OSDN Git Service

tomoyo/tomoyo-test1.git
9 years agoMerge branch 'at91/cleanup5' into next/drivers
Arnd Bergmann [Mon, 8 Dec 2014 17:27:21 +0000 (18:27 +0100)]
Merge branch 'at91/cleanup5' into next/drivers

The at91 cleanups changed a lot of files, this merges in the
latest cleanups to resolve the conflicts

Conflicts:
arch/arm/mach-at91/at91sam9260.c
arch/arm/mach-at91/at91sam9261.c
arch/arm/mach-at91/at91sam9263.c
arch/arm/mach-at91/clock.c
arch/arm/mach-at91/clock.h
drivers/rtc/Kconfig

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoMerge branch 'brcm/stb-smp-uart' into next/drivers
Arnd Bergmann [Mon, 8 Dec 2014 16:16:03 +0000 (17:16 +0100)]
Merge branch 'brcm/stb-smp-uart' into next/drivers

This resolves a nonobvious merge conflict that I got wrong the
first time.

* brcm/stb-smp-uart:
  bus: brcmstb_gisb: save and restore GISB timeout
  bus: brcmstb_gisb: register the fault code hook
  ARM: brcmstb: Kconfig: drop unneeded symbol selections
  ARM: brcmstb: reintroduce SMP support
  ARM: brcmstb: add debug UART for earlyprintk support

Conflicts:
drivers/bus/brcmstb_gisb.c

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reported-by: Florian Fainelli <f.fainelli@gmail.com>
9 years agoMerge branch 'clocksource/physical-timers' into next/drivers
Olof Johansson [Fri, 5 Dec 2014 07:32:16 +0000 (23:32 -0800)]
Merge branch 'clocksource/physical-timers' into next/drivers

* clocksource/physical-timers:
  clocksource: arch_timer: Allow the device tree to specify uninitialized timer registers
  clocksource: arch_timer: Fix code to use physical timers when requested

9 years agoclocksource: arch_timer: Allow the device tree to specify uninitialized timer registers
Doug Anderson [Wed, 8 Oct 2014 07:33:47 +0000 (00:33 -0700)]
clocksource: arch_timer: Allow the device tree to specify uninitialized timer registers

Some 32-bit (ARMv7) systems are architected like this:

* The firmware doesn't know and doesn't care about hypervisor mode and
  we don't want to add the complexity of hypervisor there.

* The firmware isn't involved in SMP bringup or resume.

* The ARCH timer come up with an uninitialized offset (CNTVOFF)
  between the virtual and physical counters.  Each core gets a
  different random offset.

* The device boots in "Secure SVC" mode.

* Nothing has touched the reset value of CNTHCTL.PL1PCEN or
  CNTHCTL.PL1PCTEN (both default to 1 at reset)

On systems like the above, it doesn't make sense to use the virtual
counter.  There's nobody managing the offset and each time a core goes
down and comes back up it will get reinitialized to some other random
value.

This adds an optional property which can inform the kernel of this
situation, and firmware is free to remove the property if it is going
to initialize the CNTVOFF registers when each CPU comes out of reset.

Currently, the best course of action in this case is to use the
physical timer, which is why it is important that CNTHCTL hasn't been
changed from its reset value and it's a reasonable assumption given
that the firmware has never entered HYP mode.

Note that it's been said that on ARMv8 systems the firmware and
kernel really can't be architected as described above.  That means
using the physical timer like this really only makes sense for ARMv7
systems.

Signed-off-by: Doug Anderson <dianders@chromium.org>
Signed-off-by: Sonny Rao <sonnyrao@chromium.org>
Reviewed-by: Mark Rutland <mark.rutland@arm.com>
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoclocksource: arch_timer: Fix code to use physical timers when requested
Sonny Rao [Mon, 24 Nov 2014 07:02:44 +0000 (23:02 -0800)]
clocksource: arch_timer: Fix code to use physical timers when requested

This is a bug fix for using physical arch timers when
the arch_timer_use_virtual boolean is false.  It restores the
arch_counter_get_cntpct() function after removal in

0d651e4e "clocksource: arch_timer: use virtual counters"

We need this on certain ARMv7 systems which are architected like this:

* The firmware doesn't know and doesn't care about hypervisor mode and
  we don't want to add the complexity of hypervisor there.

* The firmware isn't involved in SMP bringup or resume.

* The ARCH timer come up with an uninitialized offset between the
  virtual and physical counters.  Each core gets a different random
  offset.

* The device boots in "Secure SVC" mode.

* Nothing has touched the reset value of CNTHCTL.PL1PCEN or
  CNTHCTL.PL1PCTEN (both default to 1 at reset)

One example of such as system is RK3288 where it is much simpler to
use the physical counter since there's nobody managing the offset and
each time a core goes down and comes back up it will get reinitialized
to some other random value.

Fixes: 0d651e4e65e9 ("clocksource: arch_timer: use virtual counters")
Cc: stable@vger.kernel.org
Signed-off-by: Sonny Rao <sonnyrao@chromium.org>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoMerge tag 'tegra-for-3.19-iommu' of git://git.kernel.org/pub/scm/linux/kernel/git...
Arnd Bergmann [Thu, 4 Dec 2014 16:20:02 +0000 (17:20 +0100)]
Merge tag 'tegra-for-3.19-iommu' of git://git./linux/kernel/git/tegra/linux into next/drivers

Pull "ARM: tegra: IOMMU support for v3.19" from Thierry Reding:

This adds the driver pieces required for IOMMU support on Tegra30,
Tegra114 and Tegra124.

* tag 'tegra-for-3.19-iommu' of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
  memory: Add NVIDIA Tegra memory controller support
  of: Add NVIDIA Tegra memory controller binding
  ARM: tegra: Move AHB Kconfig to drivers/amba
  amba: Add Kconfig file
  clk: tegra: Implement memory-controller clock
  powerpc/iommu: Rename iommu_[un]map_sg functions
  iommu: Improve error handling when setting bus iommu
  iommu: Do more input validation in iommu_map_sg()
  iommu: Add iommu_map_sg() function

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoMerge tag 'reset-for-3.19-2' of git://git.pengutronix.de/git/pza/linux into next...
Arnd Bergmann [Thu, 4 Dec 2014 15:57:36 +0000 (16:57 +0100)]
Merge tag 'reset-for-3.19-2' of git://git.pengutronix.de/git/pza/linux into next/drivers

Pull "Reset controller changes for v3.19" from Philipp Zabel:

This adds a new driver for the sti soc family, and creates
a reset_control_status interface, which is added to the existing
drivers.

* tag 'reset-for-3.19-2' of git://git.pengutronix.de/git/pza/linux:
  reset: add socfpga_reset_status
  reset: sti: Document sti-picophyreset controllers bindings.
  reset: stih407: Add softreset, powerdown and picophy controllers
  reset: stih407: Add reset controllers DT bindings
  reset: add reset_control_status helper function

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoMerge tag 'arm-soc/for-3.19/brcmstb-drivers' of https://github.com/brcm/linux into...
Arnd Bergmann [Thu, 4 Dec 2014 15:44:05 +0000 (16:44 +0100)]
Merge tag 'arm-soc/for-3.19/brcmstb-drivers' of https://github.com/brcm/linux into next/drivers

This pull request contains the following changes to the Broadcom GISB bus
arbiter from Kevin Cernekee:

- Extend brcmstb GISB bus driver to work on MIPS (currently ARM-only) and support
  65nm and 40nm MIPS-based chips such as: BCM7038, BCM7400 and BCM7435

* tag 'arm-soc/for-3.19/brcmstb-drivers' of https://github.com/brcm/linux:
  bus: brcmstb_gisb: Add register offset tables for older chips
  bus: brcmstb_gisb: Look up register offsets in a table
  bus: brcmstb_gisb: Introduce wrapper functions for MMIO accesses
  bus: brcmstb_gisb: Make the driver buildable on MIPS

Conflicts:
drivers/bus/brcmstb_gisb.c

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agomemory: Add NVIDIA Tegra memory controller support
Thierry Reding [Wed, 16 Apr 2014 07:24:44 +0000 (09:24 +0200)]
memory: Add NVIDIA Tegra memory controller support

The memory controller on NVIDIA Tegra exposes various knobs that can be
used to tune the behaviour of the clients attached to it.

Currently this driver sets up the latency allowance registers to the HW
defaults. Eventually an API should be exported by this driver (via a
custom API or a generic subsystem) to allow clients to register latency
requirements.

This driver also registers an IOMMU (SMMU) that's implemented by the
memory controller. It is supported on Tegra30, Tegra114 and Tegra124
currently. Tegra20 has a GART instead.

The Tegra SMMU operates on memory clients and SWGROUPs. A memory client
is a unidirectional, special-purpose DMA master. A SWGROUP represents a
set of memory clients that form a logical functional unit corresponding
to a single device. Typically a device has two clients: one client for
read transactions and one client for write transactions, but there are
also devices that have only read clients, but many of them (such as the
display controllers).

Because there is no 1:1 relationship between memory clients and devices
the driver keeps a table of memory clients and the SWGROUPs that they
belong to per SoC. Note that this is an exception and due to the fact
that the SMMU is tightly integrated with the rest of the Tegra SoC. The
use of these tables is discouraged in drivers for generic IOMMU devices
such as the ARM SMMU because the same IOMMU could be used in any number
of SoCs and keeping such tables for each SoC would not scale.

Acked-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>
9 years agoARM: at91: remove unused board.h file
Arnd Bergmann [Tue, 2 Dec 2014 16:25:19 +0000 (17:25 +0100)]
ARM: at91: remove unused board.h file

All functions declared in this file are gone.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
[nicolas.ferre@atmel.com: re-order patches so modify board-dt-sam9]
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: remove unneeded header files
Nicolas Ferre [Tue, 2 Dec 2014 14:22:25 +0000 (15:22 +0100)]
ARM: at91: remove unneeded header files

These files were left behind with no reason. Remove them.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/clocksource: remove !DT PIT initializations
Arnd Bergmann [Tue, 2 Dec 2014 13:54:03 +0000 (14:54 +0100)]
ARM: at91/clocksource: remove !DT PIT initializations

As AT91 !DT code is now removed, cleanup the PIT clocksource driver.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
[nicolas.ferre@atmel.com: split patch]
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Boris BREZILLON <boris.brezillon@free-electrons.com>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
9 years agoARM: at91: at91rm9200 ST initialization is now DT only
Arnd Bergmann [Tue, 2 Dec 2014 13:26:44 +0000 (14:26 +0100)]
ARM: at91: at91rm9200 ST initialization is now DT only

As at91rm9200 is now DT only, there is no need to keep old entry point in this
at91rm9200 System Timer (ST) driver.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
[nicolas.ferre@atmel.com: split patch]
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: remove old AT91-specific drivers
Arnd Bergmann [Tue, 2 Dec 2014 11:08:27 +0000 (12:08 +0100)]
ARM: at91: remove old AT91-specific drivers

GPIO and LED drivers were replaced by generic ones for DT boards. These drivers
were remaining: delete them now. Modifications are also done on the
corresponding header files.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
[nicolas.ferre@atmel.com: split patch]
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: cleanup initilisation code by removing dead code
Arnd Bergmann [Tue, 2 Dec 2014 11:54:57 +0000 (12:54 +0100)]
ARM: at91: cleanup initilisation code by removing dead code

The AT91-specific SoC strucutre "struct at91_init_soc" was filled with specific
!DT initilisation functions. Now that we got rid of the !DT board file
description, remove unneeded functions.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
[nicolas.ferre@atmel.com: split patch]
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/Kconfig: select board files automatically
Arnd Bergmann [Tue, 2 Dec 2014 09:32:52 +0000 (10:32 +0100)]
ARM: at91/Kconfig: select board files automatically

An explicit selection option is not needed for board files so now we select the
board from SoC option.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
[nicolas.ferre@atmel.com: remove option's comments; split patch]
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agobus: brcmstb_gisb: Add register offset tables for older chips
Kevin Cernekee [Wed, 26 Nov 2014 00:49:52 +0000 (16:49 -0800)]
bus: brcmstb_gisb: Add register offset tables for older chips

This will select the appropriate register layout based on the DT
"compatible" string.

Signed-off-by: Kevin Cernekee <cernekee@gmail.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
9 years agobus: brcmstb_gisb: Look up register offsets in a table
Kevin Cernekee [Wed, 26 Nov 2014 00:49:51 +0000 (16:49 -0800)]
bus: brcmstb_gisb: Look up register offsets in a table

There are at least 4 incompatible variations of this hardware block,
so let's use the ARB_* constants as a table index instead of hardcoding
specific register offsets.  Also, allow for the possibility of adding
old devices that are missing some of the registers.

Signed-off-by: Kevin Cernekee <cernekee@gmail.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
9 years agobus: brcmstb_gisb: Introduce wrapper functions for MMIO accesses
Kevin Cernekee [Wed, 26 Nov 2014 00:49:50 +0000 (16:49 -0800)]
bus: brcmstb_gisb: Introduce wrapper functions for MMIO accesses

These will be used to abstract out chip-to-chip differences.

Signed-off-by: Kevin Cernekee <cernekee@gmail.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
9 years agobus: brcmstb_gisb: Make the driver buildable on MIPS
Kevin Cernekee [Wed, 26 Nov 2014 00:49:49 +0000 (16:49 -0800)]
bus: brcmstb_gisb: Make the driver buildable on MIPS

BCM7xxx ARM and MIPS platforms share a similar hardware block for
reporting GISB errors, so they both benefit from the use of this driver.
Conditionally compile the ARM-specific bus error handler so that the
GISB error IRQ handler works on other architectures.

Signed-off-by: Kevin Cernekee <cernekee@gmail.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
9 years agoMerge tag 'samsung-driver' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene...
Arnd Bergmann [Fri, 28 Nov 2014 12:25:21 +0000 (13:25 +0100)]
Merge tag 'samsung-driver' of git://git./linux/kernel/git/kgene/linux-samsung into next/drivers

Pull "Samsung driver update for v3.19" from Kukjin Kim:

- Change samsung serial dependencies for exynos7 (ARMv8)

* tag 'samsung-driver' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung:
  serial: samsung: Fix serial config dependencies for exynos7

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoARM: at91: remove unused IRQ function declarations
Nicolas Ferre [Thu, 27 Nov 2014 15:28:54 +0000 (16:28 +0100)]
ARM: at91: remove unused IRQ function declarations

Since 3b26f39b0ab1 (ARM: at91: make use of the new AIC driver for dt enabled
boards) the old IRQ initialisation functions aren't used anymore: remove their
declaration in generic.h.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
9 years agoARM: at91: remove legacy IRQ driver and related code
Nicolas Ferre [Thu, 27 Nov 2014 15:10:06 +0000 (16:10 +0100)]
ARM: at91: remove legacy IRQ driver and related code

Remove irc.c and associated header file. The related code was idendified by
the CONFIG_OLD_IRQ_AT91 option that was removed previously. It has been spotted
by following coccinelle semantic match:

@rule1@
expression E;
statement S;
@@
(
- if (IS_ENABLED(CONFIG_OLD_IRQ_AT91)) S
|
- if (IS_ENABLED(CONFIG_OLD_IRQ_AT91) && E) S
)

Cc: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
9 years agoARM: at91: remove old at91-specific clock driver
Nicolas Ferre [Wed, 26 Nov 2014 16:19:14 +0000 (17:19 +0100)]
ARM: at91: remove old at91-specific clock driver

This clock driver collection was specific to AT91 and only used in !DT cases.
All clocks and the clock trees for all Atmel SoCs are now described by drivers
using the common clock framework.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
9 years agoARM: at91: remove clock data in at91sam9n12.c and at91sam9x5.c files
Nicolas Ferre [Wed, 26 Nov 2014 16:25:24 +0000 (17:25 +0100)]
ARM: at91: remove clock data in at91sam9n12.c and at91sam9x5.c files

As the CONFIG_OLD_CLK_AT91 option is gone, let's completely remove the AT91
old clock driver related data.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
9 years agoARM: at91: remove all !DT related configuration options
Nicolas Ferre [Fri, 21 Nov 2014 16:10:06 +0000 (17:10 +0100)]
ARM: at91: remove all !DT related configuration options

OLD_CLK_AT91 & OLD_IRQ_AT91 were only selected by entries in Kconfig.non_dt
that are now gone. So we remove all this legacy stuff and select the proper
options in the SOC_ entries.

As USE_OF is now selected directly in arch/arm/Kconfig AT91 entry, we can
safely remove it everywhere in this file.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
9 years agoARM: at91/trivial: update Kconfig comment to mention SAMA5
Nicolas Ferre [Fri, 21 Nov 2014 15:44:31 +0000 (16:44 +0100)]
ARM: at91/trivial: update Kconfig comment to mention SAMA5

Cortex-A5 SAMA5 processors were not listed, add this in the AT91 comment.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
9 years agoARM: at91: always USE_OF from now on
Nicolas Ferre [Fri, 21 Nov 2014 15:41:50 +0000 (16:41 +0100)]
ARM: at91: always USE_OF from now on

As we always use Device Tree now, we can add the configuration here.
Also remove the condition for PINCTRL_AT91.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
9 years agoARM: at91/Kconfig: remove ARCH_AT91RM9200 option for drivers
Nicolas Ferre [Fri, 21 Nov 2014 15:34:08 +0000 (16:34 +0100)]
ARM: at91/Kconfig: remove ARCH_AT91RM9200 option for drivers

The precise selection is useless, so we simply remove these dependencies.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
9 years agoof: Add NVIDIA Tegra memory controller binding
Thierry Reding [Thu, 26 Jun 2014 19:18:53 +0000 (21:18 +0200)]
of: Add NVIDIA Tegra memory controller binding

The memory controller on NVIDIA Tegra exposes various knobs that can be
used to tune the behaviour of the clients attached to it.

In addition, the memory controller implements an SMMU (IOMMU) which can
translate I/O virtual addresses to physical addresses for clients. This
is useful for scatter-gather operation on devices that don't support it
natively and for virtualization or process separation.

Signed-off-by: Thierry Reding <treding@nvidia.com>
9 years agoARM: tegra: Move AHB Kconfig to drivers/amba
Thierry Reding [Tue, 29 Jul 2014 14:24:25 +0000 (16:24 +0200)]
ARM: tegra: Move AHB Kconfig to drivers/amba

This will allow the Kconfig option to be shared among 32-bit and 64-bit
ARM.

Signed-off-by: Thierry Reding <treding@nvidia.com>
9 years agoamba: Add Kconfig file
Thierry Reding [Thu, 2 Oct 2014 12:57:43 +0000 (14:57 +0200)]
amba: Add Kconfig file

Rather than duplicate the ARM_AMBA Kconfig symbol in both 32-bit and
64-bit ARM architectures, move the common definition to drivers/amba
where dependent drivers will be located.

Signed-off-by: Thierry Reding <treding@nvidia.com>
9 years agoclk: tegra: Implement memory-controller clock
Thierry Reding [Tue, 29 Jul 2014 08:17:53 +0000 (10:17 +0200)]
clk: tegra: Implement memory-controller clock

The memory controller clock runs either at half or the same frequency as
the EMC clock.

Reviewed-By: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Acked-by: Mike Turquette <mturquette@linaro.org>
Signed-off-by: Thierry Reding <treding@nvidia.com>
9 years agoMerge branch 'core' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Thierry Reding [Wed, 26 Nov 2014 08:41:09 +0000 (09:41 +0100)]
Merge branch 'core' of git://git./linux/kernel/git/joro/iommu

This branch contains a couple of changes that will conflict with the
Tegra SMMU driver rewrite. Since the driver is largely rewritten the
conflict resolution is non-trivial.

9 years agoARM: at91: switch configuration option to SOC_AT91RM9200
Nicolas Ferre [Fri, 21 Nov 2014 15:22:17 +0000 (16:22 +0100)]
ARM: at91: switch configuration option to SOC_AT91RM9200

As the ARCH_AT91RM9200 is removed because being !DT, we use
the SOC_AT91RM9200 variant. This option can certainly be removed
once the ST driver is reworked a bit.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
9 years agoARM: at91: remove at91rm9200 legacy board support
Nicolas Ferre [Fri, 21 Nov 2014 15:15:50 +0000 (16:15 +0100)]
ARM: at91: remove at91rm9200 legacy board support

Second part of at91rm9200 legacy !DT removal. This is the core !DT support
removal for this Atmel SoC.
Note that from now on, the Kconfig.non_dt file and its specialized options are
completely removed.
Use the Device Tree for running this board with newer kernels.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
9 years agoARM: at91: remove at91rm9200 legacy boards files
Nicolas Ferre [Fri, 21 Nov 2014 13:30:13 +0000 (14:30 +0100)]
ARM: at91: remove at91rm9200 legacy boards files

Remove old board files that use at91rm9200 Atmel SoC. The
device tree is mature on this SoCs. It must be used now.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
9 years agoserial: samsung: Fix serial config dependencies for exynos7
Pankaj Dubey [Fri, 21 Nov 2014 15:09:25 +0000 (00:09 +0900)]
serial: samsung: Fix serial config dependencies for exynos7

Exynos7 has a similar serial controller to that present in older Samsung
SoCs. To re-use the existing serial driver on Exynos7 we need to have
SERIAL_SAMSUNG_UARTS_4 and SERIAL_SAMSUNG_UARTS selected. This is not
possible because these symbols are dependent on PLAT_SAMSUNG which is
not present for the ARMv8 based exynos7.

Change the dependency of these symbols from PLAT_SAMSUNG to the serial
driver thus making it available on exynos7. As the existing platform
specific code making use of these symbols is related to uart driver this
change in dependency should not cause any issues.

Signed-off-by: Pankaj Dubey <pankaj.dubey@samsung.com>
Signed-off-by: Naveen Krishna Chatradhi <ch.naveen@samsung.com>
Signed-off-by: Abhilash Kesavan <a.kesavan@samsung.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Kukjin Kim <kgene.kim@samsung.com>
9 years agobus: brcmstb_gisb: resolve section mismatch
Florian Fainelli [Thu, 20 Nov 2014 18:14:46 +0000 (10:14 -0800)]
bus: brcmstb_gisb: resolve section mismatch

Commit f1bee783dd37 moved the call to hook_fault_code in
brcmstb_gisb_arb_probe() which now calls a function annotated with __init, so
this one must also be annotated with __init.

In order to avoid introducing another section mismatch, call
platform_driver_probe() manually and remove the .probe assignment from
brcmstb_gisb_arb_driver, this is very similar to what
drivers/pci/host/pci-imx6.c does since we basically have the same constraints
here.

Fixes: f1bee783dd37 ("bus: brcmstb_gisb: register the fault code hook")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoARM: common: edma: edma_pm_resume may be unused
Arnd Bergmann [Fri, 21 Nov 2014 10:48:54 +0000 (11:48 +0100)]
ARM: common: edma: edma_pm_resume may be unused

The recently introduced resume hook in the edma driver
is not referenced when CONFIG_PM_SLEEP is not set, which
results in a compile warning in keystone builds.

This adds an appropriate #ifdef.

Cc: Nishanth Menon <nm@ti.com>
Cc: Daniel Mack <zonque@gmail.com>
Cc: Joel Fernandes <joelf@ti.com>
Acked-by: Sekhar Nori <nsekhar@ti.com>
Fixes: a2b1175131: ("ARM: common: edma: add suspend resume hook")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoARM: at91/Kconfig: remove useless fbdev Kconfig options
Nicolas Ferre [Fri, 21 Nov 2014 11:14:08 +0000 (12:14 +0100)]
ARM: at91/Kconfig: remove useless fbdev Kconfig options

Remove Kconfig options used nowhere.
The removal of CONFIG_ARCH_AT91SAM9261 option allowed to spot this oddness.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: remove at91sam9261/at91sam9g10 legacy board support
Nicolas Ferre [Fri, 21 Nov 2014 11:09:27 +0000 (12:09 +0100)]
ARM: at91: remove at91sam9261/at91sam9g10 legacy board support

Remove legacy support for at91sam9261/at91sam9g10 boards.
This include board files removal plus all legacy code for non DT boards
support.
Use the Device Tree for running this board with newer kernels.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoMerge tag 'davinci-for-v3.19/edma' of git://git.kernel.org/pub/scm/linux/kernel/git...
Arnd Bergmann [Thu, 20 Nov 2014 16:55:58 +0000 (17:55 +0100)]
Merge tag 'davinci-for-v3.19/edma' of git://git./linux/kernel/git/nsekhar/linux-davinci into next/drivers

Pull "Adds suspend/resume support to EDMA driver" from Sekhar Nori:

* tag 'davinci-for-v3.19/edma' of git://git.kernel.org/pub/scm/linux/kernel/git/nsekhar/linux-davinci:
  ARM: common: edma: add suspend resume hook

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoMerge branch 'at91/cleanup' into next/drivers
Arnd Bergmann [Thu, 20 Nov 2014 13:02:34 +0000 (14:02 +0100)]
Merge branch 'at91/cleanup' into next/drivers

This resolves some of the obvious conflicts between the at91 cleanup and
drivers branches.

Conflicts:
arch/arm/mach-at91/at91sam9g45.c
arch/arm/mach-at91/at91sam9rl.c
drivers/rtc/Kconfig

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoMerge tag 'arm-perf-3.19' of git://git.kernel.org/pub/scm/linux/kernel/git/will/linux...
Arnd Bergmann [Thu, 20 Nov 2014 12:49:52 +0000 (13:49 +0100)]
Merge tag 'arm-perf-3.19' of git://git./linux/kernel/git/will/linux into next/drivers

Pull "ARM: perf: updates for 3.19" from Will Deacon:

This patch series takes us slightly further on the road to big.LITTLE
support in perf. The main change enabling this is moving the CCI PMU
driver away from the arm-pmu abstraction, allowing the arch code to
focus specifically on support for CPU PMUs.

* tag 'arm-perf-3.19' of git://git.kernel.org/pub/scm/linux/kernel/git/will/linux:
  arm: perf: fold hotplug notifier into arm_pmu
  arm: perf: dynamically allocate cpu hardware data
  arm: perf: fold percpu_pmu into pmu_hw_events
  arm: perf: kill get_hw_events()
  arm: perf: limit size of accounting data
  arm: perf: use IDR types for CPU PMUs
  arm: perf: make PMU probing data-driven
  arm: perf: add missing pr_info newlines
  arm: perf: factor out callchain code
  ARM: perf: use pr_* instead of printk
  ARM: perf: remove useless return and check of idx in counter handling
  bus: cci: move away from arm_pmu framework

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoMerge tag 'at91-drivers' of git://git.kernel.org/pub/scm/linux/kernel/git/nferre...
Arnd Bergmann [Thu, 20 Nov 2014 12:42:21 +0000 (13:42 +0100)]
Merge tag 'at91-drivers' of git://git./linux/kernel/git/nferre/linux-at91 into next/drivers

Pull "First batch of drivers for 3.19" from Nicolas Ferre:

It is only about a not so recent driver for old platforms: RTT as RTC driver:
- RTT as RTC driver enhancements and machine specific include files removal
- RTT as RTC driver conversion to device tree

* tag 'at91-drivers' of git://git.kernel.org/pub/scm/linux/kernel/git/nferre/linux-at91:
  rtc: at91sam9: add DT bindings documentation
  rtc: at91sam9: use clk API instead of relying on AT91_SLOW_CLOCK
  ARM: at91: add clk_lookup entry for RTT devices
  rtc: at91sam9: rework the Kconfig description
  rtc: at91sam9: make use of syscon/regmap to access GPBR registers
  rtc: at91sam9: add DT support
  rtc: at91sam9: replace devm_ioremap by devm_ioremap_resource
  rtc: at91sam9: use standard readl/writel functions instead of raw versions
  rtc: at91sam9: remove references to mach specific headers

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoMerge tag 'keystone-driver-soc' of git://git.kernel.org/pub/scm/linux/kernel/git...
Arnd Bergmann [Thu, 20 Nov 2014 12:15:25 +0000 (13:15 +0100)]
Merge tag 'keystone-driver-soc' of git://git./linux/kernel/git/ssantosh/linux-keystone into next/drivers

Pull "Keystone SOC Navigator driver non critical fixes frm Alex for 3.19" from Santosh Shilimkar:

- Use list_for_each_entry_safe to prevent use after free
- Return proper error if devm_kzalloc fails
- Use list_first_entry_or_null() at appropriate places

* tag 'keystone-driver-soc' of git://git.kernel.org/pub/scm/linux/kernel/git/ssantosh/linux-keystone:
  soc: ti: knav_qmss_queue: Use list_for_each_entry_safe to prevent use after free
  soc: ti: knav_qmss_queue: Return proper error if devm_kzalloc fails
  soc: ti: knav_qmss_queue: Fix unbalanced locking ins knav_pool_create()
  soc: ti: Use list_first_entry_or_null() at appropriate places

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoMerge tag 'omap-for-v3.19/w1-and-l3-noc' of git://git.kernel.org/pub/scm/linux/kernel...
Arnd Bergmann [Thu, 20 Nov 2014 11:03:16 +0000 (12:03 +0100)]
Merge tag 'omap-for-v3.19/w1-and-l3-noc' of git://git./linux/kernel/git/tmlind/linux-omap into next/drivers

Pull "omap driver changes for v3.19" from Tony Lindgren:

Few omap driver changes to add minimal device tree support
for the omap 1w driver, and to fix resume and interrupt issues
on the l3-noc driver.

* tag 'omap-for-v3.19/w1-and-l3-noc' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap:
  bus: omap_l3_noc: Correct returning IRQ_HANDLED unconditionally in the irq handler
  bus: omap_l3_noc: Add resume hook to restore context
  w1: omap-hdq: support device probing with device-tree

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
9 years agoARM: at91/Kconfig: remove unused config options
Nicolas Ferre [Wed, 19 Nov 2014 10:15:02 +0000 (11:15 +0100)]
ARM: at91/Kconfig: remove unused config options

When removing old board !DT support, several Kconfig options were deleted.
Propagate this removal to drivers Kconfig files.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: remove at91sam9260/at91sam9g20 legacy board support
Nicolas Ferre [Wed, 19 Nov 2014 09:52:43 +0000 (10:52 +0100)]
ARM: at91: remove at91sam9260/at91sam9g20 legacy board support

Second part of at91sam9260/at91sam9g20 legacy !DT removal. This is the core !DT
support removal for these two Atmel SoCs.
Use the Device Tree for running this board with newer kernels.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: remove at91sam9260/at91sam9g20 legacy boards files
Nicolas Ferre [Wed, 19 Nov 2014 09:30:41 +0000 (10:30 +0100)]
ARM: at91: remove at91sam9260/at91sam9g20 legacy boards files

Remove old board files that use at91sam9260 or at91sam9g20 Atmel SoCs. The
device tree is mature on these SoCs. It must be used now.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: remove at91sam9263 legacy board support
Nicolas Ferre [Wed, 19 Nov 2014 09:05:54 +0000 (10:05 +0100)]
ARM: at91: remove at91sam9263 legacy board support

Remove legacy support for at91sam9263 boards.
This include board files removal plus all legacy code for non DT boards
support.
Use the Device Tree for running this board with newer kernels.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91/at91sam9g45: remove useless header file
Nicolas Ferre [Wed, 19 Nov 2014 09:38:39 +0000 (10:38 +0100)]
ARM: at91/at91sam9g45: remove useless header file

Remove this useless cpu.h header file forgotten during the !DT support removal.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: common: edma: add suspend resume hook
Daniel Mack [Tue, 26 Aug 2014 08:52:53 +0000 (10:52 +0200)]
ARM: common: edma: add suspend resume hook

This patch makes the edma driver resume correctly after suspend. Tested
on an AM33xx platform with cyclic audio streams and omap_hsmmc.

All information can be reconstructed by already known runtime
information.

As we now use some functions that were previously only used from __init
context, annotations had to be dropped.

[nm@ti.com: added error handling for runtime + suspend_late/early_resume]
Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Daniel Mack <zonque@gmail.com>
Tested-by: Joel Fernandes <joelf@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
[nsekhar@ti.com: remove unneeded pm_runtime_get_sync() from resume]
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
9 years agopowerpc/iommu: Rename iommu_[un]map_sg functions
Joerg Roedel [Wed, 5 Nov 2014 14:28:30 +0000 (15:28 +0100)]
powerpc/iommu: Rename iommu_[un]map_sg functions

The IOMMU-API gained support for a new iommu_map_sg
function. This causes compile failures on powerpc because
the function name is already globally used there.
This patch renames adds a ppc_ prefix to these functions to
solve the compile problem.

Signed-off-by: Joerg Roedel <jroedel@suse.de>
9 years agortc: at91sam9: add DT bindings documentation
Boris BREZILLON [Tue, 23 Sep 2014 14:45:12 +0000 (16:45 +0200)]
rtc: at91sam9: add DT bindings documentation

Add RTT bindings documentation.

Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Acked-by: Johan Hovold <johan@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agortc: at91sam9: use clk API instead of relying on AT91_SLOW_CLOCK
Boris BREZILLON [Tue, 23 Sep 2014 14:41:07 +0000 (16:41 +0200)]
rtc: at91sam9: use clk API instead of relying on AT91_SLOW_CLOCK

The RTT block is using the slow clock which is accessible through the clk
API.
Use the clk API to retrieve, enable and get the slow clk rate instead of
the AT91_SLOW_CLOCK macro (which hardcodes the slow clk rate).
Doing this allows us to reference the clk thus preventing the CCF from
disabling it during the "disable unused" phase.

Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Johan Hovold <johan@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: add clk_lookup entry for RTT devices
Boris BREZILLON [Tue, 23 Sep 2014 11:18:33 +0000 (13:18 +0200)]
ARM: at91: add clk_lookup entry for RTT devices

First export the clk32k clk.
Then add clk_lookup entries for RTT devices so that rtc-at91sam9 driver
can retrieve and manipulate the slow clk.

Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Johan Hovold <johan@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agortc: at91sam9: rework the Kconfig description
Boris BREZILLON [Tue, 23 Sep 2014 11:16:05 +0000 (13:16 +0200)]
rtc: at91sam9: rework the Kconfig description

Remove all references to AT91CAP9 SoC which has been removed.
Rework help message to remove any specific references to AT91SAM9 SoCs.
State that RTC_DRV_AT91SAM9_RTT and RTC_DRV_AT91SAM9_GPBR options are only
used when booting non DT boards.

Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Acked-by: Johan Hovold <johan@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agortc: at91sam9: make use of syscon/regmap to access GPBR registers
Boris BREZILLON [Tue, 23 Sep 2014 11:14:44 +0000 (13:14 +0200)]
rtc: at91sam9: make use of syscon/regmap to access GPBR registers

The GPBR registers are not part of the RTT block and thus should not be
defined in the reg property of the rtt node.

Use syscon to provide a proper DT representation and reference the GPBR
syscon device in a new "atmel,rtt-rtc-time-reg" property which store both
the syscon device phandle and the register offset within the GPBR block.

When using non DT boards, we won't be able to retrieve the syscon regmap,
hence we need to create our own regmap using the memory region defined
in the 2nd memory resource assigned to the RTT platform device.

Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Johan Hovold <johan@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agortc: at91sam9: add DT support
Boris BREZILLON [Tue, 23 Sep 2014 11:14:24 +0000 (13:14 +0200)]
rtc: at91sam9: add DT support

Add of_match_table to the existing driver so that rtt nodes defined in at91
DTs can be attached to this driver.

Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Johan Hovold <johan@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agortc: at91sam9: replace devm_ioremap by devm_ioremap_resource
Boris BREZILLON [Tue, 23 Sep 2014 11:14:09 +0000 (13:14 +0200)]
rtc: at91sam9: replace devm_ioremap by devm_ioremap_resource

Replace devm_ioremap calls by devm_ioremap_resource which already check
resource consistency (resource != NULL) and print an error in case of
failure.

Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Johan Hovold <johan@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agortc: at91sam9: use standard readl/writel functions instead of raw versions
Boris BREZILLON [Tue, 23 Sep 2014 11:13:52 +0000 (13:13 +0200)]
rtc: at91sam9: use standard readl/writel functions instead of raw versions

Raw versions of writel and writel should not be directly used and should
be replaced by their relaxed versions (readl/writel_relaxed), which take
endianness conversion into account.

In this driver we prefer the standard readl/writel function which add the
appropriate memory barrier around the access (the performance penalty is
negligible for this kind of application).

Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Johan Hovold <johan@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agortc: at91sam9: remove references to mach specific headers
Boris BREZILLON [Tue, 23 Sep 2014 11:13:29 +0000 (13:13 +0200)]
rtc: at91sam9: remove references to mach specific headers

In order to support multi platform kernel drivers should not include
machine specific headers.
Copy RTT macros in the driver code and remove any machine specific
headers.

Signed-off-by: Boris BREZILLON <boris.brezillon@free-electrons.com>
Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Johan Hovold <johan@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: move sdramc/ddrsdr header to include/soc/at91
Alexandre Belloni [Fri, 7 Nov 2014 20:58:21 +0000 (21:58 +0100)]
ARM: at91: move sdramc/ddrsdr header to include/soc/at91

Move the (DDR) SDRAM controller headers to include/soc/at91 to remove the
dependency on mach/ headers from the at91-reset driver.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: remove CONFIG_MACH_SAMA5_DT
Alexandre Belloni [Mon, 10 Nov 2014 20:45:53 +0000 (21:45 +0100)]
ARM: at91: remove CONFIG_MACH_SAMA5_DT

CONFIG_MACH_SAMA5_DT is useless as the only way to boot on sama5 based boards is
to use device tree.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com>
[nicolas.ferre@atmel.com: adapt on top of cleanup branch]
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: remove unused CONFIG_ARCH_AT91SAM9G45 option
Nicolas Ferre [Mon, 10 Nov 2014 11:14:28 +0000 (12:14 +0100)]
ARM: at91: remove unused CONFIG_ARCH_AT91SAM9G45 option

The CONFIG_ARCH_AT91SAM9G45 config option was removed by ("ARM: at91: remove
at91sam9g45/9m10 legacy board support") so cleanup the use of it.

Reported-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com>
9 years agobus: omap_l3_noc: Correct returning IRQ_HANDLED unconditionally in the irq handler
Keerthy [Mon, 10 Nov 2014 18:19:48 +0000 (23:49 +0530)]
bus: omap_l3_noc: Correct returning IRQ_HANDLED unconditionally in the irq handler

Correct returning IRQ_HANDLED unconditionally in the irq handler.
Return IRQ_NONE for some interrupt which we do not expect to be
handled in this handler. This prevents kernel stalling with back
to back spurious interrupts.

Fixes: 2722e56de6 ("OMAP4: l3: Introduce l3-interconnect error handling driver")
Acked-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Keerthy <j-keerthy@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
9 years agobus: omap_l3_noc: Add resume hook to restore context
Keerthy [Mon, 10 Nov 2014 18:19:47 +0000 (23:49 +0530)]
bus: omap_l3_noc: Add resume hook to restore context

On certain SoCs such as AM437x SoC, L3_noc error registers are
maintained in power domain such as per domain which looses context as part
of low power state such as RTC+DDR mode. On these platforms when we
mask interrupts which we cannot handle, the source of these interrupts
still remain on resume, however, the flag mux registers now contain
their reset value (unmasked) - this breaks the system with infinite
interrupts since we do not these interrupts to take place ever again.

To handle this: restore the masking of interrupts which we have
already recorded in the system as ones we cannot handle.

Fixes: 2100b595b7 ("bus: omap_l3_noc: ignore masked out unclearable targets")
Acked-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Keerthy <j-keerthy@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
9 years agow1: omap-hdq: support device probing with device-tree
NeilBrown [Fri, 7 Nov 2014 23:30:32 +0000 (10:30 +1100)]
w1: omap-hdq: support device probing with device-tree

This driver has no 'compatible' string and so is not found when
using device-tree.

Add one with value to match
hdqw1w: 1w@480b2000 {
device in omap3.dtsi.

Signed-off-by: NeilBrown <neilb@suse.de>
Acked-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Tony Lindgren <tony@atomide.com>
9 years agoARM: at91: remove useless init_time for DT-only SoCs
Boris Brezillon [Mon, 10 Nov 2014 09:32:12 +0000 (10:32 +0100)]
ARM: at91: remove useless init_time for DT-only SoCs

init_time is only needed when booting non-DT boards, we can thus safely
remove init_time functions.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: fix build breakage due to legacy board removals
Olof Johansson [Sun, 9 Nov 2014 06:39:51 +0000 (22:39 -0800)]
ARM: at91: fix build breakage due to legacy board removals

Fixes the following missing includes:

arch/arm/mach-at91/at91sam9g45.c: In function 'at91sam9g45_init_time':
arch/arm/mach-at91/at91sam9g45.c:39:23: error: 'NR_IRQS_LEGACY' undeclared (first use in this function)
  at91sam926x_pit_init(NR_IRQS_LEGACY + AT91_ID_SYS);
                       ^
arch/arm/mach-at91/at91sam9g45.c:39:23: note: each undeclared identifier is reported only once for each function it appears in
make[3]: *** [arch/arm/mach-at91/at91sam9g45.o] Error 1
arch/arm/mach-at91/at91sam9rl.c: In function 'at91sam9rl_init_time':
arch/arm/mach-at91/at91sam9rl.c:51:23: error: 'NR_IRQS_LEGACY' undeclared (first use in this function)
  at91sam926x_pit_init(NR_IRQS_LEGACY + AT91_ID_SYS);

Cc: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoMerge tag 'berlin-driver-3.19-1' of git://git.infradead.org/users/hesselba/linux...
Olof Johansson [Sun, 9 Nov 2014 00:41:37 +0000 (16:41 -0800)]
Merge tag 'berlin-driver-3.19-1' of git://git.infradead.org/users/hesselba/linux-berlin into next/drivers

Merge "ARM: berlin: Driver changes for v3.19 (round 1)" from Sebastian
Hesselbarth:

Berlin Driver changes for v3.19 (round 1)
- Reset driver, Acked by Philipp Zabel

* tag 'berlin-driver-3.19-1' of git://git.infradead.org/users/hesselba/linux-berlin:
  reset: add the Berlin reset controller driver

Signed-off-by: Olof Johansson <olof@lixom.net>
9 years agoiommu: Improve error handling when setting bus iommu
Heiko Stübner [Wed, 29 Oct 2014 00:22:56 +0000 (01:22 +0100)]
iommu: Improve error handling when setting bus iommu

When some part of bus_set_iommu fails it should undo any made changes
and not simply leave everything as is.

This includes unregistering the bus notifier in iommu_bus_init when
add_iommu_group fails and also setting the bus->iommu_ops back to NULL.

Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
9 years agoiommu: Do more input validation in iommu_map_sg()
Joerg Roedel [Tue, 4 Nov 2014 13:53:51 +0000 (14:53 +0100)]
iommu: Do more input validation in iommu_map_sg()

The IOMMU-API works on page boundarys, unlike the DMA-API
which can work with sub-page buffers. The sg->offset
field does not make sense on the IOMMU level, so force it to
be 0. Do some error-path consolidation while at it.

Signed-off-by: Joerg Roedel <jroedel@suse.de>
9 years agoiommu: Add iommu_map_sg() function
Olav Haugan [Sat, 25 Oct 2014 16:55:16 +0000 (09:55 -0700)]
iommu: Add iommu_map_sg() function

Mapping and unmapping are more often than not in the critical path.
map_sg allows IOMMU driver implementations to optimize the process
of mapping buffers into the IOMMU page tables.

Instead of mapping a buffer one page at a time and requiring potentially
expensive TLB operations for each page, this function allows the driver
to map all pages in one go and defer TLB maintenance until after all
pages have been mapped.

Additionally, the mapping operation would be faster in general since
clients does not have to keep calling map API over and over again for
each physically contiguous chunk of memory that needs to be mapped to a
virtually contiguous region.

Signed-off-by: Olav Haugan <ohaugan@codeaurora.org>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
9 years agoreset: add socfpga_reset_status
Dinh Nguyen [Mon, 3 Nov 2014 22:33:05 +0000 (16:33 -0600)]
reset: add socfpga_reset_status

Populate the reset_status callback for SOCFPGA.

Signed-off-by: Alan Tull <atull@opensource.altera.com>
Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com>
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
9 years agosoc: ti: knav_qmss_queue: Use list_for_each_entry_safe to prevent use after free
Axel Lin [Mon, 3 Nov 2014 18:21:54 +0000 (10:21 -0800)]
soc: ti: knav_qmss_queue: Use list_for_each_entry_safe to prevent use after free

list_for_each_entry_safe() is necessary if list objects are deleted from
the list while traversing it.

Signed-off-by: Axel Lin <axel.lin@ingics.com>
Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
9 years agoARM: at91: remove at91sam9rl legacy board support
Nicolas Ferre [Tue, 21 Oct 2014 13:09:21 +0000 (15:09 +0200)]
ARM: at91: remove at91sam9rl legacy board support

Remove legacy support for at91sam9rl boards.
This include board files removal plus all legacy code for non DT boards
support.
Use the Device Tree for running this board with newer kernels.

Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
[nicolas.ferre@atmel.com: Add sentence about DT]
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: remove at91sam9g45/9m10 legacy board support
Nicolas Ferre [Tue, 21 Oct 2014 12:55:46 +0000 (14:55 +0200)]
ARM: at91: remove at91sam9g45/9m10 legacy board support

Remove legacy support for at91sam9g45/9m10 boards.
This include board files removal plus all legacy code for non DT boards
support (i.e. at91sam9g45.c and at91sam9g45_devices.c).
Use the Device Tree for running this board with newer kernels.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
[nicolas.ferre@atmel.com: Add sentence about DT, removed defconfig]
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
9 years agoARM: at91: remove no-MMU at91x40 support
Nicolas Ferre [Tue, 21 Oct 2014 12:16:54 +0000 (14:16 +0200)]
ARM: at91: remove no-MMU at91x40 support

As there is currently no-one to take care of this old !MMU target and as its
support in recent kernels is a bit rotten, remove this at91x40 support and the
board file associated with it (at91eb01).
There are modern ARM !MMU in Mainline now so this target is not interesting for
building tests anymore. It would be better to start from these modern ARM !MMU
platforms to reintroduce at91x40 support if needed.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Greg Ungerer <gerg@uclinux.org>
Acked-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
9 years agoMerge tag 'sti-reset-for-v3.19-1' of git://git.stlinux.com/devel/kernel/linux-sti...
Philipp Zabel [Mon, 3 Nov 2014 08:50:27 +0000 (09:50 +0100)]
Merge tag 'sti-reset-for-v3.19-1' of git://git.stlinux.com/devel/kernel/linux-sti into reset/for_v3.19

STi reset controller updates for v3.19.

This tag adds support to STiH407 reset controllers:
  1. st,stih407-powerdown
  2. st,stih407-softreset
  3. st,stih407-picophyreset

9 years agoLinux 3.18-rc3 v3.18-rc3
Linus Torvalds [Sun, 2 Nov 2014 23:01:51 +0000 (15:01 -0800)]
Linux 3.18-rc3

9 years agoMerge tag 'for-linus-20141102' of git://git.infradead.org/linux-mtd
Linus Torvalds [Sun, 2 Nov 2014 22:45:52 +0000 (14:45 -0800)]
Merge tag 'for-linus-20141102' of git://git.infradead.org/linux-mtd

Pull MTD fixes from Brian Norris:
 "Three main MTD fixes for 3.18:

   - A regression from 3.16 which was noticed in 3.17.  With the
     restructuring of the m25p80.c driver and the SPI NOR library
     framework, we omitted proper listing of the SPI device IDs.  This
     means m25p80.c wouldn't auto-load (modprobe) properly when built as
     a module.  For now, we duplicate the device IDs into both modules.

   - The OMAP / ELM modules were depending on an implicit link ordering.
     Use deferred probing so that the new link order (in 3.18-rc) can
     still allow for successful probing.

   - Fix suspend/resume support for LH28F640BF NOR flash"

* tag 'for-linus-20141102' of git://git.infradead.org/linux-mtd:
  mtd: cfi_cmdset_0001.c: fix resume for LH28F640BF chips
  mtd: omap: fix mtd devices not showing up
  mtd: m25p80,spi-nor: Fix module aliases for m25p80
  mtd: spi-nor: make spi_nor_scan() take a chip type name, not spi_device_id
  mtd: m25p80: get rid of spi_get_device_id

9 years agoMerge tag 'scsi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb...
Linus Torvalds [Sun, 2 Nov 2014 22:39:35 +0000 (14:39 -0800)]
Merge tag 'scsi-for-linus' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "This is a set of six patches consisting of:
   - two MAINTAINER updates
   - two scsi-mq fixs for the old parallel interface (not every request
     is tagged and we need to set the right flags to populate the SPI
     tag message)
   - a fix for a memory leak in scatterlist traversal caused by a
     preallocation update in 3.17
   - an ipv6 fix for cxgbi"

[ The scatterlist fix also came in separately through the block layer tree ]

* tag 'scsi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  MAINTAINERS: ufs - remove self
  MAINTAINERS: change hpsa and cciss maintainer
  libcxgbi : support ipv6 address host_param
  scsi: set REQ_QUEUE for the blk-mq case
  Revert "block: all blk-mq requests are tagged"
  lib/scatterlist: fix memory leak with scsi-mq

9 years agoMerge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Linus Torvalds [Sun, 2 Nov 2014 22:27:30 +0000 (14:27 -0800)]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux

Pull drm fixes from Dave Airlie:
 "Nothing too astounding or major: radeon, i915, vmwgfx, armada and
  exynos.

  Biggest ones:
   - vmwgfx has one big locking regression fix
   - i915 has come displayport fixes
   - radeon has some stability and a memory alloc failure
   - armada and exynos have some vblank fixes"

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: (24 commits)
  drm/exynos: correct connector->dpms field before resuming
  drm/exynos: enable vblank after DPMS on
  drm/exynos: init kms poll at the end of initialization
  drm/exynos: propagate plane initialization errors
  drm/exynos: vidi: fix build warning
  drm/exynos: remove explicit encoder/connector de-initialization
  drm/exynos: init vblank with real number of crtcs
  drm/vmwgfx: Filter out modes those cannot be supported by the current VRAM size.
  drm/vmwgfx: Fix hash key computation
  drm/vmwgfx: fix lock breakage
  drm/i915/dp: only use training pattern 3 on platforms that support it
  drm/radeon: remove some buggy dead code
  drm/i915: Ignore VBT backlight check on Macbook 2, 1
  drm/radeon: remove invalid pci id
  drm/radeon: dpm fixes for asrock systems
  radeon: clean up coding style differences in radeon_get_bios()
  drm/radeon: Use drm_malloc_ab instead of kmalloc_array
  drm/radeon/dpm: disable ulv support on SI
  drm/i915: Fix GMBUSFREQ on vlv/chv
  drm/i915: Ignore long hpds on eDP ports
  ...

9 years agoMerge branch 'fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-arm
Linus Torvalds [Sun, 2 Nov 2014 20:56:20 +0000 (12:56 -0800)]
Merge branch 'fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-arm

Pull ARM fixes from Russell King:
 - add the new bpf syscall to ARM.
 - drop a redundant return statement in __iommu_alloc_remap()
 - fix a performance issue noticed by Thomas Petazzoni with
   kmap_atomic().
 - fix an issue with the L2 cache OF parsing code which caused it to
   incorrectly print warnings on each boot, and make the warning text
   more consistent with the rest of the code

* 'fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-arm:
  ARM: 8180/1: mm: implement no-highmem fast path in kmap_atomic_pfn()
  ARM: 8183/1: l2c: Improve l2c310_of_parse() error message
  ARM: 8181/1: Drop extra return statement
  ARM: 8182/1: l2c: Make l2x0_cache_size_of_parse() return 'int'
  ARM: enable bpf syscall

9 years agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Linus Torvalds [Sun, 2 Nov 2014 20:31:02 +0000 (12:31 -0800)]
Merge tag 'for-linus' of git://git./virt/kvm/kvm

Pull kvm fixes from Paolo Bonzini:
 "A small set of x86 fixes.  The most serious is an SRCU lockdep fix.

  A bit late - needed some time to test the SRCU fix, which only came in
  on Friday"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: vmx: defer load of APIC access page address during reset
  KVM: nVMX: Disable preemption while reading from shadow VMCS
  KVM: x86: Fix far-jump to non-canonical check
  KVM: emulator: fix execution close to the segment limit
  KVM: emulator: fix error code for __linearize

9 years agoMerge branch 'exynos-drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git...
Dave Airlie [Sun, 2 Nov 2014 19:23:17 +0000 (05:23 +1000)]
Merge branch 'exynos-drm-fixes' of git://git./linux/kernel/git/daeinki/drm-exynos into drm-fixes

This pull-request includes some bug fixes and code cleanups.
Especially, this fixes the bind failure issue occurred when it tries
to re-bind Exynos drm driver after unbound, and the modetest failure
issue incurred by not having a pair to vblank on and off requests.

* 'exynos-drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos:
  drm/exynos: correct connector->dpms field before resuming
  drm/exynos: enable vblank after DPMS on
  drm/exynos: init kms poll at the end of initialization
  drm/exynos: propagate plane initialization errors
  drm/exynos: vidi: fix build warning
  drm/exynos: remove explicit encoder/connector de-initialization
  drm/exynos: init vblank with real number of crtcs

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Linus Torvalds [Sun, 2 Nov 2014 18:28:43 +0000 (10:28 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/viro/vfs

Pull VFS fixes from Al Viro:
 "A bunch of assorted fixes, most of them followups to overlayfs merge"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  ovl: initialize ->is_cursor
  Return short read or 0 at end of a raw device, not EIO
  isofs: don't bother with ->d_op for normal case
  isofs_cmp(): we'll never see a dentry for . or ..
  overlayfs: fix lockdep misannotation
  ovl: fix check for cursor
  overlayfs: barriers for opening upper-layer directory
  rcu: Provide counterpart to rcu_dereference() for non-RCU situations
  staging: android: logger: Fix log corruption regression

9 years agoirda: stop calling sk_prot->disconnect() on connection failure
Linus Torvalds [Sun, 2 Nov 2014 18:20:26 +0000 (10:20 -0800)]
irda: stop calling sk_prot->disconnect() on connection failure

The sk_prot is irda's own set of protocol handlers, so irda should
statically know what that function is anyway, without using an indirect
pointer.  And as it happens, we know *exactly* what that pointer is
statically: it's NULL, because irda doesn't define a disconnect
operation.

So calling that function is doubly wrong, and will just cause an oops.

Reported-by: Martin Lang <mlg.hessigheim@gmail.com>
Cc: Samuel Ortiz <samuel@sortiz.org>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agodrm/exynos: correct connector->dpms field before resuming
Andrzej Hajda [Fri, 10 Oct 2014 12:31:56 +0000 (14:31 +0200)]
drm/exynos: correct connector->dpms field before resuming

During system suspend after connector switch off its dpms field
is set to connector previous dpms state. To properly resume dpms field
should be set to its actual state (off) before resuming to previous dpms state.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: enable vblank after DPMS on
Andrzej Hajda [Fri, 10 Oct 2014 12:31:55 +0000 (14:31 +0200)]
drm/exynos: enable vblank after DPMS on

Before DPMS off driver disables vblank.
It should be balanced by vblank enable after DPMS on.
The patch fixes issue with page_flip ioctl not being able
to acquire vblank counter introduced by patch:
drm: Always reject drm_vblank_get() after drm_vblank_off()

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: init kms poll at the end of initialization
Andrzej Hajda [Fri, 10 Oct 2014 12:31:54 +0000 (14:31 +0200)]
drm/exynos: init kms poll at the end of initialization

HPD events can be generated by components even if drm_dev is not fully
initialized, to skip such events kms poll initialization should
be performed at the end of load callback followed directly by forced
connection detection.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: propagate plane initialization errors
Andrzej Hajda [Fri, 10 Oct 2014 12:31:53 +0000 (14:31 +0200)]
drm/exynos: propagate plane initialization errors

In case of error during plane initialization load callback
incorrectly return success, this patch fixes it.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: vidi: fix build warning
Inki Dae [Tue, 7 Oct 2014 15:16:34 +0000 (00:16 +0900)]
drm/exynos: vidi: fix build warning

encoder object isn't used anymore so remove it.

Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: remove explicit encoder/connector de-initialization
Andrzej Hajda [Mon, 22 Sep 2014 09:30:48 +0000 (11:30 +0200)]
drm/exynos: remove explicit encoder/connector de-initialization

All KMS objects are destroyed by drm_mode_config_cleanup in proper order
so component drivers should not care about it.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: init vblank with real number of crtcs
Andrzej Hajda [Tue, 7 Oct 2014 13:09:14 +0000 (22:09 +0900)]
drm/exynos: init vblank with real number of crtcs

Initialization of vblank with MAX_CRTC caused attempts
to disabling vblanks for non-existing crtcs in case
drm used fewer crtcs. The patch fixes it.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agoKVM: vmx: defer load of APIC access page address during reset
Paolo Bonzini [Sun, 2 Nov 2014 06:54:30 +0000 (07:54 +0100)]
KVM: vmx: defer load of APIC access page address during reset

Most call paths to vmx_vcpu_reset do not hold the SRCU lock.  Defer loading
the APIC access page to the next vmentry.

This avoids the following lockdep splat:

[ INFO: suspicious RCU usage. ]
3.18.0-rc2-test2+ #70 Not tainted
-------------------------------
include/linux/kvm_host.h:474 suspicious rcu_dereference_check() usage!

other info that might help us debug this:

rcu_scheduler_active = 1, debug_locks = 0
1 lock held by qemu-system-x86/2371:
 #0:  (&vcpu->mutex){+.+...}, at: [<ffffffffa037d800>] vcpu_load+0x20/0xd0 [kvm]

stack backtrace:
CPU: 4 PID: 2371 Comm: qemu-system-x86 Not tainted 3.18.0-rc2-test2+ #70
Hardware name: Dell Inc. OptiPlex 9010/0M9KCM, BIOS A12 01/10/2013
 0000000000000001 ffff880209983ca8 ffffffff816f514f 0000000000000000
 ffff8802099b8990 ffff880209983cd8 ffffffff810bd687 00000000000fee00
 ffff880208a2c000 ffff880208a10000 ffff88020ef50040 ffff880209983d08
Call Trace:
 [<ffffffff816f514f>] dump_stack+0x4e/0x71
 [<ffffffff810bd687>] lockdep_rcu_suspicious+0xe7/0x120
 [<ffffffffa037d055>] gfn_to_memslot+0xd5/0xe0 [kvm]
 [<ffffffffa03807d3>] __gfn_to_pfn+0x33/0x60 [kvm]
 [<ffffffffa0380885>] gfn_to_page+0x25/0x90 [kvm]
 [<ffffffffa038aeec>] kvm_vcpu_reload_apic_access_page+0x3c/0x80 [kvm]
 [<ffffffffa08f0a9c>] vmx_vcpu_reset+0x20c/0x460 [kvm_intel]
 [<ffffffffa039ab8e>] kvm_vcpu_reset+0x15e/0x1b0 [kvm]
 [<ffffffffa039ac0c>] kvm_arch_vcpu_setup+0x2c/0x50 [kvm]
 [<ffffffffa037f7e0>] kvm_vm_ioctl+0x1d0/0x780 [kvm]
 [<ffffffff810bc664>] ? __lock_is_held+0x54/0x80
 [<ffffffff812231f0>] do_vfs_ioctl+0x300/0x520
 [<ffffffff8122ee45>] ? __fget+0x5/0x250
 [<ffffffff8122f0fa>] ? __fget_light+0x2a/0xe0
 [<ffffffff81223491>] SyS_ioctl+0x81/0xa0
 [<ffffffff816fed6d>] system_call_fastpath+0x16/0x1b

Reported-by: Takashi Iwai <tiwai@suse.de>
Reported-by: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Reviewed-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Tested-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Fixes: 38b9917350cb2946e368ba684cfc33d1672f104e
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoKVM: nVMX: Disable preemption while reading from shadow VMCS
Jan Kiszka [Wed, 8 Oct 2014 16:05:39 +0000 (18:05 +0200)]
KVM: nVMX: Disable preemption while reading from shadow VMCS

In order to access the shadow VMCS, we need to load it. At this point,
vmx->loaded_vmcs->vmcs and the actually loaded one start to differ. If
we now get preempted by Linux, vmx_vcpu_put and, on return, the
vmx_vcpu_load will work against the wrong vmcs. That can cause
copy_shadow_to_vmcs12 to corrupt the vmcs12 state.

Fix the issue by disabling preemption during the copy operation.
copy_vmcs12_to_shadow is safe from this issue as it is executed by
vmx_vcpu_run when preemption is already disabled before vmentry.

This bug is exposed by running Jailhouse within KVM on CPUs with
shadow VMCS support.  Jailhouse never expects an interrupt pending
vmexit, but the bug can cause it if, after copy_shadow_to_vmcs12
is preempted, the active VMCS happens to have the virtual interrupt
pending flag set in the CPU-based execution controls.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>