12 years agosh: Move out cayman-specific panic handler code to its own file.
Paul Mundt [Fri, 8 May 2009 08:41:59 +0000]
sh: Move out cayman-specific panic handler code to its own file.

This moves out the cayman-specific panic handler code to a better
location, and leaves the generic implementation a simple stub that is
still used under emulation.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Kill off unused sh64 debug code.
Paul Mundt [Fri, 8 May 2009 08:36:57 +0000]
sh: Kill off unused sh64 debug code.

None of the print_page() code and associated helpers are presently used
by anything in-tree, so just kill it off.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Kill off sh64's hand-rolled syscall tracer.
Paul Mundt [Fri, 8 May 2009 08:25:35 +0000]
sh: Kill off sh64's hand-rolled syscall tracer.

This is no longer necessary, as there are now sufficient generic
alternatives available.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Kill off the global rtc_lock with extreme prejudice.
Paul Mundt [Fri, 8 May 2009 07:59:05 +0000]
sh: Kill off the global rtc_lock with extreme prejudice.

Now that all of the possible users for rtc_lock have gone away, it is no
longer necessary to keep this lock definition around.

This follows several other architectures that have either recently
dropped it or never supported it in the first place.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: mach-sh03: Give the sh03 rtc its own spinlock.
Paul Mundt [Fri, 8 May 2009 07:57:35 +0000]
sh: mach-sh03: Give the sh03 rtc its own spinlock.

This converts the sh03 rtc code off of using the global rtc_lock and on
to its own spinlock. There are no other possible users of the rtc_lock,
so serializing with it is not necessary.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Kill off dead timer sysclass pm hooks.
Paul Mundt [Fri, 8 May 2009 07:47:48 +0000]
sh: Kill off dead timer sysclass pm hooks.

With the conversion to generic clockevents these are completely unused,
so just kill it off.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Kill off dead handle_timer_tick() code.
Paul Mundt [Fri, 8 May 2009 07:44:00 +0000]
sh: Kill off dead handle_timer_tick() code.

Nothing is using this anymore now that we have fully converted to generic
time, so kill it off completely.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Wire up GENERIC_CMOS_UPDATE for the platforms that need it.
Paul Mundt [Fri, 8 May 2009 07:36:13 +0000]
sh: Wire up GENERIC_CMOS_UPDATE for the platforms that need it.

Now that everything has converted over to generic timekeeping, we need an
alternate method for keeping the RTC updated for those platforms that are
still using the rtc_sh_get/set_time pairs, presently limited to SH-03 and
the Dreamcast. This wires up the GENERIC_CMOS_UPDATE hooks for those to
maintain the same behaviour.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Rename arch/sh/kernel/time_32.c to arch/sh/kernel/time.c.
Paul Mundt [Fri, 8 May 2009 07:17:36 +0000]
sh: Rename arch/sh/kernel/time_32.c to arch/sh/kernel/time.c.

This is now fully generic, and used both by _32 and _64 variants.
Rename it accordingly.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Finish the sh64 migration off of ARCH_USES_GETTIMEOFFSET.
Paul Mundt [Fri, 8 May 2009 07:12:17 +0000]
sh: Finish the sh64 migration off of ARCH_USES_GETTIMEOFFSET.

This adds sh_tmu support to the SH-5 subtypes, which subsequently allows
us to kill off time_64.c and use the now generic time_32.c. As a bonus,
SH-5 now supports highres timers and tickless for the first time.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Move out rtc-sh registration from time_64.c to setup-sh5.c
Paul Mundt [Fri, 8 May 2009 06:39:25 +0000]
sh: Move out rtc-sh registration from time_64.c to setup-sh5.c

Now that the onchip_remap() mess is sorted out, the rtc-sh support code
for SH-5 can follow the same approach as the other CPUs.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Enable new TMU driver support for all SH-3 and SH-4 CPUs.
Paul Mundt [Fri, 8 May 2009 06:28:15 +0000]
sh: Enable new TMU driver support for all SH-3 and SH-4 CPUs.

The TMU block is supported on all SH-3 and SH-4 subtypes, so just select
it there, rather than conditionalizing it per subtype.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: TMU platform data for sh7785
Magnus Damm [Thu, 7 May 2009 10:55:37 +0000]
sh: TMU platform data for sh7785

This patch adds TMU platform data for sh7785. Both clockevent
and clocksource support is enabled.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: r7785 highlander clock fixes
Magnus Damm [Thu, 7 May 2009 10:44:55 +0000]
sh: r7785 highlander clock fixes

Update the r7785 highlander defconfig to fix
PCLK value and that mode4 is set high.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: call clock framework init() callback once
Magnus Damm [Thu, 7 May 2009 10:31:39 +0000]
sh: call clock framework init() callback once

Make sure that clk->ops->init() only gets called once in
the case of CLK_ALWAYS_ENABLED. Without this patch the
init() callback may be called multiple times.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: sh7785 early scif fix
Magnus Damm [Thu, 7 May 2009 10:17:44 +0000]
sh: sh7785 early scif fix

This patch moves the SH4 case of EARLY_SCIF_CONSOLE_PORT
so the SH7785 default value gets used. Without this patch
the value for SH7785 is set to 0xffe80000.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Kill off unused SH-5 irq_describe cruft.
Paul Mundt [Thu, 7 May 2009 09:17:20 +0000]
sh: Kill off unused SH-5 irq_describe cruft.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Integrate the SH-5 onchip_remap() more coherently.
Paul Mundt [Thu, 7 May 2009 09:10:27 +0000]
sh: Integrate the SH-5 onchip_remap() more coherently.

Presently this is special-cased for early initialization. While there are
situations where these static early initializations are still necessary,
with minor changes it is possible to use this for the regular ioremap
implementation as well. This allows us to kill off the special-casing for
the remap completely and to start tidying up all of the SH-5
special-casing in drivers.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: sh64 still needs ARCH_USES_GETTIMEOFFSET temporarily.
Paul Mundt [Thu, 7 May 2009 08:55:09 +0000]
sh: sh64 still needs ARCH_USES_GETTIMEOFFSET temporarily.

sh64 is still using this, so re-enable it temporarily while SH-5 gets
converted to use the generic code.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Handle shm_align_mask also for HAVE_ARCH_UNMAPPED_AREA_TOPDOWN.
Paul Mundt [Thu, 7 May 2009 07:38:16 +0000]
sh: Handle shm_align_mask also for HAVE_ARCH_UNMAPPED_AREA_TOPDOWN.

Presently shm_align_mask is only looked at for the bottom up case, but we
still want this for proper colouring constraints in the topdown case.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Flag IRQSTACKS as BROKEN for now.
Paul Mundt [Thu, 7 May 2009 06:24:36 +0000]
sh: Flag IRQSTACKS as BROKEN for now.

There still seems to be some stack corruption to sort out here, so flag
this as BROKEN until this issue is sorted out.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Move dummy clockevents broadcast timer to its new home.
Paul Mundt [Sun, 3 May 2009 09:29:27 +0000]
sh: Move dummy clockevents broadcast timer to its new home.

The old arch/sh/kernel/timers/ directly will be going away completely
once the rest of the TMU users are migrated, so move the dummy broadcast
driver up a level in preparation.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Kill off the now unused ARCH_USES_GETTIMEOFFSET code.
Paul Mundt [Sun, 3 May 2009 09:18:14 +0000]
sh: Kill off the now unused ARCH_USES_GETTIMEOFFSET code.

Now that the stragglers (MTU2/CMT/etc.) have been rewritten and we are
selecting both GENERIC_TIME and GENERIC_CLOCKEVENTS, the get_offset()
timer op is completely unused. As a result, we are now able to kill off
the ARCH_USES_GETTIMEOFFSET references.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: select both GENERIC_TIME and GENERIC_CLOCKEVENTS.
Paul Mundt [Sun, 3 May 2009 09:12:26 +0000]
sh: select both GENERIC_TIME and GENERIC_CLOCKEVENTS.

Now that the rest of the timers that didn't support clockevents have been
rewritten, both of these can be enabled by default.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agoclocksource: sh_mtu2/cmt_register() should be static.
Paul Mundt [Sun, 3 May 2009 09:05:42 +0000]
clocksource: sh_mtu2/cmt_register() should be static.

Neither of these need to be exported, so just make them static.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: Consolidate MTU2/CMT/TMU timer platform data.
Paul Mundt [Sun, 3 May 2009 08:57:17 +0000]
sh: Consolidate MTU2/CMT/TMU timer platform data.

All of the SH timers use a roughly identical structure for platform data,
which presently is broken out for each block. Consolidate all of these
definitions, as there is no reason for them to be broken out in the first
place.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: TMU platform data for sh7722
Magnus Damm [Fri, 1 May 2009 06:58:52 +0000]
sh: TMU platform data for sh7722

This patch adds TMU platform data for sh7722. Only clockevent
mode is enabled for now, clocksource requires this patch:
"clocksource: setup mult_orig in clocksource_enable()"

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agoclocksource: SuperH TMU Timer driver
Magnus Damm [Fri, 1 May 2009 06:51:00 +0000]
clocksource: SuperH TMU Timer driver

This patch adds a TMU driver for the SuperH architecture.

The TMU driver is a platform driver with early platform
support to allow using a TMU channel as clockevent or
clocksource during system bootup or later.

Clocksource or clockevent can be selected.
Both periodic and oneshot clockevents are supported.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: remove old MTU2 driver
Magnus Damm [Thu, 30 Apr 2009 07:12:09 +0000]
sh: remove old MTU2 driver

This patch removes the old MTU2 driver (CONFIG_SH_MTU2/timer-mtu2.c)

As replacement, select the sh_cmt driver with CONFIG_SH_TIMER_MTU2
and configure timer channel using platform data.

If multiple MTU channels are enabled using platform data, use the
earlytimer parameter on the kernel command line to select channel.
For instance, use "earlytimer=sh_mtu2.0" to select the first channel.

To verify which timer is being used, look at printouts or the timer
irq count in /proc/interrupts.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: sh2a MTU2 platform data
Magnus Damm [Thu, 30 Apr 2009 07:06:26 +0000]
sh: sh2a MTU2 platform data

This patch adds MTU2 platform data for the following cpus:
 - sh7201 (3/5 channels)
 - sh7203/sh7263 (2/4 channels)
 - sh7206 (3/5 channels)
 - MXG (3/5 channels)

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agoclocksource: SuperH MTU2 Timer driver
Magnus Damm [Thu, 30 Apr 2009 07:02:49 +0000]
clocksource: SuperH MTU2 Timer driver

This patch adds a MTU2 driver for the SuperH architecture.

The MTU2 driver is a platform driver with early platform
support to allow using a MTU2 channel as only clockevent
during system bootup.

Clocksource on sh2a is currently unsupported due to code
generation issues with 64-bit math, so at this point only
periodic clockevent support is in place.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agotime: sh: convert to use arch_getoffset() infrastructure
john stultz [Fri, 1 May 2009 20:10:28 +0000]
time: sh: convert to use arch_getoffset() infrastructure

Convert sh to use GENERIC_TIME via the arch_getoffset() infrastructure.

Signed-off-by: John Stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: remove obsolete hw_interrupt_type
Thomas Gleixner [Sat, 2 May 2009 20:01:20 +0000]
sh: remove obsolete hw_interrupt_type

Impact: cleanup

Convert the last remaining users to struct irq_chip and remove the
define.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: remove obsolete no_irq_type
Thomas Gleixner [Sat, 2 May 2009 20:00:44 +0000]
sh: remove obsolete no_irq_type

Impact: cleanup

convert the last remaining users to no_irq_chip

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agoMerge branch 'timers/clocksource' of git://git.kernel.org/pub/scm/linux/kernel/git...
Paul Mundt [Sun, 3 May 2009 08:24:55 +0000]
Merge branch 'timers/clocksource' of git://git./linux/kernel/git/x86/linux-2.6-tip

12 years agotimekeeping: create arch_gettimeoffset infrastructure
john stultz [Fri, 1 May 2009 20:10:26 +0000]
timekeeping: create arch_gettimeoffset infrastructure

Some arches don't supply their own clocksource. This is mainly the
case in architectures that get their inter-tick times by reading the
counter on their interval timer.  Since these timers wrap every tick,
they're not really useful as clocksources.  Wrapping them to act like
one is possible but not very efficient. So we provide a callout these
arches can implement for use with the jiffies clocksource to provide
finer then tick granular time.

[ Impact: ease the migration to generic time keeping ]

Signed-off-by: John Stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

12 years agoclocksource: setup mult_orig in clocksource_enable()
Magnus Damm [Fri, 1 May 2009 05:45:46 +0000]
clocksource: setup mult_orig in clocksource_enable()

Setup clocksource mult_orig in clocksource_enable().

Clocksource drivers can save power by using keeping the
device clock disabled while the clocksource is unused.

In practice this means that the enable() and disable()
callbacks perform clk_enable() and clk_disable().

The enable() callback may also use clk_get_rate() to get
the clock rate from the clock framework. This information
can then be used to calculate the shift and mult variables.

Currently the mult_orig variable is setup from mult at
registration time only. This is conflicting with the above
case since the clock is disabled and the mult variable is
not yet calculated at the time of registration.

Moving the mult_orig setup code to clocksource_enable()
allows us to both handle the common case with no enable()
callback and the mult-changed-after-enable() case.

[ Impact: allow dynamic clock source usage ]

Signed-off-by: Magnus Damm <damm@igel.co.jp>
LKML-Reference: <20090501054546.8193.10688.sendpatchset@rx1.opensource.se>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

12 years agoLinux 2.6.30-rc4
Linus Torvalds [Thu, 30 Apr 2009 04:48:16 +0000]
Linux 2.6.30-rc4

12 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ecryptfs...
Linus Torvalds [Thu, 30 Apr 2009 04:25:53 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/ecryptfs/ecryptfs-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ecryptfs/ecryptfs-2.6:
  eCryptfs: Fix min function comparison warning
  ecryptfs: fix printk format warning

12 years agoMerge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab...
Linus Torvalds [Thu, 30 Apr 2009 04:25:39 +0000]
Merge branch 'for_linus' of git://git./linux/kernel/git/mchehab/linux-2.6

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6:
  V4L/DVB (11652): au0828: fix kernel oops regression on USB disconnect.
  V4L/DVB (11626): cx23885: Two fixes for DViCO FusionHDTV DVB-T Dual Express
  V4L/DVB (11612): mx3_camera: Fix compilation with CONFIG_PM
  V4L/DVB (11570): patch: s2255drv: fix race condition on set mode
  V4L/DVB (11568): cx18: Fix the handling of i2c bus registration error
  V4L/DVB (11561a): move media after i2c
  V4L/DVB (11516): drivers/media/video/saa5246a.c: fix use-after-free
  V4L/DVB (11515): drivers/media/video/saa5249.c: fix use-after-free and leak
  V4L/DVB (11494a): cx231xx Kconfig fixes
  V4L/DVB (11494): cx18: Send correct input routing value to external audio multiplexers

12 years agosh: remove old CMT driver
Magnus Damm [Thu, 30 Apr 2009 04:09:26 +0000]
sh: remove old CMT driver

This patch removes the old CMT driver (CONFIG_SH_CMT/timer-cmt.c)

As replacement, select the sh_cmt driver with CONFIG_SH_TIMER_CMT
and configure timer channel using platform data.

If multiple CMT channels are enabled using platform data, use the
earlytimer parameter on the kernel command line to select channel.
For instance, use "earlytimer=sh_cmt.0" to select the first channel.

To verify which timer is being used, look at printouts or the timer
irq count in /proc/interrupts.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: sh2/sh2a 16-bit CMT platform data
Magnus Damm [Thu, 30 Apr 2009 04:08:18 +0000]
sh: sh2/sh2a 16-bit CMT platform data

This patch adds 16-bit cmt platform data for the following cpus:
 - sh7619 (2 channels)
 - sh7203/sh7263 (2 channels)
 - sh7206 (2 channels)

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agoclocksource: sh_cmt 16-bit fixes
Magnus Damm [Wed, 29 Apr 2009 14:50:37 +0000]
clocksource: sh_cmt 16-bit fixes

This patch contains various fixes for 16-bit cmt hardware.
With this applied periodic clockevents work fine on sh7203.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agosh: pass through ioremap() for non-mmu processors.
Magnus Damm [Thu, 30 Apr 2009 03:56:37 +0000]
sh: pass through ioremap() for non-mmu processors.

All 32-bit SuperH processors currently go through __ioremap_mode()
and check for IO_TRAPPED and directly mapped segments. With this
patch we simplify the MMU less case with a pass through version of
 __ioremap_mode() which just returns the physical address.

The effects of this is change are:
 - fix non-MMU ioremap() of high address hardware blocks (sh7203 CMT)
 - make sure IO_TRAPPED is not selected

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agoV4L/DVB (11652): au0828: fix kernel oops regression on USB disconnect.
Devin Heitmueller [Tue, 28 Apr 2009 16:14:07 +0000]
V4L/DVB (11652): au0828: fix kernel oops regression on USB disconnect.

A regression was introduced in hg changeset 33810c734a0d, which resulted in
a kernel panic whenever the device was disconnected from USB.  The call to
4l2_device_register() was overwriting the pointer for usb_set_intfdata(), so
when au0828_usb_disconnect() was called, the usb_get_intfdata() returned a
pointer to the v4l2_device instead of the au0828_dev structure.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

12 years agoV4L/DVB (11626): cx23885: Two fixes for DViCO FusionHDTV DVB-T Dual Express
Christopher Pascoe [Mon, 27 Apr 2009 14:27:04 +0000]
V4L/DVB (11626): cx23885: Two fixes for DViCO FusionHDTV DVB-T Dual Express

Two fixes for DViCO FusionHDTV DVB-T Dual Express:

 * Reset correct tuner when reinitializing xc3028.
 * Disable the I2C gate control to avoid locking up the I2C bus.

Tested-by: John Knops <jknops@australiaonline.net.au>
Reviewed-by: Steven Toth <stoth@linuxtv.org>
Signed-off-by: Christopher Pascoe <linuxdvb@itee.uq.edu.au>
Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

12 years agoV4L/DVB (11612): mx3_camera: Fix compilation with CONFIG_PM
Sascha Hauer [Fri, 24 Apr 2009 15:58:24 +0000]
V4L/DVB (11612): mx3_camera: Fix compilation with CONFIG_PM

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

12 years agoV4L/DVB (11570): patch: s2255drv: fix race condition on set mode
Dean Anderson [Mon, 20 Apr 2009 22:07:44 +0000]
V4L/DVB (11570): patch: s2255drv: fix race condition on set mode

set_modeready flag must be set before command sent to USB in
s2255_write_config.

Signed-off-by: Dean Anderson <dean@sensoray.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

12 years agoV4L/DVB (11568): cx18: Fix the handling of i2c bus registration error
Jean Delvare [Fri, 17 Apr 2009 13:56:51 +0000]
V4L/DVB (11568): cx18: Fix the handling of i2c bus registration error

* Return actual error values as returned by the i2c subsystem, rather
  than 0 or 1.
* If the registration of the second bus fails, unregister the first one
  before exiting, otherwise we are leaking resources.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

12 years agoV4L/DVB (11561a): move media after i2c
Guennadi Liakhovetski [Tue, 21 Apr 2009 07:22:38 +0000]
V4L/DVB (11561a): move media after i2c

Currently drivers/media drivers are linked very early - directly after
base, block, misc, and mfd and before ata, scsi, ide, input, firewire,
usb, and i2c. This breaks static build of video4linux drivers, that use
generic CPU i2c adapter drivers and the v4l2-subdev subsystem, because
during video4linux probing the v4l2-subdev core requires a struct
i2c_adapter context, which cannot be satisfied before the i2c subsystem is
initialised. Moving drivers/media after drivers/i2c fixes this problem.

The best way to trigger action is by submitting a patch:-) So, let's see
what comes out of it - on the one hand I don't see any reason why media
has to be linked this early, and nobody was able to give me one yesterday
as this problem has been discussed on linux-media, OTOH, maybe indeed it
would be better to move i2c the whole way up above media, but that'd be
much bigger of a change, I think.
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Acked-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

12 years agoV4L/DVB (11516): drivers/media/video/saa5246a.c: fix use-after-free
Dan Carpenter [Tue, 14 Apr 2009 22:51:30 +0000]
V4L/DVB (11516): drivers/media/video/saa5246a.c: fix use-after-free

I lowered the kfree(t) down a couple lines and removed the superflous
"t->vdev = NULL;"

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

12 years agoV4L/DVB (11515): drivers/media/video/saa5249.c: fix use-after-free and leak
Dan Carpenter [Tue, 14 Apr 2009 22:50:33 +0000]
V4L/DVB (11515): drivers/media/video/saa5249.c: fix use-after-free and leak

I moved the kfree() down a couple lines.  t->vdev is going to be in freed
memory so there is no point setting it to NULL.  I added a kfree(t) on a

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

12 years agoV4L/DVB (11494a): cx231xx Kconfig fixes
Mauro Carvalho Chehab [Tue, 14 Apr 2009 16:56:58 +0000]
V4L/DVB (11494a): cx231xx Kconfig fixes

selecting ALSA module breaks if !SND. Just remove select.

While here, let's fix the whitespacing at the Kconfig.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

12 years agoV4L/DVB (11494): cx18: Send correct input routing value to external audio multiplexers
Andy Walls [Tue, 14 Apr 2009 00:43:31 +0000]
V4L/DVB (11494): cx18: Send correct input routing value to external audio multiplexers

A late v4l2_subdev framework change accidentally sent the audio input
routing value to the external multiplexer, instead of the muxer input routing
value to the external multiplexer.  This change corrects that error.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

12 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Wed, 29 Apr 2009 14:55:45 +0000]
Merge git://git./linux/kernel/git/davem/net-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (24 commits)
  e100: do not go D3 in shutdown unless system is powering off
  netfilter: revised locking for x_tables
  Bluetooth: Fix connection establishment with low security requirement
  Bluetooth: Add different pairing timeout for Legacy Pairing
  Bluetooth: Ensure that HCI sysfs add/del is preempt safe
  net: Avoid extra wakeups of threads blocked in wait_for_packet()
  net: Fix typo in net_device_ops description.
  ipv4: Limit size of route cache hash table
  Add reference to CAPI 2.0 standard
  Documentation/isdn/INTERFACE.CAPI
  update Documentation/isdn/00-INDEX
  ixgbe: Fix WoL functionality for 82599 KX4 devices
  veth: prevent oops caused by netdev destructor
  xfrm: wrong hash value for temporary SA
  forcedeth: tx timeout fix
  net: Fix LL_MAX_HEADER for CONFIG_TR_MODULE
  mlx4_en: Handle page allocation failure during receive
  mlx4_en: Fix cleanup flow on cq activation
  vlan: update vlan carrier state for admin up/down
  netfilter: xt_recent: fix stack overread in compat code
  ...

12 years agoe100: do not go D3 in shutdown unless system is powering off
Thadeu Lima de Souza Cascardo [Wed, 29 Apr 2009 05:42:39 +0000]
e100: do not go D3 in shutdown unless system is powering off

After experimenting with kexec with the last merges after 2.6.29, I've
had some problems when probing e100.  It would not read the eeprom.  After
some bisects, I realized this has been like that since forever (at least
2.6.18).  The problem is that shutdown is doing the same thing that
suspend does and puts the device in D3 state.  I couldn't find a way to
get the device back to a sane state in the probe function.  So, based on
some similar patches from Rafael J. Wysocki for e1000, e1000e, and ixgbe,
I wrote this one for e100.

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Acked-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

12 years agoMerge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/holtmann/bluet...
David S. Miller [Wed, 29 Apr 2009 05:40:49 +0000]
Merge branch 'master' of git://git./linux/kernel/git/holtmann/bluetooth-2.6

12 years agonetfilter: revised locking for x_tables
Stephen Hemminger [Wed, 29 Apr 2009 05:36:33 +0000]
netfilter: revised locking for x_tables

The x_tables are organized with a table structure and a per-cpu copies
of the counters and rules. On older kernels there was a reader/writer
lock per table which was a performance bottleneck. In 2.6.30-rc, this
was converted to use RCU and the counters/rules which solved the performance
problems for do_table but made replacing rules much slower because of
the necessary RCU grace period.

This version uses a per-cpu set of spinlocks and counters to allow to
table processing to proceed without the cache thrashing of a global
reader lock and keeps the same performance for table updates.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

12 years agoMerge branch 'drm-intel-next' of git://git.kernel.org/pub/scm/linux/kernel/git/anholt...
Linus Torvalds [Wed, 29 Apr 2009 00:21:20 +0000]
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:
  drm/i915: fix up error path leak in i915_cmdbuffer
  drm/i915: fix unpaired i915 device mutex on entervt failure.
  drm/i915: add support for G41 chipset
  drm/i915: Enable ASLE if present
  drm/i915: Unregister ACPI video driver when exiting
  drm/i915: Register ACPI video even when not modesetting
  drm/i915: fix transition to I915_TILING_NONE
  drm/i915: Don't let an oops get triggered from irq_emit without dma init.
  drm/i915: allow tiled front buffers on 965+

12 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland...
Linus Torvalds [Wed, 29 Apr 2009 00:21:07 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/roland/infiniband

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (21 commits)
  RDMA/nes: Update iw_nes version
  RDMA/nes: Fix error path in nes_accept()
  RDMA/nes: Fix hang issues for large cluster dynamic connections
  RDMA/nes: Increase rexmit timeout interval
  RDMA/nes: Check for sequence number wrap-around
  RDMA/nes: Do not set apbvt entry for loopback
  RDMA/nes: Fix unused variable compile warning when INFINIBAND_NES_DEBUG=n
  RDMA/nes: Fix fw_ver in /sys
  RDMA/nes: Set trace length to 1 inch for SFP_D
  RDMA/nes: Enable repause timer for port 1
  RDMA/nes: Correct CDR loop filter setting for port 1
  RDMA/nes: Modify thermo mitigation to flip SerDes1 ref clk to internal
  RDMA/nes: Fix resource issues in nes_create_cq() and nes_destroy_cq()
  RDMA/nes: Remove root_256()'s unused pbl_count_256 parameter
  mlx4_core: Fix memory leak in mlx4_enable_msi_x()
  IB/mthca: Fix timeout for INIT_HCA and a few other commands
  RDMA/cxgb3: Don't zero QP attrs when moving to IDLE
  RDMA/nes: Fix bugs in nes_reg_phys_mr()
  RDMA/nes: Fix compiler warning at nes_verbs.c:1955
  IPoIB: Disable NAPI while CQ is being drained
  ...

12 years agom68k: arch/m68k/kernel/sun3-head.S needs <linux/init.h>
Geert Uytterhoeven [Tue, 28 Apr 2009 19:20:11 +0000]
m68k: arch/m68k/kernel/sun3-head.S needs <linux/init.h>

Fix this:

  arch/m68k/kernel/sun3-head.S: Assembler messages:
  arch/m68k/kernel/sun3-head.S:32: Error: Unknown operator -- statement `__head' ignored

Introduced by commit 6f335cab0431d5df4995bcd4fd952d4c746d5a86 ("m68k:
convert to use __HEAD and HEAD_TEXT macros."), which started using
__HEAD without adding the appropriate include.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

12 years agoMerge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block
Linus Torvalds [Tue, 28 Apr 2009 23:06:48 +0000]
Merge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block

* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
  bio: fix memcpy corruption in bio_copy_user_iov()
  hd: fix locking
  mg_disk: fix CONFIG_LBD=y warning
  mg_disk: fix locking

12 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/lrg/voltage-2.6
Linus Torvalds [Tue, 28 Apr 2009 23:06:33 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/lrg/voltage-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/lrg/voltage-2.6:
  regulator: Fix default constraints for fixed voltage regulators
  regulator/bq24022: fix bug in is_enabled function
  regulator/virtual: fix strings compare predicates
  regulator core: fix double-free in regulator_register() error path
  drivers/regulator: fix when type is different from REGULATOR_VOLTAGE or REGULATOR_CURRENT
  unreachable code in drms_uA_update()
  regulator: fix header file missing kernel-doc

12 years agoMerge branches 'cxgb3', 'ipoib', 'mthca', 'mlx4' and 'nes' into for-linus
Roland Dreier [Tue, 28 Apr 2009 23:01:31 +0000]
Merge branches 'cxgb3', 'ipoib', 'mthca', 'mlx4' and 'nes' into for-linus

12 years agoMerge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc
Linus Torvalds [Tue, 28 Apr 2009 22:55:32 +0000]
Merge branch 'merge' of git://git./linux/kernel/git/paulus/powerpc

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc:
  powerpc/ps3: Fix build error on UP
  powerpc/cell: Select PCI for IBM_CELL_BLADE AND CELLEB
  powerpc: ppc32 needs elf_read_implies_exec()
  powerpc/86xx: Add device_type entry to soc for ppc9a
  powerpc/44x: Correct memory size calculation for denali-based boards
  maintainers: Fix PowerPC 4xx git tree
  powerpc: fix for long standing bug noticed by gcc 4.4.0
  Revert "powerpc: Add support for early tlbilx opcode"

12 years agopowerpc: Revert switch to TEXT_TEXT in linker script
Tim Abbott [Tue, 28 Apr 2009 15:47:15 +0000]
powerpc: Revert switch to TEXT_TEXT in linker script

Commit edada399 broke the build on 64-bit powerpc because it moved the
__ftr_alt_* sections of a file away from the .text section, causing
link failures due to relative conditional branch targets being too far
away from the branch instructions.  This happens on pretty much all
64-bit powerpc configs.

This change reverts commit edada399 while preserving the update from
the *.refok sections to .ref.text that has happened since.

Signed-off-by: Tim Abbott <tabbott@mit.edu>
Requested-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

12 years agobio: fix memcpy corruption in bio_copy_user_iov()
FUJITA Tomonori [Tue, 28 Apr 2009 18:24:29 +0000]
bio: fix memcpy corruption in bio_copy_user_iov()

st driver uses blk_rq_map_user() in order to just build a request out
of page frames. In this case, map_data->offset is a non zero value and
iov[0].iov_base is NULL. We need to increase nr_pages for that.

Cc: stable@kernel.org
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

12 years agohd: fix locking
Tejun Heo [Tue, 28 Apr 2009 03:38:33 +0000]
hd: fix locking

hd dance around local irq and HD_IRQ enable without achieving much.
It ends up transferring data from irq handler with both local irq and
HD_IRQ disabled.  The only place it actually does something is while
transferring the first block of a request which it does with HD_IRQ
disabled but local irq enabled.

Unfortunately, the dancing is horribly broken from locking POV.  IRQ
and timeout handlers access block queue without grabbing the queue
lock and running the driver in SMP configuration crashes the whole
machine pretty quickly.

Remove meaningless irq enable/disable dancing and add proper locking
in issue, irq and timeout paths.

Signed-off-by: Tejun Heo <tj@kernel.org>

12 years agomg_disk: fix CONFIG_LBD=y warning
Bartlomiej Zolnierkiewicz [Tue, 28 Apr 2009 03:38:33 +0000]
mg_disk: fix CONFIG_LBD=y warning

drivers/block/mg_disk.c: In function ‘mg_dump_status’:
drivers/block/mg_disk.c:265: warning: format ‘%ld’ expects type ‘long int’, but
argument 2 has type ‘sector_t’

[ Impact: kill build warning ]

Cc: unsik Kim <donari75@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>

12 years agomg_disk: fix locking
Tejun Heo [Tue, 28 Apr 2009 03:38:32 +0000]
mg_disk: fix locking

IRQ and timeout handlers call functions which expect locked queue lock
without locking it.  Fix it.

While at it, convert 0s used as null pointer constant to NULLs.

[ Impact: fix locking, cleanup ]

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: unsik Kim <donari75@gmail.com>

12 years agoregulator: Fix default constraints for fixed voltage regulators
Mark Brown [Tue, 28 Apr 2009 10:09:38 +0000]
regulator: Fix default constraints for fixed voltage regulators

Default voltage constraints were being provided for fixed voltage
regulator where board constraints were not provided but these constraints
used INT_MIN as the default minimum voltage which is not a valid value
since it is less than zero. Use 1uV instead.

Also set the default values we set in the constraints themselves since
otherwise the max_uV constraint we determine will not be stored in the
actual constraint strucutre and will therefore not be used.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>

12 years agoregulator/bq24022: fix bug in is_enabled function
Philipp Zabel [Tue, 28 Apr 2009 11:34:14 +0000]
regulator/bq24022: fix bug in is_enabled function

This seems to be fallout from last October's regulator core rework.
It got noticed only because of recent regulator framework changes.

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>

12 years agoregulator/virtual: fix strings compare predicates
Mike Rapoport [Sun, 26 Apr 2009 08:49:30 +0000]
regulator/virtual: fix strings compare predicates

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>

12 years agoregulator core: fix double-free in regulator_register() error path
Paul Walmsley [Sat, 25 Apr 2009 11:28:36 +0000]
regulator core: fix double-free in regulator_register() error path

During regulator registration, any error after device_register() will
cause a double-free on the struct regulator_dev 'rdev'.  The bug is in
drivers/regulator/core.c:regulator_register():

...
scrub:
device_unregister(&rdev->dev);
clean:
kfree(rdev);                           <---
rdev = ERR_PTR(ret);
goto out;
...

device_unregister() calls regulator_dev_release() which frees rdev.  The
subsequent kfree corrupts memory and causes some OMAP3 systems to oops on
boot in regulator_get().

Applies against 2.6.30-rc3.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>

12 years agodrivers/regulator: fix when type is different from REGULATOR_VOLTAGE or REGULATOR_CURRENT
Diego Liziero [Tue, 14 Apr 2009 01:04:47 +0000]
drivers/regulator: fix when type is different from REGULATOR_VOLTAGE or REGULATOR_CURRENT

When regulator_desc->type is something different from REGULATOR_VOLTAGE or REGULATOR_CURRENT
the if should probably return ERR_PTR(-EINVAL)

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

@@ expression E; constant C; @@
(
- !E == C
+ E != C
)

Signed-off-by: Diego Liziero <diegoliz@gmail.com>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>

12 years agounreachable code in drms_uA_update()
Dan Carpenter [Wed, 8 Apr 2009 10:52:39 +0000]
unreachable code in drms_uA_update()

I removed the extra semi-colon and indented the return statement.

The unreachable code was found by smatch (http://repo.or.cz/w/smatch.git).
The patch was compile tested.

regards,
dan carpenter

Signed-off-by: Dan Carpenter <error27@gmail.com>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>

12 years agoregulator: fix header file missing kernel-doc
Randy Dunlap [Sat, 4 Apr 2009 04:31:30 +0000]
regulator: fix header file missing kernel-doc

Add regulator header file missing kernel-doc:

Warning(include/linux/regulator/driver.h:117): No description found for parameter 'set_mode'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
cc: Liam Girdwood <lrg@slimlogic.co.uk>
cc: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>

12 years agoBluetooth: Fix connection establishment with low security requirement
Marcel Holtmann [Tue, 28 Apr 2009 16:04:55 +0000]
Bluetooth: Fix connection establishment with low security requirement

The Bluetooth 2.1 specification introduced four different security modes
that can be mapped using Legacy Pairing and Simple Pairing. With the
usage of Simple Pairing it is required that all connections (except
the ones for SDP) are encrypted. So even the low security requirement
mandates an encrypted connection when using Simple Pairing. When using
Legacy Pairing (for Bluetooth 2.0 devices and older) this is not required
since it causes interoperability issues.

To support this properly the low security requirement translates into
different host controller transactions depending if Simple Pairing is
supported or not. However in case of Simple Pairing the command to
switch on encryption after a successful authentication is not triggered
for the low security mode. This patch fixes this and actually makes
the logic to differentiate between Simple Pairing and Legacy Pairing
a lot simpler.

Based on a report by Ville Tervo <ville.tervo@nokia.com>

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

12 years agoBluetooth: Add different pairing timeout for Legacy Pairing
Marcel Holtmann [Sun, 26 Apr 2009 18:01:22 +0000]
Bluetooth: Add different pairing timeout for Legacy Pairing

The Bluetooth stack uses a reference counting for all established ACL
links and if no user (L2CAP connection) is present, the link will be
terminated to save power. The problem part is the dedicated pairing
when using Legacy Pairing (Bluetooth 2.0 and before). At that point
no user is present and pairing attempts will be disconnected within
10 seconds or less. In previous kernel version this was not a problem
since the disconnect timeout wasn't triggered on incoming connections
for the first time. However this caused issues with broken host stacks
that kept the connections around after dedicated pairing. When the
support for Simple Pairing got added, the link establishment procedure
needed to be changed and now causes issues when using Legacy Pairing

When using Simple Pairing it is possible to do a proper reference
counting of ACL link users. With Legacy Pairing this is not possible
since the specification is unclear in some areas and too many broken
Bluetooth devices have already been deployed. So instead of trying to
deal with all the broken devices, a special pairing timeout will be
introduced that increases the timeout to 60 seconds when pairing is
triggered.

If a broken devices now puts the stack into an unforeseen state, the
worst that happens is the disconnect timeout triggers after 120 seconds
instead of 4 seconds. This allows successful pairings with legacy and
broken devices now.

Based on a report by Johan Hedberg <johan.hedberg@nokia.com>

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

12 years agoBluetooth: Ensure that HCI sysfs add/del is preempt safe
Roger Quadros [Thu, 23 Apr 2009 11:50:54 +0000]
Bluetooth: Ensure that HCI sysfs add/del is preempt safe

Use a different work_struct variables for add_conn() and del_conn() and
use single work queue instead of two for adding and deleting connections.

It eliminates the following error on a preemptible kernel:

[  204.358032] Unable to handle kernel NULL pointer dereference at virtual address 0000000c
[  204.370697] pgd = c0004000
[  204.373443] [0000000c] *pgd=00000000
[  204.378601] Internal error: Oops: 17 [#1] PREEMPT
[  204.383361] Modules linked in: vfat fat rfcomm sco l2cap sd_mod scsi_mod iphb pvr2d drm omaplfb ps
[  204.438537] CPU: 0    Not tainted  (2.6.28-maemo2 #1)
[  204.443664] PC is at klist_put+0x2c/0xb4
[  204.447601] LR is at klist_put+0x18/0xb4
[  204.451568] pc : [<c0270f08>]    lr : [<c0270ef4>]    psr: a0000113
[  204.451568] sp : cf1b3f10  ip : cf1b3f10  fp : cf1b3f2c
[  204.463104] r10: 00000000  r9 : 00000000  r8 : bf08029c
[  204.468353] r7 : c7869200  r6 : cfbe2690  r5 : c78692c8  r4 : 00000001
[  204.474945] r3 : 00000001  r2 : cf1b2000  r1 : 00000001  r0 : 00000000
[  204.481506] Flags: NzCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM Segment kernel
[  204.488861] Control: 10c5387d  Table: 887fc018  DAC: 00000017
[  204.494628] Process btdelconn (pid: 515, stack limit = 0xcf1b22e0)

Signed-off-by: Roger Quadros <ext-roger.quadros@nokia.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

12 years agosh: register the rtc-generic platform device properly.
Paul Mundt [Tue, 28 Apr 2009 14:12:10 +0000]
sh: register the rtc-generic platform device properly.

The device registration was accidentally omitted, add it back in. Do some
basic device probing as well, so this doesn't show up for platforms that
tie in to the RTC interface properly.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agonet: Avoid extra wakeups of threads blocked in wait_for_packet()
Eric Dumazet [Tue, 28 Apr 2009 09:24:21 +0000]
net: Avoid extra wakeups of threads blocked in wait_for_packet()

In 2.6.25 we added UDP mem accounting.

This unfortunatly added a penalty when a frame is transmitted, since
we have at TX completion time to call sock_wfree() to perform necessary
memory accounting. This calls sock_def_write_space() and utimately
scheduler if any thread is waiting on the socket.
Thread(s) waiting for an incoming frame was scheduled, then had to sleep
again as event was meaningless.

(All threads waiting on a socket are using same sk_sleep anchor)

This adds lot of extra wakeups and increases latencies, as noted
by Christoph Lameter, and slows down softirq handler.

Reference : http://marc.info/?l=linux-netdev&m=124060437012283&w=2

Fortunatly, Davide Libenzi recently added concept of keyed wakeups
into kernel, and particularly for sockets (see commit
37e5540b3c9d838eb20f2ca8ea2eb8072271e403
epoll keyed wakeups: make sockets use keyed wakeups)

Davide goal was to optimize epoll, but this new wakeup infrastructure
can help non epoll users as well, if they care to setup an appropriate
handler.

This patch introduces new DEFINE_WAIT_FUNC() helper and uses it
in wait_for_packet(), so that only relevant event can wakeup a thread
blocked in this function.

Trace of function calls from bnx2 TX completion bnx2_poll_work() is :
__kfree_skb()
 skb_release_head_state()
  sock_wfree()
   sock_def_write_space()
    __wake_up_sync_key()
     __wake_up_common()
      receiver_wake_function() : Stops here since thread is waiting for an INPUT

Reported-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

12 years agosh: setup timers in late_time_init()
Magnus Damm [Tue, 28 Apr 2009 08:19:50 +0000]
sh: setup timers in late_time_init()

This patch moves the SuperH timer setup code from time_init()
to late_time_init(). Good things about this change:
 - interrupts: they are enabled at late_time_init()
 - mm: regular kmalloc() can be used at late_time_init()

Together with moving to late_time_init() this patch changes
the sh_cmt driver to always allocate with kmalloc(). This
simplifies the code a bit and also fixes section mismatches.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agoclocksource: improve sh_cmt clocksource overflow handling
Magnus Damm [Tue, 28 Apr 2009 08:17:54 +0000]
clocksource: improve sh_cmt clocksource overflow handling

This patch improves the sh_cmt clocksource handling.

Currently the counter value is ignored in the case of
overflow. With this patch the overflow flag is read
before and after reading the counter, removing any
counter value and overflow flag mismatch issues.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

12 years agopowerpc/ps3: Fix build error on UP
Geoff Levand [Thu, 16 Apr 2009 09:05:37 +0000]
powerpc/ps3: Fix build error on UP

A non-SMP version of smp_send_stop() is now included in smp.h.
Remove the unneeded definition in the PS3 smp.c.

Fixes build errors like these when CONFIG_SMP=n:

  arch/powerpc/platforms/ps3/setup.c:49: error: redefinition of 'smp_send_stop'
  include/linux/smp.h:125: error: previous definition of 'smp_send_stop' was here

Reported-by: Subrata Modak <subrata@linux.vnet.ibm.com>
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>

12 years agopowerpc/cell: Select PCI for IBM_CELL_BLADE AND CELLEB
Michael Ellerman [Wed, 22 Apr 2009 22:43:03 +0000]
powerpc/cell: Select PCI for IBM_CELL_BLADE AND CELLEB

Currently PPC_CELL_NATIVE selects PPC_OF_PLATFORM_PCI, but does not
select PCI. This can lead to a config with the former and the latter
disabled, which does not build.

To fix this PPC_CELL_NATIVE should select PCI. However, that would
force PCI on for QPACE, which also selects PPC_CELL_NATIVE. So
instead move the select of PPC_OF_PLATFORM_PCI and PCI under both
IBM_CELL_BLADE and CELLEB.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>

12 years agopowerpc: ppc32 needs elf_read_implies_exec()
Kumar Gala [Fri, 17 Apr 2009 08:03:01 +0000]
powerpc: ppc32 needs elf_read_implies_exec()

On ppc64 we implemented elf_read_implies_exec() for 32-bit binaries
because old toolchains had bugs where they didn't mark program
segments executable that needed to be.  For some reason we didn't do
this on ppc32 builds.  This hadn't been an issue until commit 8d30c14c
("powerpc/mm: Rework I$/D$ coherency (v3)"), which had as a side
effect that we are now enforcing execute permissions to some extent on
32-bit 4xx and Book E processors.

This fixes it by defining elf_read_implies_exec on 32-bit to turn on
the read-implies-exec behaviour on programs that are sufficiently old
that they don't have a PT_GNU_STACK program header.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>

12 years agoRemove unused support code for refok sections.
Tim Abbott [Mon, 27 Apr 2009 18:02:27 +0000]
Remove unused support code for refok sections.

The old refok sections

  .text.init.refok
  .data.init.refok
  .exit.text.refok

have been deprecated since commit
312b1485fb509c9bc32eda28ad29537896658cb8.  After the other patches in
this patch series nothing is put in these sections, so clean things up
by eliminating all the remaining references to them.

Signed-off-by: Tim Abbott <tabbott@mit.edu>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

12 years agosparc: cleanup references to deprecated .text.init* sections.
Tim Abbott [Mon, 27 Apr 2009 18:02:26 +0000]
sparc: cleanup references to deprecated .text.init* sections.

The section .text.init.refok is deprecated and __REF (.ref.text)
should be used in assembly files instead.  This patch cleans up a few
uses of .text.init.refok in the sparc architecture.

Also fix a reference to .text.init in a comment that wasn't updated to
.init.text.

Signed-off-by: Tim Abbott <tabbott@mit.edu>
Cc: David S. Miller <davem@davemloft.net>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

12 years agosh: Use __INIT macro instead of .text.init.
Tim Abbott [Mon, 27 Apr 2009 18:02:25 +0000]
sh: Use __INIT macro instead of .text.init.

The sh architecture has some code in the .text.init section, but it
does not reference that section in its linker scripts.

This change moves this code from the .text.init section to the
.init.text section, which is presumably where it belongs.

Signed-off-by: Tim Abbott <tabbott@mit.edu>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Acked-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

12 years agopowerpc: Use __REF macro instead of old .text.init.refok.
Tim Abbott [Mon, 27 Apr 2009 18:02:24 +0000]
powerpc: Use __REF macro instead of old .text.init.refok.

The section .text.init.refok is deprecated and __REF (.ref.text)
should be used in assembly files instead.  This patch cleans up a few
uses of .text.init.refok in the powerpc architecture.

Signed-off-by: Tim Abbott <tabbott@mit.edu>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

12 years agopowerpc: Use TEXT_TEXT macro in linker script.
Tim Abbott [Mon, 27 Apr 2009 18:02:23 +0000]
powerpc: Use TEXT_TEXT macro in linker script.

Rather than adding .ref.text to the powerpc linker script so that we
can use __REF on the powerpc architecture, it seems simpler to switch
to using the generic TEXT_TEXT macro.

Signed-off-by: Tim Abbott <tabbott@mit.edu>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

12 years agoarm: Use __INIT macro instead of .text.init.
Tim Abbott [Mon, 27 Apr 2009 18:02:22 +0000]
arm: Use __INIT macro instead of .text.init.

arm is placing some code in the .text.init section, but it does not
reference that section in its linker scripts.

This change moves this code from the .text.init section to the
.init.text section, which is presumably where it belongs.

Signed-off-by: Tim Abbott <tabbott@mit.edu>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

12 years agoFRV: Use __INIT macro instead of .text.init.
Tim Abbott [Mon, 27 Apr 2009 23:37:58 +0000]
FRV: Use __INIT macro instead of .text.init.

FRV is placing some code in the .text.init section but does not reference that
section in its linker scripts.

This change moves this code from the .text.init section to the .init.text
section, which is presumably where it belongs.

Signed-off-by: Tim Abbott <tabbott@mit.edu>
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

12 years agoFRV: Remove unused header asm/init.h.
Tim Abbott [Mon, 27 Apr 2009 23:37:53 +0000]
FRV: Remove unused header asm/init.h.

It seems nothing has included the frv asm/init.h header for some time, and its
actual contents are out of date with include/linux/init.h.  So just delete it.

Signed-off-by: Tim Abbott <tabbott@mit.edu>
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

12 years agoMerge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/jwboyer/powerpc...
Paul Mackerras [Tue, 28 Apr 2009 00:47:16 +0000]
Merge branch 'merge' of git://git./linux/kernel/git/jwboyer/powerpc-4xx into merge

12 years agoRDMA/nes: Update iw_nes version
Chien Tung [Mon, 27 Apr 2009 20:46:29 +0000]
RDMA/nes: Update iw_nes version

Update version number to 1.5.0.0

Signed-off-by: Chien Tung <chien.tin.tung@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

12 years agoRDMA/nes: Fix error path in nes_accept()
Faisal Latif [Mon, 27 Apr 2009 20:45:19 +0000]
RDMA/nes: Fix error path in nes_accept()

If reg_phys_mem() fails, we need to free memory allocated for MPA
frame with private data before returning the error. Also move
nes_add_ref() after the reg_phys_mem() is successful.

Signed-off-by: Faisal Latif <faisal.latif@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

12 years agoRDMA/nes: Fix hang issues for large cluster dynamic connections
Faisal Latif [Mon, 27 Apr 2009 20:41:06 +0000]
RDMA/nes: Fix hang issues for large cluster dynamic connections

Running large cluster setup, we are hanging after many hours of
testing.  Fixing this required going over the code and making sure the
rexmit entry was properly removed based on the cm_node's state and
packet received.  Also when receiving a FIN packet, check seq# and
make sure there were no errors before calling handle_fin().

Following are the changes done in nes_cm.c:

* handle_ack_pkt() needs to return error value, so in case of error,
  handle_fin() is not called. Some cleanup done while going over the code.

* handle_rst_pkt(), handling of cm_node's NES_CM_STATE_LAST_ACK is missing.

* process_packet(), in case of FIN only packet is received, call
  check_seq() before processing.

* in handle_fin_pkt(), we are calling cleanup_retrans_entry() for all
  conditions, even if the packets need to be dropped.

Signed-off-by: Faisal Latif <faisal.latif@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

12 years agoRDMA/nes: Increase rexmit timeout interval
Faisal Latif [Mon, 27 Apr 2009 20:39:36 +0000]
RDMA/nes: Increase rexmit timeout interval

Under heavy load with large cluster testing, it may take longer to
receive a response to MPA requests.  Change the driver to wait longer
after each rexmit to max time value.

Signed-off-by: Faisal Latif <faisal.latif@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>