GitHub/moto-9609/android_kernel_motorola_exynos9610.git
13 years agodrm/i915/ringbuffer: Simplify the ring irq refcounting
Chris Wilson [Tue, 4 Jan 2011 22:22:56 +0000 (22:22 +0000)]
drm/i915/ringbuffer: Simplify the ring irq refcounting

... and move it under the spinlock to gain the appropriate memory
barriers.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=32752
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915/debugfs: Show the per-ring IMR
Chris Wilson [Tue, 4 Jan 2011 22:22:17 +0000 (22:22 +0000)]
drm/i915/debugfs: Show the per-ring IMR

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: Mask USER interrupts on gen6 (until required)
Chris Wilson [Tue, 4 Jan 2011 17:35:21 +0000 (17:35 +0000)]
drm/i915: Mask USER interrupts on gen6 (until required)

Otherwise we may consume 20% of the CPU just handling IRQs whilst
rendering. Ouch.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: Handle ringbuffer stalls when flushing
Chris Wilson [Tue, 4 Jan 2011 17:34:02 +0000 (17:34 +0000)]
drm/i915: Handle ringbuffer stalls when flushing

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: Enforce write ordering through the GTT
Chris Wilson [Tue, 4 Jan 2011 18:42:07 +0000 (18:42 +0000)]
drm/i915: Enforce write ordering through the GTT

We need to ensure that writes through the GTT land before any
modification to the MMIO registers and so must impose a mandatory write
barrier when flushing the GTT domain. This was revealed by relaxing the
write ordering by experimentally mapping the registers and the GATT as
write-combining.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: Remove impossible test
Chris Wilson [Mon, 3 Jan 2011 16:39:21 +0000 (16:39 +0000)]
drm/i915: Remove impossible test

As has_gem is unconditionally set to true, the conditional immediately
following that assignment is superfluous.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: avoid reading non-existent PLL reg on Ironlake+
Jesse Barnes [Thu, 30 Dec 2010 17:36:39 +0000 (09:36 -0800)]
drm/i915: avoid reading non-existent PLL reg on Ironlake+

These functions need to be reworked for Ironlake and above, but until
then at least avoid reading non-existent registers.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
[ickle: combine with a gratuitous tidy]
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: add 'reset' parameter
Chris Wilson [Thu, 23 Dec 2010 13:33:15 +0000 (13:33 +0000)]
drm/i915: add 'reset' parameter

When bringing up new hardware, or otherwise experimenting, GPU hangs are
a way of life. However, the automatic GPU reset can do more harm than
good under these circumstances, as we may wish to capture a full trace for
debugging.

Based on a patch by Zhenyu Wang.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: fix the wrong latency value while computing wm0
Yuanhan Liu [Thu, 23 Dec 2010 08:35:40 +0000 (16:35 +0800)]
drm/i915: fix the wrong latency value while computing wm0

On Ironlake, the LP0 latency is hardcoded and in ns unit, while on
Sandybridge, it comes from a register and with unit 0.1 us. So, fix
the wrong latency value while computing wm0 on Ironlake and Sandybridge.

Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: support low power watermarks on Ironlake
Jesse Barnes [Tue, 21 Dec 2010 21:10:23 +0000 (13:10 -0800)]
drm/i915: support low power watermarks on Ironlake

This patch actually makes the watermark code even uglier (if that's
possible), but has the advantage of sharing code between SNB and ILK at
least.  Longer term we should refactor the watermark stuff into its own
file and clean it up now that we know how it's supposed to work.

Supporting WM2 on my Vaio reduced power consumption by around 0.5W, so
this patch is definitely worthwhile (though it also needs lots of test
coverage).

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
[ickle: pass the watermark structs arounds]
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agoagp/intel: Flush the chipset write buffers when changing GTT base
Chris Wilson [Thu, 23 Dec 2010 10:40:38 +0000 (10:40 +0000)]
agp/intel: Flush the chipset write buffers when changing GTT base

Flush the chipset write buffers before and after adjusting the GTT base
register, just in case. We only modify this value upon initialisation
(boot and resume) so there should be no outstanding writes, however
there are always those persistent PGTBL_ER that keep getting reported
upon resume.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: Workaround erratum on i830 for TAIL pointer within last 2 cachelines
Chris Wilson [Wed, 22 Dec 2010 14:04:47 +0000 (14:04 +0000)]
drm/i915: Workaround erratum on i830 for TAIL pointer within last 2 cachelines

On i830 if the tail pointer is set to within 2 cachelines of the end of
the buffer, the chip may hang. So instead if the tail were to land in
that location, we pad the end of the buffer with NOPs, and start again
at the beginning.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: Use the mappable sizes determined by GTT for consistency.
Chris Wilson [Wed, 22 Dec 2010 14:07:12 +0000 (14:07 +0000)]
drm/i915: Use the mappable sizes determined by GTT for consistency.

There should be no difference, but we can eliminate redundant code.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: support overclocking on Sandy Bridge
Jesse Barnes [Mon, 20 Dec 2010 19:34:20 +0000 (11:34 -0800)]
drm/i915: support overclocking on Sandy Bridge

In some configuration, the PCU may allow us to overclock the GPU.
Check for this case and adjust the max frequency as appropriate.  Also
initialize the min/max frequencies to default values as indicated by
hardware.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915/lvds: Always use 0 to disable the pfit controller
Chris Wilson [Tue, 11 Jan 2011 18:09:58 +0000 (18:09 +0000)]
drm/i915/lvds: Always use 0 to disable the pfit controller

... and just any combination of bits & ~PFIT_ENABLE. This way we do not
attempt disable to the panel fitter controller uselessly upon
intel_lvds_disable().

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915/panel: Only record the backlight level when it is enabled
Chris Wilson [Tue, 11 Jan 2011 17:06:04 +0000 (17:06 +0000)]
drm/i915/panel: Only record the backlight level when it is enabled

By tracking the current status of the backlight we can prevent recording
the value of the current backlight when we have disabled it. And so
prevent restoring it to 'off' after an unbalanced sequence of
intel_lvds_disable/enable.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=22672
Tested-by: Alex Riesen <raa.lkml@gmail.com>
Tested-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@kernel.org
13 years agodrm/i915/sdvo: Defer detection of output capabilities until probing
Chris Wilson [Tue, 4 Jan 2011 20:10:52 +0000 (20:10 +0000)]
drm/i915/sdvo: Defer detection of output capabilities until probing

Alex Fiestas reported an issue with his HDMI connector being misdetected
as DVI unless he had something connected upon boot. By moving the
decision as to whether to use HDMI or DVI encoding for the HDMI capable
output until we probe the monitor means that we should avoid sending a
HDMI signal to a DVI monitor and also correctly detect hardware like
Alex's.

However, to really determine what connector is soldered onto the wire we
need to inspect the VBT sdvo child devices - but can we trust it?

Reported-by: Alex Fiestas <alex@eyeos.org>
Tested-by: Alex Fiestas <alex@eyeos.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=32828
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@kernel.org
13 years agodrm/i915: fix calculation of eDP signal levels on Sandybridge
Yuanhan Liu [Thu, 6 Jan 2011 10:26:08 +0000 (18:26 +0800)]
drm/i915: fix calculation of eDP signal levels on Sandybridge

Some voltage swing/pre-emphasis level use the same value on eDP
Sandybridge, like 400mv_0db and 600mv_0db are with the same value
of (0x0 << 22). So, fix them, and point out the value if it isn't
a supported voltage swing/pre-emphasis level.

Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@kernel.org
13 years agodrm/i915/crt: Check for a analog monitor in case of DVI-I
David Müller [Thu, 6 Jan 2011 12:29:32 +0000 (12:29 +0000)]
drm/i915/crt: Check for a analog monitor in case of DVI-I

Since Linux 2.6.36 the digital output on my system (855GME + DVI-I) is
not working any longer. The analog output is always activated
regardless of the type of monitor attached.

The culprit seems to be intel_crt_detect_ddc(), which returns true as
soon as an ACK from the EDID device is received. Obviously this
approach does not work with DVI-I where the analog and digital outputs
share a common DDC bus.

In a similar manner to the shared DDC wire, ala the "Mac Mini Hack", we
need an additional check to make sure that there really is an analog
device attached to the DDC.

Signed-off-by: David Müller <d.mueller@elsoft.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@kernel.org
13 years agodrm/i915: make DP training try a little harder
Jesse Barnes [Wed, 5 Jan 2011 22:45:24 +0000 (14:45 -0800)]
drm/i915: make DP training try a little harder

When trying to do channel equalization, we need to make sure we still
have clock recovery on all lanes while training.  We also need to try
clock recovery again if we lose the clock or if channel eq fails 5
times.  We'll try clock recovery up to 5 more times before giving up
entirely.

Gets suspend/resume working on my Vaio again and brings us back into
compliance with the DP training sequence spec.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
13 years agodrm/i915: check eDP encoder correctly when setting modes
Jesse Barnes [Tue, 4 Jan 2011 18:46:49 +0000 (10:46 -0800)]
drm/i915: check eDP encoder correctly when setting modes

We were using a stale pointer in the check which caused us to use CPU
attached DP params when we should have been using PCH attached params.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=31988
Tested-by: Jan-Hendrik Zab <jan@jhz.name>
Tested-by: Christoph Lukas <christoph.lukas@gmx.net>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@kernel.org
13 years agoMerge branch 'drm-core-next' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied...
Linus Torvalds [Tue, 11 Jan 2011 01:11:39 +0000 (17:11 -0800)]
Merge branch 'drm-core-next' of git://git./linux/kernel/git/airlied/drm-2.6

* 'drm-core-next' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6: (390 commits)
  drm/radeon/kms: disable underscan by default
  drm/radeon/kms: only enable hdmi features if the monitor supports audio
  drm: Restore the old_fb upon modeset failure
  drm/nouveau: fix hwmon device binding
  radeon: consolidate asic-specific function decls for pre-r600
  vga_switcheroo: comparing too few characters in strncmp()
  drm/radeon/kms: add NI pci ids
  drm/radeon/kms: don't enable pcie gen2 on NI yet
  drm/radeon/kms: add radeon_asic struct for NI asics
  drm/radeon/kms/ni: load default sclk/mclk/vddc at pm init
  drm/radeon/kms: add ucode loader for NI
  drm/radeon/kms: add support for DCE5 display LUTs
  drm/radeon/kms: add ni_reg.h
  drm/radeon/kms: add bo blit support for NI
  drm/radeon/kms: always use writeback/events for fences on NI
  drm/radeon/kms: adjust default clock/vddc tracking for pm on DCE5
  drm/radeon/kms: add backend map workaround for barts
  drm/radeon/kms: fill gpu init for NI asics
  drm/radeon/kms: add disabled vbios accessor for NI asics
  drm/radeon/kms: handle NI thermal controller
  ...

13 years agoMerge branch 'i2c-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jdelvar...
Linus Torvalds [Tue, 11 Jan 2011 01:09:13 +0000 (17:09 -0800)]
Merge branch 'i2c-for-linus' of git://git./linux/kernel/git/jdelvare/staging

* 'i2c-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jdelvare/staging:
  i2c: Constify i2c_client where possible
  i2c-algo-bit: Complain about masters which can't read SCL
  i2c-algo-bit: Refactor adapter registration
  i2c: Add generic I2C multiplexer using GPIO API
  i2c-nforce2: Remove unnecessary cast of pci_get_drvdata
  i2c-i801: Include <linux/slab.h>

13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/vapier...
Linus Torvalds [Tue, 11 Jan 2011 01:06:08 +0000 (17:06 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/vapier/blackfin

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/vapier/blackfin: (52 commits)
  Blackfin: encode cpu-rev into uImage name
  Blackfin: bf54x: don't ack GPIO ints when unmasking them
  Blackfin: sram_free_with_lsl: do not ignore return value of sram_free
  Blackfin: boards: add missing "static" to peripheral lists
  Blackfin: DNP5370: new board port
  Blackfin: bf518f-ezbrd: fix dsa resources
  Blackfin: move "-m elf32bfin" to general LDFLAGS
  Blackfin: kgdb_test: make sure to initialize num2
  Blackfin: kgdb: disable preempt schedule when running single step in kgdb
  Blackfin: kgdb: disable interrupt when single stepping in ADEOS
  Blackfin: SMP: kgdb: apply anomaly 257 work around
  Blackfin: fix building IPIPE code when XIP is enabled
  Blackfin: SMP: kgdb: flush core internal write buffer before flushinv
  Blackfin: sport_uart resources: remove unused secondary RX/TX pins
  Blackfin: tll6527m: fix spelling in unused code (struct name)
  Blackfin: bf527-ezkit: add adau1373 chip address
  Blackfin: no-mpu: fix masking of small uncached dma region
  Blackfin: pm: drop irq save/restore in standby and suspend to mem callback
  MAINTAINERS: update Analog Devices support info
  Blackfin: dpmc.h: pull in new pll.h
  ...

13 years agoMerge branch 'driver-core-next' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Tue, 11 Jan 2011 00:10:33 +0000 (16:10 -0800)]
Merge branch 'driver-core-next' of git://git./linux/kernel/git/gregkh/driver-core-2.6

* 'driver-core-next' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core-2.6:
  driver core: Document that device_rename() is only for networking
  sysfs: remove useless test from sysfs_merge_group
  driver-core: merge private parts of class and bus
  driver core: fix whitespace in class_attr_string

13 years agoMerge branch 'staging-next' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Tue, 11 Jan 2011 00:04:53 +0000 (16:04 -0800)]
Merge branch 'staging-next' of git://git./linux/kernel/git/gregkh/staging-2.6

* 'staging-next' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging-2.6: (510 commits)
  staging: speakup: fix failure handling
  staging: usbip: remove double giveback of URB
  Staging: batman-adv: Remove batman-adv from staging
  Staging: hv: Use only one txf buffer per channel and kmalloc/GFP_KERNEL on initialize
  staging: hv: remove unneeded osd_schedule_callback
  staging: hv: convert channel_mgmt.c to not call osd_schedule_callback
  staging: hv: convert vmbus_on_msg_dpc to not call osd_schedule_callback
  staging: brcm80211: Fix WL_<type> logging macros
  Staging: IIO: DDS: AD9833 / AD9834 driver
  Staging: IIO: dds.h convenience macros
  Staging: IIO: Direct digital synthesis abi documentation
  staging: brcm80211: Convert ETHER_TYPE_802_1X to ETH_P_PAE
  staging: brcm80211: Remove unused ETHER_TYPE_<foo> #defines
  staging: brcm80211: Remove ETHER_HDR_LEN, use ETH_HLEN
  staging: brcm80211: Convert ETHER_ADDR_LEN to ETH_ALEN
  staging: brcm80211: Convert ETHER_IS<FOO> to is_<foo>_ether_addr
  staging: brcm80211: Remove unused ether_<foo> #defines and struct
  staging: brcm80211: Convert ETHER_IS_MULTI to is_multicast_ether_addr
  staging: brcm80211: Remove unused #defines ETHER_<foo>_LOCALADDR
  Staging: comedi: Fix checkpatch.pl issues in file s526.c
  ...

Fix up trivial conflict in drivers/video/udlfb.c

13 years agoMerge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mjg59/platf...
Linus Torvalds [Mon, 10 Jan 2011 23:39:48 +0000 (15:39 -0800)]
Merge branch 'for_linus' of git://git./linux/kernel/git/mjg59/platform-drivers-x86

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mjg59/platform-drivers-x86: (36 commits)
  sony-laptop: support new hotkeys on the P, Z and EC series
  platform/x86: Consistently select LEDS Kconfig options
  sony-laptop: fix sparse non-ANSI function warning
  intel_ips: fix sparse non-ANSI function warning
  Support KHLB2 in the compal laptop driver
  acer-wmi: Enabled Acer Launch Manager mode
  [PATCH] intel_pmic_gpio: modify EOI handling following change of kernel irq subsystem
  ACPI Thinkpad: We must always call va_end() after va_start() but do not do so in thinkpad_acpi.c::acpi_evalf()
  acer-wmi: Initialize wlan/bluetooth/wwan rfkill software block state
  acer-wmi: Detect the WiFi/Bluetooth/3G devices available
  acer-wmi: Add 3G rfkill sysfs file
  acer-wmi: Add acer wmi hotkey events support
  platform/x86: Kconfig: Replace select by depends on ACPI_WMI
  ideapad: pass ideapad_priv as argument (part 2)
  ideapad: pass ideapad_priv as argument (part 1)
  ideapad: add markups, unify comments and return result when init
  ideapad: add hotkey support
  ideapad: let camera power control entry under platform driver
  ideapad: add platform driver for ideapad
  fujitsu-laptop: fix compiler warning on pnp_ids
  ...

13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux...
Linus Torvalds [Mon, 10 Jan 2011 23:39:23 +0000 (15:39 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/geert/linux-m68k

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k:
  m68knommu: Need to check __get_user()/__put_user() result
  m68knommu: signal.c __user annotations
  m68knommu: Equivalent of "m68k: handle new gcc's"
  m68knommu: f_pcr has been gone since headers' merge
  m68knommu: Don't lose state if sigframe setup fails
  m68knommu: Handle multiple pending signals
  m68knommu: Switch to saner sigsuspend
  m68knommu: Don't bother with SA_ONESHOT
  m68k: Check __get_user()/__put_user() return value
  m68k: Missing syscall_trace() on sigreturn
  m68k: Fix stack mangling logics in sigreturn
  m68k: If we fail to set sigframe up, just leave regs alone...
  m68k: Don't lose state if sigframe setup fails
  m68k: Simplify the singlestepping handling in signals
  m68k: Switch to saner sigsuspend()
  m68k: Resetting sa_handler in local copy of k_sigaction is pointless
  m68k/sun3: Kill pte_unmap() warnings

13 years agoMerge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6
Linus Torvalds [Mon, 10 Jan 2011 22:52:44 +0000 (14:52 -0800)]
Merge branch 'release' of git://git./linux/kernel/git/aegl/linux-2.6

* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6:
  [IA64] Avoid array overflow if there are too many cpus in SRAT table
  [IA64] Remove unlikely from cpu_is_offline
  [IA64] irq_ia64, use set_irq_chip
  [IA64] perfmon: Change vmalloc to vzalloc and drop memset.
  [IA64] eliminate race condition in smp_flush_tlb_mm

13 years agoMerge branch 'for-torvalds' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw...
Linus Torvalds [Mon, 10 Jan 2011 22:50:38 +0000 (14:50 -0800)]
Merge branch 'for-torvalds' of git://git./linux/kernel/git/linusw/linux-stericsson

* 'for-torvalds' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-stericsson:
  ux500: allow 5500 and 8500 to be built together
  ux500: modem_irq is only for 5500
  ux500: dynamic SOC detection
  ux500: rename MOP board Kconfig
  ux500: remove build-time changing macros

13 years agoMerge branch 'msm-smp' of git://codeaurora.org/quic/kernel/davidb/linux-msm
Linus Torvalds [Mon, 10 Jan 2011 22:50:08 +0000 (14:50 -0800)]
Merge branch 'msm-smp' of git://codeaurora.org/quic/kernel/davidb/linux-msm

* 'msm-smp' of git://codeaurora.org/quic/kernel/davidb/linux-msm:
  msm: add SMP support for msm
  msm: hotplug: support cpu hotplug on msm
  msm: timer: SMP timer support for msm
  msm: scm-boot: Support for setting cold/warm boot addresses
  msm: Secure Channel Manager (SCM) support

13 years agoi2c: Constify i2c_client where possible
Jean Delvare [Mon, 10 Jan 2011 21:11:23 +0000 (22:11 +0100)]
i2c: Constify i2c_client where possible

Helper functions for I2C and SMBus transactions don't modify the
i2c_client that is passed to them, so it can be marked const.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
13 years agoi2c-algo-bit: Complain about masters which can't read SCL
Jean Delvare [Mon, 10 Jan 2011 21:11:23 +0000 (22:11 +0100)]
i2c-algo-bit: Complain about masters which can't read SCL

The I2C specification explicitly describes both SDA and SCL as
bidirectional lines. An I2C master with a read-only SCL is thus not
compliant. If a slow slave stretches the clock, errors will happen,
so the bus can't be considered as reliable.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
13 years agoi2c-algo-bit: Refactor adapter registration
Jean Delvare [Mon, 10 Jan 2011 21:11:23 +0000 (22:11 +0100)]
i2c-algo-bit: Refactor adapter registration

Use a function pointer to decide whether to call i2c_add_adapter or
i2c_add_numbered_adapter. This makes the code more compact than the
current strategy of having the common code in a separate function.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
13 years agoi2c: Add generic I2C multiplexer using GPIO API
Peter Korsgaard [Mon, 10 Jan 2011 21:11:23 +0000 (22:11 +0100)]
i2c: Add generic I2C multiplexer using GPIO API

Add an i2c mux driver providing access to i2c bus segments using a
hardware MUX sitting on a master bus and controlled through gpio pins.

E.G. something like:

  ----------              ----------  Bus segment 1   - - - - -
 |          | SCL/SDA    |          |-------------- |           |
 |          |------------|          |
 |          |            |          | Bus segment 2 |           |
 |  Linux   | GPIO 1..N  |   MUX    |---------------   Devices
 |          |------------|          |               |           |
 |          |            |          | Bus segment M
 |          |            |          |---------------|           |
  ----------              ----------                  - - - - -

SCL/SDA of the master I2C bus is multiplexed to bus segment 1..M
according to the settings of the GPIO pins 1..N.

Signed-off-by: Peter Korsgaard <peter.korsgaard@barco.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
13 years agoi2c-nforce2: Remove unnecessary cast of pci_get_drvdata
Joe Perches [Mon, 10 Jan 2011 21:11:22 +0000 (22:11 +0100)]
i2c-nforce2: Remove unnecessary cast of pci_get_drvdata

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
13 years agoi2c-i801: Include <linux/slab.h>
Ben Hutchings [Mon, 10 Jan 2011 21:11:22 +0000 (22:11 +0100)]
i2c-i801: Include <linux/slab.h>

Commit 5a0e3ad6af8660be21ca98a971cd00f331318c05 added direct inclusion
of <linux/slab.h> to those source files that appeared to need it, but
somehow missed this.  On most architectures <linux/slab.h> is still
indirectly included, but there are exceptions such as alpha.

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
13 years agostaging: speakup: fix failure handling
William Hubbs [Thu, 16 Dec 2010 19:26:58 +0000 (13:26 -0600)]
staging: speakup: fix failure handling

fix the failure handling in kobjects and the main function so that we
release the virtual keyboard if we exit due to another failure.

Signed-off-by: William Hubbs <w.d.hubbs@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agostaging: usbip: remove double giveback of URB
Márton Németh [Mon, 13 Dec 2010 20:59:09 +0000 (21:59 +0100)]
staging: usbip: remove double giveback of URB

In the vhci_urb_dequeue() function the TCP connection is checked twice.
Each time when the TCP connection is closed the URB is unlinked and given
back. Remove the second attempt of unlinking and giving back of the URB completely.

This patch fixes the bug described at https://bugzilla.kernel.org/show_bug.cgi?id=24872 .

Signed-off-by: Márton Németh <nm127@freemail.hu>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris...
Linus Torvalds [Mon, 10 Jan 2011 19:18:59 +0000 (11:18 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/jmorris/security-testing-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6: (30 commits)
  MAINTAINERS: Add tomoyo-dev-en ML.
  SELinux: define permissions for DCB netlink messages
  encrypted-keys: style and other cleanup
  encrypted-keys: verify datablob size before converting to binary
  trusted-keys: kzalloc and other cleanup
  trusted-keys: additional TSS return code and other error handling
  syslog: check cap_syslog when dmesg_restrict
  Smack: Transmute labels on specified directories
  selinux: cache sidtab_context_to_sid results
  SELinux: do not compute transition labels on mountpoint labeled filesystems
  This patch adds a new security attribute to Smack called SMACK64EXEC. It defines label that is used while task is running.
  SELinux: merge policydb_index_classes and policydb_index_others
  selinux: convert part of the sym_val_to_name array to use flex_array
  selinux: convert type_val_to_struct to flex_array
  flex_array: fix flex_array_put_ptr macro to be valid C
  SELinux: do not set automatic i_ino in selinuxfs
  selinux: rework security_netlbl_secattr_to_sid
  SELinux: standardize return code handling in selinuxfs.c
  SELinux: standardize return code handling in selinuxfs.c
  SELinux: standardize return code handling in policydb.c
  ...

13 years agoux500: allow 5500 and 8500 to be built together
Rabin Vincent [Wed, 8 Dec 2010 05:38:01 +0000 (11:08 +0530)]
ux500: allow 5500 and 8500 to be built together

Signed-off-by: Rabin Vincent <rabin.vincent@stericsson.com>
Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
13 years agoux500: modem_irq is only for 5500
Rabin Vincent [Wed, 8 Dec 2010 05:38:00 +0000 (11:08 +0530)]
ux500: modem_irq is only for 5500

Signed-off-by: Rabin Vincent <rabin.vincent@stericsson.com>
[Modified to hit the right file]
Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
13 years agoux500: dynamic SOC detection
Rabin Vincent [Wed, 8 Dec 2010 05:37:59 +0000 (11:07 +0530)]
ux500: dynamic SOC detection

Dynamically detect the DBx500 SOC an revision based on the ASIC ID.

Signed-off-by: Rabin Vincent <rabin.vincent@stericsson.com>
Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
13 years agoux500: rename MOP board Kconfig
Rabin Vincent [Wed, 8 Dec 2010 05:37:58 +0000 (11:07 +0530)]
ux500: rename MOP board Kconfig

Rename the MOP board Kconfig entries to the same name as the machine
type, so that the machine_is_*() macros work correctly.

Signed-off-by: Rabin Vincent <rabin.vincent@stericsson.com>
[Updated to match changes in the tree]
Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
13 years agoux500: remove build-time changing macros
Rabin Vincent [Wed, 8 Dec 2010 05:37:57 +0000 (11:07 +0530)]
ux500: remove build-time changing macros

To allow the possiblity of building U8500 and U5500 support in the same
image.

Signed-off-by: Rabin Vincent <rabin.vincent@stericsson.com>
[Rebased to latest changes in Russells tree]
Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
13 years agosony-laptop: support new hotkeys on the P, Z and EC series
Mattia Dongili [Sat, 8 Jan 2011 09:47:29 +0000 (18:47 +0900)]
sony-laptop: support new hotkeys on the P, Z and EC series

Add new mappings for assist, VAIO, zoom and eject buttons present on
refurbished P, Z and EC models.

Reported-by: Gyorgy Jeney <nog.lkml@gmail.com>
Reported-by: Stephan Mueller <smueller@chronox.de>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Cc: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Mattia Dongili <malattia@linux.it>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
13 years agoMerge branch 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groec...
Linus Torvalds [Mon, 10 Jan 2011 16:57:46 +0000 (08:57 -0800)]
Merge branch 'hwmon-for-linus' of git://git./linux/kernel/git/groeck/staging

* 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/staging: (44 commits)
  hwmon: Support for Dallas Semiconductor DS620
  hwmon: driver for Sensirion SHT21 humidity and temperature sensor
  hwmon: Add humidity attribute to sysfs ABI
  hwmon: sysfs ABI updates
  hwmon: (via-cputemp) sync hotplug handling with coretemp/pkgtemp
  hwmon: (lm95241) Rewrite to avoid using macros
  hwmon: (applesmc) Fix checkpatch errors and fix value range checks
  hwmon: (applesmc) Update copyright information
  hwmon: (applesmc) Silence driver
  hwmon: (applesmc) Simplify feature sysfs handling
  hwmon: (applesmc) Dynamic creation of fan files
  hwmon: (applesmc) Extract all features generically
  hwmon: (applesmc) Handle new temperature format
  hwmon: (applesmc) Dynamic creation of temperature files
  hwmon: (applesmc) Introduce a register lookup table
  hwmon: (applesmc) Use pr_fmt and pr_<level>
  hwmon: (applesmc) Relax the severity of device init failure
  hwmon: (applesmc) Add MacBookAir3,1(3,2) support
  hwmon: (w83627hf) Use pr_fmt and pr_<level>
  hwmon: (w83627ehf) Use pr_fmt and pr_<level>
  ...

13 years agoMerge branch 'next-devicetree' of git://git.secretlab.ca/git/linux-2.6
Linus Torvalds [Mon, 10 Jan 2011 16:57:03 +0000 (08:57 -0800)]
Merge branch 'next-devicetree' of git://git.secretlab.ca/git/linux-2.6

* 'next-devicetree' of git://git.secretlab.ca/git/linux-2.6: (29 commits)
  of/flattree: forward declare struct device_node in of_fdt.h
  ipmi: explicitly include of_address.h and of_irq.h
  sparc: explicitly cast negative phandle checks to s32
  powerpc/405: Fix missing #{address,size}-cells in i2c node
  powerpc/5200: dts: refactor dts files
  powerpc/5200: dts: Change combatible strings on localbus
  powerpc/5200: dts: remove unused properties
  powerpc/5200: dts: rename nodes to prepare for refactoring dts files
  of/flattree: Update dtc to current mainline.
  of/device: Don't register disabled devices
  powerpc/dts: fix syntax bugs in bluestone.dts
  of: Fixes for OF probing on little endian systems
  of: make drivers depend on CONFIG_OF instead of CONFIG_PPC_OF
  of/flattree: Add of_flat_dt_match() helper function
  of_serial: explicitly include of_irq.h
  of/flattree: Refactor unflatten_device_tree and add fdt_unflatten_tree
  of/flattree: Reorder unflatten_dt_node
  of/flattree: Refactor unflatten_dt_node
  of/flattree: Add non-boottime device tree functions
  of/flattree: Add Kconfig for EARLY_FLATTREE
  ...

Fix up trivial conflict in arch/sparc/prom/tree_32.c as per Grant.

13 years agoplatform/x86: Consistently select LEDS Kconfig options
Mark Brown [Mon, 10 Jan 2011 14:37:02 +0000 (14:37 +0000)]
platform/x86: Consistently select LEDS Kconfig options

Currently the x86 platform devices are not consistent about selecting
or depending on the LEDs Kconfig variables, and this inconsistency
leads to Kconfig getting upset and refusing to offer LEDs (even on
non-x86 platforms):

drivers/platform/x86/Kconfig:422:error: recursive dependency detected!
drivers/platform/x86/Kconfig:422:       symbol EEEPC_WMI depends on ACPI_WMI
drivers/platform/x86/Kconfig:438:       symbol ACPI_WMI is selected by ACER_WMI
drivers/platform/x86/Kconfig:18:        symbol ACER_WMI depends on LEDS_CLASS
drivers/leds/Kconfig:10:        symbol LEDS_CLASS is selected by EEEPC_WMI

Fix this by always selecting rather than depending on the symbols as
slightly more drivers use this approach already and it seems more
user friendly.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
13 years agosony-laptop: fix sparse non-ANSI function warning
Randy Dunlap [Sun, 9 Jan 2011 03:56:44 +0000 (19:56 -0800)]
sony-laptop: fix sparse non-ANSI function warning

Fix sparse warning for non-ANSI function declaration:

drivers/platform/x86/sony-laptop.c:1134:35: warning: non-ANSI function declaration of function 'sony_nc_rfkill_update'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Matthew Garrett <mjg@redhat.com>
Cc: Mattia Dongili <malattia@linux.it>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
13 years agointel_ips: fix sparse non-ANSI function warning
Randy Dunlap [Sun, 9 Jan 2011 03:55:40 +0000 (19:55 -0800)]
intel_ips: fix sparse non-ANSI function warning

Fix sparse warning for non-ANSI function declaration:

drivers/platform/x86/intel_ips.c:1477:25: warning: non-ANSI function declaration of function 'ips_link_to_i915_driver'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
13 years agoheaders: kobject.h redux
Alexey Dobriyan [Mon, 10 Jan 2011 06:18:25 +0000 (08:18 +0200)]
headers: kobject.h redux

Remove kobject.h from files which don't need it, notably,
sched.h and fs.h.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoheaders: path.h redux
Alexey Dobriyan [Mon, 10 Jan 2011 06:17:10 +0000 (08:17 +0200)]
headers: path.h redux

Remove path.h from sched.h and other files.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoMerge branch 'stable/bug-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Mon, 10 Jan 2011 16:48:46 +0000 (08:48 -0800)]
Merge branch 'stable/bug-fixes' of git://git./linux/kernel/git/konrad/xen

* 'stable/bug-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
  xen/event: validate irq before get evtchn by irq
  xen/fb: fix potential memory leak
  xen/fb: fix xenfb suspend/resume race.
  xen: disable ACPI NUMA for PV guests
  xen/irq: Cleanup the find_unbound_irq

13 years agoMerge branch 'stable/generic' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen
Linus Torvalds [Mon, 10 Jan 2011 16:48:29 +0000 (08:48 -0800)]
Merge branch 'stable/generic' of git://git./linux/kernel/git/konrad/xen

* 'stable/generic' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
  xen: HVM X2APIC support
  apic: Move hypervisor detection of x2apic to hypervisor.h

13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg...
Linus Torvalds [Mon, 10 Jan 2011 16:38:01 +0000 (08:38 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/penberg/slab-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6:
  slub: Fix a crash during slabinfo -v
  tracing/slab: Move kmalloc tracepoint out of inline code
  slub: Fix slub_lock down/up imbalance
  slub: Fix build breakage in Documentation/vm
  slub tracing: move trace calls out of always inlined functions to reduce kernel code size
  slub: move slabinfo.c to tools/slub/slabinfo.c

13 years agoMerge branch 'misc' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6
Linus Torvalds [Mon, 10 Jan 2011 16:29:25 +0000 (08:29 -0800)]
Merge branch 'misc' of git://git./linux/kernel/git/mmarek/kbuild-2.6

* 'misc' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6:
  Documentation/kbuild: add info that 'choice' can have a symbol name
  kbuild: add numeric --set-val option to scripts/config
  headers_check: Fix warning text
  headers_check: better search for functions in headers
  scripts/coccinelle: update for compatability with Coccinelle 0.2.4
  tags: put function prototypes back!
  Kconfig: fix single letter command in scripts/config
  gitignore: add scripts/recordmcount

13 years agoMerge branch 'packaging' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek...
Linus Torvalds [Mon, 10 Jan 2011 16:28:38 +0000 (08:28 -0800)]
Merge branch 'packaging' of git://git./linux/kernel/git/mmarek/kbuild-2.6

* 'packaging' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6:
  deb-pkg: s/hdr/kernel_headers_/ && s/header/libc_headers_/
  deb-pkg: Make deb-pkg generate a seperate linux-libc-dev deb
  kbuild: create linux-headers package in deb-pkg
  kbuild, deb-pkg: support overriding userland architecture
  kbuild, deb-pkg: select userland architecture based on UTS_MACHINE
  kbuild, deb-pkg: Fix build with paranoid umask

13 years agoMerge branch 'kconfig' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6
Linus Torvalds [Mon, 10 Jan 2011 16:28:17 +0000 (08:28 -0800)]
Merge branch 'kconfig' of git://git./linux/kernel/git/mmarek/kbuild-2.6

* 'kconfig' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6:
  nconf: handle comment entries within choice/endchoice
  kconfig: fix warning
  kconfig: Make expr_copy() take a const argument
  kconfig: simplify select-with-unmet-direct-dependency warning
  kconfig: add more S_INT and S_HEX consistency checks
  kconfig: fix `zconfdebug' extern declaration
  kconfig/conf: merge duplicate switch's case
  kconfig: fix typos
  kbuild/gconf: add dummy inline for bind_textdomain_codeset()
  kbuild/nconf: fix spaces damage
  kconfig: nuke second argument of conf_write_symbol()
  kconfig: do not define AUTOCONF_INCLUDED
  kconfig: the day kconfig warns about "select"-abuse has come

13 years agoMerge branch 'kbuild' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6
Linus Torvalds [Mon, 10 Jan 2011 16:27:52 +0000 (08:27 -0800)]
Merge branch 'kbuild' of git://git./linux/kernel/git/mmarek/kbuild-2.6

* 'kbuild' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6:
  mkuboot.sh: Fail if mkimage is missing
  gen_init_cpio: checkpatch fixes
  gen_init_cpio: Avoid race between call to stat() and call to open()
  modpost: Fix address calculation in reloc_location()
  Make fixdep error handling more explicit
  checksyscalls: Fix stand-alone usage
  modpost: Put .zdebug* section on white list
  kbuild: fix interaction of CONFIG_IKCONFIG and KCONFIG_CONFIG
  kbuild: export linux/{a.out,kvm,kvm_para}.h on headers_install_all
  kbuild: introduce HDR_ARCH_LIST for headers_install_all
  headers_install: check exit status of unifdef
  gen_init_cpio: remove leading `/' from file names
  scripts/genksyms: fix header usage
  fixdep: use hash table instead of a single array

13 years agoMerge branch 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzi...
Linus Torvalds [Mon, 10 Jan 2011 16:22:33 +0000 (08:22 -0800)]
Merge branch 'upstream-linus' of git://git./linux/kernel/git/jgarzik/libata-dev

* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev:
  pata_platform: Remove CONFIG_HAVE_PATA_PLATFORM's dependencies.
  pata_hpt37x: actually limit HPT370 to UltraDMA/66
  pata_hpt3x2n: coding style cleanup
  pata_hpt37x: coding style cleanup
  pata_hpt366: coding style cleanup
  pata_hpt3x2n: calculate average f_CNT
  pata_hpt3x2n: clarify about HPT371N support
  pata_hpt{37x|3x2n}: SATA mode filtering
  [libata] avoid needlessly passing around ptr to SCSI completion func
  [libata] new driver acard_ahci, for ATP8620 host controller

13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid
Linus Torvalds [Mon, 10 Jan 2011 16:15:37 +0000 (08:15 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/jikos/hid

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid: (34 commits)
  HID: roccat: Update sysfs attribute doc
  HID: roccat: don't use #pragma pack
  HID: roccat: Add support for Roccat Kone[+] v2
  HID: roccat: reduce number of functions in kone and pyra drivers
  HID: roccat: declare meaning of pack pragma usage in driver headers
  HID: roccat: use class for char device for sysfs attribute creation
  sysfs: Introducing binary attributes for struct class
  HID: hidraw: add compatibility ioctl() for 32-bit applications.
  HID: hid-picolcd: Fix memory leak in picolcd_debug_out_report()
  HID: picolcd: fix misuse of logical operation in place of bitop
  HID: usbhid: base runtime PM on modern API
  HID: replace offsets values with their corresponding BTN_* defines
  HID: hid-mosart: support suspend/resume
  HID: hid-mosart: ignore buttons report
  HID: hid-picolcd: don't use flush_scheduled_work()
  HID: simplify an index check in hid_lookup_collection
  HID: Hoist assigns from ifs
  HID: Remove superfluous __inline__
  HID: Use vzalloc for vmalloc/memset(,0...)
  HID: Add and use hid_<level>: dev_<level> equivalents
  ...

13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael...
Linus Torvalds [Mon, 10 Jan 2011 16:14:53 +0000 (08:14 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/rafael/suspend-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/suspend-2.6:
  spi / PM: Support dev_pm_ops
  PM: Prototype the pm_generic_ operations
  PM / Runtime: Generic resume shouldn't set RPM_ACTIVE unconditionally
  PM: Use dev_name() in core device suspend and resume routines
  PM: Permit registration of parentless devices during system suspend
  PM: Replace the device power.status field with a bit field
  PM: Remove redundant checks from core device resume routines
  PM: Use a different list of devices for each stage of device suspend
  PM: Avoid compiler warning in pm_noirq_op()
  PM: Use pm_wakeup_pending() in __device_suspend()
  PM / Wakeup: Replace pm_check_wakeup_events() with pm_wakeup_pending()
  PM: Prevent dpm_prepare() from returning errors unnecessarily
  PM: Fix references to basic-pm-debugging.txt in drivers-testing.txt
  PM / Runtime: Add synchronous runtime interface for interrupt handlers (v3)
  PM / Hibernate: When failed, in_suspend should be reset
  PM / Hibernate: hibernation_ops->leave should be checked too
  Freezer: Fix a race during freezing of TASK_STOPPED tasks
  PM: Use proper ccflag flag in kernel/power/Makefile
  PM / Runtime: Fix comments to match runtime callback code

13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394...
Linus Torvalds [Mon, 10 Jan 2011 16:14:02 +0000 (08:14 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/ieee1394/linux1394-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6:
  firewire: use split transaction timeout only for split transactions
  firewire: ohci: consolidate context status flags
  firewire: ohci: cache the context run bit
  firewire: ohci: flush AT contexts after bus reset - addendum
  firewire: ohci: flush AT contexts after bus reset for OHCI 1.2
  firewire: net: set carrier state at ifup
  firewire: net: add carrier detection
  firewire: net: ratelimit error messages
  firewire: ohci: restart iso DMA contexts on resume from low power mode
  firewire: ohci: restore GUID on resume.
  firewire: ohci: use common buffer for self IDs and AR descriptors
  firewire: ohci: optimize iso context checks in the interrupt handler
  firewire: make PHY packet header format consistent
  firewire: ohci: properly clear posted write errors
  firewire: ohci: flush MMIO writes in the interrupt handler
  firewire: ohci: fix AT context initialization error handling
  firewire: ohci: Asynchronous Reception rewrite
  firewire: core: Update WARN uses
  firewire: nosy: char device is not seekable

13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ryusuke...
Linus Torvalds [Mon, 10 Jan 2011 15:50:40 +0000 (07:50 -0800)]
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: unfold nilfs_dat_inode function
  nilfs2: do not pass sbi to functions which can get it from inode
  nilfs2: get rid of nilfs_mount_options structure
  nilfs2: simplify nilfs_mdt_freeze_buffer
  nilfs2: get rid of loaded flag from nilfs object
  nilfs2: fix a checkpatch error in page.c
  nilfs2: fiemap support
  nilfs2: mark buffer heads as delayed until the data is written to disk
  nilfs2: call nilfs_error inside bmap routines
  fs/nilfs2/super.c: Use printf extension %pV
  MAINTAINERS: add nilfs2 git tree entry

13 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6
Linus Torvalds [Mon, 10 Jan 2011 15:47:11 +0000 (07:47 -0800)]
Merge git://git./linux/kernel/git/sfrench/cifs-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
  cifs: use CreationTime like an i_generation field
  cifs: switch cifs_open and cifs_create to use CIFSSMBUnixSetFileInfo
  cifs: show "acl" in DebugData Features when it's compiled in
  cifs: move "ntlmssp" and "local_leases" options out of experimental code
  cifs: replace some hardcoded values with preprocessor constants
  cifs: remove unnecessary locking around sequence_number
  [CIFS] Fix minor merge conflict in fs/cifs/dir.c
  CIFS: Simplify cifs_open code
  CIFS: Simplify non-posix open stuff (try #2)
  CIFS: Add match_port check during looking for an existing connection (try #4)
  CIFS: Simplify ipv*_connect functions into one (try #4)
  cifs: Support NTLM2 session security during NTLMSSP authentication [try #5]
  cifs: don't overwrite dentry name in d_revalidate

13 years agoxen/event: validate irq before get evtchn by irq
Joe Jin [Fri, 7 Jan 2011 06:50:12 +0000 (14:50 +0800)]
xen/event: validate irq before get evtchn by irq

When retrieving the event channel number from irq, the irq
number may not be valid under some conditions.

So far that can be when we suspend/resume and irq ends with -1.
Validate and return sanitized irq and provide diagnostics information.

[v3: added unlikely on the WARN path]
[v2: reworded the commit message]
Signed-off-by: Joe Jin <joe.jin@oracle.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Tested-by: Gurudas Pai <gurudas.pai@oracle.com>
Cc: Ian Campbell <ian.campbell@citrix.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
13 years agoxen/fb: fix potential memory leak
Joe Jin [Fri, 7 Jan 2011 10:20:54 +0000 (18:20 +0800)]
xen/fb: fix potential memory leak

This patch fixes a potential memory leak when xenfb connect to
the backend fails.

Thanks for Ian's review and comments.

[v2: reworded the commit message a bit]

Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Joe Jin <joe.jin@oracle.com>
Tested-by: Gurudas Pai <gurudas.pai@oracle.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
13 years agoxen/fb: fix xenfb suspend/resume race.
Joe Jin [Fri, 7 Jan 2011 10:17:17 +0000 (18:17 +0800)]
xen/fb: fix xenfb suspend/resume race.

When migrating guests over a long period we hit this:

<1>BUG: unable to handle kernel paging request at 0000000b819fdb98
<1>IP: [<ffffffff812a588f>] notify_remote_via_irq+0x13/0x34
<4>PGD 94b10067 PUD 0
<0>Oops: 0000 [#1] SMP
.. snip..
Call Trace:
 [<ffffffff812712c9>] xenfb_send_event+0x5c/0x5e
 [<ffffffff8100ea5f>] ? xen_restore_fl_direct_end+0x0/0x1
 [<ffffffff81438d80>] ? _spin_unlock_irqrestore+0x16/0x18
 [<ffffffff812714ee>] xenfb_refresh+0x1b1/0x1d7
 [<ffffffff81270568>] ? sys_imageblit+0x1ac/0x458
 [<ffffffff81271786>] xenfb_imageblit+0x2f/0x34
 [<ffffffff8126a3e5>] soft_cursor+0x1b5/0x1c8
 [<ffffffff8126a137>] bit_cursor+0x4b6/0x4d7
 [<ffffffff8100ea5f>] ? xen_restore_fl_direct_end+0x0/0x1
 [<ffffffff81438d80>] ? _spin_unlock_irqrestore+0x16/0x18
 [<ffffffff81269c81>] ? bit_cursor+0x0/0x4d7
 [<ffffffff812656b7>] fb_flashcursor+0xff/0x111
 [<ffffffff812655b8>] ? fb_flashcursor+0x0/0x111
 [<ffffffff81071812>] worker_thread+0x14d/0x1ed
 [<ffffffff81075a8c>] ? autoremove_wake_function+0x0/0x3d
 [<ffffffff81438d80>] ? _spin_unlock_irqrestore+0x16/0x18
 [<ffffffff810716c5>] ? worker_thread+0x0/0x1ed
 [<ffffffff810756e3>] kthread+0x6e/0x76
 [<ffffffff81012dea>] child_rip+0xa/0x20
 [<ffffffff81011fd1>] ? int_ret_from_sys_call+0x7/0x1b
 [<ffffffff8101275d>] ? retint_restore_args+0x5/0x6
 [<ffffffff81012de0>] ? child_rip+0x0/0x20
Code: 6b ff 0c 8b 87 a4 db 9f 81 66 85 c0 74 08 0f b7 f8 e8 3b ff ff ff c9
c3 55 48 89 e5 48 83 ec 10 0f 1f 44 00 00 89 ff 48 6b ff 0c <8b> 87 a4 db 9f
81 66 85 c0 74 14 48 8d 75 f0 0f b7 c0 bf 04 00
RIP  [<ffffffff812a588f>] notify_remote_via_irq+0x13/0x34
 RSP <ffff8800e7bf7bd0>
CR2: 0000000b819fdb98
---[ end trace 098b4b74827595d0 ]---

The root cause of the panic is the race between the resume and reconnect to the backend.

Clearing the 'update_wanted' flag of xenfb before disconnecting from the
backend fixes this issue.

Signed-off-by: Joe Jin <joe.jin@oracle.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Tested-by: Gurudas Pai <gurudas.pai@oracle.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
13 years agoxen: disable ACPI NUMA for PV guests
Ian Campbell [Fri, 3 Dec 2010 09:54:03 +0000 (09:54 +0000)]
xen: disable ACPI NUMA for PV guests

Xen does not currently expose PV-NUMA information to PV
guests. Therefore disable NUMA for the time being to prevent the
kernel picking up on an host-level NUMA information which it might
come across in the firmware.

[ Added comment - Jeremy ]

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/teigland/dlm
Linus Torvalds [Mon, 10 Jan 2011 15:46:26 +0000 (07:46 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/teigland/dlm

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/teigland/dlm:
  dlm: sanitize work_start() in lowcomms.c
  dlm: reduce cond_resched during send
  dlm: use TCP_NODELAY
  dlm: Use cmwq for send and receive workqueues
  dlm: Handle application limited situations properly.

13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi...
Linus Torvalds [Mon, 10 Jan 2011 15:43:54 +0000 (07:43 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/mszeredi/fuse

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  fuse: fix ioctl ABI
  fuse: allow batching of FORGET requests
  fuse: separate queue for FORGET requests
  fuse: ioctl cleanup

Fix up trivial conflict in fs/fuse/inode.c due to RCU lookup having done
the RCU-freeing of the inode in fuse_destroy_inode().

13 years agopipe_fs_i.h: fix kernel-doc warning
Randy Dunlap [Sun, 9 Jan 2011 03:40:33 +0000 (19:40 -0800)]
pipe_fs_i.h: fix kernel-doc warning

Fix kernel-doc notation warnings in pipe_fs_i.h:

  Warning(include/linux/pipe_fs_i.h:58): No description found for parameter 'buffers'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agohrtimer.h: fix kernel-doc warning
Randy Dunlap [Sun, 9 Jan 2011 03:38:02 +0000 (19:38 -0800)]
hrtimer.h: fix kernel-doc warning

Fix new kernel-doc notation warning in hrtimer.h:

  Warning(include/linux/hrtimer.h:150): Excess struct/union/enum/typedef member 'first' description in 'hrtimer_clock_base'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agofs: fix dcache.h kernel-doc notation
Randy Dunlap [Sun, 9 Jan 2011 03:37:20 +0000 (19:37 -0800)]
fs: fix dcache.h kernel-doc notation

Fix new kernel-doc notation warning in dcache.h:

  Warning(include/linux/dcache.h:316): Excess function parameter 'Returns' description in '__d_rcu_to_refcount'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Nick Piggin <npiggin@kernel.dk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agofs: fix namei.c kernel-doc notation
Randy Dunlap [Sun, 9 Jan 2011 03:36:21 +0000 (19:36 -0800)]
fs: fix namei.c kernel-doc notation

Fix new kernel-doc notation warnings in fs/namei.c and spell
ECHILD correctly.

  Warning(fs/namei.c:218): No description found for parameter 'flags'
  Warning(fs/namei.c:425): Excess function parameter 'Returns' description in 'nameidata_drop_rcu'
  Warning(fs/namei.c:478): Excess function parameter 'Returns' description in 'nameidata_dentry_drop_rcu'
  Warning(fs/namei.c:540): Excess function parameter 'Returns' description in 'nameidata_drop_rcu_last'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Nick Piggin <npiggin@kernel.dk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoBlackfin: encode cpu-rev into uImage name
Mike Frysinger [Fri, 7 Jan 2011 22:44:06 +0000 (22:44 +0000)]
Blackfin: encode cpu-rev into uImage name

Encoding the cpu family name apparently confuses people when they try to
boot an image on a sub-variant, so encode the specific cpu name and the
silicon rev instead.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: bf54x: don't ack GPIO ints when unmasking them
Mike Frysinger [Fri, 10 Dec 2010 20:42:59 +0000 (20:42 +0000)]
Blackfin: bf54x: don't ack GPIO ints when unmasking them

When unmasking a GPIO interrupt on a BF54x part, the code will implicitly
ack any pending interrupts.  This is not what unmasking should do and can
cause people to miss interrupts from their devices, so punt the code.

Reported-by: Rutger Hofman <rutger@cs.vu.nl>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: sram_free_with_lsl: do not ignore return value of sram_free
Mike Frysinger [Mon, 6 Dec 2010 21:12:23 +0000 (21:12 +0000)]
Blackfin: sram_free_with_lsl: do not ignore return value of sram_free

If there was an error in the lower free functions, we need to pass that
back up so the calling process is able to check things.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: boards: add missing "static" to peripheral lists
Mike Frysinger [Wed, 24 Nov 2010 09:23:04 +0000 (09:23 +0000)]
Blackfin: boards: add missing "static" to peripheral lists

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: DNP5370: new board port
Andreas Schallenberg [Wed, 24 Nov 2010 22:15:16 +0000 (22:15 +0000)]
Blackfin: DNP5370: new board port

Signed-off-by: Andreas Schallenberg <Andreas.Schallenberg@3alitydigital.de>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: bf518f-ezbrd: fix dsa resources
Sonic Zhang [Tue, 23 Nov 2010 10:00:15 +0000 (10:00 +0000)]
Blackfin: bf518f-ezbrd: fix dsa resources

On bf518-ezbrd with DSA Ethernet switch enabled, on chip MAC is
connecting to phy at address 3.  If allowing the netdevice framework to
probe both 3 phys of the switch when registering MAC phy, phy at address
1 is checked for link active status other than phy at address 3.  If
connecting a cable to port 2 and leave port 1 open, link status in phy
2 and 3 are online, while that in phy 1 is offline.  So, the phy layer
sets wrong offline status to net device on port 3.  In this case, no data
can be transferred via ethernet port 2.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: move "-m elf32bfin" to general LDFLAGS
Mike Frysinger [Wed, 17 Nov 2010 07:34:58 +0000 (07:34 +0000)]
Blackfin: move "-m elf32bfin" to general LDFLAGS

Common code now invokes the linker directly which causes build failures
when using an FDPIC toolchain.  So move the emulation setting out of the
module-specific LDFLAGS and into the common LDFLAGS.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: kgdb_test: make sure to initialize num2
Vivi Li [Wed, 17 Nov 2010 04:26:19 +0000 (04:26 +0000)]
Blackfin: kgdb_test: make sure to initialize num2

We check its value at runtime, so we want to avoid garbage across runs.

Signed-off-by: Vivi Li <vivi.li@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: kgdb: disable preempt schedule when running single step in kgdb
Sonic Zhang [Wed, 17 Nov 2010 06:52:39 +0000 (06:52 +0000)]
Blackfin: kgdb: disable preempt schedule when running single step in kgdb

Otherwise, gdb continue operation after a breakpoint is hit may trap
into endless breakpoint.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: kgdb: disable interrupt when single stepping in ADEOS
Sonic Zhang [Wed, 17 Nov 2010 06:19:28 +0000 (06:19 +0000)]
Blackfin: kgdb: disable interrupt when single stepping in ADEOS

When ADEOS and kgdb are both enabled, single step in linux kernel may be
scheduled to Xenomai core after return from interrupt handlers.  This
blocks gdb continue operation after a break point is hit.  So, disable
interrupt when running gdb single step.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: SMP: kgdb: apply anomaly 257 work around
Sonic Zhang [Tue, 16 Nov 2010 09:41:07 +0000 (09:41 +0000)]
Blackfin: SMP: kgdb: apply anomaly 257 work around

When run kgdb testing, it looks like coreb hangs in single step or trap
exception without handling anomaly 05000257 properly on bf561 v0.5.  But,
the anomaly list says it apply to bf561 v0.4 and bellow.  Apply its work
around to 0.5 temporarily until the behavior and the root cause can be
confirmed by the hardware team.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: fix building IPIPE code when XIP is enabled
Mike Frysinger [Sun, 14 Nov 2010 00:02:53 +0000 (00:02 +0000)]
Blackfin: fix building IPIPE code when XIP is enabled

The low level assembly needs to use the pseudo_long_call helper so that
we use the right call insn when doing kernel XIP.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: SMP: kgdb: flush core internal write buffer before flushinv
Sonic Zhang [Fri, 12 Nov 2010 05:54:32 +0000 (05:54 +0000)]
Blackfin: SMP: kgdb: flush core internal write buffer before flushinv

KGDB single step in SMP kernel may hang forever in flushinv without a
CSYNC ahead.  This is because the core internal write buffers need to
be flushed before invalidating the data cache to make sure the insn
fetch is not out of sync.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: sport_uart resources: remove unused secondary RX/TX pins
Sonic Zhang [Fri, 12 Nov 2010 02:45:38 +0000 (02:45 +0000)]
Blackfin: sport_uart resources: remove unused secondary RX/TX pins

The SPORT/UART driver doesn't use the secondary channel pins, so don't
try and request them thus keeping other drivers from using them.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: tll6527m: fix spelling in unused code (struct name)
Stefan Weil [Sun, 7 Nov 2010 21:14:53 +0000 (22:14 +0100)]
Blackfin: tll6527m: fix spelling in unused code (struct name)

platfrom -> platform

This changes a struct name. The related code is conditionally
compiled and won't work because the include file linux/gpio-decoder.h
is missing, so removing this code would be an even better solution.

If the missing include file is added, it must fix the spelling, too.

Signed-off-by: Stefan Weil <weil@mail.berlios.de>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: bf527-ezkit: add adau1373 chip address
Sonic Zhang [Fri, 5 Nov 2010 11:29:46 +0000 (11:29 +0000)]
Blackfin: bf527-ezkit: add adau1373 chip address

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: no-mpu: fix masking of small uncached dma region
Sonic Zhang [Thu, 4 Nov 2010 08:01:37 +0000 (08:01 +0000)]
Blackfin: no-mpu: fix masking of small uncached dma region

When using an uncached DMA region less than 1 MiB, we try to mask off
the whole last 1 MiB for it.  Unfortunately, this fails as we forgot
to subtract one from the calculated mask, leading to the region still
be marked as cacheable.

Reported-by: Andrew Rook <andrew.rook@speakerbus.co.uk>
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: pm: drop irq save/restore in standby and suspend to mem callback
Steven Miao [Wed, 3 Nov 2010 09:25:09 +0000 (09:25 +0000)]
Blackfin: pm: drop irq save/restore in standby and suspend to mem callback

Since linux-2.6.31, the kernel suspend framework will do disable_irq/enable_irq,
so save/restore irq in standby and suspend to mem callback should be dropped.
Otherwise the common code notices things are enabled and complains.

Signed-off-by: Steven Miao <realmz6@gmail.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoMAINTAINERS: update Analog Devices support info
Michael Hennerich [Thu, 28 Oct 2010 11:31:28 +0000 (11:31 +0000)]
MAINTAINERS: update Analog Devices support info

Add more Analog Devices driver entries and refine the differences between
Blackfin-specific code and processor independent drivers.

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: dpmc.h: pull in new pll.h
Mike Frysinger [Thu, 28 Oct 2010 19:34:09 +0000 (15:34 -0400)]
Blackfin: dpmc.h: pull in new pll.h

Any consumer of dpmc.h expects to use VR_CTL, so also pull in the new
mach/pll.h header for those functions.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: bf54x: add MMR layout for PINT
Mike Frysinger [Wed, 27 Oct 2010 20:21:51 +0000 (20:21 +0000)]
Blackfin: bf54x: add MMR layout for PINT

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: bf561: SMP: add multicore pll handlers
Mike Frysinger [Wed, 27 Oct 2010 20:32:24 +0000 (16:32 -0400)]
Blackfin: bf561: SMP: add multicore pll handlers

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: unify pll.h headers
Mike Frysinger [Wed, 27 Oct 2010 19:29:26 +0000 (15:29 -0400)]
Blackfin: unify pll.h headers

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
13 years agoBlackfin: SMP: rewrite IPI handling to avoid memory allocation
Yi Li [Thu, 17 Dec 2009 08:20:32 +0000 (08:20 +0000)]
Blackfin: SMP: rewrite IPI handling to avoid memory allocation

Currently, sending an interprocessor interrupt (IPI) requires building up
a message dynamically which means memory allocation.  But often times, we
will want to send an IPI in low level contexts where allocation is not
possible which may lead to a panic().  So create a per-cpu static array
for the message queue and use that instead.

Further, while we have two supplemental interrupts, we are currently only
using one of them.  So use the second one for the most common IPI message
of all -- smp_send_reschedule().  This avoids ugly contention for locks
which in turn would require an IPI message ...

In general, this improves SMP performance, and in some cases allows the
SMP port to work in places it wouldn't before.  Such as the PREEMPT_RT
state where the slab is protected by a per-cpu spin lock.  If the slab
kmalloc/kfree were to put the task to sleep, and that task was actually
the IPI handler, then the system falls down yet again.

After running some various stress tests on the system, the static limit
of 5 messages seems to work.  On the off chance even this overflows, we
simply panic(), and we can review that scenario to see if the limit needs
to be increased a bit more.

Signed-off-by: Yi Li <yi.li@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>