  ffffffff81ea80e0 ffff880440fadda0 ffffffff8185a230 0000000000000000
[  204.390196]  000000000000001e 0000000000000002 0000000000000002 0000000000000000
[  204.398856] Call Trace:
[  204.401683]  [<ffffffff814aa307>] cpu_hotplug_notify+0x57/0x70
[  204.408638]  [<ffffffff8185a230>] notifier_call_chain+0x100/0x150
[  204.415553]  [<ffffffff810a7dae>] __raw_notifier_call_chain+0xe/0x10
[  204.422772]  [<ffffffff81072163>] cpu_notify+0x23/0x50
[  204.428616]  [<ffffffff810723b2>] _cpu_up+0x132/0x1a0
[  204.434361]  [<ffffffff8107249d>] cpu_up+0x7d/0xa0
[  204.439819]  [<ffffffff81836c9c>] cpu_subsys_online+0x3c/0x90
[  204.446345]  [<ffffffff81554625>] device_online+0x45/0xa0
[  204.452471]  [<ffffffff815546ce>] online_store+0x4e/0x80
[  204.458511]  [<ffffffff815519a8>] dev_attr_store+0x18/0x30
[  204.464744]  [<ffffffff812a68f1>] sysfs_write_file+0x151/0x1c0
[  204.471681]  [<ffffffff81217ef1>] vfs_write+0xe1/0x160
[  204.477524]  [<ffffffff8121889c>] SyS_write+0x4c/0x90
[  204.483270]  [<ffffffff8185f2ed>] system_call_fastpath+0x1a/0x1f
[  204.490081] Code: 41 54 41 89 fc 8b 3d 48 25 85 01 53 48 8b 1d 30 25 85 01 48 03 1c c5 40 90 fb 81 48 8b 05 19 25 85 01 c7 43 0c 01 00 00 00 66 90 <48> 83 78 48 00 74 4f 41 83 c0 01 41 39 f0 7e 10 48 c7 c7 38 79
[  204.515723] RIP  [<ffffffff814aa1c0>] intel_idle_cpu_init+0x40/0x130
[  204.522996]  RSP <ffff880440fadd28>
[  204.526976] CR2: ffffffff8227cce8
[  204.530766] ---[ end trace 336f56cc3d1cfc8c ]---

Fixes: 9d046ccb98085 (intel_idle: mark states tables with __initdata tag)
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
10 years agoALSA: Merge memalloc code into snd-pcm module
Takashi Iwai [Thu, 9 Jan 2014 06:48:12 +0000 (07:48 +0100)] 
ALSA: Merge memalloc code into snd-pcm module

Instead of keeping a separate snd-page-alloc module, merge into the
core snd-pcm module, as we don't need to keep it as an individual
module due to the drop of page reservation.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoALSA: Remove superfluous header inclusions in memalloc.c
Takashi Iwai [Thu, 9 Jan 2014 06:44:11 +0000 (07:44 +0100)] 
ALSA: Remove superfluous header inclusions in memalloc.c

After cutting off the proc and page reservation codes, we don't need
many headers any longer.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoALSA: Remove memory reservation code from memalloc helper
Takashi Iwai [Wed, 8 Jan 2014 15:12:25 +0000 (16:12 +0100)] 
ALSA: Remove memory reservation code from memalloc helper

Nowadays we have CMA for obtaining the contiguous memory pages
efficiently.  Let's kill the old kludge for reserving the memory pages
for large buffers.  It was rarely useful (only for preserving pages
among module reloading or a little help by an early boot scripting),
used only by a couple of drivers, and yet it gives too much ugliness
than its benefit.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoALSA: Remove memory accounting in memalloc helper
Takashi Iwai [Wed, 8 Jan 2014 15:09:31 +0000 (16:09 +0100)] 
ALSA: Remove memory accounting in memalloc helper

It's almost superfluous, and doesn't help much for real uses.
Let's reduce the layer size.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoALSA: hda/hdmi - apply all Haswell fix-ups to Broadwell display codec
Mengdong Lin [Wed, 8 Jan 2014 20:55:32 +0000 (15:55 -0500)] 
ALSA: hda/hdmi - apply all Haswell fix-ups to Broadwell display codec

Broadwell and Haswell have the same behavior on display audio. So this patch
defines is_haswell_plus() to include codecs for both Haswell and its successor
Broadwell, and apply all Haswell fix-ups to Broadwell.

Signed-off-by: Mengdong Lin <mengdong.lin@intel.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoALSA: hda - add codec ID for Broadwell display audio codec
Mengdong Lin [Wed, 8 Jan 2014 20:55:24 +0000 (15:55 -0500)] 
ALSA: hda - add codec ID for Broadwell display audio codec

This patch adds codec ID (0x80862808) and module alias for Broadwell
display codec.

Signed-off-by: Mengdong Lin <mengdong.lin@intel.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoALSA: hda - add device ID for Broadwell display audio controller
Mengdong Lin [Wed, 8 Jan 2014 20:55:14 +0000 (15:55 -0500)] 
ALSA: hda - add device ID for Broadwell display audio controller

This patch adds the device ID for Intel Broadwell display HD-Audio controller,
and applies Haswell properties to this device.

Signed-off-by: Mengdong Lin <mengdong.lin@intel.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoMerge branch 'parisc-3.13' of git://git.kernel.org/pub/scm/linux/kernel/git/deller...
Linus Torvalds [Thu, 9 Jan 2014 01:09:26 +0000 (09:09 +0800)] 
Merge branch 'parisc-3.13' of git://git./linux/kernel/git/deller/parisc-linux

Pull parisc fix from Helge Deller:
 "This patch fixes the kmap/kunmap implementation on parisc and finally
  makes AIO work on parisc"

* 'parisc-3.13' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
  parisc: Ensure full cache coherency for kmap/kunmap

10 years agoMerge branch 'for-3.13-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj...
Linus Torvalds [Thu, 9 Jan 2014 01:08:23 +0000 (09:08 +0800)] 
Merge branch 'for-3.13-fixes' of git://git./linux/kernel/git/tj/libata

Pull libata fixes from Tejun Heo:
 "Late fixes for libata.  Nothing too interesting.  Adding missing PM
  callbacks to satat_sis and an additional PCI ID for ahci"

* 'for-3.13-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata:
  sata_sis: missing PM support
  ahci: add PCI ID for Marvell 88SE9170 SATA controller

10 years agoparisc: Ensure full cache coherency for kmap/kunmap
John David Anglin [Mon, 6 Jan 2014 02:25:00 +0000 (21:25 -0500)] 
parisc: Ensure full cache coherency for kmap/kunmap

Helge Deller noted a few weeks ago problems with the AIO support on
parisc. This change is the result of numerous iterations on how best to
deal with this problem.

The solution adopted here is to provide full cache coherency in a
uniform manner on all parisc systems. This involves calling
flush_dcache_page() on kmap operations and flush_kernel_dcache_page() on
kunmap operations. As a result, the copy_user_page() and
clear_user_page() functions can be removed and the overall code is
simpler.

The change ensures that both userspace and kernel aliases to a mapped
page are invalidated and flushed. This is necessary for the correct
operation of PA8800 and PA8900 based systems which do not support
inequivalent aliases.

With this change, I have observed no cache related issues on c8000 and
rp3440. It is now possible for example to do kernel builds with "-j64"
on four way systems.

On systems using XFS file systems, the patch recently posted by Mikulas
Patocka to "fix crash using XFS on loopback" is needed to avoid a hang
caused by an uninitialized lock passed to flush_dcache_page() in the
page struct.

Signed-off-by: John David Anglin <dave.anglin@bell.net>
Cc: stable@vger.kernel.org # v3.9+
Signed-off-by: Helge Deller <deller@gmx.de>
10 years agoASoC: bcm: Remove obsoleted Kconfig dependency
Takashi Iwai [Wed, 8 Jan 2014 09:37:07 +0000 (10:37 +0100)] 
ASoC: bcm: Remove obsoleted Kconfig dependency

CONFIG_SND_SOC_DMAENGINE_PCM was renamed to CONFIG_SND_DMAENGINE_PCM
recently.  And yet we don't have to select it since
CONFIG_SND_GENERIC_DMAENGINE_PCM selects the dependency by itself, so
just rip it off.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Acked-by: Florian Meier <florian.meier@koalo.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoMerge tag 'nfc-fixes-3.13-1' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo...
John W. Linville [Wed, 8 Jan 2014 18:36:17 +0000 (13:36 -0500)] 
Merge tag 'nfc-fixes-3.13-1' of git://git./linux/kernel/git/sameo/nfc-fixes

Samuel Ortiz <sameo@linux.intel.com> says:

"This is the first NFC fixes pull request for 3.13.

It only contains one fix for a regression introduced with commit
e29a9e2ae165620d. Without this fix, we can not establish a p2p link in
target mode. Only initiator mode works."

Signed-off-by: John W. Linville <linville@tuxdriver.com>
10 years agoASoC: wm_adsp: Start DSP booting earlier in the DAPM process
Charles Keepax [Wed, 8 Jan 2014 17:42:19 +0000 (17:42 +0000)] 
ASoC: wm_adsp: Start DSP booting earlier in the DAPM process

Move the start of booting the DSP to earlier in the DAPM process, and
move the final starting of the DSP to later in the DAPM process. This
allows us to overlap some of the processing with other components of the
system being brought up.

Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: wm_adsp: Factor out ADSP2 boot proceedure
Charles Keepax [Wed, 8 Jan 2014 17:42:18 +0000 (17:42 +0000)] 
ASoC: wm_adsp: Factor out ADSP2 boot proceedure

Move the ADSP2 boot proceedure into a work structure in preparation for
running it asynchronously with the reset of the audio path bring up.

Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl-ssi: Drop ac97 specific trigger function
Markus Pargmann [Fri, 20 Dec 2013 13:11:35 +0000 (14:11 +0100)] 
ASoC: fsl-ssi: Drop ac97 specific trigger function

The normal trigger function can now be used for AC97. Drop AC97 trigger
function.

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl-ssi: Move RX/TX configuration to seperate functions
Markus Pargmann [Fri, 20 Dec 2013 13:11:34 +0000 (14:11 +0100)] 
ASoC: fsl-ssi: Move RX/TX configuration to seperate functions

This patch defines the appropriate register values for different
oparation modes and IP versions. We have to handle DMA/FIQ, AC97,
DEBUG-IRQs and offline/online configuration support.

With this patch we cleanup some driver code that was not reference
manual conform and try to cleanup the whole trigger function to seperate
the actual register values from the enable/disable logic, which is now
hidden in fsl_ssi_config helpers.

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl-ssi: Add configuration helper functions
Markus Pargmann [Fri, 20 Dec 2013 13:11:33 +0000 (14:11 +0100)] 
ASoC: fsl-ssi: Add configuration helper functions

This patch adds a struct 'fsl_ssi_rxtx_reg_val' which holds register
values necessary to enable rx/tx. Based on those preset register values,
the added configuration functions will cleanly enable/disable different
parts of the SSI IP while supporting online/offline configuration.
Different operating modes can be setup directly as different register
values in fsl_ssi_reg_val.

These functions and structs will help to cleanup and simplify the
trigger function to support many different IP versions (online/offline
configuration) and different operating modes.

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl-ssi: Add offline_config flag
Markus Pargmann [Fri, 20 Dec 2013 13:11:32 +0000 (14:11 +0100)] 
ASoC: fsl-ssi: Add offline_config flag

imx50-ssi and later versions of this IP support online reconfiguration
of all registers. The reference manual does not list any registers that
can only be configured while the SSI unit is disabled. This patch
introduces the flag for later use.

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl-ssi: Fix interrupt stats for imx
Markus Pargmann [Fri, 20 Dec 2013 13:11:31 +0000 (14:11 +0100)] 
ASoC: fsl-ssi: Fix interrupt stats for imx

irqs should only be requested/released with enabled DMA. Previously
interrupt statistics where disabled for IMX Processors because of
different writeable SISR bits.

This patch introduces support for irqstats on imx processors again by
creating a sisr writeback mask and introducing a imx35-ssi compatible.

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl-ssi: Add imx51-ssi and of_device_id matching
Markus Pargmann [Fri, 20 Dec 2013 13:11:30 +0000 (14:11 +0100)] 
ASoC: fsl-ssi: Add imx51-ssi and of_device_id matching

There is a small but significant difference between imx21-ssi and
imx51-ssi and above. imx21-ssi does not allow online reconfiguration of
some registers. They have to be configured at the beginning.

imx51-ssi has to reconfigure the SSI unit while it is running. Otherwise
it would not be possible to have two streams in parallel. The new SDMA
unit in imx51 and above has to be configured before the first DMA
request arrives. Therefor we need to setup TDMAE/RDMAE just before
starting the stream.

This patch introduces distinct imx51-ssi as compatible and adds
of_device_id matching in the probe function.

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl-ssi: Move sysfs stats to debugfs
Markus Pargmann [Fri, 20 Dec 2013 13:11:29 +0000 (14:11 +0100)] 
ASoC: fsl-ssi: Move sysfs stats to debugfs

Interrupt statistics of fsl_ssi are mainly for debugging purpose. Most
of those interrupts show error states, e.g. under/overflow.
The stats should be exposed via debugfs instead of sysfs.

This patch moves the statistics file to debugfs.

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl-ssi: Fix probe error handling
Markus Pargmann [Fri, 20 Dec 2013 13:11:28 +0000 (14:11 +0100)] 
ASoC: fsl-ssi: Fix probe error handling

This patch fixes the error handling in the fsl-ssi probe function.

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoclk: clk-divider: fix divisor > 255 bug
James Hogan [Mon, 16 Dec 2013 10:41:38 +0000 (10:41 +0000)] 
clk: clk-divider: fix divisor > 255 bug

Commit 6d9252bd9a4bb (clk: Add support for power of two type dividers)
merged in v3.6 added the _get_val function to convert a divisor value to
a register field value depending on the flags. However it used the type
u8 for the div field, causing divisors larger than 255 to be masked
and the resultant clock rate to be too high.

E.g. in my case an 11bit divider was supposed to divide 24.576 MHz down
to 32.768KHz. The divisor was correctly calculated as 750 (0x2ee). This
was masked to 238 (0xee) resulting in a frequency of 103.26KHz.

Signed-off-by: James Hogan <james.hogan@imgtec.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: stable@vger.kernel.org
Signed-off-by: Mike Turquette <mturquette@linaro.org>
10 years agoALSA: hda - Split Thinkpad ACPI-related code
Takashi Iwai [Wed, 8 Jan 2014 10:44:21 +0000 (11:44 +0100)] 
ALSA: hda - Split Thinkpad ACPI-related code

Both patch_realtek.c and patch_conexant.c contain the fairy same code
snippet for supporting Thinkpad ACPI LED controls.  Split them into
thinkpad_helper.c and include it from both places.  Although this
isn't the best approach from the code size POV, the probability for
coexistence of both Realtek and Conexant codecs on a single machine is
pretty low, thus it'll end up with less memory footprint than
splitting to yet another module.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoASoC: ssm2602: add 16kHz sampling rate support
Andreas Pretzsch [Tue, 7 Jan 2014 21:34:41 +0000 (22:34 +0100)] 
ASoC: ssm2602: add 16kHz sampling rate support

SSM260x also supports 16kHz with external master clocks of 12.000MHz,
12.288MHz and 18.432MHz.
Add matching coefficients, update constraints and announced rates.

Signed-off-by: Andreas Pretzsch <apr@cn-eng.de>
Acked-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: twl4030: Pass the twl4030_priv directly to twl4030_can_write_to_chip()
Peter Ujfalusi [Fri, 3 Jan 2014 13:27:56 +0000 (15:27 +0200)] 
ASoC: twl4030: Pass the twl4030_priv directly to twl4030_can_write_to_chip()

To avoid another lookup for the twl4030_priv in there.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: twl4030: Move the ctl cache update local to twl4030_write() function
Peter Ujfalusi [Fri, 3 Jan 2014 13:27:55 +0000 (15:27 +0200)] 
ASoC: twl4030: Move the ctl cache update local to twl4030_write() function

There's no other users of this functionality, the code can be moved inside
of twl4030_write.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: twl4030: Parameter alignment fixes (for code consistency)
Peter Ujfalusi [Fri, 3 Jan 2014 13:27:54 +0000 (15:27 +0200)] 
ASoC: twl4030: Parameter alignment fixes (for code consistency)

Over time the multi line alignment got messed up. Correct them in one go
so the code will look consistent.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: twl4030: Remove local reg cache
Peter Ujfalusi [Fri, 3 Jan 2014 13:27:53 +0000 (15:27 +0200)] 
ASoC: twl4030: Remove local reg cache

Depend on the regmap reg cache implementation for register caching done in
the twl-core driver.
The local register cache can be removed and we can keep only shadow copies
of certain ctl registers for pop noise reduction.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: twl4030: Introduce local ctl register cache
Peter Ujfalusi [Fri, 3 Jan 2014 13:27:52 +0000 (15:27 +0200)] 
ASoC: twl4030: Introduce local ctl register cache

Few registers need to be cached in the codec driver level. These registers
should only be written when the path is active to avoid pop noise on the
given path.
This patch adds an array which covers the range where the sensitive registers
are located and uppon loadinf the driver the ctl cache will be initialized.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: twl4030: Remove reset registers functionality
Peter Ujfalusi [Fri, 3 Jan 2014 13:27:51 +0000 (15:27 +0200)] 
ASoC: twl4030: Remove reset registers functionality

The register states now tracked by the regmap implementation in the core which
makes the reset registers functionality 'redundant' since we know the state
of the registers now all the time.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: twl4030: Remove check defaults functionality
Peter Ujfalusi [Fri, 3 Jan 2014 13:27:50 +0000 (15:27 +0200)] 
ASoC: twl4030: Remove check defaults functionality

No need to keep the check defaults functionality anymore.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: twl4030: Separate write condition checking from I/O function
Peter Ujfalusi [Fri, 3 Jan 2014 13:27:49 +0000 (15:27 +0200)] 
ASoC: twl4030: Separate write condition checking from I/O function

Simplifies the code a bit and prepares it to the removal of local caching.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoMerge tag 'ib-asoc-3.14.2' of git://git.linaro.org/people/ljones/mfd into asoc-tlw4030
Mark Brown [Wed, 8 Jan 2014 12:46:36 +0000 (12:46 +0000)] 
Merge tag 'ib-asoc-3.14.2' of git://git.linaro.org/people/ljones/mfd into asoc-tlw4030

Immutable branch between MFD and ASoC due for the v3.14 merge window

10 years agoASoC: fsl-sai: Clean up the code
Xiubo Li [Wed, 8 Jan 2014 08:13:05 +0000 (16:13 +0800)] 
ASoC: fsl-sai: Clean up the code

Makes the code slightly shorter.

Signed-off-by: Xiubo Li <Li.Xiubo@freescale.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: ux500: Fix sparse non static symbol warning
Wei Yongjun [Wed, 8 Jan 2014 10:13:35 +0000 (18:13 +0800)] 
ASoC: ux500: Fix sparse non static symbol warning

Fixes the following sparse warning:

sound/soc/ux500/ux500_msp_i2s.c:649:5: warning:
 symbol 'ux500_msp_i2s_of_init_msp' was not declared. Should it be static?

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: adau1701: Fix ADAU1701_SEROCTL_WORD_LEN_16 constant
Lars-Peter Clausen [Wed, 8 Jan 2014 10:22:25 +0000 (11:22 +0100)] 
ASoC: adau1701: Fix ADAU1701_SEROCTL_WORD_LEN_16 constant

The driver defines ADAU1701_SEROCTL_WORD_LEN_16 as 0x10 while it should be b10,
so 0x2. This patch fixes it.

Reported-by: Magnus Reftel <magnus.reftel@lockless.no>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Mark Brown <broonie@linaro.org>
Cc: stable@vger.kernel.org
10 years agoASoC: sapm: Automatically connect DAI link widgets in DAPM graph.
Liam Girdwood [Wed, 8 Jan 2014 10:40:19 +0000 (10:40 +0000)] 
ASoC: sapm: Automatically connect DAI link widgets in DAPM graph.

Connect the DAPM graph through each BE DAI link to the componnent(s) on the
other side of the BE DAI link. This allows the graph to be walked on
both sides of the link when graph changes are made.

Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: utils: Add internal call to determine if DAI is dummy.
Liam Girdwood [Wed, 8 Jan 2014 10:40:18 +0000 (10:40 +0000)] 
ASoC: utils: Add internal call to determine if DAI is dummy.

Provide a quick way to tell if a DAI is a dummy DAI or a regular DAI.
This is for internal DAPM usage only and is used to determine whether to
insert a DAI link connection into the DAPM graph.

Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agomfd: twl-core: Enable regcache for audio registers
Peter Ujfalusi [Fri, 3 Jan 2014 13:27:48 +0000 (15:27 +0200)] 
mfd: twl-core: Enable regcache for audio registers

Enable regmap's regcache for the audio registers:
i2c address 0x49, register range 0x01 - 0x49
Mark all other registers as volatile to avoid any side effect for the non
audio functions behind 0x49 i2c address.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
10 years agomfd: twl-core: API to set the regcache bypass for a given regmap in twl
Peter Ujfalusi [Fri, 3 Jan 2014 13:27:47 +0000 (15:27 +0200)] 
mfd: twl-core: API to set the regcache bypass for a given regmap in twl

If the regcache is enabled on the regmap module drivers might need to access
to HW register(s) in certain cases in cache bypass mode.
As an example of this is the audio block's ANAMICL register. In normal
operation the content can be cached but during initialization one bit from
the register need to be monitored. With the twl_set_regcache_bypass() the
client driver can switch regcache bypass on and off when it is needed so
we can utilize the regcache for more registers.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
10 years agomfd: twl-core: Simplify IO wrapper functions by moving common code out
Peter Ujfalusi [Fri, 3 Jan 2014 13:27:46 +0000 (15:27 +0200)] 
mfd: twl-core: Simplify IO wrapper functions by moving common code out

The new twl_get_regmap() function will return a pointer to the regmap needed
for the given module.

Since both read and write function were using the same code to do the lookup
we can reuse this in both places to simplify the code.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
10 years agoALSA: hda - Don't create duplicated ctls for loopback paths
Takashi Iwai [Tue, 7 Jan 2014 17:11:44 +0000 (18:11 +0100)] 
ALSA: hda - Don't create duplicated ctls for loopback paths

AD1986A mic pins (0x1d and 0x1f) share the same widget for controlling
the loopback volume/mute, but the generic parser didn't check it.
This ended up with the duplicated controls for the same effect.

This patch adds the check of the duplication for avoiding it.

After this fix, there will be only one control although it affects
both paths; this remaining issue should be fixed later in a different
patch.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=66621
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoALSA: hda - Correct AD1986A 3stack pin configs
Takashi Iwai [Tue, 7 Jan 2014 16:48:11 +0000 (17:48 +0100)] 
ALSA: hda - Correct AD1986A 3stack pin configs

The 3stack pin configs for AD1986A codec had incorrect values that
resulted in broken mic and line-in.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=66621
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoALSA: hda - Add consistent tag names for firmware patch
Takashi Iwai [Tue, 7 Jan 2014 14:28:51 +0000 (15:28 +0100)] 
ALSA: hda - Add consistent tag names for firmware patch

Some tags used in the firmware patch file are inconsistent with hwdep
sysfs file names, such as, the firmware patch takes [hint] tag while
sysfs file is */hints.  This makes even me referring back to the
document often.  Let's provide the same tag names as sysfs for
reducing confusions.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoALSA: hda - firmware patch code cleanup
Takashi Iwai [Tue, 7 Jan 2014 14:23:44 +0000 (15:23 +0100)] 
ALSA: hda - firmware patch code cleanup

Just a code refactoring: the need_codec flag in hda_patch_item struct
can be removed by checking the current mode instead.  No functional
change.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoALSA: hda - Increment default stream numbers for AMD HDMI controllers
Takashi Iwai [Mon, 9 Dec 2013 09:18:09 +0000 (10:18 +0100)] 
ALSA: hda - Increment default stream numbers for AMD HDMI controllers

It turned out that some AMD HDMI controllers still don't provide
proper values in GCAP register (all zero), and the driver assigns only
one stream in that case, although the connected codec chip supports
more than one stream.

In this patch, the default max number of streams for AMD HDMI
controllers is increased to 8, which  should suffice for most use
cases.  The overhead by this increase is more azx_dev struct and BDL
allocations, so it's negligible.  Of course, if the controller
provides a proper GCAP register, the register value would be used.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoALSA: hda - Minor code optimization for patch_realtek.c
Takashi Iwai [Tue, 7 Jan 2014 17:22:49 +0000 (18:22 +0100)] 
ALSA: hda - Minor code optimization for patch_realtek.c

No functional change.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoMerge branch 'drm-nouveau-next' of git://anongit.freedesktop.org/nouveau/linux-2...
Dave Airlie [Wed, 8 Jan 2014 07:57:45 +0000 (17:57 +1000)] 
Merge branch 'drm-nouveau-next' of git://anongit.freedesktop.org/nouveau/linux-2.6 into drm-fixes

misc fixes for nouveau, one more msi rearm, regression fix for old bioses
crash and leak fixes.

* 'drm-nouveau-next' of git://anongit.freedesktop.org/nouveau/linux-2.6:
  drm/nouveau/nouveau: fix memory leak in nouveau_crtc_page_flip()
  drm/nouveau/bios: fix offset calculation for BMPv1 bioses
  drm/nouveau: return offset of allocated notifier
  drm/nouveau/bios: make jump conditional
  drm/nvce/mc: fix msi rearm on GF114
  drm/nvc0/gr: fix mthd data submission
  drm/nouveau: populate master subdev pointer only when fully constructed

10 years agoMerge tag 'drm-intel-fixes-2014-01-08' of git://people.freedesktop.org/~danvet/drm...
Dave Airlie [Wed, 8 Jan 2014 07:55:44 +0000 (17:55 +1000)] 
Merge tag 'drm-intel-fixes-2014-01-08' of git://people.freedesktop.org/~danvet/drm-intel into drm-fixes

Just a revert (gen4 backlight seems a lost cause) and a tlb coherency fix
for bdw, plus the patch to sign up Jani for co-maintainer. Thanks to Ben
for taking care of -fixes while I've enjoyed a bit of vacation.

* tag 'drm-intel-fixes-2014-01-08' of git://people.freedesktop.org/~danvet/drm-intel:
  MAINTAINERS: Updates for drm/i915
  Revert "drm/i915: assume all GM45 Acer laptops use inverted backlight PWM"
  drm/i915/bdw: Flush system agent on gen8 also

10 years agodrm/nouveau/nouveau: fix memory leak in nouveau_crtc_page_flip()
Christian Engelmayer [Sun, 29 Dec 2013 22:08:54 +0000 (23:08 +0100)] 
drm/nouveau/nouveau: fix memory leak in nouveau_crtc_page_flip()

Fix a memory leak in the nouveau_crtc_page_flip() error handling path.

Signed-off-by: Christian Engelmayer <cengelma@gmx.at>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
10 years agodrm/nouveau/bios: fix offset calculation for BMPv1 bioses
Ilia Mirkin [Tue, 7 Jan 2014 17:33:59 +0000 (12:33 -0500)] 
drm/nouveau/bios: fix offset calculation for BMPv1 bioses

The only BIOS on record that needs the 14 offset has a bios major
version 2 but BMP version 1.01. Another bunch of BIOSes that need the 18
offset have BMP version 2.01 or 5.01 or higher. So instead of looking at the
bios major version, look at the BMP version. BIOSes with BMP version 0
do not contain a detectable script, so always return 0 for them.

See https://bugs.freedesktop.org/show_bug.cgi?id=68835

Reported-by: Mauro Molinari <mauromol@tiscali.it>
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
10 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf
David S. Miller [Tue, 7 Jan 2014 23:38:03 +0000 (18:38 -0500)] 
Merge branch 'master' of git://git./linux/kernel/git/pablo/nf

Pablo Neira Ayuso says:

====================
The following patchset contains two patches:

* fix the IRC NAT helper which was broken when adding (incomplete) IPv6
  support, from Daniel Borkmann.

* Refine the previous bugtrap that Jesper added to catch problems for the
  usage of the sequence adjustment extension in IPVs in Dec 16th, it may
  spam messages in case of finding a real bug.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agotipc: correctly unlink packets from deferred packet queue
Erik Hugne [Tue, 7 Jan 2014 20:51:36 +0000 (15:51 -0500)] 
tipc: correctly unlink packets from deferred packet queue

When we pull a received packet from a link's 'deferred packets' queue
for processing, its 'next' pointer is not cleared, and still refers to
the next packet in that queue, if any. This is incorrect, but caused
no harm before commit 40ba3cdf542a469aaa9083fa041656e59b109b90 ("tipc:
message reassembly using fragment chain") was introduced. After that
commit, it may sometimes lead to the following oops:

general protection fault: 0000 [#1] SMP DEBUG_PAGEALLOC
Modules linked in: tipc
CPU: 4 PID: 0 Comm: swapper/4 Tainted: G        W 3.13.0-rc2+ #6
Hardware name: Bochs Bochs, BIOS Bochs 01/01/2007
task: ffff880017af4880 ti: ffff880017aee000 task.ti: ffff880017aee000
RIP: 0010:[<ffffffff81710694>]  [<ffffffff81710694>] skb_try_coalesce+0x44/0x3d0
RSP: 0018:ffff880016603a78  EFLAGS: 00010212
RAX: 6b6b6b6bd6d6d6d6 RBX: ffff880013106ac0 RCX: ffff880016603ad0
RDX: ffff880016603ad7 RSI: ffff88001223ed00 RDI: ffff880013106ac0
RBP: ffff880016603ab8 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000000 R12: ffff88001223ed00
R13: ffff880016603ad0 R14: 000000000000058c R15: ffff880012297650
FS:  0000000000000000(0000) GS:ffff880016600000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 000000000805b000 CR3: 0000000011f5d000 CR4: 00000000000006e0
Stack:
 ffff880016603a88 ffffffff810a38ed ffff880016603aa8 ffff88001223ed00
 0000000000000001 ffff880012297648 ffff880016603b68 ffff880012297650
 ffff880016603b08 ffffffffa0006c51 ffff880016603b08 00ffffffa00005fc
Call Trace:
 <IRQ>
 [<ffffffff810a38ed>] ? trace_hardirqs_on+0xd/0x10
 [<ffffffffa0006c51>] tipc_link_recv_fragment+0xd1/0x1b0 [tipc]
 [<ffffffffa0007214>] tipc_recv_msg+0x4e4/0x920 [tipc]
 [<ffffffffa00016f0>] ? tipc_l2_rcv_msg+0x40/0x250 [tipc]
 [<ffffffffa000177c>] tipc_l2_rcv_msg+0xcc/0x250 [tipc]
 [<ffffffffa00016f0>] ? tipc_l2_rcv_msg+0x40/0x250 [tipc]
 [<ffffffff8171e65b>] __netif_receive_skb_core+0x80b/0xd00
 [<ffffffff8171df94>] ? __netif_receive_skb_core+0x144/0xd00
 [<ffffffff8171eb76>] __netif_receive_skb+0x26/0x70
 [<ffffffff8171ed6d>] netif_receive_skb+0x2d/0x200
 [<ffffffff8171fe70>] napi_gro_receive+0xb0/0x130
 [<ffffffff815647c2>] e1000_clean_rx_irq+0x2c2/0x530
 [<ffffffff81565986>] e1000_clean+0x266/0x9c0
 [<ffffffff81985f7b>] ? notifier_call_chain+0x2b/0x160
 [<ffffffff8171f971>] net_rx_action+0x141/0x310
 [<ffffffff81051c1b>] __do_softirq+0xeb/0x480
 [<ffffffff819817bb>] ? _raw_spin_unlock+0x2b/0x40
 [<ffffffff810b8c42>] ? handle_fasteoi_irq+0x72/0x100
 [<ffffffff81052346>] irq_exit+0x96/0xc0
 [<ffffffff8198cbc3>] do_IRQ+0x63/0xe0
 [<ffffffff81981def>] common_interrupt+0x6f/0x6f
 <EOI>

This happens when the last fragment of a message has passed through the
the receiving link's 'deferred packets' queue, and at least one other
packet was added to that queue while it was there. After the fragment
chain with the complete message has been successfully delivered to the
receiving socket, it is released. Since 'next' pointer of the last
fragment in the released chain now is non-NULL, we get the crash shown
above.

We fix this by clearing the 'next' pointer of all received packets,
including those being pulled from the 'deferred' queue, before they
undergo any further processing.

Fixes: 40ba3cdf542a4 ("tipc: message reassembly using fragment chain")
Signed-off-by: Erik Hugne <erik.hugne@ericsson.com>
Reported-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoipv6: pcpu_tstats.syncp should be initialised in ip6_vti.c
Li RongQing [Tue, 7 Jan 2014 07:39:43 +0000 (15:39 +0800)] 
ipv6: pcpu_tstats.syncp should be initialised in ip6_vti.c

initialise pcpu_tstats.syncp to kill the calltrace
[   11.973950] Call Trace:
[   11.973950]  [<819bbaff>] dump_stack+0x48/0x60
[   11.973950]  [<819bbaff>] dump_stack+0x48/0x60
[   11.973950]  [<81078dcf>] __lock_acquire.isra.22+0x1bf/0xc10
[   11.973950]  [<81078dcf>] __lock_acquire.isra.22+0x1bf/0xc10
[   11.973950]  [<81079fa7>] lock_acquire+0x77/0xa0
[   11.973950]  [<81079fa7>] lock_acquire+0x77/0xa0
[   11.973950]  [<817ca7ab>] ? dev_get_stats+0xcb/0x130
[   11.973950]  [<817ca7ab>] ? dev_get_stats+0xcb/0x130
[   11.973950]  [<8183862d>] ip_tunnel_get_stats64+0x6d/0x230
[   11.973950]  [<8183862d>] ip_tunnel_get_stats64+0x6d/0x230
[   11.973950]  [<817ca7ab>] ? dev_get_stats+0xcb/0x130
[   11.973950]  [<817ca7ab>] ? dev_get_stats+0xcb/0x130
[   11.973950]  [<811cf8c1>] ? __nla_reserve+0x21/0xd0
[   11.973950]  [<811cf8c1>] ? __nla_reserve+0x21/0xd0
[   11.973950]  [<817ca7ab>] dev_get_stats+0xcb/0x130
[   11.973950]  [<817ca7ab>] dev_get_stats+0xcb/0x130
[   11.973950]  [<817d5409>] rtnl_fill_ifinfo+0x569/0xe20
[   11.973950]  [<817d5409>] rtnl_fill_ifinfo+0x569/0xe20
[   11.973950]  [<810352e0>] ? kvm_clock_read+0x20/0x30
[   11.973950]  [<810352e0>] ? kvm_clock_read+0x20/0x30
[   11.973950]  [<81008e38>] ? sched_clock+0x8/0x10
[   11.973950]  [<81008e38>] ? sched_clock+0x8/0x10
[   11.973950]  [<8106ba45>] ? sched_clock_local+0x25/0x170
[   11.973950]  [<8106ba45>] ? sched_clock_local+0x25/0x170
[   11.973950]  [<810da6bd>] ? __kmalloc+0x3d/0x90
[   11.973950]  [<810da6bd>] ? __kmalloc+0x3d/0x90
[   11.973950]  [<817b8c10>] ? __kmalloc_reserve.isra.41+0x20/0x70
[   11.973950]  [<817b8c10>] ? __kmalloc_reserve.isra.41+0x20/0x70
[   11.973950]  [<810da81a>] ? slob_alloc_node+0x2a/0x60
[   11.973950]  [<810da81a>] ? slob_alloc_node+0x2a/0x60
[   11.973950]  [<817b919a>] ? __alloc_skb+0x6a/0x2b0
[   11.973950]  [<817b919a>] ? __alloc_skb+0x6a/0x2b0
[   11.973950]  [<817d8795>] rtmsg_ifinfo+0x65/0xe0
[   11.973950]  [<817d8795>] rtmsg_ifinfo+0x65/0xe0
[   11.973950]  [<817cbd31>] register_netdevice+0x531/0x5a0
[   11.973950]  [<817cbd31>] register_netdevice+0x531/0x5a0
[   11.973950]  [<81892b87>] ? ip6_tnl_get_cap+0x27/0x90
[   11.973950]  [<81892b87>] ? ip6_tnl_get_cap+0x27/0x90
[   11.973950]  [<817cbdb6>] register_netdev+0x16/0x30
[   11.973950]  [<817cbdb6>] register_netdev+0x16/0x30
[   11.973950]  [<81f574a6>] vti6_init_net+0x1c4/0x1d4
[   11.973950]  [<81f574a6>] vti6_init_net+0x1c4/0x1d4
[   11.973950]  [<81f573af>] ? vti6_init_net+0xcd/0x1d4
[   11.973950]  [<81f573af>] ? vti6_init_net+0xcd/0x1d4
[   11.973950]  [<817c16df>] ops_init.constprop.11+0x17f/0x1c0
[   11.973950]  [<817c16df>] ops_init.constprop.11+0x17f/0x1c0
[   11.973950]  [<817c1779>] register_pernet_operations.isra.9+0x59/0x90
[   11.973950]  [<817c1779>] register_pernet_operations.isra.9+0x59/0x90
[   11.973950]  [<817c18d1>] register_pernet_device+0x21/0x60
[   11.973950]  [<817c18d1>] register_pernet_device+0x21/0x60
[   11.973950]  [<81f574b6>] ? vti6_init_net+0x1d4/0x1d4
[   11.973950]  [<81f574b6>] ? vti6_init_net+0x1d4/0x1d4
[   11.973950]  [<81f574c7>] vti6_tunnel_init+0x11/0x68
[   11.973950]  [<81f574c7>] vti6_tunnel_init+0x11/0x68
[   11.973950]  [<81f572a1>] ? mip6_init+0x73/0xb4
[   11.973950]  [<81f572a1>] ? mip6_init+0x73/0xb4
[   11.973950]  [<81f0cba4>] do_one_initcall+0xbb/0x15b
[   11.973950]  [<81f0cba4>] do_one_initcall+0xbb/0x15b
[   11.973950]  [<811a00d8>] ? sha_transform+0x528/0x1150
[   11.973950]  [<811a00d8>] ? sha_transform+0x528/0x1150
[   11.973950]  [<81f0c544>] ? repair_env_string+0x12/0x51
[   11.973950]  [<81f0c544>] ? repair_env_string+0x12/0x51
[   11.973950]  [<8105c30d>] ? parse_args+0x2ad/0x440
[   11.973950]  [<8105c30d>] ? parse_args+0x2ad/0x440
[   11.973950]  [<810546be>] ? __usermodehelper_set_disable_depth+0x3e/0x50
[   11.973950]  [<810546be>] ? __usermodehelper_set_disable_depth+0x3e/0x50
[   11.973950]  [<81f0cd27>] kernel_init_freeable+0xe3/0x182
[   11.973950]  [<81f0cd27>] kernel_init_freeable+0xe3/0x182
[   11.973950]  [<81f0c532>] ? do_early_param+0x7a/0x7a
[   11.973950]  [<81f0c532>] ? do_early_param+0x7a/0x7a
[   11.973950]  [<819b5b1b>] kernel_init+0xb/0x100
[   11.973950]  [<819b5b1b>] kernel_init+0xb/0x100
[   11.973950]  [<819cebf7>] ret_from_kernel_thread+0x1b/0x28
[   11.973950]  [<819cebf7>] ret_from_kernel_thread+0x1b/0x28
[   11.973950]  [<819b5b10>] ? rest_init+0xc0/0xc0
[   11.973950]  [<819b5b10>] ? rest_init+0xc0/0xc0

Before 469bdcefdc ("ipv6: fix the use of pcpu_tstats in ip6_vti.c"),
the pcpu_tstats.syncp is not used to pretect the 64bit elements of
pcpu_tstats, so not appear this calltrace.

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Li RongQing <roy.qing.li@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
10 years agoASoC: dpcm: Explicitly set BE DAI link supported stream directions
Liam Girdwood [Tue, 7 Jan 2014 17:51:42 +0000 (17:51 +0000)] 
ASoC: dpcm: Explicitly set BE DAI link supported stream directions

Some BE DAIs can be "dummy" (when the DSP is controlling the DAI) and as such
wont have set a minimum number of playback or capture channels required for BE
DAI registration (to establish supported stream directions).

Force machine drivers to explicitly set whether they support playback and capture
stream directions for every BE DAIs.

Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: docs: Update the Overview document
Liam Girdwood [Tue, 7 Jan 2014 17:51:41 +0000 (17:51 +0000)] 
ASoC: docs: Update the Overview document

Update the ASoC overview to bring it up to date with the current code base
and include multi-component.

Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoALSA: compress: add num_sample_rates in snd_codec_desc
Vinod Koul [Tue, 7 Jan 2014 16:25:42 +0000 (21:55 +0530)] 
ALSA: compress: add num_sample_rates in snd_codec_desc

this gives ability to convey the valid values of supported rates in
sample_rates array

Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
10 years agoASoC: pcm: Fix lack of platform bespoke_trigger() call
Jean-Francois Moine [Fri, 3 Jan 2014 08:19:18 +0000 (09:19 +0100)] 
ASoC: pcm: Fix lack of platform bespoke_trigger() call

When the platform driver has no ops, the platform function
bespoke_trigger() is no more called.

The problem was introduced by the commit c5914b0aaea6494aaa9e415cbd32f8b7eb604af0
"ASoC: pcm: Check for ops before deferencing them"

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: simple-card: keep the property's name the same pattern
Xiubo Li [Tue, 7 Jan 2014 01:15:16 +0000 (09:15 +0800)] 
ASoC: simple-card: keep the property's name the same pattern

Even though we might not have rigor rule for the simple card property names,
according to the existing ones, they are all in a same pattern:

[simple-audio-card,]XXX;

Rename simple-audio-routing to simple-audio-card,routing, and make the simple
card's properties has one unified name.

Signed-off-by: Xiubo Li <Li.Xiubo@freescale.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: simple-card: fix the DAPM routes map parsing
Xiubo Li [Tue, 7 Jan 2014 01:13:42 +0000 (09:13 +0800)] 
ASoC: simple-card: fix the DAPM routes map parsing

The simple-card's DAPM route maping is optional.

Signed-off-by: Xiubo Li <Li.Xiubo@freescale.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: ux500: Dramatically reduce the size of the DAI driver data struct
Lee Jones [Thu, 19 Dec 2013 15:55:08 +0000 (15:55 +0000)] 
ASoC: ux500: Dramatically reduce the size of the DAI driver data struct

We no longer have a means to differentiate between MSP devices at probe
time, mainly because we don't really have to. So rather than have an over-
sized static data structure in place, where the only difference between
devices is the ID and name (which are unused), we'll just create one
succinct, statically assigned and shared one instead.

Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: ux500_pcm: Differentiate between pdata and DT initialisation
Lee Jones [Thu, 19 Dec 2013 15:55:07 +0000 (15:55 +0000)] 
ASoC: ux500_pcm: Differentiate between pdata and DT initialisation

If booting with full DT support (i.e. DMA too, the last piece of the
puzzle), then we don't need to use the compatible_request_channel call
back or require some of the historical bumph which probably isn't
required by a platform data start-up now either. This will also be
ripped out in upcoming commits.

Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: ux500_pcm: Take out pointless dev_dbg() call
Lee Jones [Thu, 19 Dec 2013 15:55:06 +0000 (15:55 +0000)] 
ASoC: ux500_pcm: Take out pointless dev_dbg() call

Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: ux500: Store DMA data in the DAI differently in the pdata and DT case
Lee Jones [Thu, 19 Dec 2013 15:55:05 +0000 (15:55 +0000)] 
ASoC: ux500: Store DMA data in the DAI differently in the pdata and DT case

In this patch we do two things. Firstly, instead of open coding the
store of DMA data in to the DAI for later use, we use the API provided.
Secondly we create and store similar DMA data for the DT case, only
this time we use 'struct snd_dmaengine_dai_dma_data' which is provided
by the core for this very reason.

Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: ux500_pcm: Extend Device Tree support to deal with DMA data
Lee Jones [Thu, 19 Dec 2013 15:55:04 +0000 (15:55 +0000)] 
ASoC: ux500_pcm: Extend Device Tree support to deal with DMA data

Soon we will strip out pdata support from the Ux500 set of ASoC drivers.
When this happens it will have to supply a DMA slave_config to the
dmaengine. At the moment a great deal of this comes from pdata via
AUXDATA. We need to become independent of this soon. This patch starts
the process by allocating memory for the associated data structures and
fetches the MSP id used for const struct indexing.

Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: ux500_pcm: Expect different saved DMA data when obtaining from DAI store
Lee Jones [Thu, 19 Dec 2013 15:55:03 +0000 (15:55 +0000)] 
ASoC: ux500_pcm: Expect different saved DMA data when obtaining from DAI store

In preparation for full Device Tree enablement we must differentiate
between the two varying ways DMA data can be held in the DAI store. If
we're booting with Device Tree the provided 'snd_dmaengine_dai_dma_data'
data structure shall be used, whereas in order to avoid breaking legacy
platform data we also need to be able to translate DMA data stored using
the UX500 specific 'ux500_msp_dma_params' method.

Once we move over to solely use Device Tree, we can enforce the use of
'snd_dmaengine_dai_dma_data' and this code can be removed altogether.

Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: ux500_pcm: Stop pretending that we support varying address widths
Lee Jones [Thu, 19 Dec 2013 15:55:02 +0000 (15:55 +0000)] 
ASoC: ux500_pcm: Stop pretending that we support varying address widths

The Slave Config's addr_width attribute is populated by data_width of
dma_cfg, which in turn is derived from dma_params' data_size attribute
and that comes from the slot_width which is always 16 bits (2 Bytes).
We're cutting out the middle man here and just setting the DMA Slave
Config directly.

Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: Ux500: Match platform by device node when booting Device Tree
Lee Jones [Thu, 19 Dec 2013 15:55:01 +0000 (15:55 +0000)] 
ASoC: Ux500: Match platform by device node when booting Device Tree

We're getting closer to fully enabling the Ux500 ASoC driver for Device
Tree. When we switch over from using AUXDATA we'll need to match platform
by only Device Tree nodes. In this patch we NULL out the platform_name,
and supply nodes for each platform device.

Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: ux500: Provide better checking for Device Tree and/or Platform Data
Lee Jones [Thu, 19 Dec 2013 15:55:00 +0000 (15:55 +0000)] 
ASoC: ux500: Provide better checking for Device Tree and/or Platform Data

These drivers will not work without platform specific data, which is
passed in via Device Tree or Platform Data. To avoid the chance of
NULL pointer dereferencing and alike, let's ensure we have at least
one of the methods in play before continuing.

Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoARM: ux500: Don't use enums for MSP IDs - for easy DT conversion
Lee Jones [Thu, 19 Dec 2013 15:54:58 +0000 (15:54 +0000)] 
ARM: ux500: Don't use enums for MSP IDs - for easy DT conversion

Signed-off-by: Lee Jones <lee.jones@linaro.org>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: mxs: Remove SND_DMAENGINE_PCM_FLAG_NO_RESIDUE flag
Lars-Peter Clausen [Mon, 6 Jan 2014 16:19:22 +0000 (17:19 +0100)] 
ASoC: mxs: Remove SND_DMAENGINE_PCM_FLAG_NO_RESIDUE flag

Since commit 7b11304 ("dma: mxs-dma: Report correct residue for cyclic DMA")
the mxs dmaengine driver has support for residue reporting. So there is no need
to specify the SND_DMAENGINE_PCM_FLAG_NO_RESIDUE flag anymore. This allows a
finer grained resolution for the PCM pointer as well as avoids the race
condition that can occur with the period counting that is used when the
dmaengine driver does not support residue reporting.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Tested-by: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: mxs: Don't set unused struct snd_pcm_hardware fields
Lars-Peter Clausen [Mon, 6 Jan 2014 13:19:12 +0000 (14:19 +0100)] 
ASoC: mxs: Don't set unused struct snd_pcm_hardware fields

The ASoC core assumes that the PCM component of the ASoC card transparently
moves data around and does not impose any restrictions on the memory layout or
the transfer speed. It ignores all fields from the snd_pcm_hardware struct for
the PCM driver that are related to this. Setting these fields in the PCM driver
might suggest otherwise though, so rather not set them.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Tested-by: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl: Don't set unused struct snd_pcm_hardware fields
Lars-Peter Clausen [Mon, 6 Jan 2014 13:19:09 +0000 (14:19 +0100)] 
ASoC: fsl: Don't set unused struct snd_pcm_hardware fields

The ASoC core assumes that the PCM component of the ASoC card transparently
moves data around and does not impose any restrictions on the memory layout or
the transfer speed. It ignores all fields from the snd_pcm_hardware struct for
the PCM driver that are related to this. Setting these fields in the PCM driver
might suggest otherwise though, so rather not set them.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Tested-by: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoASoC: fsl: fsl_ssi: Use '%ld' to print 'long int'
Fabio Estevam [Tue, 7 Jan 2014 10:00:13 +0000 (08:00 -0200)] 
ASoC: fsl: fsl_ssi: Use '%ld' to print 'long int'

Commit 6873ee464a (ASoC: fsl_ssi: Fix printing return code on clk error) caused
the following build warning:

sound/soc/fsl/fsl_ssi.c: In function 'fsl_ssi_probe':
sound/soc/fsl/fsl_ssi.c:1196:6: warning: format '%d' expects argument of type 'int', but argument 3 has type 'long int' [-Wformat]

Fix it by using '%ld' to print the 'long int' format.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
10 years agoMAINTAINERS: Updates for drm/i915
Daniel Vetter [Tue, 7 Jan 2014 08:24:31 +0000 (09:24 +0100)] 
MAINTAINERS: Updates for drm/i915

Jani for co-maintainer!

Jani has been a really active bug-scrubber in the past few months.
I've asked him whether he wants to do this in a more official capacity
and he agreed. I've already chatted with Dave and Jesse and they
support this.

Note that everyone can't now just relax because "Jani will do all the
bug scrubbing" - au contraire expect more nagging and poking now that
we have more bandwidth.

Longer-term the plan is to share more of the maintainer duties, but we
need to fix up the infrastructure a bit first (like moving the git
repo to a common location).

While at it also add the newly set-up patchwork instance.

Cc: Dave Airlie <airlied@gmail.com>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
10 years agodrm/nouveau: return offset of allocated notifier
Bob Gleitsmann [Sun, 5 Jan 2014 22:59:07 +0000 (08:59 +1000)] 
drm/nouveau: return offset of allocated notifier

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
10 years agodrm/nouveau/bios: make jump conditional
Ilia Mirkin [Mon, 6 Jan 2014 01:07:02 +0000 (20:07 -0500)] 
drm/nouveau/bios: make jump conditional

This fixes a hang in VBIOS scripts of the form "condition; jump".
The jump used to always be executed, while now it will only be
executed if the condition is true.

See https://bugs.freedesktop.org/show_bug.cgi?id=72943

Reported-by: Darcy BrĂ¡s da Silva <dardevelin@cidadecool.com>
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: stable@vger.kernel.org
10 years agodrm/nvce/mc: fix msi rearm on GF114
Sid Boyce [Sun, 5 Jan 2014 23:12:05 +0000 (09:12 +1000)] 
drm/nvce/mc: fix msi rearm on GF114

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
10 years agodrm/nvc0/gr: fix mthd data submission
Kelly Doran [Fri, 20 Dec 2013 17:07:26 +0000 (11:07 -0600)] 
drm/nvc0/gr: fix mthd data submission

If the initial data element is 0, it will never be written, even
though the value from the previous method may be there.

Signed-off-by: Kelly Doran <kel.p.doran@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
10 years agodrm/nouveau: populate master subdev pointer only when fully constructed
Ben Skeggs [Tue, 26 Nov 2013 23:46:56 +0000 (09:46 +1000)] 
drm/nouveau: populate master subdev pointer only when fully constructed

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
10 years agoMerge tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Tue, 7 Jan 2014 00:22:42 +0000 (08:22 +0800)] 
Merge tag 'ext4_for_linus_stable' of git://git./linux/kernel/git/tytso/ext4

Pull ext4 bugfix from Ted Ts'o:
 "Fix a regression introduced in v3.13-rc6"

* tag 'ext4_for_linus_stable' of http://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: fix bigalloc regression

This page took 0.104732 seconds and 5 git commands to generate.