deliverable/linux.git
11 years agoMerge tag 'rpi-for-3.7' of git://git.kernel.org/pub/scm/linux/kernel/git/swarren...
Olof Johansson [Fri, 21 Sep 2012 00:10:51 +0000 (17:10 -0700)] 
Merge tag 'rpi-for-3.7' of git://git./linux/kernel/git/swarren/linux-rpi into next/soc

ARM: add basic BCM2835 SoC and Raspberry Pi board support

The BCM2835 is an ARM SoC from Broadcom. This patch adds very basic
support for this SoC; enough to boot the system into an initrd with
UART console, interrupt controller, timers, and a stub clock driver.

Also provided is a similarly basic device tree for the Raspberry Pi
Model B board.

This series was written by Simon Arlott, Chris Boot, and Dom Cobley
downstream, with reference to a Broadcom tree, and modified for upstream
and submitted by Stephen Warren.

* tag 'rpi-for-3.7' of git://git.kernel.org/pub/scm/linux/kernel/git/swarren/linux-rpi:
  MAINTAINERS: add an entry for the BCM2835 ARM sub-architecture
  ARM: bcm2835: instantiate console UART
  ARM: bcm2835: add stub clock driver
  ARM: bcm2835: add system timer
  ARM: bcm2835: add interrupt controller driver
  ARM: add infra-structure for BCM2835 and Raspberry Pi

11 years agoARM: OMAP: AM33xx hwmod: fixup SPI after platform_data move
Kevin Hilman [Thu, 20 Sep 2012 16:38:14 +0000 (09:38 -0700)] 
ARM: OMAP: AM33xx hwmod: fixup SPI after platform_data move

AM33xx hwmod data includes "mcspi.h" which has now been moved after
the platform_data move.  Fix it.

Signed-off-by: Kevin Hilman <khilman@ti.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
11 years agoMAINTAINERS: add an entry for the BCM2835 ARM sub-architecture
Stephen Warren [Sat, 15 Sep 2012 06:18:54 +0000 (00:18 -0600)] 
MAINTAINERS: add an entry for the BCM2835 ARM sub-architecture

Add myself as the maintainer for the BCM2835 ARM support, and related
drivers. This is mainly so that the MAINTAINERS file contains some
relevant entry, and the rpi/ARM mailing lists; I'd be quite happy if
anyone else came along and wanted to maintain/co-maintain this.

Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
11 years agoARM: bcm2835: instantiate console UART
Simon Arlott [Tue, 11 Sep 2012 05:29:17 +0000 (23:29 -0600)] 
ARM: bcm2835: instantiate console UART

This patch was extracted from git://github.com/lp0/linux.git branch
rpi-split as of 2012/09/08, and modified as follows:

* s/bcm2708/bcm2835/.
* Modified device tree vendor prefix.
* Modified UART DT node to use a unit-address to create unique UART node
  names, rather than using non-type names "uart0" and "uart1".
  Note that UART 1 (the Broadcom "mini UART") is not yet present, but
  I'm naming the DT node in anticipation that it will be added.

Signed-off-by: Chris Boot <bootc@bootc.net>
Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Dom Cobley <popcornmix@gmail.com>
Signed-off-by: Dom Cobley <dc4@broadcom.com>
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
11 years agoARM: bcm2835: add stub clock driver
Simon Arlott [Tue, 11 Sep 2012 05:26:15 +0000 (23:26 -0600)] 
ARM: bcm2835: add stub clock driver

This patch adds a minimal stub clock driver for the BCM2835. Its sole
purpose is to allow the PL011 AMBA clk_get() API calls to provide
something that looks enough like a clock that the driver probes and
operates correctly.

This patch was extracted from git://github.com/lp0/linux.git branch
rpi-split as of 2012/09/08, and modified as follows:

* Reworked to call clk_register_fixed_rate(), and clk_register_clkdev()
  rather than using static data to represent the clocks.
* Moved implementation to drivers/clk/.
* Modified .dev_id for UART clocks to match UART DT node names.
* s/bcm2708/bcm2835/.
* Modified device tree vendor prefix.

Signed-off-by: Chris Boot <bootc@bootc.net>
Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Dom Cobley <popcornmix@gmail.com>
Signed-off-by: Dom Cobley <dc4@broadcom.com>
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Acked-by: Mike Turquette <mturquette@linaro.org>
11 years agoARM: bcm2835: add system timer
Simon Arlott [Tue, 11 Sep 2012 04:38:35 +0000 (22:38 -0600)] 
ARM: bcm2835: add system timer

The System Timer peripheral provides four 32-bit timer channels and a
single 64-bit free running counter. Each channel has an output compare
register, which is compared against the 32 least significant bits of the
free running counter values, and generates an interrupt.

Timer 3 is used as the Linux timer.

The BCM2835 also contains an SP804-based timer module. However, it
apparently has significant differences from the standard SP804 IP block,
and Broadcom's documentation recommends using the system timer instead.

This patch was extracted from git://github.com/lp0/linux.git branch
rpi-split as of 2012/09/08, and modified as follows:

* s/bcm2708/bcm2835/.
* Modified device tree vendor prefix.
* Moved to drivers/clocksource/. This looks like the desired location for
  such code now.
* Added DT binding docs.
* Moved struct sys_timer bcm2835_timer into time.c to encapsulate it more.
* Simplified bcm2835_time_init() to find one matching node and operate on
  it, rather than looping over all matching nodes. This seems more
  consistent with other clocksource code.
* Simplified bcm2835_time_init() using of_iomap().
* Renamed struct bcm2835_timer.index to match_mask to better represent its
  purpose.
* s/printk(PR_INFO/pr_info(/

Signed-off-by: Chris Boot <bootc@bootc.net>
Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Dom Cobley <popcornmix@gmail.com>
Signed-off-by: Dom Cobley <dc4@broadcom.com>
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
11 years agoARM: bcm2835: add interrupt controller driver
Simon Arlott [Thu, 13 Sep 2012 01:57:26 +0000 (19:57 -0600)] 
ARM: bcm2835: add interrupt controller driver

The BCM2835 contains a custom interrupt controller, which supports 72
interrupt sources using a 2-level register scheme. The interrupt
controller, or the HW block containing it, is referred to occasionally
as "armctrl" in the SoC documentation, hence the symbol naming in the
code.

This patch was extracted from git://github.com/lp0/linux.git branch
rpi-split as of 2012/09/08, and modified as follows:

* s/bcm2708/bcm2835/.
* Modified device tree vendor prefix.
* Moved implementation to drivers/irchip/.
* Added devicetree documentation, and hence removed list of IRQs from
  bcm2835.dtsi.
* Changed shift in MAKE_HWIRQ() and HWIRQ_BANK() from 8 to 5 to reduce
  the size of the hwirq space, and pass the total size of the hwirq space
  to irq_domain_add_linear(), rather than just the number of valid hwirqs;
  the two are different due to the hwirq space being sparse.
* Added the interrupt controller DT node to the top-level of the DT,
  rather than nesting it inside a /axi node. Hence, changed the reg value
  since /axi had a ranges property. This seems simpler to me, but I'm not
  sure if everyone will like this change or not.
* Don't set struct irq_domain_ops.map = irq_domain_simple_map, hence
  removing the need to patch include/linux/irqdomain.h or
  kernel/irq/irqdomain.c.
* Simplified armctrl_of_init() using of_iomap().
* Removed unused IS_VALID_BANK()/IS_VALID_IRQ() macros.
* Renamed armctrl_handle_irq() to prevent possible symbol clashes.
* Made armctrl_of_init() static.
* Removed comment "Each bank is registered as a separate interrupt
  controller" since this is no longer true.
* Removed FSF address from license header.
* Added my name to copyright header.

Signed-off-by: Chris Boot <bootc@bootc.net>
Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Dom Cobley <popcornmix@gmail.com>
Signed-off-by: Dom Cobley <dc4@broadcom.com>
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
11 years agoARM: add infra-structure for BCM2835 and Raspberry Pi
Simon Arlott [Sat, 26 May 2012 07:04:43 +0000 (01:04 -0600)] 
ARM: add infra-structure for BCM2835 and Raspberry Pi

The BCM2835 is an ARM SoC from Broadcom. This patch adds very basic
support for this SoC.

http://www.broadcom.com/products/BCM2835
http://www.raspberrypi.org/wp-content/uploads/2012/02/BCM2835-ARM-Peripherals.pdf

Note that the documentation in the latter .pdf assumes the MMU setup
that's used on the "VideoCore" companion processor, and does not document
physical peripheral addresses. Subtract 0x5e000000 to obtain the physical
addresses. This is accounted for by the ranges property in the /soc node
in the device tree.

The BCM2835 SoC is used in the Raspberry Pi. This patch also adds a
minimal device tree for this board; enough to see some very early kernel
boot messages through earlyprintk. However, this patch does not yet
provide a useful booting system.

http://www.raspberrypi.org/.

This patch was extracted from git://github.com/lp0/linux.git branch
rpi-split from 3-4 months ago, and significantly stripped down and
modified since.

Signed-off-by: Chris Boot <bootc@bootc.net>
Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Dom Cobley <popcornmix@gmail.com>
Signed-off-by: Dom Cobley <dc4@broadcom.com>
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
11 years agoMerge tag 'omap-devel-am33xx-for-v3.7' of git://git.kernel.org/pub/scm/linux/kernel...
Olof Johansson [Mon, 17 Sep 2012 03:12:26 +0000 (20:12 -0700)] 
Merge tag 'omap-devel-am33xx-for-v3.7' of git://git./linux/kernel/git/tmlind/linux-omap into next/soc

From Tony Lindgren:
From Paul Walmsley <paul@pwsan.com>:

AM33xx hwmod data and miscellaneous clock and hwmod fixes.  AM33xx
should now boot on mainline after this is applied, according to
Vaibhav.

(The shortlog makes no sense here since it contains mostly the dependent
cleanups that are part of the preceding branches).

Signed-off-by: Olof Johansson <olof@lixom.net>
11 years agoMerge tag 'devel-omap-device-for-v3.7' of git://git.kernel.org/pub/scm/linux/kernel...
Olof Johansson [Mon, 17 Sep 2012 02:24:53 +0000 (19:24 -0700)] 
Merge tag 'devel-omap-device-for-v3.7' of git://git./linux/kernel/git/tmlind/linux-omap into next/soc

Kevin Hilman <khilman@ti.com>:

Updates for omap_device layer for v3.7.

Allows omap_device layer to keep track of driver bound status in order
to make more intelligent decisions about idling unused devices.

* tag 'devel-omap-device-for-v3.7' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap:
  ARM: OMAP: omap_device: idle devices with no driver bound
  ARM: OMAP: omap_device: don't attempt late suspend if no driver bound
  ARM: OMAP: omap_device: keep track of driver bound status
  + sync to 3.6-rc5

11 years agoMerge tag 'tegra-for-3.7-cpu-hotplug' of git://git.kernel.org/pub/scm/linux/kernel...
Olof Johansson [Mon, 17 Sep 2012 01:33:23 +0000 (18:33 -0700)] 
Merge tag 'tegra-for-3.7-cpu-hotplug' of git://git./linux/kernel/git/swarren/linux-tegra into next/soc

From Stephen Warren:

ARM: tegra: implement CPU hotplug

This branch implements CPU hot-plugging support for both Tegra20 and
Tegra30. Portions of the implementation are contained in the clock
driver, hence this branch is based on the common clock conversion in
order to avoid duplicating work.

By Joseph Lo
via Stephen Warren
* tag 'tegra-for-3.7-cpu-hotplug' of git://git.kernel.org/pub/scm/linux/kernel/git/swarren/linux-tegra:
  ARM: tegra20: add CPU hotplug support
  ARM: tegra30: add CPU hotplug support
  ARM: tegra: clean up the common assembly macros into sleep.h
  ARM: tegra: replace the CPU CAR access code by tegra_cpu_car_ops
  ARM: tegra: introduce tegra_cpu_car_ops structures

11 years agoMerge tag 'tegra-for-3.7-common-clk' of git://git.kernel.org/pub/scm/linux/kernel...
Olof Johansson [Mon, 17 Sep 2012 01:25:15 +0000 (18:25 -0700)] 
Merge tag 'tegra-for-3.7-common-clk' of git://git./linux/kernel/git/swarren/linux-tegra into next/soc

From Stephen Warren:

ARM: tegra: switch to the common clock framework

This branch contains a few bug-fixes, followed by a conversion of Tegra's
clock driver to the common clock framework, followed by various bug fixes
found after the conversion.

* tag 'tegra-for-3.7-common-clk' of git://git.kernel.org/pub/scm/linux/kernel/git/swarren/linux-tegra:
  ARM: Tegra: Add smp_twd clock for Tegra20
  ARM: tegra: cpu-tegra: explicitly manage re-parenting
  ARM: tegra: fix overflow in tegra20_pll_clk_round_rate()
  ARM: tegra: Fix data type for io address
  ARM: tegra: remove tegra_timer from tegra_list_clks
  ARM: tegra30: clocks: fix the wrong tegra_audio_sync_clk_ops name
  ARM: tegra: clocks: separate tegra_clk_32k_ops from Tegra20 and Tegra30
  ARM: tegra: Remove duplicate code
  ARM: tegra: Port tegra to generic clock framework
  ARM: tegra: Add clk_tegra structure and helper functions
  ARM: tegra: Rename tegra20 clock file
  ARM: tegra20: Separate out clk ops and clk data
  ARM: tegra30: Separate out clk ops and clk data
  ARM: tegra: fix U16 divider range check
  ARM: tegra: turn on UART A clock at boot

11 years agoMerge tag 'imx-soc' of git://git.pengutronix.de/git/imx/linux-2.6 into next/soc
Olof Johansson [Mon, 17 Sep 2012 00:15:29 +0000 (17:15 -0700)] 
Merge tag 'imx-soc' of git://git.pengutronix.de/git/imx/linux-2.6 into next/soc

ARM i.MX SoC updates

* tag 'imx-soc' of git://git.pengutronix.de/git/imx/linux-2.6:
  ARM: i.MX35: Implement camera and keypad clocks
  ARM: mxc: ssi-fiq: Make ssi-fiq.S Thumb-2 compatible
  ARM i.MX53: register CAN clocks
  arm imx31: add a few pinmux settings the tt01 needs

11 years agoARM: tegra20: add CPU hotplug support
Joseph Lo [Thu, 16 Aug 2012 09:31:52 +0000 (17:31 +0800)] 
ARM: tegra20: add CPU hotplug support

Hotplug function put CPU in offline or online mode at runtime.
When the CPU been put into offline, it was been clock gated. The
offline CPU can be power gated, when the remaining CPU goes into
LP2.

Based on the worked by:
Colin Cross <ccross@android.com>
Gary King <gking@nvidia.com>

Signed-off-by: Joseph Lo <josephl@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra30: add CPU hotplug support
Joseph Lo [Thu, 16 Aug 2012 09:31:51 +0000 (17:31 +0800)] 
ARM: tegra30: add CPU hotplug support

Hotplug function put CPUs in offline or online state at runtime.
When the CPU been put in the offline state, it was been clock and
power gated. Except primary CPU other CPUs can be hotplugged.

Based on the work by:
Scott Williams <scwilliams@nvidia.com>
Colin Cross <ccross@android.com>
Gary King <gking@nvidia.com>

Signed-off-by: Joseph Lo <josephl@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra: clean up the common assembly macros into sleep.h
Joseph Lo [Thu, 16 Aug 2012 09:31:50 +0000 (17:31 +0800)] 
ARM: tegra: clean up the common assembly macros into sleep.h

There are some common macros for Tegra low-level assembly code. Clean
up them into one header file and move the definitions that will be
re-used into it as well.

Signed-off-by: Joseph Lo <josephl@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra: replace the CPU CAR access code by tegra_cpu_car_ops
Joseph Lo [Thu, 16 Aug 2012 09:31:49 +0000 (17:31 +0800)] 
ARM: tegra: replace the CPU CAR access code by tegra_cpu_car_ops

Replacing the code that directly access to CAR registers with
tegra_cpu_car_ops. This ops hides CPU CAR access inside and
provides control interface for it.

Signed-off-by: Joseph Lo <josephl@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra: introduce tegra_cpu_car_ops structures
Joseph Lo [Thu, 16 Aug 2012 09:31:48 +0000 (17:31 +0800)] 
ARM: tegra: introduce tegra_cpu_car_ops structures

The tegra_cpu_car_ops provide the interface for CPU to control
it's clock gating and reset status. The other drivers should use
this for CPU control. And should not directly access CAR registers
to control CPU.

Signed-off-by: Joseph Lo <josephl@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: Tegra: Add smp_twd clock for Tegra20
Prashant Gaikwad [Thu, 13 Sep 2012 09:34:33 +0000 (15:04 +0530)] 
ARM: Tegra: Add smp_twd clock for Tegra20

Clockevent's frequency is changed upon cpufreq change
notification. It fetches local timer's rate to update the
clockevent frequency. This patch adds local timer clock
for Tegra20.

Signed-off-by: Prashant Gaikwad <pgaikwad@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoMerge tag 'omap-devel-a2-for-3.7' of git://git.kernel.org/pub/scm/linux/kernel/git...
Tony Lindgren [Thu, 13 Sep 2012 04:29:07 +0000 (21:29 -0700)] 
Merge tag 'omap-devel-a2-for-3.7' of git://git./linux/kernel/git/pjw/omap-pending into devel-am33xx

AM33xx hwmod data and miscellaneous clock and hwmod fixes.  AM33xx
should now boot on mainline after this is applied, according to
Vaibhav.

This second version includes trailing commas at the end of structure
records at Tony's request.  It also adds a OMAP_INTC_START macro
expansion to each IRQ number to make the sparseirq conversion easier.

Basic build, boot, and PM test transcripts are here:

http://www.pwsan.com/omap/testlogs/am33xx_hwmod_clock_devel_3.7/20120912165952/

11 years agoMerge tag 'for_3.7-omap_device' of git://git.kernel.org/pub/scm/linux/kernel/git...
Tony Lindgren [Thu, 13 Sep 2012 04:17:56 +0000 (21:17 -0700)] 
Merge tag 'for_3.7-omap_device' of git://git./linux/kernel/git/khilman/linux-omap-pm into devel-omap-device

Updates for omap_device layer for v3.7.

Allows omap_device layer to keep track of driver bound status in order
to make more intelligent decisions about idling unused devices.

11 years agoARM: OMAP1: Move SoC specific headers from plat to mach for omap1
Tony Lindgren [Sat, 1 Sep 2012 00:04:35 +0000 (17:04 -0700)] 
ARM: OMAP1: Move SoC specific headers from plat to mach for omap1

There's no need to have these in plat-omap any longer. Note that these
could eventually be made local to mach-omap1 instead of being in mach.

But to do that, at least various driver access using omap7xxx.h registers
needs to be fixed first.

Cc: spi-devel-general@lists.sourceforge.net
Cc: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP2+ Move SoC specific headers to be local to mach-omap2
Tony Lindgren [Fri, 31 Aug 2012 23:08:07 +0000 (16:08 -0700)] 
ARM: OMAP2+ Move SoC specific headers to be local to mach-omap2

These can now be moved to be local headers in mach-omap2.

Note that this patch removes arch/arm/plat-omap/devices.c as it
will get removed anyways with Paul Walmsley's patch
"ARM: OMAP: split OMAP1, OMAP2+ RNG device registration".

Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP: Split plat/hardware.h, use local soc.h for omap2+
Tony Lindgren [Fri, 31 Aug 2012 17:59:07 +0000 (10:59 -0700)] 
ARM: OMAP: Split plat/hardware.h, use local soc.h for omap2+

As the plat and mach includes need to disappear for single zImage work,
we need to remove plat/hardware.h.

Do this by splitting plat/hardware.h into omap1 and omap2+ specific files.

The old plat/hardware.h already has omap1 only defines, so it gets moved
to mach/hardware.h for omap1. For omap2+, we use the local soc.h
that for now just includes the related SoC headers to keep this patch more
readable.

Note that the local soc.h still includes plat/cpu.h that can be dealt
with in later patches. Let's also include plat/serial.h from common.h for
all the board-*.c files. This allows making the include files local later
on without patching these files again.

Note that only minimal changes are done in this patch for the
drivers/watchdog/omap_wdt.c driver to keep things compiling. Further
patches are needed to eventually remove cpu_is_omap usage in the drivers.

Also only minimal changes are done to sound/soc/omap/* to remove the
unneeded includes and to define OMAP44XX_MCPDM_L3_BASE locally so there's
no need to include omap44xx.h.

While at it, also sort some of the includes in the standard way.

Cc: linux-watchdog@vger.kernel.org
Cc: alsa-devel@alsa-project.org
Cc: Peter Ujfalusi <peter.ujfalusi@ti.com>
Cc: Jarkko Nikula <jarkko.nikula@bitmer.com>
Cc: Liam Girdwood <lrg@ti.com>
Acked-by: Wim Van Sebroeck <wim@iguana.be>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP: Remove unused old gpio-switch.h
Tony Lindgren [Fri, 31 Aug 2012 17:15:41 +0000 (10:15 -0700)] 
ARM: OMAP: Remove unused old gpio-switch.h

This is no longer used anywhere.

Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP1: Move plat/irqs.h to mach/irqs.h
Tony Lindgren [Thu, 30 Aug 2012 23:57:21 +0000 (16:57 -0700)] 
ARM: OMAP1: Move plat/irqs.h to mach/irqs.h

This is now omap1 specific files.

Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP2+: Remove hardcoded IRQs and enable SPARSE_IRQ
Tony Lindgren [Tue, 28 Aug 2012 00:43:01 +0000 (17:43 -0700)] 
ARM: OMAP2+: Remove hardcoded IRQs and enable SPARSE_IRQ

Remove hardcoded IRQs in irqs.h and related files as these
are no longer needed.

Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP2+: Prepare for irqs.h removal
Tony Lindgren [Tue, 28 Aug 2012 00:43:01 +0000 (17:43 -0700)] 
ARM: OMAP2+: Prepare for irqs.h removal

As the interrupts should only be defined in the platform_data, and
eventually coming from device tree, there's no need to define them
in header files.

Let's remove the hardcoded references to irqs.h and fix up the includes
so we don't rely on headers included in irqs.h. Note that we're
defining OMAP_INTC_START as 0 to the interrupts. This will be needed
when we enable SPARSE_IRQ. For some drivers we need to add
#include <plat/cpu.h> for now until these drivers are fixed to
remove cpu_is_omapxxxx() usage.

While at it, sort som of the includes the standard way, and add
the trailing commas where they are missing in the related data
structures.

Note that for drivers/staging/tidspbridge we just define things
locally.

Cc: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoW1: OMAP HDQ1W: Remove dependencies to mach/hardware.h
Tony Lindgren [Fri, 7 Sep 2012 19:53:15 +0000 (12:53 -0700)] 
W1: OMAP HDQ1W: Remove dependencies to mach/hardware.h

No need for hardcoded IRQ here.

We can't include mach headers for ARM common zImage support.

Cc: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoInput: omap-keypad: Remove dependencies to mach includes
Tony Lindgren [Fri, 7 Sep 2012 20:27:58 +0000 (13:27 -0700)] 
Input: omap-keypad: Remove dependencies to mach includes

Remove support for omap2+ as it's no longer needed since
it's using matrix-keypad. This way we can remove depency
to plat and mach headers which is needed for ARM common
zImage support.

Also remove INT_KEYBOARD by using omap_kp->irq.

Note that this patch depends on an earlier patch
"ARM: OMAP: Move gpio.h to include/linux/platform_data".

Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Cc: linux-input@vger.kernel.org
Acked-by: Sourav Poddar <sourav.poddar@ti.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP: Move gpio.h to include/linux/platform_data
Tony Lindgren [Thu, 30 Aug 2012 22:37:24 +0000 (15:37 -0700)] 
ARM: OMAP: Move gpio.h to include/linux/platform_data

This way we can remove includes of plat/gpio.h which won't work
with the single zImage support.

Note that we also remove the cpu_class_is_omap2() check
in gpio-omap.c as the drivers should not call it as we need to
make it local to arch/arm/mach-omap2 for single zImage support.

While at it, arrange the related includes in the standard way.

Cc: Grant Likely <grant.likely@secretlab.ca>
Cc: linux-mtd@lists.infradead.org
Cc: alsa-devel@alsa-project.org
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP2+: Remove hardcoded twl4030 gpio_base, irq_base and irq_end
Tony Lindgren [Wed, 5 Sep 2012 00:43:30 +0000 (17:43 -0700)] 
ARM: OMAP2+: Remove hardcoded twl4030 gpio_base, irq_base and irq_end

We can't use hardcoded interrupts for SPARSE_IRQ, and can replace
the hardcoded gpio_base with twl_gpiochip.base after it's been
allocated.

Cc: Grant Likely <grant.likely@secretlab.ca>
Cc: Samuel Ortiz <sameo@linux.intel.com>
Cc: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP2+: Remove unused nand_irq for GPMC
Tony Lindgren [Wed, 5 Sep 2012 00:43:30 +0000 (17:43 -0700)] 
ARM: OMAP2+: Remove unused nand_irq for GPMC

This is no longer needed and assumes a fixed IRQ number
that won't work with SPARSE_IRQ.

Acked-by: Afzal Mohammed <afzal@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP2+: Make INTCPS_NR_IRQS local for mach-omap2/irq.c
Tony Lindgren [Wed, 5 Sep 2012 00:43:29 +0000 (17:43 -0700)] 
ARM: OMAP2+: Make INTCPS_NR_IRQS local for mach-omap2/irq.c

Make INTCPS_NR_IRQS local for mach-omap2/irq.c

Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP1: Define OMAP1_INT_I2C locally
Tony Lindgren [Wed, 5 Sep 2012 00:43:29 +0000 (17:43 -0700)] 
ARM: OMAP1: Define OMAP1_INT_I2C locally

This is needed to start removing hardcoded IRQs for omap2+.

Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP1: Move define of OMAP_LCD_DMA to dma.h
Tony Lindgren [Wed, 5 Sep 2012 00:43:29 +0000 (17:43 -0700)] 
ARM: OMAP1: Move define of OMAP_LCD_DMA to dma.h

This is needed to start removing hardcoded IRQs on omap2+.

Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoMerge tags 'omap-devel-gpmc-fixed-for-v3.7' and 'cleanup-omap-tags-for-v3.7' into...
Tony Lindgren [Thu, 13 Sep 2012 01:05:19 +0000 (18:05 -0700)] 
Merge tags 'omap-devel-gpmc-fixed-for-v3.7' and 'cleanup-omap-tags-for-v3.7' into cleanup-sparseirq

Changes for GPMC (General Purpose Memory Controller) that take it
closer for being just a regular device driver.

Remove the ancient omap specific atags that are no longer needed.

At some point we were planning to pass the bootloader information
with custom atags that did not work out too well.

There's no need for these any longer as the kernel has been booting
fine without them for quite some time. And Now we have device tree
support that can be used instead.

11 years agoARM: AM33XX: clock: Add dcan clock aliases for device-tree
Vaibhav Hiremath [Mon, 27 Aug 2012 10:04:02 +0000 (15:34 +0530)] 
ARM: AM33XX: clock: Add dcan clock aliases for device-tree

Currently, the device names for the dcan module follows the
format "dcan.X", where 'X' is the dcan instance number.
On other side, driver may request for clock with/without con_id
and dev_id, and it is expected that platform should respect this
request and return the requested clock handle.

Now, when using device tree, the format of the device name created
by OF layer is different, "<reg-address>.<device-name>",
assuming that the device-tree "reg" property is specified.
This causes the look-up failure for clock node in dcan driver

To fix this add new dcan clock alias for using device-tree.

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
Cc: Tony Lindgren <tony@atomide.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Benoit Cousson <b-cousson@ti.com>
Cc: Grant Likely <grant.likely@secretlab.ca>
Cc: Rob Herring <robherring2@gmail.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
11 years agoARM: OMAP2+: dpll: Add missing soc_is_am33xx() check for common functions
Vaibhav Hiremath [Fri, 24 Aug 2012 14:54:24 +0000 (20:24 +0530)] 
ARM: OMAP2+: dpll: Add missing soc_is_am33xx() check for common functions

Add missing soc_is_am33xx() check for DPLL common control & clock
related functions, without this dpll programmability would be broken
for am33xx family of devices.

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
11 years agoARM: OMAP: omap_device: idle devices with no driver bound
Kevin Hilman [Tue, 10 Jul 2012 22:06:11 +0000 (15:06 -0700)] 
ARM: OMAP: omap_device: idle devices with no driver bound

Under some circumstances, drivers may leave an omap_device enabled due
to driver programming errors, or due to a failure in the drivers
probe method.

Using the recently added omap_device driver_status field, we can
detect conditions where an omap_device is enabled but has no driver
bound and then ensure that the device is properly idled until it can
be probed again.

The goal of this feature is not only to detect and warn on these error
conditions, but also to ensure that devices are properly put in
low-power states so they do not prevent SoC-wide low-power states.

Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>
11 years agoARM: OMAP: omap_device: don't attempt late suspend if no driver bound
Kevin Hilman [Tue, 10 Jul 2012 22:29:04 +0000 (15:29 -0700)] 
ARM: OMAP: omap_device: don't attempt late suspend if no driver bound

Currently, the omap_device PM domain layer uses the late suspend and
early resume callbacks to ensure devices are in their low power
states.

However, this is attempted even in cases where a driver probe has
failed.  If a driver's ->probe() method fails, the driver is likely in
a state where it is not expecting its runtime PM callbacks to be
called, yet currently the omap_device PM domain code attempts to call
the drivers callbacks.

To fix, use the omap_device driver_status field to check whether a
driver is bound to the omap_device before attempting to trigger driver
callbacks.

Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>
11 years agoARM: OMAP: omap_device: keep track of driver bound status
Kevin Hilman [Tue, 10 Jul 2012 18:13:16 +0000 (11:13 -0700)] 
ARM: OMAP: omap_device: keep track of driver bound status

Use the bus notifier to keep track of driver bound status by adding a
new internal field to struct omap_device: _driver_status.

This will be useful for follow-up patches which need to know whether
or not a driver is bound in order to make intelligent omap_device
enable/idle decisions.

Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>
11 years agoMerge branch 'renesas/pmu' into next/soc
Olof Johansson [Wed, 12 Sep 2012 06:04:30 +0000 (23:04 -0700)] 
Merge branch 'renesas/pmu' into next/soc

* renesas/pmu:
  ARM: shmobile: emev2: enable PMU(Performance Monitoring Unit)
  ARM: shmobile: sh73a0: enable PMU(Performance Monitoring Unit)

11 years agoMerge branch 'soc' of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas...
Olof Johansson [Wed, 12 Sep 2012 05:56:33 +0000 (22:56 -0700)] 
Merge branch 'soc' of git://git./linux/kernel/git/horms/renesas into next/soc

* 'soc' of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas:
  r8a7779: add SDHI clock support

11 years agoARM: OMAP3+: hwmod: Add AM33XX HWMOD data
Vaibhav Hiremath [Wed, 25 Jul 2012 19:51:13 +0000 (13:51 -0600)] 
ARM: OMAP3+: hwmod: Add AM33XX HWMOD data

This patch adds HWMOD data for all the peripherals of
AM335X device and also hooks up to the existing OMAP framework.

hwmod data has been already been cleaned up for the recent
changes in clocktree, where all leaf nodes have been removed,
since with modulemode based control, both clock and hwmod
interface does same thing. This reduces the code size to large
extent and also avoids duplication of same control.
So instead of specifying module's leaf node as a main_clk,
now we are relying on parent clock of module's functional clock.

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
Signed-off-by: Afzal Mohammed <afzal@ti.com>
Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Benoit Cousson <b-cousson@ti.com>
Cc: Tony Lindgren <tony@atomide.com>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Rajendra Nayak <rnayak@ti.com>
[paul@pwsan.com: removed period in hwmod device names; changed mmc2 main_clk
 to mmc_clk at Vaibhav's request; added trailing commas to structure
 records at Tony's request to deal with some rmk parsing issues; added
 OMAP_INTC_START to facilitate sparse-IRQ conversion]
Signed-off-by: Paul Walmsley <paul@pwsan.com>
11 years agoARM: OMAP2+: hwmod: Hook-up am33xx support in omap_hwmod framework
Vaibhav Hiremath [Tue, 11 Sep 2012 23:18:53 +0000 (17:18 -0600)] 
ARM: OMAP2+: hwmod: Hook-up am33xx support in omap_hwmod framework

AM33XX PRCM architecture is different that any OMAP family
of devices, so it is required to have separate implementation
to handle AM33XX module enable/disable, reset assert/deassert
functionality.
This patch adds wrapper api's in omap_hwmod framework to
access prm/cm for AM33XX family of devices.

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
Cc: Paul Walmsley <paul@pwsan.com>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Tony Lindgren <tony@atomide.com>
[paul@pwsan.com: fixed checkpatch messages]
Signed-off-by: Paul Walmsley <paul@pwsan.com>
11 years agoserial/8250: Limit the omap workarounds to omap1
Tony Lindgren [Tue, 11 Sep 2012 05:31:04 +0000 (22:31 -0700)] 
serial/8250: Limit the omap workarounds to omap1

These workarounds do not apply for CONFIG_ARCH_OMAP2PLUS at all,
so let's make it just CONFIG_ARCH_OMAP1.

This is needed to for ARM common zImage changes for
omap2+ to avoid including plat and mach headers.

Signed-off-by: Tony Lindgren <tony@atomide.com>
Acked-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoARM: tegra: cpu-tegra: explicitly manage re-parenting
Stephen Warren [Mon, 10 Sep 2012 23:05:01 +0000 (17:05 -0600)] 
ARM: tegra: cpu-tegra: explicitly manage re-parenting

When changing a PLL's rate, it must have no active children. The CPU
clock cannot be stopped, and CPU clock's divider is not used. The old
clock driver used to handle this by internally reparenting the CPU clock
onto a different PLL when changing the CPU clock rate. However, the new
common-clock based clock driver does not do this, and probably cannot do
this due to the locking issues it would cause.

To solve this, have the Tegra cpufreq driver explicitly perform the
reparenting operations itself. This is probably reasonable anyway,
since such reparenting is somewhat a matter of policy (e.g. which
alternate clock source to use, whether to leave the CPU clock a child
of the alternate clock source if it's running at the desired rate),
and hence is something more appropriate for the cpufreq driver than
the core clock driver anyway.

Cc: Prashant Gaikwad <pgaikwad@nvidia.com>
Cc: Peter De Schrijver <pdeschrijver@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra: fix overflow in tegra20_pll_clk_round_rate()
Stephen Warren [Mon, 10 Sep 2012 23:02:45 +0000 (17:02 -0600)] 
ARM: tegra: fix overflow in tegra20_pll_clk_round_rate()

32-bit math isn't enough when e.g. *prate=12000000, and sel->n=1000.
Use 64-bit math to prevent this.

Cc: Prashant Gaikwad <pgaikwad@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: shmobile: emev2: enable PMU(Performance Monitoring Unit)
Tetsuyuki Kobayashi [Thu, 6 Sep 2012 11:14:07 +0000 (20:14 +0900)] 
ARM: shmobile: emev2: enable PMU(Performance Monitoring Unit)

This patch enables PMU(Performance Monitoring Unit) for emev2.

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Simon Horman <horms@verge.net.au>
11 years agoARM: shmobile: sh73a0: enable PMU(Performance Monitoring Unit)
Tetsuyuki Kobayashi [Thu, 6 Sep 2012 11:14:06 +0000 (20:14 +0900)] 
ARM: shmobile: sh73a0: enable PMU(Performance Monitoring Unit)

This patch enables PMU(Performance Monitoring Unit) for sh73a0.

Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Signed-off-by: Simon Horman <horms@verge.net.au>
11 years agoARM: OMAP: remove plat/board.h file
Igor Grinberg [Tue, 28 Aug 2012 23:18:57 +0000 (02:18 +0300)] 
ARM: OMAP: remove plat/board.h file

plat/board.h file is now empty - remove it.

Cc: Jarkko Lavinen <jarkko.lavinen@nokia.com>
Cc: Chris Ball <cjb@laptop.org>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: linux-mmc@vger.kernel.org
Cc: linux-mtd@lists.infradead.org
Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Acked-by: Chris Ball <cjb@laptop.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP: move debug_card_init() function
Igor Grinberg [Tue, 28 Aug 2012 23:18:56 +0000 (02:18 +0300)] 
ARM: OMAP: move debug_card_init() function

debug_card_init() function resides in the plat/board.h file.
Move it to a separate header file under plat/ so the board.h file can be
removed.

Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agotty vt: Fix line garbage in virtual console on command line edition
Jean-François Moine [Thu, 6 Sep 2012 17:24:13 +0000 (19:24 +0200)] 
tty vt: Fix line garbage in virtual console on command line edition

On some machines using a specific hardware for console screen output,
the update of the pixel frame buffer does not work correctly when
using command line edition. This may be due to a memory cache bug
in the machine on which the problem has been found, but an other
solution is possible.

This patch proposes to avoid touching directly the pixel frame buffer
and to rebuild each character using the standard putc() function
on command line edition.

The resulting code is smaller and not obviously slower (no read
access to the pixel frame buffer).

Tested on a Cubox (ARM Marvell Dove 88AP510 SoC).

Signed-off-by: Jean-François Moine <moinejf@free.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoARM: OMAP1: move lcd pdata out of arch/arm/*
Igor Grinberg [Tue, 28 Aug 2012 23:18:55 +0000 (02:18 +0300)] 
ARM: OMAP1: move lcd pdata out of arch/arm/*

omap1 lcd platform data resides inside plat/board.h while it
should be inside include/linux/...
Move the omap1 lcd platform data to include/linux/omapfb.h.

Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Acked-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP1: move omap1_bl pdata out of arch/arm/*
Igor Grinberg [Tue, 28 Aug 2012 23:18:54 +0000 (02:18 +0300)] 
ARM: OMAP1: move omap1_bl pdata out of arch/arm/*

omap1 backlight platform data resides inside plat/board.h while it
should be inside include/linux/...
Move the omap1 backlight platform data to
include/linux/platform_data/.

Cc: Richard Purdie <rpurdie@rpsys.net>
Cc: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Cc: linux-fbdev@vger.kernel.org
Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Acked-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP: remove the omap custom tags
Igor Grinberg [Tue, 28 Aug 2012 23:18:53 +0000 (02:18 +0300)] 
ARM: OMAP: remove the omap custom tags

The omap custom initialization tags are not used anymore (if ever)
by the mainline kernel.
Thus remove the omap custom initialization tags.

Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP1: remove the crystal type tag parsing
Igor Grinberg [Tue, 28 Aug 2012 23:18:52 +0000 (02:18 +0300)] 
ARM: OMAP1: remove the crystal type tag parsing

The omap1 crystal setting uses the OMAP custom tags.
Those tags are not used in upstream kernel and therefore the crystal
type is never set by the tag parsing code on upstream kernels.
Remove the crystal tag parsing code.

Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP: remove the sti console workaround
Igor Grinberg [Tue, 28 Aug 2012 23:18:51 +0000 (02:18 +0300)] 
ARM: OMAP: remove the sti console workaround

The sti console workaround uses the OMAP custom tags.
Those tags are not used in upstream kernel and therefore the workaround
never fires on upstream kernels.
Remove the sti console workaround tags part.
This leaves the workaround functional part intact so can be reused if
needed.

Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoARM: OMAP: omap3evm: cleanup revision bits
Igor Grinberg [Tue, 28 Aug 2012 23:18:50 +0000 (02:18 +0300)] 
ARM: OMAP: omap3evm: cleanup revision bits

The omap3evm has its revision information bits inside the plat/board.h
file. Those bits are not used anywhere in the upstream tree besides the
board-omap3evm.c file.
Move the OMAP3EVM_BOARD_GEN_* bits to the board file and remove the
get_omap3_evm_rev() function declaration.

Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agomsm_serial: fix clock rate on DMA-based uarts
David Brown [Fri, 7 Sep 2012 21:45:03 +0000 (14:45 -0700)] 
msm_serial: fix clock rate on DMA-based uarts

The driver explicitly requests a clock rate for the UART, but it is
off by a factor of four from the dividers that it programs into the
UART.  Fix this by setting the rate to 1/4 of the current value.

Signed-off-by: David Brown <davidb@codeaurora.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoserial: pl011: delete reset callback
Linus Walleij [Fri, 7 Sep 2012 08:02:48 +0000 (10:02 +0200)] 
serial: pl011: delete reset callback

Since commit 4fd0690bb0c3955983560bb2767ee82e2b197f9b
"serial: pl011: implement workaround for CTS clear event issue"
the PL011 UART is no longer at risk to hang up, so get rid
of the callback altogether.

Cc: Rajanikanth H.V <rajanikanth.hv@stericsson.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoserial: pl011: delete dangling bug flag
Linus Walleij [Fri, 7 Sep 2012 08:02:36 +0000 (10:02 +0200)] 
serial: pl011: delete dangling bug flag

The bug flag .interrupt_may_hang is not used since commit
4fd0690bb0c3955983560bb2767ee82e2b197f9b
"serial: pl011: implement workaround for CTS clear event issue"
so delete it.

Cc: Rajanikanth H.V <rajanikanth.hv@stericsson.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoTTY: serial: move the dereference below the NULL test
Wei Yongjun [Fri, 7 Sep 2012 06:53:49 +0000 (14:53 +0800)] 
TTY: serial: move the dereference below the NULL test

The dereference should be moved below the NULL test.

spatch with a semantic match is used to found this.
(http://coccinelle.lip6.fr/)

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoserial: omap: Request pins using pinctrl framework
Tony Lindgren [Fri, 7 Sep 2012 17:59:40 +0000 (10:59 -0700)] 
serial: omap: Request pins using pinctrl framework

Request pins using pinctrl framework. Only show a warning
on error as some boards set the pins in the bootloader
even if CONFIG_PINCTRL is enabled.

Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoserial: omap: fix DeviceTree boot
Felipe Balbi [Fri, 7 Sep 2012 18:10:33 +0000 (21:10 +0300)] 
serial: omap: fix DeviceTree boot

OMAP Architecture code, passes a few function
pointers for UART driver to use in order to
properly implement Power Management and Wakeup
capabilities.

The problem is that those function pointers,
which are passed (ab)using platform_data on
non-DT kernels, can't be passed down to drivers
through DT.

commit e5b57c0 (serial: omap: define helpers
for pdata function pointers) failed to take DT
kernels into consideration and caused a regression
to DT kernel boot.

Fix that by (re-)adding a check for valid pdata
pointer together with valid pdata->$FUNCTION
pointer.

Reported-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Tested-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoARM: OMAP: cleanup struct omap_board_config_kernel
Igor Grinberg [Tue, 28 Aug 2012 23:18:49 +0000 (02:18 +0300)] 
ARM: OMAP: cleanup struct omap_board_config_kernel

struct omap_board_config_kernel defined in the board files
is always empty and does not bring any added value.
Remove the struct omap_board_config_kernel instances from the board
files.
Also remove the omap_get_nr_config() macro and the omap_get_var_config()
function as both are not used for quite a long time (if ever).

Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Tony Lindgren <tony@atomide.com>
11 years agoLinux 3.6-rc5
Linus Torvalds [Sat, 8 Sep 2012 23:43:45 +0000 (16:43 -0700)] 
Linux 3.6-rc5

11 years agoMerge branch 'fixes-for-3.6' of git://git.linaro.org/people/mszyprowski/linux-dma...
Linus Torvalds [Sat, 8 Sep 2012 23:22:43 +0000 (16:22 -0700)] 
Merge branch 'fixes-for-3.6' of git://git.linaro.org/people/mszyprowski/linux-dma-mapping

Pull DMA-mapping fixes from Marek Szyprowski:
 "Another set of fixes for ARM dma-mapping subsystem.

  Commit e9da6e9905e6 replaced custom consistent buffer remapping code
  with generic vmalloc areas.  It however introduced some regressions
  caused by limited support for allocations in atomic context.  This
  series contains fixes for those regressions.

  For some subplatforms the default, pre-allocated pool for atomic
  allocations turned out to be too small, so a function for setting its
  size has been added.

  Another set of patches adds support for atomic allocations to
  IOMMU-aware DMA-mapping implementation.

  The last part of this pull request contains two fixes for Contiguous
  Memory Allocator, which relax too strict requirements."

* 'fixes-for-3.6' of git://git.linaro.org/people/mszyprowski/linux-dma-mapping:
  ARM: dma-mapping: IOMMU allocates pages from atomic_pool with GFP_ATOMIC
  ARM: dma-mapping: Introduce __atomic_get_pages() for __iommu_get_pages()
  ARM: dma-mapping: Refactor out to introduce __in_atomic_pool
  ARM: dma-mapping: atomic_pool with struct page **pages
  ARM: Kirkwood: increase atomic coherent pool size
  ARM: DMA-Mapping: print warning when atomic coherent allocation fails
  ARM: DMA-Mapping: add function for setting coherent pool size from platform code
  ARM: relax conditions required for enabling Contiguous Memory Allocator
  mm: cma: fix alignment requirements for contiguous regions

11 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Linus Torvalds [Sat, 8 Sep 2012 23:20:59 +0000 (16:20 -0700)] 
Merge branch 'for-linus' of git://git./linux/kernel/git/dtor/input

Pull input subsystem updates from Dmitry Torokhov.

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
  Input: wacom - add support for EMR on Cintiq 24HD touch
  Input: i8042 - add Gigabyte T1005 series netbooks to noloop table
  Input: imx_keypad - reset the hardware before enabling
  Input: edt-ft5x06 - fix build error when compiling wthout CONFIG_DEBUG_FS

11 years agoMerge branch 'lpc32xx/core' of git://git.antcom.de/linux-2.6 into next/soc
Olof Johansson [Fri, 7 Sep 2012 22:02:20 +0000 (15:02 -0700)] 
Merge branch 'lpc32xx/core' of git://git.antcom.de/linux-2.6 into next/soc

* 'lpc32xx/core' of git://git.antcom.de/linux-2.6:
  ARM: LPC32xx: Remove board specific GPIO init
  ARM: LPC32xx: Provide DMA filter callbacks via platform data
  ARM: LPC32xx: Use handle_edge_irq() callback on edge type irqs
  + sync to 3.6-rc4

11 years agoMerge branch 'upstream-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid
Linus Torvalds [Fri, 7 Sep 2012 19:29:38 +0000 (12:29 -0700)] 
Merge branch 'upstream-fixes' of git://git./linux/kernel/git/jikos/hid

Pull HID updates from Jiri Kosina:
 "It contains a fix for Eaton Ellipse MAX UPS from Alan Stern,
  performance improvement (not processing debug data if noone is
  interested), by Henrik Rydberg, and allowing tpkbd-driven devices to
  work even with generic driver in a crippled mode, by Andres Freund."

* 'upstream-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid:
  HID: tpkbd: work even if the new Lenovo Keyboard driver is not configured
  HID: Only dump input if someone is listening
  HID: add NOGET quirk for Eaton Ellipse MAX UPS

11 years agoserial: omap: fix compile breakage
Felipe Balbi [Fri, 7 Sep 2012 15:34:19 +0000 (18:34 +0300)] 
serial: omap: fix compile breakage

when rebasing patches on top of Greg's tty-next,
it looks like automerge broke a few things which
I didn't catch (for whatever reason I didn't
have OMAP Serial enabled on .config) so I ended
up breaking the build on Greg's tty-next branch.

Fix the breakage by re-adding the three missing
members on struct uart_omap_port.

Reported-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoARM: tegra: Fix data type for io address
Prashant Gaikwad [Fri, 7 Sep 2012 05:40:55 +0000 (11:10 +0530)] 
ARM: tegra: Fix data type for io address

Warnings were generated because following commit changed data type for
address pointer

195bbca ARM: 7500/1: io: avoid writeback addressing modes for __raw_ accessors

arch/arm/mach-tegra/tegra30_clocks.c: In function 'clk_measure_input_freq':
arch/arm/mach-tegra/tegra30_clocks.c:418:2: warning: passing argument 2 of '__raw_writel' makes pointer from integer without a cast
.../arch/arm/include/asm/io.h:88:20: note: expected 'volatile void *' but argument is of type 'unsigned int

Signed-off-by: Prashant Gaikwad <pgaikwad@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoHID: tpkbd: work even if the new Lenovo Keyboard driver is not configured
Andres Freund [Thu, 30 Aug 2012 12:37:14 +0000 (14:37 +0200)] 
HID: tpkbd: work even if the new Lenovo Keyboard driver is not configured

c1dcad2d32d0252e8a3023d20311b52a187ecda3 added a new driver configured by
HID_LENOVO_TPKBD but made the hid_have_special_driver entry non-optional which
lead to a recognized but non-working device if the new driver wasn't
configured (which is the correct default).

Signed-off-by: Andres Freund <andres@anarazel.de>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
11 years agoMerge tag 'stable/for-linus-3.6-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 7 Sep 2012 00:16:42 +0000 (17:16 -0700)] 
Merge tag 'stable/for-linus-3.6-rc4-tag' of git://git./linux/kernel/git/konrad/xen

Pull Xen bug-fixes from Konrad Rzeszutek Wilk:
 * Fix for TLB flushing introduced in v3.6
 * Fix Xen-SWIOTLB not using proper DMA mask - device had 64bit but
   in a 32-bit kernel we need to allocate for coherent pages from a
   32-bit pool.
 * When trying to re-use P2M nodes we had a one-off error and triggered
   a BUG_ON check with specific CONFIG_ option.
 * When doing FLR in Xen-PCI-backend we would first do FLR then save the
   PCI configuration space. We needed to do it the other way around.

* tag 'stable/for-linus-3.6-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
  xen/pciback: Fix proper FLR steps.
  xen: Use correct masking in xen_swiotlb_alloc_coherent.
  xen: fix logical error in tlb flushing
  xen/p2m: Fix one-off error in checking the P2M tree directory.

11 years agoMerge tag '3.6-pci-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Linus Torvalds [Fri, 7 Sep 2012 00:15:49 +0000 (17:15 -0700)] 
Merge tag '3.6-pci-fixes' of git://git./linux/kernel/git/helgaas/pci

Pull PCI updates from Bjorn Helgaas:
 "Power management
    - PCI/PM: Enable D3/D3cold by default for most devices
    - PCI/PM: Keep parent bridge active when probing device
    - PCI/PM: Fix config reg access for D3cold and bridge suspending
    - PCI/PM: Add ABI document for sysfs file d3cold_allowed
  Core
    - PCI: Don't print anything while decoding is disabled"

* tag '3.6-pci-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
  PCI: Don't print anything while decoding is disabled
  PCI/PM: Add ABI document for sysfs file d3cold_allowed
  PCI/PM: Fix config reg access for D3cold and bridge suspending
  PCI/PM: Keep parent bridge active when probing device
  PCI/PM: Enable D3/D3cold by default for most devices

11 years agotty_register_device_attr updated for tty-next
Tomas Hlavacek [Thu, 6 Sep 2012 21:17:47 +0000 (23:17 +0200)] 
tty_register_device_attr updated for tty-next

Added tty_device_create_release() and bound to dev->release in
tty_register_device_attr().
Added tty_port_register_device_attr() and used in uart_add_one_port()
instead of tty_register_device_attr().

Signed-off-by: Tomas Hlavacek <tmshlvck@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoARM: tegra: remove tegra_timer from tegra_list_clks
Stephen Warren [Mon, 27 Aug 2012 23:33:14 +0000 (16:33 -0700)] 
ARM: tegra: remove tegra_timer from tegra_list_clks

tegra_time is a struct sys_timer, not a struct clk, so can't be included
in an array of struct clk *.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra30: clocks: fix the wrong tegra_audio_sync_clk_ops name
Joseph Lo [Fri, 17 Aug 2012 06:54:10 +0000 (14:54 +0800)] 
ARM: tegra30: clocks: fix the wrong tegra_audio_sync_clk_ops name

It should use tegra30_audio_sync_clk_ops for tegra30. It will cause
the tegra30 use the wrong audio_sync_clk_ops when build a kernel with
a tegra20 and tegra30 both supported kernel. And building error when
a tegra30-only kernel.

Signed-off-by: Joseph Lo <josephl@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra: clocks: separate tegra_clk_32k_ops from Tegra20 and Tegra30
Joseph Lo [Fri, 17 Aug 2012 06:51:21 +0000 (14:51 +0800)] 
ARM: tegra: clocks: separate tegra_clk_32k_ops from Tegra20 and Tegra30

Currently the tegra20 and tegra30 share the same symbol for
tegra_clk_32k_ops. This will cause a compile error when building
a tegra20-only kernel image. Add tegra_clk_32k_ops for tegra20 and
modify tegra30_clk_32k_ops for tegra30.

Signed-off-by: Joseph Lo <josephl@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra: Remove duplicate code
Prashant Gaikwad [Mon, 6 Aug 2012 06:27:44 +0000 (11:57 +0530)] 
ARM: tegra: Remove duplicate code

Remove Tegra legacy clock framework code.

Signed-off-by: Prashant Gaikwad <pgaikwad@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra: Port tegra to generic clock framework
Prashant Gaikwad [Mon, 6 Aug 2012 06:27:43 +0000 (11:57 +0530)] 
ARM: tegra: Port tegra to generic clock framework

This patch converts tegra clock code to generic clock framework in following way:
 - Implement clk_ops as required by generic clk framework. (tegraXX_clocks.c)
 - Use platform specific struct clk_tegra in clk_ops implementation instead of struct clk.
 - Initialize all clock data statically. (tegraXX_clocks_data.c)

Legacy framework did not have recalc_rate and is_enabled functions. Implemented these functions.
Removed init function. It's functionality is splitted into recalc_rate and is_enabled.

Static initialization is used since slab is not up in .init_early and clock
is needed to be initialized before clockevent/clocksource initialization.
Macros redefined for clk_tegra.

Also, single struct clk_tegra is used for all type of clocks (PLL, peripheral etc.). This
is to move quickly to generic common clock framework so that other dependent features will
not be blocked (such as DT binding).

Enabling COMMON_CLOCK config moved to ARCH_TEGRA since it is enabled for both Tegra20
and Tegra30.

Signed-off-by: Prashant Gaikwad <pgaikwad@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra: Add clk_tegra structure and helper functions
Prashant Gaikwad [Mon, 6 Aug 2012 06:27:42 +0000 (11:57 +0530)] 
ARM: tegra: Add clk_tegra structure and helper functions

Add Tegra platform specific clock structure clk_tegra and
some helper functions for generic clock framework.

struct clk_tegra is the single strcture used for all types of
clocks. reset and cfg_ex ops moved to clk_tegra from clk_ops.

Signed-off-by: Prashant Gaikwad <pgaikwad@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra: Rename tegra20 clock file
Prashant Gaikwad [Mon, 6 Aug 2012 06:27:41 +0000 (11:57 +0530)] 
ARM: tegra: Rename tegra20 clock file

Make the name consistent with other files.
s/tegra2/tegra20

Signed-off-by: Prashant Gaikwad <pgaikwad@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra20: Separate out clk ops and clk data
Prashant Gaikwad [Mon, 6 Aug 2012 06:27:40 +0000 (11:57 +0530)] 
ARM: tegra20: Separate out clk ops and clk data

Move clock initialization data to separate file. This is
required for migrating to generic clock framework if static
initialization is used.

Signed-off-by: Prashant Gaikwad <pgaikwad@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra30: Separate out clk ops and clk data
Prashant Gaikwad [Mon, 6 Aug 2012 06:27:39 +0000 (11:57 +0530)] 
ARM: tegra30: Separate out clk ops and clk data

Move clock initialization data to separate file. This is
required for migrating to generic clock framework if static
initialization is used.

Signed-off-by: Prashant Gaikwad <pgaikwad@nvidia.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra: fix U16 divider range check
Stephen Warren [Tue, 24 Jul 2012 21:48:12 +0000 (15:48 -0600)] 
ARM: tegra: fix U16 divider range check

A U16 divider can divide a clock by 1..64K. However, the range-check
in clk_div16_get_divider() limited the range to 1..256. Fix this. NVIDIA's
downstream kernels already have the fixed range-check.

In practice this is a problem on Whistler's I2C bus, which uses a bus
clock rate of 100KHz (rather than the more common 400KHz on Tegra boards),
which requires a HW module clock of 8*100KHz. The parent clock is 216MHz,
leading to a desired divider of 270. Prior to conversion to the common
clock framework, this range error was somehow ignored/irrelevant and
caused no problems. However, the common clock framework evidently has
more rigorous error-checking, so this failure causes the I2C bus to fail
to operate correctly.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoARM: tegra: turn on UART A clock at boot
Stephen Warren [Tue, 24 Jul 2012 21:44:11 +0000 (15:44 -0600)] 
ARM: tegra: turn on UART A clock at boot

Some boards use UART D for the main serial console, and some use UART A.
UART D's clock is listed in board-dt-tegra20.c's clock table, whereas
UART A's clock is not. This causes the clock code to think UART A's
clock is unsed. The common clock framework turns off unused clocks at
boot time. This makes the kernel appear to hang. Add UART A's clock into
the clock table to prevent this. Eventually, this requirement should be
handled by the UART driver, and/or properties in a board-specific device
tree file.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
11 years agoMerge tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm...
Linus Torvalds [Thu, 6 Sep 2012 17:23:58 +0000 (10:23 -0700)] 
Merge tag 'fixes-for-linus' of git://git./linux/kernel/git/arm/arm-soc

Pull ARM SoC bug fixes from Olof Johansson:
 "Mostly Renesas and Atmel bugfixes this time, targeting boot and build
  problems.  A couple of patches for gemini and kirkwood as well.  On a
  whole nothing very controversial."

* tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
  ARM: gemini: fix the gemini build
  ARM: shmobile: armadillo800eva: enable rw rootfs mount
  ARM: Kirkwood: Fix 'SZ_1M' undeclared here for db88f6281-bp-setup.c
  ARM: shmobile: mackerel: fixup usb module order
  ARM: shmobile: armadillo800eva: fixup: sound card detection order
  ARM: shmobile: marzen: fixup smsc911x id for regulator
  ARM: at91/feature-removal-schedule: delay at91_mci removal
  ARM: mach-shmobile: armadillo800eva: Enable power button as wakeup source
  ARM: mach-shmobile: armadillo800eva: Fix GPIO buttons descriptions
  ARM: at91/dts: remove partial parameter in at91sam9g25ek.dts
  ARM: at91/clock: fix PLLA overclock warning
  ARM: at91: fix rtc-at91sam9 irq issue due to sparse irq support
  ARM: at91: fix system timer irq issue due to sparse irq support
  ARM: shmobile: sh73a0: fixup RELOC_BASE of intca_irq_pins_desc

11 years agoMerge tag 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck...
Linus Torvalds [Thu, 6 Sep 2012 16:39:47 +0000 (09:39 -0700)] 
Merge tag 'hwmon-for-linus' of git://git./linux/kernel/git/groeck/linux-staging

Pull a hwmon fix from Guenter Roeck:
 "One patch, fixing DIV_ROUND_CLOSEST to support negative dividends.

  While the changes are not in the drivers/hwmon directory, the problem
  primarily affects hwmon drivers, and it makes sense to push the patch
  through the hwmon tree."

* tag 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
  linux/kernel.h: Fix DIV_ROUND_CLOSEST to support negative dividends

11 years agoMerge branch 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild
Linus Torvalds [Thu, 6 Sep 2012 16:38:25 +0000 (09:38 -0700)] 
Merge branch 'rc-fixes' of git://git./linux/kernel/git/mmarek/kbuild

Pull kbuild fixes from Michal Marek:
 "These are two fixes that should go into 3.6.  The link-vmlinux.sh one
  is obvious.

  The other one fixes make firmware_install with certain configurations,
  where a file in the toplevel firmware tree gets installed first, and
  $(INSTALL_FW_PATH)/$$(dir <file>) results in /lib/firmware/./, which
  confuses make 3.82 for some reason."

* 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild:
  firmware: fix directory creation rule matching with make 3.82
  link-vmlinux.sh: Fix stray "echo" in error message

11 years agoRemove user-triggerable BUG from mpol_to_str
Dave Jones [Thu, 6 Sep 2012 16:01:00 +0000 (12:01 -0400)] 
Remove user-triggerable BUG from mpol_to_str

Trivially triggerable, found by trinity:

  kernel BUG at mm/mempolicy.c:2546!
  Process trinity-child2 (pid: 23988, threadinfo ffff88010197e000, task ffff88007821a670)
  Call Trace:
    show_numa_map+0xd5/0x450
    show_pid_numa_map+0x13/0x20
    traverse+0xf2/0x230
    seq_read+0x34b/0x3e0
    vfs_read+0xac/0x180
    sys_pread64+0xa2/0xc0
    system_call_fastpath+0x1a/0x1f
  RIP: mpol_to_str+0x156/0x360

Cc: stable@vger.kernel.org
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
11 years agotty: uartclk value from serial_core exposed to sysfs
Tomas Hlavacek [Thu, 6 Sep 2012 01:17:18 +0000 (03:17 +0200)] 
tty: uartclk value from serial_core exposed to sysfs

Added file /sys/devices/.../tty/ttySX/uartclk to allow reading
uartclk value in struct uart_port in serial_core via sysfs.

tty_register_device() has been generalized and refactored in order
to add support for setting drvdata and attribute_group to the device.

Signed-off-by: Tomas Hlavacek <tmshlvck@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoserial: serial_core.h needs console.h included first
Stephen Rothwell [Thu, 6 Sep 2012 05:05:04 +0000 (15:05 +1000)] 
serial: serial_core.h needs console.h included first

Fixes these build errors:

In file included from drivers/tty/serial/sccnxp.c:20:0:
include/linux/serial_core.h: In function 'uart_handle_break':
include/linux/serial_core.h:543:30: error: dereferencing pointer to incomplete type

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Acked-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoserial: mxs-auart: put the device in mxs_auart_probe()
Huang Shijie [Fri, 7 Sep 2012 02:38:41 +0000 (22:38 -0400)] 
serial: mxs-auart: put the device in mxs_auart_probe()

We call the get_device() in the mxs_auart_probe().
For the balance of the reference count, we should put the
device in the mxs_auart_remove().

Signed-off-by: Huang Shijie <shijie8@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoserial: mxs-auart: fix the wrong setting order
Huang Shijie [Fri, 7 Sep 2012 02:38:40 +0000 (22:38 -0400)] 
serial: mxs-auart: fix the wrong setting order

After set the AUART_CTRL0_CLKGATE, the UART will gate all the clocks off.
So the following line will not take effect.
       ................................................................
  writel(AUART_INTR_RXIEN | AUART_INTR_RTIEN | AUART_INTR_CTSMIEN,
  u->membase + AUART_INTR_CLR);
       ................................................................

To fix this issue, the patch moves this gate-off line to
the end of setting registers.

Signed-off-by: Huang Shijie <shijie8@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoserial_core: fix sizeof(pointer)
Fengguang Wu [Thu, 6 Sep 2012 02:27:51 +0000 (10:27 +0800)] 
serial_core: fix sizeof(pointer)

sizeof when applied to a pointer typed expression gives the
size of the pointer.

Generated by: scripts/coccinelle/misc/noderef.cocci

Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoserial: omap: enable RX and TX FIFO usage
Felipe Balbi [Thu, 6 Sep 2012 12:45:40 +0000 (15:45 +0300)] 
serial: omap: enable RX and TX FIFO usage

enable RX FIFO for 16 characters and TX FIFO
for 16 spaces.

Tested-by: Shubhrajyoti D <shubhrajyoti@ti.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
11 years agoserial: omap: move uart_omap_port definition to C file
Felipe Balbi [Thu, 6 Sep 2012 12:45:39 +0000 (15:45 +0300)] 
serial: omap: move uart_omap_port definition to C file

nobody needs to access the uart_omap_port structure
other than omap-serial.c file. Let's move that
structure definition to the C source file in order
to prevent anyone from accessing our structure.

Tested-by: Shubhrajyoti D <shubhrajyoti@ti.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This page took 0.050543 seconds and 5 git commands to generate.