firefly-linux-kernel-4.4.55.git
14 years agoMerge branches 'arm', 'at91', 'bcmring', 'ep93xx', 'mach-types', 'misc' and 'w90x900...
Russell King [Sat, 12 Sep 2009 11:01:34 +0000 (12:01 +0100)]
Merge branches 'arm', 'at91', 'bcmring', 'ep93xx', 'mach-types', 'misc' and 'w90x900' into devel

14 years ago[ARM] Update mach-types
Russell King [Sat, 12 Sep 2009 11:00:17 +0000 (12:00 +0100)]
[ARM] Update mach-types

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5686/1: at91: Correct AC97 reset line in at91sam9263ek board
Nicolas Ferre [Wed, 26 Aug 2009 14:33:16 +0000 (15:33 +0100)]
ARM: 5686/1: at91: Correct AC97 reset line in at91sam9263ek board

Board code was wrongly setting up the reset pin for AC97 on at91sam9263ek.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5640/1: This patch modifies the support of AC97 on the at91sam9263 ek board
sedji gaouaou [Thu, 6 Aug 2009 14:20:22 +0000 (15:20 +0100)]
ARM: 5640/1: This patch modifies the support of AC97 on the at91sam9263 ek board

This patch modifies the support of AC97 on the at91sam9263 ek board, so it would
share the code with AVR32.
Plus it removes a typo in at91sam9263_devices.c.

Signed-off-by: Sedji Gaouaou <sedji.gaouaou@atmel.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5684/1: Add nuc960 platform to w90x900
wanzongshun [Fri, 21 Aug 2009 06:09:03 +0000 (07:09 +0100)]
ARM: 5684/1: Add nuc960 platform to w90x900

Add nuc960 platform to w90x900.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5683/1: Add nuc950 platform to w90x900
wanzongshun [Fri, 21 Aug 2009 06:08:24 +0000 (07:08 +0100)]
ARM: 5683/1: Add nuc950 platform to w90x900

Add nuc950 platform to w90x900.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5682/1: Add cpu.c and dev.c and modify some files of w90p910 platform
wanzongshun [Fri, 21 Aug 2009 06:07:46 +0000 (07:07 +0100)]
ARM: 5682/1: Add cpu.c and dev.c and modify some files of w90p910 platform

Add the cpu.c and dev.c and modify w90p910 platform
to apply to use the common API(provided by cpu.c and dev.c)
at the same time, I renamed all w90x900 to nuc900 in every
c file of w90x900 platform and touchscreen's driver name.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5626/1: add suspend/resume functions to amba-pl011 serial driver
Leo Chen [Tue, 28 Jul 2009 22:43:33 +0000 (23:43 +0100)]
ARM: 5626/1: add suspend/resume functions to amba-pl011 serial driver

Add suspend/resume functions to the AMBA pl011 serial driver.

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5625/1: fix hard coded 4K resource size in amba bus detection
Leo Chen [Tue, 28 Jul 2009 22:34:59 +0000 (23:34 +0100)]
ARM: 5625/1: fix hard coded 4K resource size in amba bus detection

This patch modifies the amba bus detection logic in the kernel
to detect the AMBA devices using the calculated resource
size information rather than the hard coded 4K size.

It also calculates the resource size when request mem region
and release mem region.

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoMMC: MMCI: convert realview MMC to use gpiolib
Russell King [Thu, 9 Jul 2009 14:17:41 +0000 (15:17 +0100)]
MMC: MMCI: convert realview MMC to use gpiolib

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5685/1: Make MMCI driver compile without gpiolib
Linus Walleij [Wed, 26 Aug 2009 08:58:24 +0000 (09:58 +0100)]
ARM: 5685/1: Make MMCI driver compile without gpiolib

The recent addition of optional gpiolib support to check if a
card was inserted or write protected was really not optional.
It needs this ifdef to become optional so that U300 compiles,
for example.

Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: implement highpte
Russell King [Mon, 17 Aug 2009 19:02:06 +0000 (20:02 +0100)]
ARM: implement highpte

Add the ARM implementation of highpte, which allows PTE tables to be
placed in highmem.  Unfortunately, we do not offer highpte support
when support for L2 cache is enabled.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5671/1: bcmring: add maintainer entry
Leo Chen [Wed, 12 Aug 2009 21:08:48 +0000 (22:08 +0100)]
ARM: 5671/1: bcmring: add maintainer entry

add maintainer entry

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5670/1: bcmring: add default configuration for bcmring arch
Leo Chen [Wed, 12 Aug 2009 21:07:26 +0000 (22:07 +0100)]
ARM: 5670/1: bcmring: add default configuration for bcmring arch

add default configuration for bcmring arch

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5641/1: bcmring: add Kconfig and Makefile entries in arch/arm
Leo Chen [Fri, 7 Aug 2009 18:46:15 +0000 (19:46 +0100)]
ARM: 5641/1: bcmring: add Kconfig and Makefile entries in arch/arm

add bcmring option in Kconfig and add entry in Makefile
in arch/arm directory

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5642/1: bcmring: add Makefile/Kconfig/Makefile.boot in mach-bcmring
Leo Chen [Fri, 7 Aug 2009 18:46:49 +0000 (19:46 +0100)]
ARM: 5642/1: bcmring: add Makefile/Kconfig/Makefile.boot in mach-bcmring

add arch/arm/mach-bcmring directory

add Kconfig, Makefile, and Makefile.boot in mach-bcmring

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5653/1: bcmring: add Makefile of csp code
Leo Chen [Fri, 7 Aug 2009 19:04:17 +0000 (20:04 +0100)]
ARM: 5653/1: bcmring: add Makefile of csp code

add Makefile of csp code

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5661/1: bcmring: add csp security hardware headers
Leo Chen [Fri, 7 Aug 2009 19:10:24 +0000 (20:10 +0100)]
ARM: 5661/1: bcmring: add csp security hardware headers

add csp security hardware headers

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5660/1: bcmring: add csp timer block header and source files
Leo Chen [Fri, 7 Aug 2009 19:09:49 +0000 (20:09 +0100)]
ARM: 5660/1: bcmring: add csp timer block header and source files

add csp timer block header and source files

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5659/1: bcmring: add csp dmac source files
Leo Chen [Fri, 7 Aug 2009 19:08:00 +0000 (20:08 +0100)]
ARM: 5659/1: bcmring: add csp dmac source files

add csp dmac source files

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5658/1: bcmring: add csp dmac header files
Leo Chen [Fri, 7 Aug 2009 19:07:22 +0000 (20:07 +0100)]
ARM: 5658/1: bcmring: add csp dmac header files

add csp dmac header files

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5657/1: bcmring: add csp chipc block source code
Leo Chen [Fri, 7 Aug 2009 19:06:47 +0000 (20:06 +0100)]
ARM: 5657/1: bcmring: add csp chipc block source code

add csp chipc block source code

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5656/1: bcmring: add csp chipc inline functions
Leo Chen [Fri, 7 Aug 2009 19:06:13 +0000 (20:06 +0100)]
ARM: 5656/1: bcmring: add csp chipc inline functions

add csp chipc inline functions

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5655/1: bcmring: add csp chipc hardware definition file
Leo Chen [Fri, 7 Aug 2009 19:05:37 +0000 (20:05 +0100)]
ARM: 5655/1: bcmring: add csp chipc hardware definition file

add csp chipc hardware definition file

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5654/1: bcmring: add csp chipc hardware register file
Leo Chen [Fri, 7 Aug 2009 19:04:54 +0000 (20:04 +0100)]
ARM: 5654/1: bcmring: add csp chipc hardware register file

add csp chipc hardware register file

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5652/1: bcmring: add misc headers for csp code
Leo Chen [Fri, 7 Aug 2009 19:03:40 +0000 (20:03 +0100)]
ARM: 5652/1: bcmring: add misc headers for csp code

add misc wrapper headers for csp code
add ddrc register header file

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5651/1: bcmring: csp capability header files
Leo Chen [Fri, 7 Aug 2009 19:02:59 +0000 (20:02 +0100)]
ARM: 5651/1: bcmring: csp capability header files

add mach-bcmring csp capability header files

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5650/1: bcmring: add io.h, uncompress. h, and entry-macro.S
Leo Chen [Fri, 7 Aug 2009 19:02:21 +0000 (20:02 +0100)]
ARM: 5650/1: bcmring: add io.h, uncompress. h, and entry-macro.S

add remaining header files in include/mach directory
add entry-macro.S file

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5649/1: bcmring: add bcmring timer function
Leo Chen [Fri, 7 Aug 2009 19:01:47 +0000 (20:01 +0100)]
ARM: 5649/1: bcmring: add bcmring timer function

bcmring timer helper function, hardware register headers

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5648/1: bcmring: add bmcring dma.c
Leo Chen [Fri, 7 Aug 2009 19:01:13 +0000 (20:01 +0100)]
ARM: 5648/1: bcmring: add bmcring dma.c

implement dma support for bcmring

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5647/1: bcmring: add bcmring dma.h and dma_device.c
Leo Chen [Fri, 7 Aug 2009 19:00:39 +0000 (20:00 +0100)]
ARM: 5647/1: bcmring: add bcmring dma.h and dma_device.c

add bcmring dma.h and dma_device.c

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5646/1: bcmring: add mach-bcmring/mm.c and memory headers
Leo Chen [Fri, 7 Aug 2009 18:59:57 +0000 (19:59 +0100)]
ARM: 5646/1: bcmring: add mach-bcmring/mm.c and memory headers

memory map addresses
memory map description and init functions

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5645/1: bcmring: add bcmring irq.c
Leo Chen [Fri, 7 Aug 2009 18:59:04 +0000 (19:59 +0100)]
ARM: 5645/1: bcmring: add bcmring irq.c

init irq and handler
add irq controller register file and header files

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5644/1: add bcmring core.c, clock.c, clock.h
Leo Chen [Fri, 7 Aug 2009 18:58:26 +0000 (19:58 +0100)]
ARM: 5644/1: add bcmring core.c, clock.c, clock.h

add core.c, clock.c, and clock.h in mach-bcmring

implement timer init, clocksource init, amba device init
implement clock set/get enable/disable API
add dummy clkdev.h

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5643/1: bcmring: arch.c and header files
Leo Chen [Fri, 7 Aug 2009 18:56:19 +0000 (19:56 +0100)]
ARM: 5643/1: bcmring: arch.c and header files

add arch.c in mach-bcmring
add related header files in mach-bcmring

Signed-off-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5676/1: Provide more useful introduction for w90x900
wanzongshun [Fri, 14 Aug 2009 14:38:29 +0000 (15:38 +0100)]
ARM: 5676/1: Provide more useful introduction for w90x900

Provide more useful introduction for w90x900

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5675/1: The semaphore is used as mutex so make it a mutex
wanzongshun [Fri, 14 Aug 2009 14:37:42 +0000 (15:37 +0100)]
ARM: 5675/1: The semaphore is used as mutex so make it a mutex

The semaphore is used as mutex so make it a mutex.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5674/1: Add clocksource/clockevent support for w90p910 platform
wanzongshun [Fri, 14 Aug 2009 14:36:44 +0000 (15:36 +0100)]
ARM: 5674/1: Add clocksource/clockevent support for w90p910 platform

Add clocksource/clockevent support for w90p910 platform.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5677/1: ARM support for TIF_RESTORE_SIGMASK/pselect6/ppoll/epoll_pwait
Mikael Pettersson [Sat, 15 Aug 2009 11:58:11 +0000 (12:58 +0100)]
ARM: 5677/1: ARM support for TIF_RESTORE_SIGMASK/pselect6/ppoll/epoll_pwait

This patch adds support for TIF_RESTORE_SIGMASK to ARM's
signal handling, which allows to hook up the pselect6, ppoll,
and epoll_pwait syscalls on ARM.

Tested here with eabi userspace and a test program with a
deliberate race between a child's exit and the parent's
sigprocmask/select sequence. Using sys_pselect6() instead
of sigprocmask/select reliably prevents the race.

The other arch's support for TIF_RESTORE_SIGMASK has evolved
over time:

In 2.6.16:
- add TIF_RESTORE_SIGMASK which parallels TIF_SIGPENDING
- test both when checking for pending signal [changed later]
- reimplement sys_sigsuspend() to use current->saved_sigmask,
  TIF_RESTORE_SIGMASK [changed later], and -ERESTARTNOHAND;
  ditto for sys_rt_sigsuspend(), but drop private code and
  use common code via __ARCH_WANT_SYS_RT_SIGSUSPEND;
- there are now no "extra" calls to do_signal() so its oldset
  parameter is always &current->blocked so need not be passed,
  also its return value is changed to void
- change handle_signal() to return 0/-errno
- change do_signal() to honor TIF_RESTORE_SIGMASK:
  + get oldset from current->saved_sigmask if TIF_RESTORE_SIGMASK
    is set
  + if handle_signal() was successful then clear TIF_RESTORE_SIGMASK
  + if no signal was delivered and TIF_RESTORE_SIGMASK is set then
    clear it and restore the sigmask
- hook up sys_pselect6() and sys_ppoll()

In 2.6.19:
- hook up sys_epoll_pwait()

In 2.6.26:
- allow archs to override how TIF_RESTORE_SIGMASK is implemented;
  default set_restore_sigmask() sets both TIF_RESTORE_SIGMASK and
  TIF_SIGPENDING; archs need now just test TIF_SIGPENDING again
  when checking for pending signal work; some archs now implement
  TIF_RESTORE_SIGMASK as a secondary/non-atomic thread flag bit
- call set_restore_sigmask() in sys_sigsuspend() instead of setting
  TIF_RESTORE_SIGMASK

In 2.6.29-rc:
- kill sys_pselect7() which no arch wanted

So for 2.6.31-rc6/ARM this patch does the following:
- Add TIF_RESTORE_SIGMASK. Use the generic set_restore_sigmask()
  which sets both TIF_SIGPENDING and TIF_RESTORE_SIGMASK, so
  TIF_RESTORE_SIGMASK need not claim one of the scarce low thread
  flags, and existing TIF_SIGPENDING and _TIF_WORK_MASK tests need
  not be extended for TIF_RESTORE_SIGMASK.
- sys_sigsuspend() is reimplemented to use current->saved_sigmask
  and set_restore_sigmask(), making it identical to most other archs
- The private code for sys_rt_sigsuspend() is removed, instead
  generic code supplies it via __ARCH_WANT_SYS_RT_SIGSUSPEND.
- sys_sigsuspend() and sys_rt_sigsuspend() no longer need a pt_regs
  parameter, so their assembly code wrappers are removed.
- handle_signal() is changed to return 0 on success or -errno.
- The oldset parameter to do_signal() is now redundant and removed,
  and the return value is now also redundant and changed to void.
- do_signal() is changed to honor TIF_RESTORE_SIGMASK:
  + get oldset from current->saved_sigmask if TIF_RESTORE_SIGMASK
    is set
  + if handle_signal() was successful then clear TIF_RESTORE_SIGMASK
  + if no signal was delivered and TIF_RESTORE_SIGMASK is set then
    clear it and restore the sigmask
- Hook up sys_pselect6, sys_ppoll, and sys_epoll_pwait.

Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5635/1: Use DIV_ROUND_CLOSEST
Julia Lawall [Mon, 3 Aug 2009 10:57:20 +0000 (11:57 +0100)]
ARM: 5635/1: Use DIV_ROUND_CLOSEST

The kernel.h macro DIV_ROUND_CLOSEST performs the computation
(x + d/2)/d but is perhaps more readable.

The semantic patch that makes this change is as follows:
(http://www.emn.fr/x-info/coccinelle/)

// <smpl>
@haskernel@
@@

@depends on haskernel@
expression x,__divisor;
@@

- (((x) + ((__divisor) / 2)) / (__divisor))
+ DIV_ROUND_CLOSEST(x,__divisor)
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Acked-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years ago[ARM] Update mach-types
Russell King [Sat, 1 Aug 2009 15:26:03 +0000 (16:26 +0100)]
[ARM] Update mach-types

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5632/1: Board-specific data for IDE support on afeb9260
Sergey Matyukevich [Thu, 30 Jul 2009 21:29:07 +0000 (22:29 +0100)]
ARM: 5632/1: Board-specific data for IDE support on afeb9260

This patch adds board-specific data for IDE support on afeb9260.

Depends on #5631/1

Tested-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: Sergey Matyukevich <geomatsi@gmail.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Acked-by: Sergey Lapin <slapin@ossfans.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5631/1: Platform data for CF/IDE support in AT91SAM9260
Sergey Matyukevich [Thu, 30 Jul 2009 21:23:24 +0000 (22:23 +0100)]
ARM: 5631/1: Platform data for CF/IDE support in AT91SAM9260

This patch adds platform data for CF/IDE support
and SMC init code for AT91SAM9260 SoC.

Tested-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: Sergey Matyukevich <geomatsi@gmail.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5634/1: Add static setting cpu frequence for w90p910 platform
wanzongshun [Fri, 31 Jul 2009 01:30:32 +0000 (02:30 +0100)]
ARM: 5634/1: Add static setting cpu frequence for w90p910 platform

Add static setting cpu frequence for w90p910 platform.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5633/1: Add wdt resource for w90p910 platform
wanzongshun [Fri, 31 Jul 2009 01:29:45 +0000 (02:29 +0100)]
ARM: 5633/1: Add wdt resource for w90p910 platform

Add wdt resource for w90p910 platform.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5630/1: Add support for Eukrea's CPUAT91
Eric Bénard [Thu, 30 Jul 2009 06:23:42 +0000 (07:23 +0100)]
ARM: 5630/1: Add support for Eukrea's CPUAT91

CPUAT91 is based on Atmel's AT91RM9200 with up to 16MB Strataflash,
up to 128MB SDRAM and an ethernet PHY in RMII mode.

Signed-off-by: Eric Benard <ebenard@eukrea.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5629/1: Add support for Eukrea's CPU9260 & CPU9G20
Eric Bénard [Thu, 30 Jul 2009 06:20:03 +0000 (07:20 +0100)]
ARM: 5629/1: Add support for Eukrea's CPU9260 & CPU9G20

CPU9260 and CPU9G20 share the same PCB populated with either
Atmel's AT91SAM9260B or AT91SAM9G20B with up to 64MB Strataflash,
up to 128MB SDRAM, up to 2GB NAND and an ethernet PHY in RMII mode.

Signed-off-by: Eric Benard <eric@eukrea.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5628/1: ep93xx: Introduce Pulse Width Modulator (PWM) driver
Hartley Sweeten [Wed, 29 Jul 2009 21:41:06 +0000 (22:41 +0100)]
ARM: 5628/1: ep93xx: Introduce Pulse Width Modulator (PWM) driver

The EP93xx features two PWMs (one on the EP9307) with the following
features:

  * Configurable dual output
  * Separate input clocks for each PWM output
  * 16-bit resolution
  * Programmable pulse width (duty cycle), interval (frequency), and
    polarity

This adds the necessary core support as well as the driver.  A sysfs
interface is provided to control the PWM outputs.

Signed-off-by: Matthieu Crapet <mcrapet@gmail.com>
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Acked-by: Ryan Mallon <ryan@bluewatersys.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5622/1: at91: at91sam9g45 headers: DMA peripheral identifiers
Nicolas Ferre [Fri, 24 Jul 2009 11:16:57 +0000 (12:16 +0100)]
ARM: 5622/1: at91: at91sam9g45 headers: DMA peripheral identifiers

It adds DMA peripheral identifiers for hardware handshaking interface. It will
be used in platform code.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years ago[ARM] 5614/1: at91: atmel_lcdfb: add at91sam9g10 support to atmel LCD driver
Nicolas Ferre [Tue, 21 Jul 2009 10:31:29 +0000 (11:31 +0100)]
[ARM] 5614/1: at91: atmel_lcdfb: add at91sam9g10 support to atmel LCD driver

Modify atmel LCD driver: atmel_lcdfb for at91sam9g10.  This add a clock
management equivalent to at91sam9261.

Signed-off-by: Hong Xu <hong.xu@atmel.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years ago[ARM] 5619/1: Add spi mfp api for w90p910 spi driver
wanzongshun [Thu, 23 Jul 2009 07:52:51 +0000 (08:52 +0100)]
[ARM] 5619/1: Add spi mfp api for w90p910 spi driver

Add spi mfp api for w90p910 spi driver.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years ago[ARM] 5618/1: Add spi resource define for w90p910 spi driver
wanzongshun [Thu, 23 Jul 2009 07:51:46 +0000 (08:51 +0100)]
[ARM] 5618/1: Add spi resource define for w90p910 spi driver

Add spi resource define for w90p910 spi driver.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years ago[ARM] 5617/1: rename this spi device as w90p910-spi
wanzongshun [Thu, 23 Jul 2009 07:50:20 +0000 (08:50 +0100)]
[ARM] 5617/1: rename this spi device as w90p910-spi

When programming clock api, I named this spi device
as w90p910-usi, now, I think named it as w90p910-spi better

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years ago[ARM] 5616/1: Add mac resource define for w90p910 mac driver
wanzongshun [Thu, 23 Jul 2009 07:47:57 +0000 (08:47 +0100)]
[ARM] 5616/1: Add mac resource define for w90p910 mac driver

Add mac resource define for w90p910 mac driver.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years ago[ARM] 5610/1: Add fmi resource define for w90p910 platform
wanzongshun [Fri, 17 Jul 2009 13:42:33 +0000 (14:42 +0100)]
[ARM] 5610/1: Add fmi resource define for w90p910 platform

Add fmi resource define for w90p910 platform.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years ago[ARM] 5613/1: implement CALLER_ADDRESSx
Uwe Kleine-König [Tue, 21 Jul 2009 08:56:27 +0000 (09:56 +0100)]
[ARM] 5613/1: implement CALLER_ADDRESSx

From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>

As __builtin_return_address(n) doesn't work for ARM with n > 0, the
kernel needs its own implementation.

This fixes many warnings saying:

warning: unsupported argument to '__builtin_return_address'

The new methods and walk_stackframe must not be instrumented because
CALLER_ADDRESSx is used in the various tracers and tracing the tracer is
a bad idea.

What's currently missing is an implementation using unwind tables.  This
is not fatal though, it's just that the tracers don't get enough
information to be really useful.

Note that if both ARM_UNWIND and FRAME_POINTER are enabled,
walk_stackframe uses unwind information.  So in this case the same
implementation is used as when FRAME_POINTER is disabled.

Cc: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years ago[ARM] 5612/1: ep93xx: add platform LEDs
Hartley Sweeten [Mon, 20 Jul 2009 17:22:36 +0000 (18:22 +0100)]
[ARM] 5612/1: ep93xx: add platform LEDs

The EP93xx has two gpio pins specifically assigned to drive
external LEDs.  Add core support for these LEDs.

On the EDB93xx development boards, the rdled is connected to
an external reset circuit.  Turning this led on for an extended
amount of time will cause the circuit to issue a manual reset.

Refer to Cirrus App Note AN258 for more information.

http://www.cirrus.com/en/pubs/appNote/AN258REV2.pdf

This led can be safely used as the system heartbeat with the
ledtrig-heartbeat driver.

echo heartbeat > /sys/class/leds/platform:rdled/trigger

The grled can be used for any desired purpose.

Tested-by: Matthieu Crapet <mcrapet@gmail.com>
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] 5604/1: Add keypad multi-function pin api for w90p910 platform
wanzongshun [Tue, 14 Jul 2009 14:12:26 +0000 (15:12 +0100)]
[ARM] 5604/1: Add keypad multi-function pin api for w90p910 platform

Add keypad multi-function pin api for w90p910 platform.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] 5602/1: Add sub clock api for w90p910 platform
wanzongshun [Tue, 14 Jul 2009 14:10:43 +0000 (15:10 +0100)]
[ARM] 5602/1: Add sub clock api for w90p910 platform

Add sub clock api for w90p910 platform.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] 5601/1: Add HAVE_CLK depends on for w90p910 platform
wanzongshun [Tue, 14 Jul 2009 14:09:54 +0000 (15:09 +0100)]
[ARM] 5601/1: Add HAVE_CLK depends on for w90p910 platform

Add HAVE_CLK depends on for w90p910 platform.

Signed-off-by: Wan ZongShun <mcuos.com@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] 5609/1: ep93xx: add register defines for keypad support
Hartley Sweeten [Thu, 16 Jul 2009 21:46:31 +0000 (22:46 +0100)]
[ARM] 5609/1: ep93xx: add register defines for keypad support

Add missing register defines for ep93xx keypad clock support.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Acked-by: Ryan Mallon <ryan@bluewatersys.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] 5607/1: ep93xx: Use __iomem pointer on syscon write function
Ryan Mallon [Wed, 15 Jul 2009 20:51:59 +0000 (21:51 +0100)]
[ARM] 5607/1: ep93xx: Use __iomem pointer on syscon write function

Change the reg argument of the ep93xx_syscon_swlocked_write function
to be an __iomem pointer. Fixes a number of build warnings.

Signed-off-by: Ryan Mallon <ryan@bluewatersys.com>
Acked-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] 5605/1: Fix ep93xx gpio.c headers
Ryan Mallon [Wed, 15 Jul 2009 20:31:46 +0000 (21:31 +0100)]
[ARM] 5605/1: Fix ep93xx gpio.c headers

Fix a number of build errors in ep93xx gpio.c due to to missing irq.h

Signed-off-by: Ryan Mallon <ryan@bluewatersys.com>
Acked-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] 5600/1: ep93xx: core.c remove cast when copying dev_addr
Hartley Sweeten [Mon, 13 Jul 2009 18:50:10 +0000 (19:50 +0100)]
[ARM] 5600/1: ep93xx: core.c remove cast when copying dev_addr

The MAC address for the ep93xx ethernet driver can be optionally
copied from registers in the controller when booting.  Due to
[ARM] 5573/1: ep93xx: ensure typesafe io, the cast for the source
address is no longer needed.

EP93XX_ETHERNET_BASE is typed as a (void __iomem __force *) so
memcpy_fromio() needs to be used instead of memcpy().

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] 5599/1: MAINTAINERS: update for EP93XX ARM
Hartley Sweeten [Fri, 10 Jul 2009 22:02:59 +0000 (23:02 +0100)]
[ARM] 5599/1: MAINTAINERS: update for EP93XX ARM

Change the maintainer of the EP93XX ARM ARCHITECTURE.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Ryan Mallon <ryan@bluewatersys.com>
Acked-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] 5598/1: ep93xx: core.c typesafe vic_init
Hartley Sweeten [Thu, 9 Jul 2009 22:22:07 +0000 (23:22 +0100)]
[ARM] 5598/1: ep93xx: core.c typesafe vic_init

The EP93XX_VIC{1/2}_BASE defines are now typesafe due to
[ARM] 5573/1: ep93xx: ensure typesafe io.  The (void *) cast
is not needed when calling vic_init().

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Acked-by: Ryan Mallon <ryan@bluewatersys.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] pgtable: rearrange file PTE bit allocation
Russell King [Sat, 11 Jul 2009 15:57:20 +0000 (16:57 +0100)]
[ARM] pgtable: rearrange file PTE bit allocation

For future compatibility, we need to ensure that swap and file Linux
PTEs conform with the hardware PTEs "fault" encoding.  Swap PTEs
already fit in with this, but file PTEs do not.  Shift them by one
bit to ensure that they conform, using bit 2 to distinguish between
swap and file PTEs.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] remove L_PTE_BUFFERABLE and L_PTE_CACHEABLE
Russell King [Sat, 11 Jul 2009 15:51:40 +0000 (16:51 +0100)]
[ARM] remove L_PTE_BUFFERABLE and L_PTE_CACHEABLE

These old symbols are meaningless now that we have memory type
support implemented.  The entire memory type field needs to be
modified rather than just a few bits twiddled.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] remove duplicated #include
Huang Weiyi [Sat, 11 Jul 2009 01:31:58 +0000 (09:31 +0800)]
[ARM] remove duplicated #include

Remove duplicated #include('s) in
  arch/arm/mach-mx3/devices.c
  arch/arm/mach-omap1/mcbsp.c
  arch/arm/mach-omap2/mcbsp.c
  arch/arm/plat-stmp3xxx/pinmux.c

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years ago[ARM] remove duplicated #include
Huang Weiyi [Sat, 11 Jul 2009 01:33:21 +0000 (09:33 +0800)]
[ARM] remove duplicated #include

Remove duplicated #include('s) in
  drivers/spi/amba-pl022.c

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
15 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6
Linus Torvalds [Sat, 11 Jul 2009 02:19:09 +0000 (19:19 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/tiwai/sound-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6:
  ALSA: hda - targa and targa-2ch fix
  ALSA: hda - fix beep tone calculation for IDT/STAC codecs
  ALSA: hda - Missing volume controls for Intel HDA (ALC269/EeePC)
  ALSA: hda - Disable AMD SB600 64bit address support only
  ALSA: hda - Check widget types while parsing capture source in patch_via.c
  ALSA: hda - Fix capture source selection in patch_via.c
  ALSA: hda - Add missing EAPD initialization for VIA codecs
  ALSA: hda - Clean up VT170x dig-in initialization code
  ALSA: hda - Fix error path in the sanity check in azx_pcm_open()
  ALSA: hda - move 8086:fb30 quirk (stac9205) to the proper section
  ASoC: Fix wm8753 register cache size and initialization
  ASoC: add locking to mpc5200-psc-ac97 driver
  ASoC: Fix mpc5200-psc-ac97 to ensure the data ready bit is cleared
  ASoC: Fix register cache initialisation for WM8753

15 years agoisofs: fix Joliet regression
Bartlomiej Zolnierkiewicz [Fri, 10 Jul 2009 10:34:27 +0000 (12:34 +0200)]
isofs: fix Joliet regression

commit 5404ac8e4418ab3d254950ee4f9bcafc1da20b4a ("isofs: cleanup mount
option processing") missed conversion of joliet option flag resulting
in non-working Joliet support.

CC: walt <w41ter@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agotty: Fix the PL2303 private methods for sysrq
Alan Cox [Thu, 9 Jul 2009 12:36:58 +0000 (13:36 +0100)]
tty: Fix the PL2303 private methods for sysrq

PL2303 has private data shovelling methods that also have no fast path. Fix
them to work the same way as the default handler.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agotty: Fix USB kref leak
Alan Cox [Thu, 9 Jul 2009 12:36:22 +0000 (13:36 +0100)]
tty: Fix USB kref leak

The sysrq code acquired a kref leak. Fix it by passing the tty separately
from the caller (thus effectively using the callers kref which all the
callers hold anyway)

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agotty: Sort out the USB sysrq changes that wrecked performance
Alan Cox [Thu, 9 Jul 2009 12:35:52 +0000 (13:35 +0100)]
tty: Sort out the USB sysrq changes that wrecked performance

We can't go around calling all sorts of magic per character functions at
full rate 3G data speed.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agoMerge branch 'linux-next' of git://git.infradead.org/ubi-2.6
Linus Torvalds [Sat, 11 Jul 2009 02:15:34 +0000 (19:15 -0700)]
Merge branch 'linux-next' of git://git.infradead.org/ubi-2.6

* 'linux-next' of git://git.infradead.org/ubi-2.6:
  UBI: fix compilation warnings
  UBI: fix NOR flash recovery
  UBI: nicify image sequence number handling
  UBI: add image sequence number to EC header
  UBI: remove bogus debugging checks
  UBI: add empty eraseblocks verification

15 years agoMerge branch 'linux-next' of git://git.infradead.org/ubifs-2.6
Linus Torvalds [Sat, 11 Jul 2009 02:14:48 +0000 (19:14 -0700)]
Merge branch 'linux-next' of git://git.infradead.org/ubifs-2.6

* 'linux-next' of git://git.infradead.org/ubifs-2.6:
  UBIFS: fix corruption dump
  UBIFS: clean up free space checking
  UBIFS: small amendments in the LEB scanning code
  UBIFS: dump a little more in case of corruptions
  MAINTAINERS: update ahunter's e-mail address
  UBIFS: allow more than one volume to be mounted
  UBIFS: fix assertion warning
  UBIFS: minor spelling and grammar fixes
  UBIFS: fix 64-bit divisions in debug print
  UBIFS: few spelling fixes
  UBIFS: set write-buffer timout to 3-5 seconds
  UBIFS: slightly optimize write-buffer timer usage
  UBIFS: improve debugging messaged
  UBIFS: fix integer overflow warning

15 years agoMerge branch 'for-linus' of git://git390.marist.edu/pub/scm/linux-2.6
Linus Torvalds [Sat, 11 Jul 2009 02:12:51 +0000 (19:12 -0700)]
Merge branch 'for-linus' of git://git390.marist.edu/linux-2.6

* 'for-linus' of git://git390.marist.edu/pub/scm/linux-2.6:
  [S390] define KTIME_SCALAR for 32-bit s390
  [S390] add generic atomic64 support for 31 bit
  [S390] improve suspend/resume error messages
  [S390] set SCHED_OMIT_FRAME_POINTER for s390
  [S390] add __ucmpdi2() helper function
  [S390] perf_counter build fix
  [S390] shutdown actions: save/return rc from init function
  [S390] dasd: correct debugfeature sense dump
  [S390] udelay: disable lockdep to avoid false positives
  [S390] monreader: fix dev_set_drvdata conversion
  [S390] sclp: fix compile error for !SCLP_CONSOLE

15 years agoMerge branch 'for-linus' of git://git.open-osd.org/linux-open-osd
Linus Torvalds [Sat, 11 Jul 2009 02:12:24 +0000 (19:12 -0700)]
Merge branch 'for-linus' of git://git.open-osd.org/linux-open-osd

* 'for-linus' of git://git.open-osd.org/linux-open-osd:
  osdblk: Adjust queue limits to lower device's limits
  osdblk: a Linux block device for OSD objects
  MAINTAINERS: Add osd maintained files (F:)
  exofs: Avoid using file_fsync()
  exofs: Remove IBM copyrights
  exofs: Fix bio leak in error handling path (sync read)

15 years agoMerge branch 'drm-intel-next' of git://git.kernel.org/pub/scm/linux/kernel/git/anholt...
Linus Torvalds [Sat, 11 Jul 2009 02:10:59 +0000 (19:10 -0700)]
Merge branch 'drm-intel-next' of git://git./linux/kernel/git/anholt/drm-intel

* 'drm-intel-next' of git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel: (22 commits)
  drm/i915: Fix for LVDS VBT change on IGDNG
  drm/i915: Zap the GTT mapping when transitioning from untiled to tiled.
  drm/i915: Refactor calls to unmap_mapping_range
  drm/i915: Avoid saving/restore the modesetting registers twice in KMS mode
  drm: Disable the unused connectors explicitly when resuming with KMS.
  drm/i915: Restore the KMS modeset for every activated CRTC
  drm/i915: Fix harmless warning from patch merged after i2c rework.
  drm/i915: Disable GEM when a broken video BIOS takes up the whole aperture.
  drm/i915: Check the LID device to decide whether the LVDS should be initialized
  drm/i915: Move lock to more reasonable location
  drm/i915: Add gtt_offset to gem object list debugfs output
  drm/i915: Remove gtt_bound from drm_i915_gem_object
  drm/i915: Disable VGA output when doing DRM_MODE_DPMS_OFF.
  drm/i915: crt fetch EDID by DVI-I converter on G4x platform
  drm/i915: Don't update display FIFO watermark on IGDNG
  drm/i915: Adjust DisplayPort clocks to use 96MHz reference
  drm/i915: Make driver less chatty
  drm/i915: fix up a raw 64bit divide
  drm/i915: enable sdvo lvds scaling function.
  drm/i915: Set SSC frequency for 8xx chips correctly
  ...

15 years agoDoc: move Documentation/exception.txt into x86 subdir
Amerigo Wang [Fri, 10 Jul 2009 22:02:44 +0000 (15:02 -0700)]
Doc: move Documentation/exception.txt into x86 subdir

exception.txt only explains the code on x86, so it's better to
move it into Documentation/x86 directory.

And also rename it to exception-tables.txt which looks much
more reasonable.

This patch is on top of the previous one.

Signed-off-by: WANG Cong <amwang@redhat.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agoDoc: update Documentation/exception.txt
Amerigo Wang [Fri, 10 Jul 2009 22:02:41 +0000 (15:02 -0700)]
Doc: update Documentation/exception.txt

Update Documentation/exception.txt.
Remove trailing whitespaces in it.

Signed-off-by: WANG Cong <amwang@redhat.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agofuse: Fix build error
Larry Finger [Sat, 11 Jul 2009 01:06:42 +0000 (20:06 -0500)]
fuse: Fix build error

When building v2.6.31-rc2-344-g69ca06c, the following build errors are
found due to missing includes:

 CC [M]  fs/fuse/dev.o
fs/fuse/dev.c: In function ‘request_end’:
fs/fuse/dev.c:289: error: ‘BLK_RW_SYNC’ undeclared (first use in this function)
...
fs/nfs/write.c: In function ‘nfs_set_page_writeback’:
fs/nfs/write.c:207: error: ‘BLK_RW_ASYNC’ undeclared (first use in this function)

Signed-off-by: Larry Finger@lwfinger.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agoMerge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block
Linus Torvalds [Fri, 10 Jul 2009 21:29:58 +0000 (14:29 -0700)]
Merge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block

* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
  cfq-iosched: reset oom_cfqq in cfq_set_request()
  block: fix sg SG_DXFER_TO_FROM_DEV regression
  block: call blk_scsi_ioctl_init()
  Fix congestion_wait() sync/async vs read/write confusion

15 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ryusuke...
Linus Torvalds [Fri, 10 Jul 2009 21:27:21 +0000 (14:27 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/ryusuke/nilfs2

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ryusuke/nilfs2:
  nilfs2: fix disorder in cp count on error during deleting checkpoints
  nilfs2: fix lockdep warning between regular file and inode file
  nilfs2: fix incorrect KERN_CRIT messages in case of write failures
  nilfs2: fix hang problem of log writer which occurs after write failures
  nilfs2: remove unlikely directive causing mis-conversion of error code

15 years agoMerge branch 'core-fixes-for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 10 Jul 2009 21:25:59 +0000 (14:25 -0700)]
Merge branch 'core-fixes-for-linus-2' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'core-fixes-for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  dma-debug: Fix the overlap() function to be correct and readable
  oprofile: reset bt_lost_no_mapping with other stats
  x86/oprofile: rename kernel parameter for architectural perfmon to arch_perfmon
  signals: declare sys_rt_tgsigqueueinfo in syscalls.h
  rcu: Mark Hierarchical RCU no longer experimental
  dma-debug: Put all hash-chain locks into the same lock class
  dma-debug: fix off-by-one error in overlap function

15 years agoMerge branch 'perfcounters-fixes-for-linus' of git://git.kernel.org/pub/scm/linux...
Linus Torvalds [Fri, 10 Jul 2009 21:25:03 +0000 (14:25 -0700)]
Merge branch 'perfcounters-fixes-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'perfcounters-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (50 commits)
  perf report: Add "Fractal" mode output - support callchains with relative overhead rate
  perf_counter tools: callchains: Manage the cumul hits on the fly
  perf report: Change default callchain parameters
  perf report: Use a modifiable string for default callchain options
  perf report: Warn on callchain output request from non-callchain file
  x86: atomic64: Inline atomic64_read() again
  x86: atomic64: Clean up atomic64_sub_and_test() and atomic64_add_negative()
  x86: atomic64: Improve atomic64_xchg()
  x86: atomic64: Export APIs to modules
  x86: atomic64: Improve atomic64_read()
  x86: atomic64: Code atomic(64)_read and atomic(64)_set in C not CPP
  x86: atomic64: Fix unclean type use in atomic64_xchg()
  x86: atomic64: Make atomic_read() type-safe
  x86: atomic64: Reduce size of functions
  x86: atomic64: Improve atomic64_add_return()
  x86: atomic64: Improve cmpxchg8b()
  x86: atomic64: Improve atomic64_read()
  x86: atomic64: Move the 32-bit atomic64_t implementation to a .c file
  x86: atomic64: The atomic64_t data type should be 8 bytes aligned on 32-bit too
  perf report: Annotate variable initialization
  ...

15 years agosched: optimize cond_resched()
Peter Zijlstra [Fri, 10 Jul 2009 12:57:57 +0000 (14:57 +0200)]
sched: optimize cond_resched()

Optimize cond_resched() by removing one conditional.

Currently cond_resched() checks system_state ==
SYSTEM_RUNNING in order to avoid scheduling before the
scheduler is running.

We can however, as per suggestion of Matt, use
PREEMPT_ACTIVE to accomplish that very same.

Suggested-by: Matt Mackall <mpm@selenic.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Matt Mackall <mpm@selenic.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agosched: INIT_PREEMPT_COUNT
Peter Zijlstra [Fri, 10 Jul 2009 12:57:56 +0000 (14:57 +0200)]
sched: INIT_PREEMPT_COUNT

Pull the initial preempt_count value into a single
definition site.

Maintainers for: alpha, ia64 and m68k, please have a look,
your arch code is funny.

The header magic is a bit odd, but similar to the KERNEL_DS
one, CPP waits with expanding these macros until the
INIT_THREAD_INFO macro itself is expanded, which is in
arch/*/kernel/init_task.c where we've already included
sched.h so we're good.

Cc: tony.luck@intel.com
Cc: rth@twiddle.net
Cc: geert@linux-m68k.org
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Matt Mackall <mpm@selenic.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
15 years agodrm/i915: Fix for LVDS VBT change on IGDNG
Zhenyu Wang [Fri, 10 Jul 2009 06:39:59 +0000 (14:39 +0800)]
drm/i915: Fix for LVDS VBT change on IGDNG

IGDNG mobile chip's LVDS data block removes panel fitting
register definition. So this fixes offset for LVDS timing
block parsing. Thanks for Michael Fu to catch this.

Signed-off-by: Zhenyu Wang <zhenyu.z.wang@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agodrm/i915: Zap the GTT mapping when transitioning from untiled to tiled.
Eric Anholt [Fri, 10 Jul 2009 20:02:26 +0000 (13:02 -0700)]
drm/i915: Zap the GTT mapping when transitioning from untiled to tiled.

As of 52dc7d32b88156248167864f77a9026abe27b432, we could leave an old
linear GTT mapping in place, so that apps trying to GTT-mapped write in
tiled data wouldn't get the fence added, and garbage would get displayed.

Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agodrm/i915: Refactor calls to unmap_mapping_range
Chris Wilson [Fri, 10 Jul 2009 07:18:50 +0000 (08:18 +0100)]
drm/i915: Refactor calls to unmap_mapping_range

As we call unmap_mapping_range() twice in identical fashion, refactor
and attempt to explain why we need to call unmap_mapping_range().

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agodma-debug: Fix the overlap() function to be correct and readable
Ingo Molnar [Fri, 10 Jul 2009 19:38:02 +0000 (21:38 +0200)]
dma-debug: Fix the overlap() function to be correct and readable

Linus noticed how unclean and buggy the overlap() function is:

 - It uses convoluted (and bug-causing) positive checks for
   range overlap - instead of using a more natural negative
   check.

 - Even the positive checks are buggy: a positive intersection
   check has four natural cases while we checked only for three,
   missing the (addr < start && addr2 == end) case for example.

 - The variables are mis-named, making it non-obvious how the
   check was done.

 - It needlessly uses u64 instead of unsigned long. Since these
   are kernel memory pointers and we explicitly exclude highmem
   ranges anyway we cannot ever overflow 32 bits, even if we
   could. (and on 64-bit it doesnt matter anyway)

All in one, this function needs a total revamp. I used Linus's
suggestions minus the paranoid checks (we cannot overflow really
because if we get totally bad DMA ranges passed far more things
break in the systems than just DMA debugging). I also fixed a
few other small details i noticed.

Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
15 years agodrm/i915: Avoid saving/restore the modesetting registers twice in KMS mode
Zhao Yakui [Wed, 8 Jul 2009 06:13:14 +0000 (14:13 +0800)]
drm/i915: Avoid saving/restore the modesetting registers twice in KMS mode

In KMS mode we now use the normal mode-setting paths to set the modes
back to the current configuration, so we don't need to also run the more
limited non-KMS implementation of modesetting for resume.

Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agodrm: Disable the unused connectors explicitly when resuming with KMS.
Zhao Yakui [Wed, 8 Jul 2009 06:13:13 +0000 (14:13 +0800)]
drm: Disable the unused connectors explicitly when resuming with KMS.

Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Acked-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agodrm/i915: Restore the KMS modeset for every activated CRTC
Zhao Yakui [Wed, 8 Jul 2009 06:13:12 +0000 (14:13 +0800)]
drm/i915: Restore the KMS modeset for every activated CRTC

Restore the modeset for every activated CRTC in course of resume.
This is realized by calling the function of drm_helper_resume_force_mode.
Note: it is meaningful only for the KMS mode.

https://bugs.freedesktop.org/show_bug.cgi?id=21719
https://bugs.freedesktop.org/show_bug.cgi?id=21708
https://bugs.freedesktop.org/show_bug.cgi?id=22285
https://bugs.freedesktop.org/show_bug.cgi?id=22263

Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agodrm/i915: Fix harmless warning from patch merged after i2c rework.
Eric Anholt [Fri, 10 Jul 2009 19:28:30 +0000 (12:28 -0700)]
drm/i915: Fix harmless warning from patch merged after i2c rework.

Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agodrm/i915: Disable GEM when a broken video BIOS takes up the whole aperture.
Eric Anholt [Thu, 2 Jul 2009 16:30:50 +0000 (09:30 -0700)]
drm/i915: Disable GEM when a broken video BIOS takes up the whole aperture.

This is seen on some G41 systems, where the BIOS will consume all but
a few KB of the aperture.  This should be bad for all operating systems, as
it means that the OS can't dynamically manage memory between graphics and
the rest of the system, and OSes that did static memory management
statically add memory in addition to the BIOS allocation anyway.  So, instead
of working around it, just fail out verbosely.

fd.o bug #21574

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
15 years agoMerge branch 'tracing-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 10 Jul 2009 18:41:41 +0000 (11:41 -0700)]
Merge branch 'tracing-fixes-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'tracing-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  tracing: Fix trace_print_seq()
  kprobes: No need to unlock kprobe_insn_mutex
  tracing/fastboot: Document the need of initcall_debug
  trace_export: Repair missed fields
  tracing: Fix stack tracer sysctl handling

15 years agocfq-iosched: reset oom_cfqq in cfq_set_request()
Vivek Goyal [Thu, 9 Jul 2009 20:13:16 +0000 (22:13 +0200)]
cfq-iosched: reset oom_cfqq in cfq_set_request()

In case memory is scarce, we now default to oom_cfqq. Once memory is
available again, we should allocate a new cfqq and stop using oom_cfqq for
a particular io context.

Once a new request comes in, check if we are using oom_cfqq, and if yes,
try to allocate a new cfqq.

Tested the patch by forcing the use of oom_cfqq and upon next request thread
realized that it was using oom_cfqq and it allocated a new cfqq.

Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>