cascardo/linux.git
9 years agoMerge tag 'ux500-devicetree-v3.16-1' of git://git.kernel.org/pub/scm/linux/kernel...
Olof Johansson [Sun, 13 Jul 2014 04:36:50 +0000 (21:36 -0700)]
Merge tag 'ux500-devicetree-v3.16-1' of git://git./linux/kernel/git/linusw/linux-stericsson into next/dt

Merge "Ux500 devicetree changes for v3.17" from Linus Walleij:

Ux500 device tree patches for v3.17:
- Add regulators to STMPE expanders
- Add proper DMA channels for all SD/MMC blocks
- Add sensors to the device tree

* tag 'ux500-devicetree-v3.16-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-stericsson:
  ARM: ux500: add misc sensors to the device trees
  ARM: ux500: add some DB8500 DMA channel info
  ARM: ux500: add VCC and VIO regulators to STMPE IC
  + Linux 3.16-rc4

Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoMerge tag 'at91-dt' of git://github.com/at91linux/linux-at91 into next/dt
Olof Johansson [Sun, 13 Jul 2014 04:34:22 +0000 (21:34 -0700)]
Merge tag 'at91-dt' of git://github.com/at91linux/linux-at91 into next/dt

Merge "at91: dt for 3.17 #1" from Nicolas Ferre:

First DT update for 3.17:
- move of crystals DT definitions to the /clocks node
- addition of clock entries for sound for CCF enabled platforms
- addition of DMA and DMA + nand on at91sam9rl
- move to CCF for all not-converted-yet AT91 SoCs: at91rm9200, at91sam9260/9g20,
  at91sam9g45 family and at91sam9263

* tag 'at91-dt' of git://github.com/at91linux/linux-at91: (43 commits)
  ARM: at91/dt: usb_a9263: define crystals frequencies
  ARM: at91/dt: tny_a9263: define crystals frequencies
  ARM: at91/dt: sam9263ek: define crystals frequencies
  ARM: at91: move at91sam9263 SoC to the CCF
  ARM: at91/dt: sam9263: define clocks
  ARM: at91: prepare common clk transition for sam9263
  ARM: at91/dt: cosino define crystals frequencies
  ARM: at91/dt: pm9g45: crystals frequencies
  ARM: at91/dt: sam9m10g45ek: define crystals frequencies
  ARM: at91: move at91sam9g45 SoC to the CCF
  ARM: at91/dt: sam9g45: define clocks
  ARM: at91: prepare common clk transition for sam9g45
  ARM: at91/dt: kizbox: define main crystal frequency
  ARM: at91/dt: animeo_ip: define crystals frequencies
  ARM: at91/dt: ethernut5: define crystals frequencies
  ARM: at91/dt: evk-pro3: define slow crytal frequency
  ARM: at91/dt: aks-cdu: define slow crytal frequency
  ARM: at91/dt: ge863-pro3: define main crystal frequency
  ARM: at91/dt: mpa1600: define crytals frequencies
  ARM: at91/dt: qil_a9260: define crystals frequencies
  ...

Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoMerge tag 'renesas-dt2-for-v3.17' of git://git.kernel.org/pub/scm/linux/kernel/git...
Olof Johansson [Sat, 12 Jul 2014 16:34:38 +0000 (09:34 -0700)]
Merge tag 'renesas-dt2-for-v3.17' of git://git./linux/kernel/git/horms/renesas into next/dt

Merge "Second Round of Renesas ARM Based SoC DT Updates for v3.17" from Simon
Horman:

- Extend hardware coverage
  * Add DVC support for sound nodes on r8a7791 and r8a7790
  * Enable internal PCI on r8a7790/lager

* tag 'renesas-dt2-for-v3.17' of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas:
  ARM: shmobile: r8a7791: add DVC support for sound node on DTSI
  ARM: shmobile: r8a7790: add DVC support for sound node on DTSI
  ARM: shmobile: lager: enable internal PCI
  ARM: shmobile: r8a7790: add internal PCI bridge nodes

Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoMerge tag 'CCF-for-3.17' of git://github.com/at91linux/linux-at91 into at91-3.17-dt
Nicolas Ferre [Wed, 9 Jul 2014 17:14:31 +0000 (19:14 +0200)]
Merge tag 'CCF-for-3.17' of git://github.com/at91linux/linux-at91 into at91-3.17-dt

Pull "CCF migration for 3.17" from Alexandre Belloni:
"This is the switch to CCF for all the remaining at91 SoCs and boards"

9 years agoARM: at91/dt: usb_a9263: define crystals frequencies
Alexandre Belloni [Tue, 24 Jun 2014 06:23:42 +0000 (08:23 +0200)]
ARM: at91/dt: usb_a9263: define crystals frequencies

Define Calao USB-A9263 main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: tny_a9263: define crystals frequencies
Alexandre Belloni [Tue, 24 Jun 2014 06:22:51 +0000 (08:22 +0200)]
ARM: at91/dt: tny_a9263: define crystals frequencies

Define Calao TNY-A9263 main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: sam9263ek: define crystals frequencies
Alexandre Belloni [Tue, 24 Jun 2014 06:21:46 +0000 (08:21 +0200)]
ARM: at91/dt: sam9263ek: define crystals frequencies

Define at91sam9263ek main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: move at91sam9263 SoC to the CCF
Alexandre Belloni [Tue, 24 Jun 2014 06:09:44 +0000 (08:09 +0200)]
ARM: at91: move at91sam9263 SoC to the CCF

This patch removes the selection of AT91_USE_OLD_CLK when selecting
at91sam9263 SoC support. This will automatically enable COMMON_CLK_AT91
option and add support for at91 common clock implementation.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: sam9263: define clocks
Alexandre Belloni [Mon, 23 Jun 2014 04:03:37 +0000 (06:03 +0200)]
ARM: at91/dt: sam9263: define clocks

Define the at91sam9263 clocks in the SoC dtsi file.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: prepare common clk transition for sam9263
Alexandre Belloni [Thu, 10 Apr 2014 18:19:05 +0000 (20:19 +0200)]
ARM: at91: prepare common clk transition for sam9263

Enclose the sam9263 old clk registration in "#if defined(CONFIG_OLD_CLK_AT91)
 #endif"

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: cosino define crystals frequencies
Alexandre Belloni [Sat, 14 Jun 2014 00:14:32 +0000 (02:14 +0200)]
ARM: at91/dt: cosino define crystals frequencies

Define Cosino boards main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: pm9g45: crystals frequencies
Alexandre Belloni [Sat, 14 Jun 2014 00:13:16 +0000 (02:13 +0200)]
ARM: at91/dt: pm9g45: crystals frequencies

Define Ronetix pm9g45 main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: sam9m10g45ek: define crystals frequencies
Alexandre Belloni [Fri, 13 Jun 2014 18:01:51 +0000 (20:01 +0200)]
ARM: at91/dt: sam9m10g45ek: define crystals frequencies

Define at91sam9m10g45ek main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: move at91sam9g45 SoC to the CCF
Alexandre Belloni [Wed, 11 Jun 2014 20:35:59 +0000 (22:35 +0200)]
ARM: at91: move at91sam9g45 SoC to the CCF

This patch removes the selection of AT91_USE_OLD_CLK when selecting at91sam9g45
SoC support. This will automatically enable COMMON_CLK_AT91 option and add
support for at91 common clock implementation.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: sam9g45: define clocks
Alexandre Belloni [Wed, 11 Jun 2014 20:39:06 +0000 (22:39 +0200)]
ARM: at91/dt: sam9g45: define clocks

Define the at91sam9g45 clocks in the SoC dtsi file.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: prepare common clk transition for sam9g45
Alexandre Belloni [Thu, 10 Apr 2014 18:19:33 +0000 (20:19 +0200)]
ARM: at91: prepare common clk transition for sam9g45

Enclose the sam9g45 old clk registration in "#if defined(CONFIG_OLD_CLK_AT91)
 #endif"

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: kizbox: define main crystal frequency
Alexandre Belloni [Wed, 18 Jun 2014 19:05:05 +0000 (21:05 +0200)]
ARM: at91/dt: kizbox: define main crystal frequency

Define kizbox board's main crystal frequency.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: animeo_ip: define crystals frequencies
Alexandre Belloni [Wed, 18 Jun 2014 19:13:54 +0000 (21:13 +0200)]
ARM: at91/dt: animeo_ip: define crystals frequencies

Define Somfy Animeo IP main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: ethernut5: define crystals frequencies
Alexandre Belloni [Wed, 18 Jun 2014 19:09:19 +0000 (21:09 +0200)]
ARM: at91/dt: ethernut5: define crystals frequencies

Define egnite Ethernut 5 main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: evk-pro3: define slow crytal frequency
Alexandre Belloni [Wed, 9 Jul 2014 13:23:05 +0000 (15:23 +0200)]
ARM: at91/dt: evk-pro3: define slow crytal frequency

Define Telit EVK-PRO3 slow crystal frequency

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: aks-cdu: define slow crytal frequency
Alexandre Belloni [Wed, 9 Jul 2014 13:22:15 +0000 (15:22 +0200)]
ARM: at91/dt: aks-cdu: define slow crytal frequency

Define AK signal CDU slow crystal frequency

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: ge863-pro3: define main crystal frequency
Alexandre Belloni [Wed, 18 Jun 2014 19:07:22 +0000 (21:07 +0200)]
ARM: at91/dt: ge863-pro3: define main crystal frequency

Define Telit GE863-PRO3 main crystal frequency.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: mpa1600: define crytals frequencies
Alexandre Belloni [Wed, 18 Jun 2014 19:03:19 +0000 (21:03 +0200)]
ARM: at91/dt: mpa1600: define crytals frequencies

Define Phontech MPA 1600 main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: qil_a9260: define crystals frequencies
Alexandre Belloni [Wed, 18 Jun 2014 19:12:26 +0000 (21:12 +0200)]
ARM: at91/dt: qil_a9260: define crystals frequencies

Define Calao QIL-A9260 main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: tny_a9260: define crystals frequencies
Alexandre Belloni [Wed, 18 Jun 2014 19:01:51 +0000 (21:01 +0200)]
ARM: at91/dt: tny_a9260: define crystals frequencies

Define Calao TNY-A9260 and TNY-A9G20 main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: usb_a9260: define crystals frequencies
Alexandre Belloni [Mon, 16 Jun 2014 18:12:19 +0000 (20:12 +0200)]
ARM: at91/dt: usb_a9260: define crystals frequencies

Define Calao USB-A9260, USB-A9G20 and USB-A9G20-LPW main and slow crystals
frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: foxg20: define crystals frequencies
Alexandre Belloni [Tue, 17 Jun 2014 09:12:11 +0000 (11:12 +0200)]
ARM: at91/dt: foxg20: define crystals frequencies

Define Acme Systems srl Fox G20 main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: at91sam9g20ek: define crystals frequencies
Alexandre Belloni [Wed, 18 Jun 2014 19:10:25 +0000 (21:10 +0200)]
ARM: at91/dt: at91sam9g20ek: define crystals frequencies

Define at91sam9g20ek main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: move at91sam9260 SoCs to the CCF
Alexandre Belloni [Thu, 29 May 2014 18:42:30 +0000 (20:42 +0200)]
ARM: at91: move at91sam9260 SoCs to the CCF

This patch removes the selection of AT91_USE_OLD_CLK when selecting at91sam9260
SoCs support. This will automatically enable COMMON_CLK_AT91 option and add
support for at91 common clk implementation.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: sam9g20: define clocks
Alexandre Belloni [Mon, 16 Jun 2014 17:22:40 +0000 (19:22 +0200)]
ARM: at91/dt: sam9g20: define clocks

Define the at91sam9g20 clocks that differ from at91sam9260 in the SoC dtsi file.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: sam9260: define clocks
Alexandre Belloni [Fri, 18 Apr 2014 10:44:20 +0000 (12:44 +0200)]
ARM: at91/dt: sam9260: define clocks

Define the at91sam9260 clocks in the SoC dtsi file.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: prepare common clk transition for sam9260
Alexandre Belloni [Thu, 10 Apr 2014 18:18:47 +0000 (20:18 +0200)]
ARM: at91: prepare common clk transition for sam9260

Enclose the sam9260 old clk registration in "#if defined(CONFIG_OLD_CLK_AT91)
 #endif"

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: at91rm9200ek: define crystals frequencies
Alexandre Belloni [Wed, 18 Jun 2014 19:11:37 +0000 (21:11 +0200)]
ARM: at91/dt: at91rm9200ek: define crystals frequencies

Define at91rm9200ek main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: move at91rm9200 SoC to the CCF
Alexandre Belloni [Thu, 29 May 2014 18:44:44 +0000 (20:44 +0200)]
ARM: at91: move at91rm9200 SoC to the CCF

This patch removes the selection of AT91_USE_OLD_CLK when selecting at91rm9200
SoC support. This will automatically enable COMMON_CLK_AT91 option and add
support for at91 common clk implementation.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: rm9200: define clocks
Alexandre Belloni [Fri, 2 May 2014 22:46:36 +0000 (00:46 +0200)]
ARM: at91/dt: rm9200: define clocks

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: prepare common clk transition for rm9200
Alexandre Belloni [Thu, 10 Apr 2014 18:18:19 +0000 (20:18 +0200)]
ARM: at91: prepare common clk transition for rm9200

Enclose the rm9200 old clk registration in "#if defined(CONFIG_OLD_CLK_AT91)
 #endif"

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: ariag25: define crystals frequencies
Alexandre Belloni [Sat, 14 Jun 2014 00:18:24 +0000 (02:18 +0200)]
ARM: at91/dt: ariag25: define crystals frequencies

Define Acme Systems Aria G25 board main and slow crystals frequencies.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoDocumentation: dt: document all the atmel pmc compatibles
Alexandre Belloni [Tue, 1 Jul 2014 13:44:01 +0000 (15:44 +0200)]
Documentation: dt: document all the atmel pmc compatibles

Documentation for atmel-pmc only list one compatible, add the remaining
compatible strings.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
9 years agoARM: at91/dt: add NAND + DMA property for at91sam9rl
Nicolas Ferre [Thu, 15 May 2014 15:09:28 +0000 (17:09 +0200)]
ARM: at91/dt: add NAND + DMA property for at91sam9rl

Add the "atmel,nand-has-dma" property to NAND node

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
9 years agoARM: at91/dt: add DMA controller node to at91sam9rl
Nicolas Ferre [Thu, 15 May 2014 15:08:32 +0000 (17:08 +0200)]
ARM: at91/dt: add DMA controller node to at91sam9rl

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
9 years agoARM: at91: at91sam9n12ek: switch sound to CCF
Bo Shen [Mon, 9 Jun 2014 03:31:47 +0000 (11:31 +0800)]
ARM: at91: at91sam9n12ek: switch sound to CCF

As the at91sam9n12ek has switch to CCF, so add clock for wm8904

Signed-off-by: Bo Shen <voice.shen@atmel.com>
Reviwed-by: Mark Brown <broonie@linaro.org>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: sama5d3xek: switch sound to CCF
Bo Shen [Mon, 9 Jun 2014 03:31:46 +0000 (11:31 +0800)]
ARM: at91: sama5d3xek: switch sound to CCF

As the sama5d3xek board has switch to CCF, so add clock for wm8904

Signed-off-by: Bo Shen <voice.shen@atmel.com>
Reviwed-by: Mark Brown <broonie@linaro.org>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: sama5d3 crystals under the clocks node
Alexandre Belloni [Tue, 17 Jun 2014 13:30:20 +0000 (15:30 +0200)]
ARM: at91/dt: sama5d3 crystals under the clocks node

Having clocks grouped in a subnode is common practice, so move the crystals
under a clocks node for the sama5d3 SoC and sama5d3 based boards.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: sam9x5 crystals under the clocks node
Alexandre Belloni [Tue, 17 Jun 2014 13:30:19 +0000 (15:30 +0200)]
ARM: at91/dt: sam9x5 crystals under the clocks node

Having clocks grouped in a subnode is common practice, so move the crystals and
the ADC clock under a clocks node for the at91sam9x5 SoC and at91sam9x5 based
boards.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: sam9rl crystals under the clocks node
Alexandre Belloni [Tue, 17 Jun 2014 13:30:19 +0000 (15:30 +0200)]
ARM: at91/dt: sam9rl crystals under the clocks node

Having clocks grouped in a subnode is common practice, so move the crystals
under a clocks node for the at91sam9rl SoC and at91sam9rl based boards.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: sam9n12 crystals under the clocks node
Alexandre Belloni [Tue, 17 Jun 2014 13:30:18 +0000 (15:30 +0200)]
ARM: at91/dt: sam9n12 crystals under the clocks node

Having clocks grouped in a subnode is common practice, so move the crystals
under a clocks node for the at91sam9n12 SoC and at91sam9n12 based boards.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/dt: sam9261 crystals under the clocks node
Alexandre Belloni [Tue, 17 Jun 2014 13:30:18 +0000 (15:30 +0200)]
ARM: at91/dt: sam9261 crystals under the clocks node

Having clocks grouped in a subnode is common practice, so move the crystals
under a clocks node for the at91sam9261 SoC and at91sam9261 based boards.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoMerge tag 'berlin-dt-3.17-1' of git://git.infradead.org/users/hesselba/linux-berlin...
Olof Johansson [Tue, 8 Jul 2014 05:39:47 +0000 (22:39 -0700)]
Merge tag 'berlin-dt-3.17-1' of git://git.infradead.org/users/hesselba/linux-berlin into next/dt

Merge "Berlin DT changes for v3.17" from Sebastian Hesselbarth:

- L2CC latency properties for BG2Q
- DW i2c nodes for BG2Q and corresponding dev board
- SMP related nodes for BG2 and BG2Q

* tag 'berlin-dt-3.17-1' of git://git.infradead.org/users/hesselba/linux-berlin:
  ARM: dts: berlin: add SMP related nodes and properties for BG2Q
  ARM: dts: berlin: add SMP related nodes and properties for BG2
  Documentation: bindings: add the marvell,berlin-smp CPU enable method
  Documentation: bindings: add the Berlin CPU control doc
  ARM: dts: berlin: enable i2c0 and i2c2
  ARM: dts: berlin: add I2C nodes for BG2Q
  ARM: dts: berlin2q: set L2CC tag and data latency to 2 cycles

Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoARM: ux500: add misc sensors to the device trees
Linus Walleij [Sat, 12 Apr 2014 15:13:23 +0000 (17:13 +0200)]
ARM: ux500: add misc sensors to the device trees

This adds the STMicroelectonics MEMS sensor devices to the Ux500
family device trees:

- Accelerometer
- Magnetometer
- Gyroscope
- Pressure (barometer)

Cc: Lee Jones <lee.jones@linaro.org>
Cc: Denis CIOCCA <denis.ciocca@st.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
9 years agoARM: ux500: add some DB8500 DMA channel info
Linus Walleij [Wed, 11 Jun 2014 08:45:50 +0000 (10:45 +0200)]
ARM: ux500: add some DB8500 DMA channel info

This adds some missing DMA channel information to the disabled
MMC/SD/SDIO blocks number 3 and 5, and notes that the assignment
of MSP channels vary with ASIC variant.

Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
9 years agoARM: ux500: add VCC and VIO regulators to STMPE IC
Linus Walleij [Tue, 15 Apr 2014 15:31:00 +0000 (17:31 +0200)]
ARM: ux500: add VCC and VIO regulators to STMPE IC

Add a VCC and VIO regulator supplies to the the STMPE expander
found on the STUIB UIB variants.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
9 years agoMerge tag 'renesas-dt-for-v3.17' of git://git.kernel.org/pub/scm/linux/kernel/git...
Olof Johansson [Mon, 7 Jul 2014 00:22:26 +0000 (17:22 -0700)]
Merge tag 'renesas-dt-for-v3.17' of git://git./linux/kernel/git/horms/renesas into next/dt

Merge "Renesas ARM Based SoC DT Updates for v3.17" from Simon Horman:

Increased hardware coverage:
- Add R-Car sounds support to r8a7790 SoC
- Add PCIe support to r8a7790 and r8a7791 SoCs
- Increase I2C support of Henninger and lager boards
- DVFS support to Koelsch board
- Add SYS-DMAC clocks to r8a7791 SoCs
- Add USB 3.0 clocks to r8a7791 and r8a7790 SoCs
- Add LED labels to armadillo800eva board

Cleanup:
- Remove early_printk from marzen command line

* tag 'renesas-dt-for-v3.17' of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas: (29 commits)
  ARM: shmobile: r8a7791: add R-Car sound support on DTSI
  ARM: shmobile: henninger: Enable PCIe Controller & PCIe bus clock
  ARM: shmobile: koelsch: Enable PCIe Controller & PCIe bus clock
  ARM: shmobile: r8a7791: Add PCIe Controller device node
  ARM: shmobile: r8a7791: Add default PCIe bus clock
  ARM: shmobile: r8a7791: Add PCIEC clock device tree node
  ARM: shmobile: r8a7790: Add PCIe Controller device node
  ARM: shmobile: r8a7790: Add default PCIe bus clock
  ARM: shmobile: r8a7790: Add PCIEC clock device tree node
  ARM: shmobile: r8a7791: add MSTP10 support on DTSI
  ARM: shmobile: r8a7791: add audio clock on DTSI
  ARM: shmobile: r8a7790: add R-Car sound support on DTSI
  ARM: shmobile: r8a7790: add MSTP10 support on DTSI
  ARM: shmobile: henninger: add I2C2 DT support
  ARM: shmobile: koelsch: Remove duplicate i2c6 nodes
  ARM: shmobile: lager: Remove duplicate i2c3 nodes
  ARM: shmobile: Lager memory map update
  ARM: shmobile: lager: Move i2c[12]_pins nodes to pfc node
  ARM: shmobile: lager: add i2c1, i2c2 pins
  ARM: shmobile: lager: enable i2c devices
  ...

Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoLinux 3.16-rc4 v3.16-rc4
Linus Torvalds [Sun, 6 Jul 2014 19:37:51 +0000 (12:37 -0700)]
Linux 3.16-rc4

9 years agoMerge tag 'dt-for-linus' of git://git.secretlab.ca/git/linux
Linus Torvalds [Sun, 6 Jul 2014 19:11:57 +0000 (12:11 -0700)]
Merge tag 'dt-for-linus' of git://git.secretlab.ca/git/linux

Pull devicetree bugfix from Grant Likely:
 "Important bug fix for parsing 64-bit addresses on 32-bit platforms.
  Without this patch the kernel will try to use memory ranges that
  cannot be reached"

* tag 'dt-for-linus' of git://git.secretlab.ca/git/linux:
  of: Check for phys_addr_t overflows in early_init_dt_add_memory_arch

9 years agoMerge tag 'scsi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb...
Linus Torvalds [Sun, 6 Jul 2014 19:08:30 +0000 (12:08 -0700)]
Merge tag 'scsi-for-linus' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "This is a set of 13 fixes, a MAINTAINERS update and a sparse update.
  The fixes are mostly correct value initialisations, avoiding NULL
  derefs and some uninitialised pointer avoidance.

  All the patches have been incubated in -next for a few days.  The
  final patch (use the scsi data buffer length to extract transfer size)
  has been rebased to add a cc to stable, but only the commit message
  has changed"

* tag 'scsi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  [SCSI] use the scsi data buffer length to extract transfer size
  virtio-scsi: fix various bad behavior on aborted requests
  virtio-scsi: avoid cancelling uninitialized work items
  ibmvscsi: Add memory barriers for send / receive
  ibmvscsi: Abort init sequence during error recovery
  qla2xxx: Fix sparse warning in qla_target.c.
  bnx2fc: Improve stats update mechanism
  bnx2fc: do not scan uninitialized lists in case of error.
  fc: ensure scan_work isn't active when freeing fc_rport
  pm8001: Fix potential null pointer dereference and memory leak.
  MAINTAINERS: Update LSILOGIC MPT FUSION DRIVERS (FC/SAS/SPI) maintainers Email IDs
  be2iscsi: remove potential junk pointer free
  be2iscsi: add an missing goto in error path
  scsi_error: set DID_TIME_OUT correctly
  scsi_error: fix invalid setting of host byte

9 years agoMerge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Linus Torvalds [Sun, 6 Jul 2014 00:13:46 +0000 (17:13 -0700)]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux

Pull drm fixes from Dave Airlie:
 "i915, tda998x and vmwgfx fixes,

  The main one is i915 fix for missing VGA connectors, along with some
  fixes for the tda998x from Russell fixing some modesetting problems.

  (still on holidays, but got a spare moment to find these)"

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
  drm/vmwgfx: Fix incorrect write to read-only register v2:
  drm/i915: Drop early VLV WA to fix Voltage not getting dropped to Vmin
  drm/i915: only apply crt_present check on VLV
  drm/i915: Wait for vblank after enabling the primary plane on BDW
  drm/i2c: tda998x: add some basic mode validation
  drm/i2c: tda998x: faster polling for edid
  drm/i2c: tda998x: move drm_i2c_encoder_destroy call

9 years agoMerge tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm...
Linus Torvalds [Sat, 5 Jul 2014 23:57:12 +0000 (16:57 -0700)]
Merge tag 'fixes-for-linus' of git://git./linux/kernel/git/arm/arm-soc

Pull ARM SoC fixes from Olof Johansson:
 "This week's arm-soc fixes:

   - A set of of OMAP patches that we had missed Tony's pull request of:
     * Reset fix for am43xx
     * Proper OPP table for omap5
     * Fix for SoC detection of one of the DRA7 SoCs
     * hwmod updates to get SATA and OCP to work on omap5 (drivers
       merged in 3.16)
     * ... plus a handful of smaller fixes
   - sunxi needed to re-add machine specific restart code that was
     removed in anticipation of a watchdog driver being merged for 3.16,
     and it didn't make it in.
   - Marvell fixes for PCIe on SMP and a big-endian fix.
   - A trivial defconfig update to make my capri test board boot with
     bcm_defconfig again.

  ... and a couple of MAINTAINERS updates, one to claim new Keystone
  drivers that have been merged, and one to merge MXS and i.MX (both
  Freescale platforms).

  The largest diffs come from the hwmod code for omap5 and the re-add of
  the restart code on sunxi.  The hwmod stuff is quite late at this
  point but it slipped through cracks repeatedly while coming up the
  maintainer chain and only affects the one SoC so risk is low"

* tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
  MAINTAINERS: Add few more Keystone drivers
  MAINTAINERS: merge MXS entry into IMX one
  ARM: sunxi: Reintroduce the restart code for A10/A20 SoCs
  ARM: mvebu: fix cpuidle implementation to work on big-endian systems
  ARM: mvebu: update L2/PCIe deadlock workaround after L2CC cleanup
  ARM: mvebu: move Armada 375 external abort logic as a quirk
  ARM: bcm: Fix bcm and multi_v7 defconfigs
  ARM: dts: dra7-evm: remove interrupt binding
  ARM: OMAP2+: Fix parser-bug in platform muxing code
  ARM: DTS: dra7/dra7xx-clocks: ATL related changes
  ARM: OMAP2+: drop unused function
  ARM: dts: am43x-epos-evm: Add Missing cpsw-phy-sel for am43x-epos-evm
  ARM: dts: omap5: Update CPU OPP table as per final production Manual
  ARM: DRA722: add detection of SoC information
  ARM: dts: Enable twl4030 off-idle configuration for selected omaps
  ARM: OMAP5: hwmod: Add ocp2scp3 and sata hwmods
  ARM: OMAP2+: hwmod: Change hardreset soc_ops for AM43XX

9 years agoMerge branch 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sat, 5 Jul 2014 23:56:14 +0000 (16:56 -0700)]
Merge branch 'irq-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull irq fixes from Thomas Gleixner:
 "A few minor fixlets in ARM SoC irq drivers and a fix for a memory leak
  which I introduced in the last round of cleanups :("

* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  genirq: Fix memory leak when calling irq_free_hwirqs()
  irqchip: spear_shirq: Fix interrupt offset
  irqchip: brcmstb-l2: Level-2 interrupts are edge sensitive
  irqchip: armada-370-xp: Mask all interrupts during initialization.

9 years agoMerge tag 'drm-intel-fixes-2014-07-03' of git://anongit.freedesktop.org/drm-intel
Dave Airlie [Sat, 5 Jul 2014 21:49:59 +0000 (07:49 +1000)]
Merge tag 'drm-intel-fixes-2014-07-03' of git://anongit.freedesktop.org/drm-intel

Fixes for 3.16-rc3; most importantly Jesse brings back VGA he took away
on a bunch of machines. Also a vblank fix for BDW and a power workaround
fix for VLV.

* tag 'drm-intel-fixes-2014-07-03' of git://anongit.freedesktop.org/drm-intel:
  drm/i915: Drop early VLV WA to fix Voltage not getting dropped to Vmin
  drm/i915: only apply crt_present check on VLV
  drm/i915: Wait for vblank after enabling the primary plane on BDW

9 years agoMerge branch 'vmwgfx-fixes-3.16' of git://people.freedesktop.org/~thomash/linux
Dave Airlie [Sat, 5 Jul 2014 21:49:28 +0000 (07:49 +1000)]
Merge branch 'vmwgfx-fixes-3.16' of git://people.freedesktop.org/~thomash/linux

fix to a 3.15 commit.

* 'vmwgfx-fixes-3.16' of git://people.freedesktop.org/~thomash/linux:
  drm/vmwgfx: Fix incorrect write to read-only register v2:

9 years agoMerge branch 'tda998x-fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-cubox
Dave Airlie [Sat, 5 Jul 2014 21:48:26 +0000 (07:48 +1000)]
Merge branch 'tda998x-fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-cubox

mode fixes for tda998x.

* 'tda998x-fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-cubox:
  drm/i2c: tda998x: add some basic mode validation
  drm/i2c: tda998x: faster polling for edid
  drm/i2c: tda998x: move drm_i2c_encoder_destroy call

9 years agogenirq: Fix memory leak when calling irq_free_hwirqs()
Keith Busch [Mon, 30 Jun 2014 22:24:44 +0000 (16:24 -0600)]
genirq: Fix memory leak when calling irq_free_hwirqs()

irq_free_hwirqs() always calls irq_free_descs() with a cnt == 0
which makes it a no-op since the interrupt count to free is
decremented in itself.

Fixes: 7b6ef1262549f6afc5c881aaef80beb8fd15f908

Signed-off-by: Keith Busch <keith.busch@intel.com>
Acked-by: David Rientjes <rientjes@google.com>
Link: http://lkml.kernel.org/r/1404167084-8070-1-git-send-email-keith.busch@intel.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
9 years agoMerge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Linus Torvalds [Sat, 5 Jul 2014 17:12:52 +0000 (10:12 -0700)]
Merge tag 'arm64-fixes' of git://git./linux/kernel/git/arm64/linux

Pull ARM64 fixes from Catalin Marinas:
 - Exception level check at boot time (for completeness, not triggering
   any bug before)
 - I/D-cache synchronisation logic for huge pages
 - Config symbol typo

* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64: fix el2_setup check of CurrentEL
  arm64: mm: Make icache synchronisation logic huge page aware
  arm64: mm: Fix horrendous config typo

9 years agoMAINTAINERS: Add few more Keystone drivers
Santosh Shilimkar [Sun, 22 Jun 2014 20:06:22 +0000 (16:06 -0400)]
MAINTAINERS: Add few more Keystone drivers

Update MAINTAINERS file for recently added reset controller, AEMIF
and clocksource driver for Keystone SOCs.

The EMIF memory controller driver is also added along with AEMIF.

Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Olof Johansson <olof@lixom.net>
Cc: Kevin Hilman <khilman@linaro.org>
Cc: Mike Turquette <mturquette@linaro.org>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoMAINTAINERS: merge MXS entry into IMX one
Shawn Guo [Wed, 2 Jul 2014 07:37:10 +0000 (15:37 +0800)]
MAINTAINERS: merge MXS entry into IMX one

The mach-mxs platform is actually co-maintained by myself and
pengutronix folks.  Also it's hosted in the same kernel tree as IMX.
So let's merge the entry into IMX one.

Signed-off-by: Shawn Guo <shawn.guo@freescale.com>
Acked-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoMerge tag 'mvebu-fixes-3.16-2' of git://git.infradead.org/linux-mvebu into fixes
Olof Johansson [Sat, 5 Jul 2014 04:51:19 +0000 (21:51 -0700)]
Merge tag 'mvebu-fixes-3.16-2' of git://git.infradead.org/linux-mvebu into fixes

mvebu fixes for v3.16 (round #2)

 - mvebu
    - Fix PCIe deadlock now that SMP is enabled
    - Fix cpuidle for big-endian systems

* tag 'mvebu-fixes-3.16-2' of git://git.infradead.org/linux-mvebu:
  ARM: mvebu: fix cpuidle implementation to work on big-endian systems
  ARM: mvebu: update L2/PCIe deadlock workaround after L2CC cleanup
  ARM: mvebu: move Armada 375 external abort logic as a quirk

Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoARM: sunxi: Reintroduce the restart code for A10/A20 SoCs
Maxime Ripard [Sun, 29 Jun 2014 13:48:53 +0000 (15:48 +0200)]
ARM: sunxi: Reintroduce the restart code for A10/A20 SoCs

This partly reverts commits 553600502b84 (ARM: sunxi: Remove reset code from
the platform) and 5e669ec583e2 (ARM: sunxi: Remove init_machine callback) for
the sun4i, sun5i and sun7i families.

This is needed because the watchdog counterpart of these commits was dropped,
and didn't make it into 3.16. In order to still be able to reboot the board, we
need to reintroduce that code. Of course, the long term view is still to get
rid of that code in mach-sunxi.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoMerge tag 'omap-for-v3.16/fixes-against-rc1' of git://git.kernel.org/pub/scm/linux...
Olof Johansson [Sat, 5 Jul 2014 04:45:38 +0000 (21:45 -0700)]
Merge tag 'omap-for-v3.16/fixes-against-rc1' of git://git./linux/kernel/git/tmlind/linux-omap into fixes

Merge OMAP fixes from Tony Lindgren:

Fixes for omaps for issues discovered during the merge window and
enabling of a few features that had to wait for the driver
dependencies to clear.

The fixes included are:

- Fix am43xx hard reset flags
- Fix SoC detection for DRA722
- Fix CPU OPP table for omap5
- Fix legacy mux parser bug if requested muxname is a prefix of
  multiple mux entries
- Fix qspi interrupt binding that relies on the irq crossbar
  that has not yet been enabled
- Add missing phy_sel for am43x-epos-evm
- Drop unused gic_init_irq() that is no longer needed

And the enabling of features that had driver dependencies are:

- Change dra7 to use Audio Tracking Logic clock instead of a fixed
  clock now that the clock driver for it has been merged

- Enable off idle configuration for selected omaps as all the kernel
  dependencies for device tree based booting are finally merged as
  this is needed to get the automated PM tests working finally with
  device tree based booting

- Add hwmod entry for ocp2scp3 for omap5 to get sata working as
  all the driver dependencies are now in the kernel and this patch
  fell through the cracks during the merge window

* tag 'omap-for-v3.16/fixes-against-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap:
  ARM: dts: dra7-evm: remove interrupt binding
  ARM: OMAP2+: Fix parser-bug in platform muxing code
  ARM: DTS: dra7/dra7xx-clocks: ATL related changes
  ARM: OMAP2+: drop unused function
  ARM: dts: am43x-epos-evm: Add Missing cpsw-phy-sel for am43x-epos-evm
  ARM: dts: omap5: Update CPU OPP table as per final production Manual
  ARM: DRA722: add detection of SoC information
  ARM: dts: Enable twl4030 off-idle configuration for selected omaps
  ARM: OMAP5: hwmod: Add ocp2scp3 and sata hwmods
  ARM: OMAP2+: hwmod: Change hardreset soc_ops for AM43XX

9 years agoMerge tag 'md/3.16-fixes' of git://neil.brown.name/md
Linus Torvalds [Fri, 4 Jul 2014 16:37:43 +0000 (09:37 -0700)]
Merge tag 'md/3.16-fixes' of git://neil.brown.name/md

Pull md bugfixes from Neil Brown:
 "Two minor bugfixes for md in 3.16"

* tag 'md/3.16-fixes' of git://neil.brown.name/md:
  md: flush writes before starting a recovery.
  md: make sure GET_ARRAY_INFO ioctl reports correct "clean" status

9 years agoMerge tag 'sound-3.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai...
Linus Torvalds [Fri, 4 Jul 2014 15:56:57 +0000 (08:56 -0700)]
Merge tag 'sound-3.16-rc4' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "This contains a few fixes for HD-audio: yet another Dell headset pin
  quirk, a fixup for Thinkpad T540P, and an improved fix for
  Haswell/Broadwell HDMI clock setup"

* tag 'sound-3.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda - restore BCLK M/N value as per CDCLK for HSW/BDW display HDA controller
  drm/i915: provide interface for audio driver to query cdclk
  ALSA: hda - Add a fixup for Thinkpad T540p
  ALSA: hda - Add another headset pin quirk for some Dell machines

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux...
Linus Torvalds [Fri, 4 Jul 2014 15:53:53 +0000 (08:53 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/mason/linux-btrfs

Pull btrfs fixes from Chris Mason:
 "We've queued up a few fixes in my for-linus branch"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
  Btrfs: fix crash when starting transaction
  Btrfs: fix btrfs_print_leaf for skinny metadata
  Btrfs: fix race of using total_bytes_pinned
  btrfs: use E2BIG instead of EIO if compression does not help
  btrfs: remove stale comment from btrfs_flush_all_pending_stuffs
  Btrfs: fix use-after-free when cloning a trailing file hole
  btrfs: fix null pointer dereference in btrfs_show_devname when name is null
  btrfs: fix null pointer dereference in clone_fs_devices when name is null
  btrfs: fix nossd and ssd_spread mount option regression
  Btrfs: fix race between balance recovery and root deletion
  Btrfs: atomically set inode->i_flags in btrfs_update_iflags
  btrfs: only unlock block in verify_parent_transid if we locked it
  Btrfs: assert send doesn't attempt to start transactions
  btrfs compression: reuse recently used workspace
  Btrfs: fix crash when mounting raid5 btrfs with missing disks
  btrfs: create sprout should rename fsid on the sysfs as well
  btrfs: dev replace should replace the sysfs entry
  btrfs: dev add should add its sysfs entry
  btrfs: dev delete should remove sysfs entry
  btrfs: rename add_device_membership to btrfs_kobj_add_device

9 years agoarm64: fix el2_setup check of CurrentEL
Marc Zyngier [Fri, 6 Jun 2014 13:16:21 +0000 (14:16 +0100)]
arm64: fix el2_setup check of CurrentEL

The CurrentEL system register reports the Current Exception Level
of the CPU. It doesn't say anything about the stack handling, and
yet we compare it to PSR_MODE_EL2t and PSR_MODE_EL2h.

It works by chance because PSR_MODE_EL2t happens to match the right
bits, but that's otherwise a very bad idea. Just check for the EL
value instead.

Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
[catalin.marinas@arm.com: fixed arch/arm64/kernel/efi-entry.S]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
9 years agoarm64: mm: Make icache synchronisation logic huge page aware
Steve Capper [Wed, 2 Jul 2014 10:46:23 +0000 (11:46 +0100)]
arm64: mm: Make icache synchronisation logic huge page aware

The __sync_icache_dcache routine will only flush the dcache for the
first page of a compound page, potentially leading to stale icache
data residing further on in a hugetlb page.

This patch addresses this issue by taking into consideration the
order of the page when flushing the dcache.

Reported-by: Mark Brown <broonie@linaro.org>
Tested-by: Mark Brown <broonie@linaro.org>
Signed-off-by: Steve Capper <steve.capper@linaro.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: <stable@vger.kernel.org> # v3.11+
9 years agoarm64: mm: Fix horrendous config typo
Steve Capper [Wed, 25 Jun 2014 07:41:45 +0000 (08:41 +0100)]
arm64: mm: Fix horrendous config typo

The define ARM64_64K_PAGES is tested for rather than
CONFIG_ARM64_64K_PAGES. Correct that typo here.

Signed-off-by: Steve Capper <steve.capper@linaro.org>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
9 years agoALSA: hda - restore BCLK M/N value as per CDCLK for HSW/BDW display HDA controller
Mengdong Lin [Thu, 3 Jul 2014 09:02:23 +0000 (17:02 +0800)]
ALSA: hda - restore BCLK M/N value as per CDCLK for HSW/BDW display HDA controller

For HSW/BDW display HD-A controller, hda_set_bclk() is defined to set BCLK
by programming the M/N values as per the core display clock (CDCLK) queried from
i915 display driver.

And the audio driver will also set BCLK in azx_first_init() since the display
driver can turn off the shared power in boot phase if only eDP is connected
and M/N values will be lost and must be reprogrammed.

Signed-off-by: Mengdong Lin <mengdong.lin@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
9 years agodrm/i915: provide interface for audio driver to query cdclk
Jani Nikula [Fri, 4 Jul 2014 02:00:37 +0000 (10:00 +0800)]
drm/i915: provide interface for audio driver to query cdclk

For Haswell and Broadwell, if the display power well has been disabled,
the display audio controller divider values EM4 M VALUE and EM5 N VALUE
will have been lost. The CDCLK frequency is required for reprogramming them
to generate 24MHz HD-A link BCLK. So provide a private interface for the
audio driver to query CDCLK.

This is a stopgap solution until a more generic interface between audio
and display drivers has been implemented.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Mengdong Lin <mengdong.lin@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
9 years agoMerge tag 'usb-3.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Linus Torvalds [Fri, 4 Jul 2014 02:12:58 +0000 (19:12 -0700)]
Merge tag 'usb-3.16-rc4' of git://git./linux/kernel/git/gregkh/usb

Pull USB bugfixes from Greg KH:
 "Here's a round of USB bugfixes, quirk additions, and new device ids
  for 3.16-rc4.  Nothing major in here at all, just a bunch of tiny
  changes.  All have been in linux-next with no reported issues"

* tag 'usb-3.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (33 commits)
  usb: chipidea: udc: delete td from req's td list at ep_dequeue
  usb: Kconfig: make EHCI_MSM selectable for QCOM SOCs
  usb-storage/SCSI: Add broken_fua blacklist flag
  usb: musb: dsps: fix the base address for accessing the mode register
  tools: ffs-test: fix header values endianess
  usb: phy: msm: Do not do runtime pm if the phy is not idle
  usb: musb: Ensure that cppi41 timer gets armed on premature DMA TX irq
  usb: gadget: gr_udc: Fix check for invalid number of microframes
  usb: musb: Fix panic upon musb_am335x module removal
  usb: gadget: f_fs: resurect usb_functionfs_descs_head structure
  Revert "tools: ffs-test: convert to new descriptor format fixing compilation error"
  xhci: Fix runtime suspended xhci from blocking system suspend.
  xhci: clear root port wake on bits if controller isn't wake-up capable
  xhci: correct burst count field for isoc transfers on 1.0 xhci hosts
  xhci: Use correct SLOT ID when handling a reset device command
  MAINTAINERS: update e-mail address
  usb: option: add/modify Olivetti Olicard modems
  USB: ftdi_sio: fix null deref at port probe
  MAINTAINERS: drop two usb-serial subdriver entries
  USB: option: add device ID for SpeedUp SU9800 usb 3g modem
  ...

9 years agoMerge tag 'staging-3.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Fri, 4 Jul 2014 02:11:48 +0000 (19:11 -0700)]
Merge tag 'staging-3.16-rc4' of git://git./linux/kernel/git/gregkh/staging

Pull staging driver bugfixes from Greg KH:
 "Nothing major here, just 4 small bugfixes that resolve some issues
  reported for the IIO (staging and non-staging) and the tidspbridge
  driver"

* tag 'staging-3.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  staging: tidspbridge: fix an erroneous removal of parentheses
  iio: of_iio_channel_get_by_name() returns non-null pointers for error legs
  staging: iio/ad7291: fix error code in ad7291_probe()
  iio:adc:ad799x: Fix reading and writing of event values, apply shift

9 years agoMerge tag 'driver-core-3.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 4 Jul 2014 01:53:13 +0000 (18:53 -0700)]
Merge tag 'driver-core-3.16-rc4' of git://git./linux/kernel/git/gregkh/driver-core

Pull driver core fixes from Greg KH:
 "Well, one drivercore fix for kernfs to resolve a reported issue with
  sysfs files being updated from atomic contexts, and another lz4 bugfix
  for testing potential buffer overflows"

* tag 'driver-core-3.16-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
  lz4: add overrun checks to lz4_uncompress_unknownoutputsize()
  kernfs: kernfs_notify() must be useable from non-sleepable contexts

9 years agoMerge tag 'trace-fixes-v3.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 4 Jul 2014 01:37:25 +0000 (18:37 -0700)]
Merge tag 'trace-fixes-v3.16-rc3' of git://git./linux/kernel/git/rostedt/linux-trace

Pull tracing fixes from Steven Rostedt:
 "Oleg Nesterov found and fixed a bug in the perf/ftrace/uprobes code
  where running:

    # perf probe -x /lib/libc.so.6 syscall
    # echo 1 >> /sys/kernel/debug/tracing/events/probe_libc/enable
    # perf record -e probe_libc:syscall whatever

  kills the uprobe.  Along the way he found some other minor bugs and
  clean ups that he fixed up making it a total of 4 patches.

  Doing unrelated work, I found that the reading of the ftrace trace
  file disables all function tracer callbacks.  This was fine when
  ftrace was the only user, but now that it's used by perf and kprobes,
  this is a bug where reading trace can disable kprobes and perf.  A
  very unexpected side effect and should be fixed"

* tag 'trace-fixes-v3.16-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  tracing: Remove ftrace_stop/start() from reading the trace file
  tracing/uprobes: Fix the usage of uprobe_buffer_enable() in probe_event_enable()
  tracing/uprobes: Kill the bogus UPROBE_HANDLER_REMOVE code in uprobe_dispatcher()
  uprobes: Change unregister/apply to WARN() if uprobe/consumer is gone
  tracing/uprobes: Revert "Support mix of ftrace and perf"

9 years agoMerge branch 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild
Linus Torvalds [Fri, 4 Jul 2014 01:34:00 +0000 (18:34 -0700)]
Merge branch 'rc-fixes' of git://git./linux/kernel/git/mmarek/kbuild

Pull kbuild fix from Michal Marek:
 "There is one more fix for the relative paths series from -rc1: Print
  the path to the build directory at the start of the build, so that
  editors and IDEs can match the relative paths to source files"

* 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild:
  kbuild: Print the name of the build directory

9 years agoMerge branch 'for-3.16' of git://linux-nfs.org/~bfields/linux
Linus Torvalds [Fri, 4 Jul 2014 01:33:22 +0000 (18:33 -0700)]
Merge branch 'for-3.16' of git://linux-nfs.org/~bfields/linux

Pull nfsd bugfixes from Bruce Fields:
 "By coincidence, two NFSv4 symlink bugs, one introduced in the 3.16 xdr
  encoding rewrite, the other a decoding bug that I think we've had
  since the start but that just doesn't trigger very often"

* 'for-3.16' of git://linux-nfs.org/~bfields/linux:
  nfs: fix nfs4d readlink truncated packet
  nfsd: fix rare symlink decoding bug

9 years agoptrace,x86: force IRET path after a ptrace_stop()
Tejun Heo [Thu, 3 Jul 2014 19:43:15 +0000 (15:43 -0400)]
ptrace,x86: force IRET path after a ptrace_stop()

The 'sysret' fastpath does not correctly restore even all regular
registers, much less any segment registers or reflags values.  That is
very much part of why it's faster than 'iret'.

Normally that isn't a problem, because the normal ptrace() interface
catches the process using the signal handler infrastructure, which
always returns with an iret.

However, some paths can get caught using ptrace_event() instead of the
signal path, and for those we need to make sure that we aren't going to
return to user space using 'sysret'.  Otherwise the modifications that
may have been done to the register set by the tracer wouldn't
necessarily take effect.

Fix it by forcing IRET path by setting TIF_NOTIFY_RESUME from
arch_ptrace_stop_needed() which is invoked from ptrace_stop().

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Andy Lutomirski <luto@amacapital.net>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: stable@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agolz4: add overrun checks to lz4_uncompress_unknownoutputsize()
Greg Kroah-Hartman [Thu, 3 Jul 2014 23:06:57 +0000 (16:06 -0700)]
lz4: add overrun checks to lz4_uncompress_unknownoutputsize()

Jan points out that I forgot to make the needed fixes to the
lz4_uncompress_unknownoutputsize() function to mirror the changes done
in lz4_decompress() with regards to potential pointer overflows.

The only in-kernel user of this function is the zram code, which only
takes data from a valid compressed buffer that it made itself, so it's
not a big issue.  But due to external kernel modules using this
function, it's better to be safe here.

Reported-by: Jan Beulich <JBeulich@suse.com>
Cc: "Don A. Bailey" <donb@securitymouse.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
9 years agoMerge remote-tracking branch 'scsi-queue/drivers-for-3.16' into for-linus
James Bottomley [Thu, 3 Jul 2014 18:04:27 +0000 (11:04 -0700)]
Merge remote-tracking branch 'scsi-queue/drivers-for-3.16' into for-linus

9 years ago[SCSI] use the scsi data buffer length to extract transfer size
Martin K. Petersen [Tue, 24 Jun 2014 14:59:35 +0000 (16:59 +0200)]
[SCSI] use the scsi data buffer length to extract transfer size

Commit 8846bab180fa introduced a helper that can be used to query the
wire transfer size for a SCSI command taking protection information into
account.

However, some commands do not have a 1:1 mapping between the block range
they work on and the payload size (discard, write same). After the
scatterlist has been set up these requests use __data_len to store the
number of bytes to report completion on. This means that callers of
scsi_transfer_length() would get the wrong byte count for these types of
requests.

To overcome this we make scsi_transfer_length() use the scatterlist
length in the scsi_data_buffer as basis for the wire transfer
calculation instead of __data_len.

Reported-by: Christoph Hellwig <hch@infradead.org>
Debugged-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Fixes: d77e65350f2d82dfa0557707d505711f5a43c8fd
Cc: stable@vger.kernel.org
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
9 years agoMerge branch 'akpm' (patches from Andrew Morton)
Linus Torvalds [Thu, 3 Jul 2014 16:22:00 +0000 (09:22 -0700)]
Merge branch 'akpm' (patches from Andrew Morton)

Merge fixes from Andrew Morton:
 "14 fixes"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  shmem: fix init_page_accessed use to stop !PageLRU bug
  kernel/printk/printk.c: revert "printk: enable interrupts before calling console_trylock_for_printk()"
  tools/testing/selftests/ipc/msgque.c: improve error handling when not running as root
  fs/seq_file: fallback to vmalloc allocation
  /proc/stat: convert to single_open_size()
  hwpoison: fix the handling path of the victimized page frame that belong to non-LRU
  mm:vmscan: update the trace-vmscan-postprocess.pl for event vmscan/mm_vmscan_lru_isolate
  msync: fix incorrect fstart calculation
  zram: revalidate disk after capacity change
  tools: memory-hotplug fix unexpected operator error
  tools: cpu-hotplug fix unexpected operator error
  autofs4: fix false positive compile error
  slub: fix off by one in number of slab tests
  mm: page_alloc: fix CMA area initialisation when pageblock > MAX_ORDER

9 years agoshmem: fix init_page_accessed use to stop !PageLRU bug
Hugh Dickins [Wed, 2 Jul 2014 22:22:38 +0000 (15:22 -0700)]
shmem: fix init_page_accessed use to stop !PageLRU bug

Under shmem swapping load, I sometimes hit the VM_BUG_ON_PAGE(!PageLRU)
in isolate_lru_pages() at mm/vmscan.c:1281!

Commit 2457aec63745 ("mm: non-atomically mark page accessed during page
cache allocation where possible") looks like interrupted work-in-progress.

mm/filemap.c's call to init_page_accessed() is fine, but not mm/shmem.c's
- shmem_write_begin() is clearly wrong to use it after shmem_getpage(),
when the page is always visible in radix_tree, and often already on LRU.

Revert change to shmem_write_begin(), and use init_page_accessed() or
mark_page_accessed() appropriately for SGP_WRITE in shmem_getpage_gfp().

SGP_WRITE also covers shmem_symlink(), which did not mark_page_accessed()
before; but since many other filesystems use [__]page_symlink(), which did
and does mark the page accessed, consider this as rectifying an oversight.

Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Prabhakar Lad <prabhakar.csengg@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agokernel/printk/printk.c: revert "printk: enable interrupts before calling console_tryl...
Andrew Morton [Wed, 2 Jul 2014 22:22:38 +0000 (15:22 -0700)]
kernel/printk/printk.c: revert "printk: enable interrupts before calling console_trylock_for_printk()"

Revert commit 939f04bec1a4 ("printk: enable interrupts before calling
console_trylock_for_printk()").

Andreas reported:

: None of the post 3.15 kernel boot for me. They all hang at the GRUB
: screen telling me it loaded and started the kernel, but the kernel
: itself stops before it prints anything (or even replaces the GRUB
: background graphics).

939f04bec1a4 is modest latency reduction.  Revert it until we understand
the reason for these failures.

Reported-by: Andreas Bombe <aeb@debian.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agotools/testing/selftests/ipc/msgque.c: improve error handling when not running as...
Shuah Khan [Wed, 2 Jul 2014 22:22:37 +0000 (15:22 -0700)]
tools/testing/selftests/ipc/msgque.c: improve error handling when not running as root

The test fails in the middle when it is not run as root while accessing
/proc/sys/kernel/msg_next_id.  Changed it to check for root at the
beginning of the test and exit if not root.

Signed-off-by: Shuah Khan <shuah.kh@samsung.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Davidlohr Bueso <davidlohr@hp.com>
Cc: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agofs/seq_file: fallback to vmalloc allocation
Heiko Carstens [Wed, 2 Jul 2014 22:22:37 +0000 (15:22 -0700)]
fs/seq_file: fallback to vmalloc allocation

There are a couple of seq_files which use the single_open() interface.
This interface requires that the whole output must fit into a single
buffer.

E.g.  for /proc/stat allocation failures have been observed because an
order-4 memory allocation failed due to memory fragmentation.  In such
situations reading /proc/stat is not possible anymore.

Therefore change the seq_file code to fallback to vmalloc allocations
which will usually result in a couple of order-0 allocations and hence
also work if memory is fragmented.

For reference a call trace where reading from /proc/stat failed:

  sadc: page allocation failure: order:4, mode:0x1040d0
  CPU: 1 PID: 192063 Comm: sadc Not tainted 3.10.0-123.el7.s390x #1
  [...]
  Call Trace:
    show_stack+0x6c/0xe8
    warn_alloc_failed+0xd6/0x138
    __alloc_pages_nodemask+0x9da/0xb68
    __get_free_pages+0x2e/0x58
    kmalloc_order_trace+0x44/0xc0
    stat_open+0x5a/0xd8
    proc_reg_open+0x8a/0x140
    do_dentry_open+0x1bc/0x2c8
    finish_open+0x46/0x60
    do_last+0x382/0x10d0
    path_openat+0xc8/0x4f8
    do_filp_open+0x46/0xa8
    do_sys_open+0x114/0x1f0
    sysc_tracego+0x14/0x1a

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Tested-by: David Rientjes <rientjes@google.com>
Cc: Ian Kent <raven@themaw.net>
Cc: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Cc: Thorsten Diehl <thorsten.diehl@de.ibm.com>
Cc: Andrea Righi <andrea@betterlinux.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Stefan Bader <stefan.bader@canonical.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years ago/proc/stat: convert to single_open_size()
Heiko Carstens [Wed, 2 Jul 2014 22:22:37 +0000 (15:22 -0700)]
/proc/stat: convert to single_open_size()

These two patches are supposed to "fix" failed order-4 memory
allocations which have been observed when reading /proc/stat.  The
problem has been observed on s390 as well as on x86.

To address the problem change the seq_file memory allocations to
fallback to use vmalloc, so that allocations also work if memory is
fragmented.

This approach seems to be simpler and less intrusive than changing
/proc/stat to use an interator.  Also it "fixes" other users as well,
which use seq_file's single_open() interface.

This patch (of 2):

Use seq_file's single_open_size() to preallocate a buffer that is large
enough to hold the whole output, instead of open coding it.  Also
calculate the requested size using the number of online cpus instead of
possible cpus, since the size of the output only depends on the number
of online cpus.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Ian Kent <raven@themaw.net>
Cc: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Cc: Thorsten Diehl <thorsten.diehl@de.ibm.com>
Cc: Andrea Righi <andrea@betterlinux.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Stefan Bader <stefan.bader@canonical.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agohwpoison: fix the handling path of the victimized page frame that belong to non-LRU
Chen Yucong [Wed, 2 Jul 2014 22:22:37 +0000 (15:22 -0700)]
hwpoison: fix the handling path of the victimized page frame that belong to non-LRU

Until now, the kernel has the same policy to handle victimized page
frames that belong to kernel-space(reserved/slab-subsystem) or
non-LRU(unknown page state).  In other word, the result of handling
either of these victimized page frames is (IGNORED | FAILED), and the
return value of memory_failure() is -EBUSY.

This patch is to avoid that memory_failure() returns very soon due to
the "true" value of (!PageLRU(p)), and it also ensures that
action_result() can report more precise information("reserved kernel",
"kernel slab", and "unknown page state") instead of "non LRU",
especially for memory errors which are detected by memory-scrubbing.

Andi said:

: While running the mcelog test suite on 3.14 I hit the following VM_BUG_ON:
:
: soft_offline: 0x56d4: unknown non LRU page type 3ffff800008000
: page:ffffea000015b400 count:3 mapcount:2097169 mapping:          (null) index:0xffff8800056d7000
: page flags: 0x3ffff800004081(locked|slab|head)
: ------------[ cut here ]------------
: kernel BUG at mm/rmap.c:1495!
:
: I think what happened is that a LRU page turned into a slab page in
: parallel with offlining.  memory_failure initially tests for this case,
: but doesn't retest later after the page has been locked.
:
: ...
:
: I ran this patch in a loop over night with some stress plus
: the mcelog test suite running in a loop. I cannot guarantee it hit it,
: but it should have given it a good beating.
:
: The kernel survived with no messages, although the mcelog test suite
: got killed at some point because it couldn't fork anymore. Probably
: some unrelated problem.
:
: So the patch is ok for me for .16.

Signed-off-by: Chen Yucong <slaoub@gmail.com>
Acked-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Reported-by: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomm:vmscan: update the trace-vmscan-postprocess.pl for event vmscan/mm_vmscan_lru_isolate
Chen Yucong [Wed, 2 Jul 2014 22:22:36 +0000 (15:22 -0700)]
mm:vmscan: update the trace-vmscan-postprocess.pl for event vmscan/mm_vmscan_lru_isolate

When using trace-vmscan-postprocess.pl for checking the file/anon rate
of scanning, we can find that it can not be performed.  At the same
time, the following message will be reported:

  WARNING: Format not as expected for event vmscan/mm_vmscan_lru_isolate
  'file' != 'contig_taken' Fewer fields than expected in format at
  ./trace-vmscan-postprocess.pl line 171, <FORMAT> line 76.

In trace-vmscan-postprocess.pl, (contig_taken, contig_dirty, and
contig_failed) are be associated respectively to (nr_lumpy_taken,
nr_lumpy_dirty, and nr_lumpy_failed) for lumpy reclaim.  Via commit
c53919adc045 ("mm: vmscan: remove lumpy reclaim"), lumpy reclaim had
already been removed by Mel, but the update for
trace-vmscan-postprocess.pl was missed.

Signed-off-by: Chen Yucong <slaoub@gmail.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agomsync: fix incorrect fstart calculation
Namjae Jeon [Wed, 2 Jul 2014 22:22:36 +0000 (15:22 -0700)]
msync: fix incorrect fstart calculation

Fix a regression caused by 7fc34a62ca44 ("mm/msync.c: sync only the
requested range in msync()").

xfstests generic/075 fail occured on ext4 data=journal mode because the
intended range was not syncing due to wrong fstart calculation.

Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Ashish Sangwan <a.sangwan@samsung.com>
Reported-by: Eric Whitney <enwlinux@gmail.com>
Tested-by: Eric Whitney <enwlinux@gmail.com>
Acked-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Reviewed-by: Lukas Czerner <lczerner@redhat.com>
Tested-by: Lukas Czerner <lczerner@redhat.com>
Reviewed-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agozram: revalidate disk after capacity change
Minchan Kim [Wed, 2 Jul 2014 22:22:36 +0000 (15:22 -0700)]
zram: revalidate disk after capacity change

Alexander reported mkswap on /dev/zram0 is failed if other process is
opening the block device file.

Step is as follows,

0. Reset the unused zram device.
1. Use a program that opens /dev/zram0 with O_RDWR and sleeps
   until killed.
2. While that program sleeps, echo the correct value to
   /sys/block/zram0/disksize.
3. Verify (e.g. in /proc/partitions) that the disk size is applied
   correctly. It is.
4. While that program still sleeps, attempt to mkswap /dev/zram0.
   This fails: mkswap: error: swap area needs to be at least 40 KiB

When I investigated, the size get by ioctl(fd, BLKGETSIZE64, xxx) on
mkswap to get a size of blockdev was zero although zram0 has right size by
2.

The reason is zram didn't revalidate disk after changing capacity so that
size of blockdev's inode is not uptodate until all of file is close.

This patch should fix the BUG.

Signed-off-by: Minchan Kim <minchan@kernel.org>
Reported-by: Alexander E. Patrakov <patrakov@gmail.com>
Tested-by: Alexander E. Patrakov <patrakov@gmail.com>
Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Acked-by: Jerome Marchand <jmarchan@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agotools: memory-hotplug fix unexpected operator error
Shuah Khan [Wed, 2 Jul 2014 22:22:36 +0000 (15:22 -0700)]
tools: memory-hotplug fix unexpected operator error

on-off-test uses "$UID != 0" to test for root, but $UID is a construct
specific to bash.  Using /bin/sh that isn't bash results in the
following error (due to the "$UID" part expanding to nothing):

  ./on-off-test.sh: 9: [: !=: unexpected operator

Change Makefile to use bash instead.

Signed-off-by: Shuah Khan <shuah.kh@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agotools: cpu-hotplug fix unexpected operator error
Shuah Khan [Wed, 2 Jul 2014 22:22:35 +0000 (15:22 -0700)]
tools: cpu-hotplug fix unexpected operator error

on-off-test uses "$UID != 0" to test for root, but $UID is a construct
specific to bash.  Using /bin/sh that isn't bash results in the
following error (due to the "$UID" part expanding to nothing):

  ./on-off-test.sh: 9: [: !=: unexpected operator

Change Makefile to use bash instead.

Signed-off-by: Shuah Khan <shuah.kh@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoautofs4: fix false positive compile error
Ian Kent [Wed, 2 Jul 2014 22:22:35 +0000 (15:22 -0700)]
autofs4: fix false positive compile error

On strict build environments we can see:

  fs/autofs4/inode.c: In function 'autofs4_fill_super':
  fs/autofs4/inode.c:312: error: 'pgrp' may be used uninitialized in this function
  make[2]: *** [fs/autofs4/inode.o] Error 1
  make[1]: *** [fs/autofs4] Error 2
  make: *** [fs] Error 2
  make: *** Waiting for unfinished jobs....

This is due to the use of pgrp_set being used to indicate pgrp has has
been set rather than initializing pgrp itself.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoslub: fix off by one in number of slab tests
Joonsoo Kim [Wed, 2 Jul 2014 22:22:35 +0000 (15:22 -0700)]
slub: fix off by one in number of slab tests

min_partial means minimum number of slab cached in node partial list.
So, if nr_partial is less than it, we keep newly empty slab on node
partial list rather than freeing it.  But if nr_partial is equal or
greater than it, it means that we have enough partial slabs so should
free newly empty slab.  Current implementation missed the equal case so
if we set min_partial is 0, then, at least one slab could be cached.
This is critical problem to kmemcg destroying logic because it doesn't
works properly if some slabs is cached.  This patch fixes this problem.

Fixes 91cb69620284 ("slub: make dead memcg caches discard free slabs
immediately").

Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>