pandora-kernel.git
13 years agoMerge branch 'hwmod' into tmp-merge
Tony Lindgren [Tue, 6 Mar 2012 22:32:56 +0000 (14:32 -0800)]
Merge branch 'hwmod' into tmp-merge

Conflicts:
arch/arm/plat-omap/omap_device.c

13 years agoMerge branch 'cleanup-pm' into tmp-merge
Tony Lindgren [Tue, 6 Mar 2012 22:26:03 +0000 (14:26 -0800)]
Merge branch 'cleanup-pm' into tmp-merge

13 years agoMerge branch 'fixes' into tmp-merge
Tony Lindgren [Tue, 6 Mar 2012 22:22:01 +0000 (14:22 -0800)]
Merge branch 'fixes' into tmp-merge

Conflicts:
arch/arm/mach-omap2/id.c

13 years agoMerge branch 'soc' into tmp-merge
Tony Lindgren [Tue, 6 Mar 2012 22:19:27 +0000 (14:19 -0800)]
Merge branch 'soc' into tmp-merge

13 years agoMerge branch 'fixes-non-critical-part2' into tmp-merge
Tony Lindgren [Tue, 6 Mar 2012 21:57:55 +0000 (13:57 -0800)]
Merge branch 'fixes-non-critical-part2' into tmp-merge

Conflicts:
arch/arm/mach-omap2/io.c

13 years agoMerge branch 'for_3.4/cleanup/pm' of git://git.kernel.org/pub/scm/linux/kernel/git...
Tony Lindgren [Tue, 6 Mar 2012 21:32:47 +0000 (13:32 -0800)]
Merge branch 'for_3.4/cleanup/pm' of git://git./linux/kernel/git/khilman/linux-omap-pm into cleanup-pm

13 years agoMerge branch 'fixes-non-critical' into tmp-merge
Tony Lindgren [Tue, 6 Mar 2012 21:29:32 +0000 (13:29 -0800)]
Merge branch 'fixes-non-critical' into tmp-merge

Conflicts:
arch/arm/mach-omap2/Kconfig

13 years agoARM: OMAP2+: PM: fix wakeupgen warning when hotplug disabled
Kevin Hilman [Tue, 6 Mar 2012 20:00:25 +0000 (12:00 -0800)]
ARM: OMAP2+: PM: fix wakeupgen warning when hotplug disabled

When CONFIG_HOTPLUG_CPU=n, there are unused functions in wakeupgen:

arch/arm/mach-omap2/omap-wakeupgen.c:181: warning: 'wakeupgen_irqmask_all' defined but not used

Fix this by moving all the functions only used when CONFIG_HOTPLUG_CPU=y
together and wrapping in an #ifdef.

No functional changes.

Reported-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: PM: fix section mismatch with omap2_init_processor_devices()
Kevin Hilman [Tue, 6 Mar 2012 19:38:01 +0000 (11:38 -0800)]
ARM: OMAP2+: PM: fix section mismatch with omap2_init_processor_devices()

Fix the below warning by making omap2_init_processor_devices() __init.
It is called by an __init function and calls only __init functions, so
it should also be init.

WARNING: arch/arm/mach-omap2/built-in.o(.text+0x183c): Section mismatch in reference from the function omap2_init_processor_devices() to the function .init.text:_init_omap_device()
The function omap2_init_processor_devices() references
the function __init _init_omap_device().
This is often because omap2_init_processor_devices lacks a __init
annotation or the annotation of _init_omap_device is wrong.

Reported-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2: Fix section warning for n8x0 when CONFIG_MMC_OMAP is not set
Tony Lindgren [Tue, 6 Mar 2012 19:49:28 +0000 (11:49 -0800)]
ARM: OMAP2: Fix section warning for n8x0 when CONFIG_MMC_OMAP is not set

Otherwise we get:

arch/arm/mach-omap2/board-n8x0.c:39:12: warning:
'slot1_cover_open' defined but not used [-Wunused-variable]
arch/arm/mach-omap2/board-n8x0.c:40:12: warning:
'slot2_cover_open' defined but not used [-Wunused-variable]
arch/arm/mach-omap2/board-n8x0.c:41:23: warning:
'mmc_device' defined but not used [-Wunused-variable]

Reported-by: Russell King <linux@arm.linux.org.uk>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: Fix omap24xx_io_desc warning if SoC subtypes are not selected
Tony Lindgren [Tue, 6 Mar 2012 19:49:22 +0000 (11:49 -0800)]
ARM: OMAP2+: Fix omap24xx_io_desc warning if SoC subtypes are not selected

Otherwise we get the following warning:

arch/arm/mach-omap2/io.c:53:24: warning:
'omap24xx_io_desc' defined but not used [-Wunused-variable]

Reported-by: Russell King <linux@arm.linux.org.uk>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP1: Fix section mismatch for omap1_init_early()
Tony Lindgren [Tue, 6 Mar 2012 00:11:04 +0000 (16:11 -0800)]
ARM: OMAP1: Fix section mismatch for omap1_init_early()

Fix the following warning:

WARNING: vmlinux.o(.text+0x1286c): Section mismatch in reference
from the function omap1_init_early() to the function .init.text:omap1_clk_init()
The function omap1_init_early() references
the function __init omap1_clk_init().
This is often because omap1_init_early lacks a __init
annotation or the annotation of omap1_clk_init is wrong.

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP1: Fix typo in lcd_dma.c
Masanari Iida [Tue, 6 Mar 2012 00:11:04 +0000 (16:11 -0800)]
ARM: OMAP1: Fix typo in lcd_dma.c

Correct spelling "resulotion" to "resolution" in
arch/arm/mach-omap1/lcd_dma.c

Signed-off-by: Masanari Iida <standby24x7@gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: mailbox: trivial whitespace fix
Felipe Contreras [Tue, 6 Mar 2012 00:11:03 +0000 (16:11 -0800)]
ARM: OMAP: mailbox: trivial whitespace fix

Trivial whitespace fix

Signed-off-by: Felipe Contreras <felipe.contreras@nokia.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: Remove definition cpu_is_omap4430()
Jon Hunter [Tue, 6 Mar 2012 00:11:03 +0000 (16:11 -0800)]
ARM: OMAP: Remove definition cpu_is_omap4430()

The definition cpu_is_omap4430() always returns 0 even when CONFIG_ARCH_OMAP4
is enabled. This macro should be removed and the macro cpu_is_omap443x() should
be used where needed for OMAP4430 devices.

Signed-off-by: Jon Hunter <jon-hunter@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: included some headers twice
Danny Kukawka [Tue, 6 Mar 2012 00:11:03 +0000 (16:11 -0800)]
ARM: OMAP2+: included some headers twice

arch/arm/mach-omap2/: included some headers tiwce:
- arch/arm/mach-omap2/board-ldp.c: 'linux/gpio.h'
- arch/arm/mach-omap2/io.c: 'common.h'
- arch/arm/mach-omap2/omap_hwmod_44xx_data.c: 'plat/i2c.h'

Remove the duplicates.

Signed-off-by: Danny Kukawka <danny.kukawka@bisect.de>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: clock.c: included linux/debugfs.h twice
Danny Kukawka [Tue, 6 Mar 2012 00:11:03 +0000 (16:11 -0800)]
ARM: OMAP: clock.c: included linux/debugfs.h twice

arch/arm/plat-omap/clock.c: included 'linux/debugfs.h' twice,
remove the duplicate.

Signed-off-by: Danny Kukawka <danny.kukawka@bisect.de>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: don't build hwspinlock in vain
Ohad Ben-Cohen [Tue, 6 Mar 2012 00:11:02 +0000 (16:11 -0800)]
ARM: OMAP: don't build hwspinlock in vain

Cleanup: don't build mach-omap2/hwspinlock.c if the OMAP hwspinlock
driver isn't configured.

This will both shorten build time and avoid registering a device
which isn't needed.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: ads7846_init: put gpio_pendown into pdata if it's provided
Ilya Yanok [Tue, 6 Mar 2012 00:11:02 +0000 (16:11 -0800)]
ARM: OMAP2+: ads7846_init: put gpio_pendown into pdata if it's provided

If platform data is provided by the caller gpio_pendown is put into
unused static ads7846_config structure and effectively has no effect.
Of course caller can set gpio_pendown field in platform data himself
but it seems natural to do this in ads7846_init to remove duplication.

Signed-off-by: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Ilya Yanok <yanok@emcraft.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: omap: pandora: fix usbhs platform data
Grazvydas Ignotas [Tue, 6 Mar 2012 00:11:02 +0000 (16:11 -0800)]
ARM: omap: pandora: fix usbhs platform data

It turned out wrong OMAP HSUSB port was configured on pandora,
but still managed to work somehow. This was noticed after enabling
in-kernel mux, where USB muxing was causing other devices not to work,
because hsusb1 pins (instead of hsusb2) were wrongly remuxed, which
are used for other things on pandora.

Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: sram: Add am33xx SRAM support (minimal)
Vaibhav Bedia [Tue, 6 Mar 2012 00:11:01 +0000 (16:11 -0800)]
ARM: OMAP: sram: Add am33xx SRAM support (minimal)

Update SRAM start & size for am33xx SoC's.

Note: cpu_is_34xx() is true for am33xx also. Doing
cpu_is_am33xx() check after cpu_is_34xx() will not
achieve what we want due to the above reason.
Hence cpu_is_am33xx() is done before cpu_is_34xx()

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
Signed-off-by: Afzal Mohammed <afzal@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: id: Add am33xx SoC type detection
Afzal Mohammed [Tue, 6 Mar 2012 00:11:01 +0000 (16:11 -0800)]
ARM: OMAP2+: id: Add am33xx SoC type detection

Determine SoC type, i.e. whether GP or HS

Note: cpu_is_34xx() is true for am33xx also. Doing
cpu_is_am33xx() check after cpu_is_34xx() will not
achieve what we want due to the above reason.
Hence cpu_is_am33xx() is done before cpu_is_34xx()

Signed-off-by: Afzal Mohammed <afzal@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: GPMC: Export gpmc_enable_hwecc and gpmc_calculate_ecc
Bernhard Walle [Tue, 6 Mar 2012 00:11:01 +0000 (16:11 -0800)]
ARM: OMAP2+: GPMC: Export gpmc_enable_hwecc and gpmc_calculate_ecc

To be able to compile kernel/drivers/mtd/nand/omap2.ko as module, that
two symbols need to be exported. Otherwise, I get following error
message

    ERROR: "gpmc_calculate_ecc" [drivers/mtd/nand/omap2.ko] undefined!
    ERROR: "gpmc_enable_hwecc" [drivers/mtd/nand/omap2.ko] undefined!

Signed-off-by: Bernhard Walle <walle@corscience.de>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: dmtimer: fix missing content/correction in low-power mode support
Tarun Kanti DebBarma [Tue, 6 Mar 2012 00:11:00 +0000 (16:11 -0800)]
ARM: OMAP: dmtimer: fix missing content/correction in low-power mode support

Since omap_dm_timer_write_reg/__omap_dm_timer_write is now modified
to use timer->func_base OCP_CFG should not use this wrapper anymore.
Instead use __raw_writel() directly and use timer->io_base instead
to write to OCP_CFG.

The timer->sys_stat is valid only if timer->revision is 1. In the
context restore function make this correction.

Save the contexts and loss count when timer is stopped.
Also, disable the clock. Else, clock usecount would become imbalanced.

Signed-off-by: Tarun Kanti DebBarma <tarun.kanti@ti.com>
Tested-by: Omar Ramirez Luna <omar.ramirez@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoRevert "ARM: OMAP2+: Fix multiple randconfig errors with SOC_OMAP and SOC_OMAP_NOOP"
Tony Lindgren [Tue, 6 Mar 2012 00:02:18 +0000 (16:02 -0800)]
Revert "ARM: OMAP2+: Fix multiple randconfig errors with SOC_OMAP and SOC_OMAP_NOOP"

This reverts commit c295fb633e321a7df3b8846c4eaddc5da3e0aaed.

This makes existing .config files bloated by selecting in all
omaps as noted by Russell King - ARM Linux <linux@arm.linux.org.uk>.

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: PM: share some suspend-related functions across OMAP2, 3, 4
Paul Walmsley [Thu, 2 Feb 2012 09:30:50 +0000 (02:30 -0700)]
ARM: OMAP2+: PM: share some suspend-related functions across OMAP2, 3, 4

The platform_suspend_ops can be shared across OMAP2, 3, and 4, along
with all of the functions referenced in that structure.  This patch
shares them.  It also removes the suspend_state file-scoped variable
in the OMAP2 and 3 PM code; it does not appear to be actually needed
by anything.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Santosh Shilimkar <santosh.shilimkar@ti.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Reviewed-by: Kevin Hilman <khilman@ti.com>
[khilman@ti.com: minor rework needed due to rebase/merge with conflicting changes]
Signed-off-by: Kevin Hilman <khilman@ti.com>
13 years agoARM: OMAP2+: omap_device: call all suspend, resume callbacks when OMAP_DEVICE_NO_IDLE...
Paul Walmsley [Sat, 3 Mar 2012 20:15:33 +0000 (13:15 -0700)]
ARM: OMAP2+: omap_device: call all suspend, resume callbacks when OMAP_DEVICE_NO_IDLE_ON_SUSPEND is set

During system suspend, when OMAP_DEVICE_NO_IDLE_ON_SUSPEND is set on
an omap_device, call the corresponding driver's ->suspend() and
->suspend_noirq() callbacks (if present).  Similarly, during resume,
the driver's ->resume() and ->resume_noirq() callbacks must both be
called, if present.  (The previous code only called ->suspend_noirq()
and ->resume_noirq().)

If all of these callbacks aren't called, some important driver
suspend/resume code may not get executed.

In current mainline, the bug fixed by this patch is only a problem
under the following conditions:

- the kernel is running on an OMAP4

- an OMAP UART is used as a console

- the kernel command line parameter 'no_console_suspend' is specified

- and the system enters suspend ("echo mem > /sys/power/state").

Under this combined circumstance, the system cannot be awakened via
the serial port after commit be4b0281956c5cae4f63f31f11d07625a6988766c
("tty: serial: OMAP: block idle while the UART is transferring data in
PIO mode").  This is because the OMAP UART driver's ->suspend()
callback is never called.  The ->suspend() callback would have called
uart_suspend_port() which in turn would call enable_irq_wake().  Since
enable_irq_wake() isn't called for the UART's IRQ, check_wakeup_irqs()
would mask off the UART IRQ in the GIC.

On v3.3 kernels prior to the above commit, serial resume from suspend
presumably occurred via the PRCM interrupt.  The UART was in
smart-idle mode, so it was able to send a PRCM wakeup which in turn
would be converted into a PRCM interrupt to the GIC, waking up the
kernel.  But after the above commit, when the system is suspended in
the middle of a UART transmit, the UART IP block would be in no-idle
mode.  In no-idle mode, the UART won't generate wakeups to the PRCM
when incoming characters are received; only GIC interrupts.  But since
the UART driver's ->suspend() callback is never called,
uart_suspend_port() and enable_irq_wake() is never called; so the UART
interrupt is masked by check_wakeup_irqs() and the UART can't wake up
the MPU.

The remaining mechanism that could have awakened the system would have
been I/O chain wakeups.  These wouldn't be active because the console
UART's clocks are never disabled when no_console_suspend is used,
preventing the full chip from idling.  Also, current mainline doesn't
yet support full chip idle states for OMAP4, so I/O chain wakeups are
not enabled.

This patch is the result of a collaboration.  John Stultz
<johnstul@us.ibm.com> and Andy Green <andy.green@linaro.org> reported
the serial wakeup problem that led to the discovery of this problem.
Kevin Hilman <khilman@ti.com> narrowed the problem down to the use of
no_console_suspend.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: John Stultz <johnstul@us.ibm.com>
Cc: Andy Green <andy.green@linaro.org>
Reviewed-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>
13 years agoARM: OMAP: omap_device: remove omap_device_parent
Kevin Hilman [Wed, 15 Feb 2012 19:47:45 +0000 (11:47 -0800)]
ARM: OMAP: omap_device: remove omap_device_parent

Currently all omap_devices are forced to have the dummy device
'omap_device_parent' as a parent.  This was used to distinguish
omap_devices from "normal" platform_devices in the OMAP PM core code.

Now that we implement the PM core using PM domains, this is no longer
needed, and is removed.

This also frees up omap_devices to have a more complex parent/child
relationships that model actual device relationships.

The only in-tree user of omap_device_parent was the OMAP PM layer to
handle lost-context count for omap_devices.  That is now converted to
use the presence of the omap_device_pm_domain instead.

Signed-off-by: Kevin Hilman <khilman@ti.com>
13 years agoARM: OMAP2+: PM debug: fix the use of debugfs_create_* API
Jean Pihet [Wed, 1 Feb 2012 08:37:31 +0000 (09:37 +0100)]
ARM: OMAP2+: PM debug: fix the use of debugfs_create_* API

Check the return code pointer value from debugfs_create_dir for error
or NULL.
Also added an additional check to prevent the creation of a 'suspend'
entry at the debugfs root in case a power domain directory cannot be
created.

Signed-off-by: Jean Pihet <j-pihet@ti.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>
13 years agoARM: OMAP2+: PM: share clkdms_setup() across OMAP2, 3, 4
Paul Walmsley [Thu, 2 Feb 2012 09:38:50 +0000 (02:38 -0700)]
ARM: OMAP2+: PM: share clkdms_setup() across OMAP2, 3, 4

clkdms_setup() is identical across OMAP2, 3, and 4, so share it.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Tested-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>
13 years agoARM: OMAP2+: PM: clean up omap_set_pwrdm_state()
Paul Walmsley [Mon, 30 Jan 2012 09:47:24 +0000 (02:47 -0700)]
ARM: OMAP2+: PM: clean up omap_set_pwrdm_state()

Clean up a few different parts of omap_set_pwrdm_state():

- Remove a superfluous call to pwrdm_state_switch().  Not needed
  unless LOWPOWERSTATECHANGE is used, because the state switch code is
  called by either clkdm_sleep() or clkdm_allow_idle().

- Add code to wait for the power state transition in the OMAP4+ low
  power state change.  This is speculative, so I would particularly
  appreciate feedback on this part.

- Remove a superfluous call to pwrdm_read_pwrst().

- Update variable names to be more meaningful (hopefully) and precise.

- Fix an error path bug that would not place the clockdomain back into
  hardware-supervised idle or sleep mode if the power state could not
  be programmed.

The documentation for this function still needs major improvements;
that's left for a later patch.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Tested-by: Tero Kristo <t-kristo@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>
13 years agoARM: OMAP3: PM: remove superfluous calls to pwrdm_clear_all_prev_pwrst()
Paul Walmsley [Mon, 30 Jan 2012 09:47:24 +0000 (02:47 -0700)]
ARM: OMAP3: PM: remove superfluous calls to pwrdm_clear_all_prev_pwrst()

Remove some superfluous calls to pwrdm_clear_all_prev_pwrst().
pwrdm_pre_transition(), which appears a few lines after these calls,
invokes pwrdm_clear_all_prev_pwrst() on each powerdomain -- there's no
need to do it twice.

N.B.: some of us have observed that accesses to the previous
powerstate registers seem to be quite slow.  Although the writes
removed by this patch should be buffered by the write buffer, there is
a read to a PRM register immediately afterwards.  That will block the
OMAP3 MPU until all of those writes complete.  So this patch should
result in a minor performance improvement during idle entry.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Tero Kristo <t-kristo@ti.com>
[khilman@ti.com: removed a couple more for OMAP4]
Tested-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Kevin Hilman <khilman@ti.com>
13 years agoMerge remote-tracking branch 'omap/hsmmc' into for_3.4/cleanup/pm-base
Kevin Hilman [Mon, 5 Mar 2012 23:37:04 +0000 (15:37 -0800)]
Merge remote-tracking branch 'omap/hsmmc' into for_3.4/cleanup/pm-base

13 years agoARM: OMAP2+: Fix module build errors with CONFIG_OMAP4_ERRATA_I688
R Sricharan [Fri, 2 Mar 2012 11:01:18 +0000 (16:31 +0530)]
ARM: OMAP2+: Fix module build errors with CONFIG_OMAP4_ERRATA_I688

While building modules with randconfig the below errors are observed.

ERROR: "omap_bus_sync" [drivers/watchdog/sp805_wdt.ko] undefined!
ERROR: "omap_bus_sync" [drivers/watchdog/dw_wdt.ko] undefined!
ERROR: "omap_bus_sync" [drivers/virtio/virtio_ring.ko] undefined!
ERROR: "omap_bus_sync" [drivers/video/sm501fb.ko] undefined!
ERROR: "omap_bus_sync" [drivers/usb/mon/usbmon.ko] undefined!
ERROR: "omap_bus_sync" [drivers/usb/host/sl811-hcd.ko] undefined!
ERROR: "omap_bus_sync" [drivers/usb/host/ohci-hcd.ko] undefined!
ERROR: "omap_bus_sync" [drivers/usb/host/isp1760.ko] undefined!
ERROR: "omap_bus_sync" [drivers/usb/host/isp1362-hcd.ko] undefined!
ERROR: "omap_bus_sync" [drivers/usb/host/isp116x-hcd.ko] undefined!
ERROR: "omap_bus_sync" [drivers/usb/core/usbcore.ko] undefined!
ERROR: "omap_bus_sync" [drivers/tty/serial/altera_uart.ko] undefined!
ERROR: "omap_bus_sync" [drivers/tty/serial/altera_jtaguart.ko] undefined!
ERROR: "omap_bus_sync" [drivers/tty/serial/8250/8250_dw.ko] undefined!
ERROR: "omap_bus_sync" [drivers/ssb/ssb.ko] undefined!
ERROR: "omap_bus_sync" [drivers/rtc/rtc-cmos.ko] undefined!
ERROR: "omap_bus_sync" [drivers/rtc/rtc-bq4802.ko] undefined!
ERROR: "omap_bus_sync" [drivers/mtd/nand/tmio_nand.ko] undefined!
ERROR: "omap_bus_sync" [drivers/mtd/nand/omap2.ko] undefined!

Signed-off-by: R Sricharan <r.sricharan@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: id: Add missing break statement in omap3xxx_check_revision
Vaibhav Hiremath [Mon, 23 Jan 2012 07:56:47 +0000 (13:26 +0530)]
ARM: OMAP: id: Add missing break statement in omap3xxx_check_revision

Add missing break statement in the function omap3xxx_check_revision.

The commit id 4390f5b2cb1f56 [ARM: OMAP: TI814X: Add cpu type macros
and detection support], removed the 'break' statement from the function
omap3xxx_check_revision(), resulting into wrong omap/cpu_revision
initialization for AM335x devices.

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
[tony@atomide.com: refreshed to apply after changes to cpu_rev]
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: Fix L4_EMU_34XX_BASE error after iomap changes
Tony Lindgren [Wed, 29 Feb 2012 01:28:09 +0000 (17:28 -0800)]
ARM: OMAP2+: Fix L4_EMU_34XX_BASE error after iomap changes

With the introduction of iomap changes platform init code
fails for emu.c if CONFIG_OMAP3_EMU is selected:

arch/arm/mach-omap2/emu.c:35:8: error:
'L4_EMU_34XX_BASE' undeclared here (not in a function)

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoMerge branch 'fixes-non-critical' into tmp-merge
Tony Lindgren [Wed, 29 Feb 2012 00:45:52 +0000 (16:45 -0800)]
Merge branch 'fixes-non-critical' into tmp-merge

Conflicts:
arch/arm/mach-omap2/Kconfig

13 years agoMerge branch 'dt' into tmp-merge
Tony Lindgren [Wed, 29 Feb 2012 00:44:50 +0000 (16:44 -0800)]
Merge branch 'dt' into tmp-merge

13 years agoMerge branch 'hsmmc' into fixes-non-critical
Tony Lindgren [Wed, 29 Feb 2012 00:15:06 +0000 (16:15 -0800)]
Merge branch 'hsmmc' into fixes-non-critical

13 years agoARM: OMAP2+: Remove apply_uV constraints for fixed regulator
Rajendra Nayak [Tue, 28 Feb 2012 09:32:34 +0000 (15:02 +0530)]
ARM: OMAP2+: Remove apply_uV constraints for fixed regulator

VUSB is a fixed regulator, so get rid of the apply_uV constraint
for it, which fixes the following error seen at boot on omap4
SDP and PANDA boards.

machine_constraints_voltage: VUSB: failed to apply 3300000uV constraint
twl_reg twl_reg.46: can't register VUSB, -22
twl_reg: probe of twl_reg.46 failed with error -22

Signed-off-by: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: irqs: Fix NR_IRQS value to handle PRCM interrupts
Cousson, Benoit [Tue, 28 Feb 2012 13:10:09 +0000 (14:10 +0100)]
ARM: OMAP: irqs: Fix NR_IRQS value to handle PRCM interrupts

The following commit: 2f31b51659c2d8315ea2888ba5b93076febe672b
Author: Tero Kristo <t-kristo@ti.com>
Date:   Fri Dec 16 14:37:00 2011 -0700

    ARM: OMAP4: PRM: use PRCM interrupt handler

introduced the PRCM interrupt handler and thus the need
for 64 more interrupts. Since SPARSE_IRQ is still not fully
functional on OMAP, the NR_IRQS needs to be updated to avoid
the failure that happen during irq_alloc_descs call inside
the PRCM driver:

[    0.208221] PRCM: failed to allocate irq descs: -12

Later the mux framework is then unable to request an IRQ from
the PRCM interrupt handler.

[    1.802795] mux: Failed to setup hwmod io irq -22

Fix that by adding 64 more interrupts for OMAP2PLUS config.

Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Cc: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoMerge branch 'randconfig' into fixes-non-critical
Tony Lindgren [Tue, 28 Feb 2012 22:43:18 +0000 (14:43 -0800)]
Merge branch 'randconfig' into fixes-non-critical

13 years agoOMAP4: dma: Correct CPU version check for dma_common_ch_end
Peter Ujfalusi [Thu, 23 Feb 2012 07:44:41 +0000 (09:44 +0200)]
OMAP4: dma: Correct CPU version check for dma_common_ch_end

CCDN is the last common channel register in all OMAP4 versions. Use
cpu_is_omap44xx() instead of the cpu_is_omap4430().
cpu_is_omap4430() returns 0 unconditionally. This causes that the
dma_common_ch_end register variable is not configured correctly on OMAP4, not
even for OMAP4430.
Because of this, registers between CCFN - CCDN will be not cleard in the
omap2_clear_dma function in OMAP4.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: Fix section mismatch warning for platform_cpu_die()
Santosh Shilimkar [Wed, 22 Feb 2012 10:40:04 +0000 (16:10 +0530)]
ARM: OMAP: Fix section mismatch warning for platform_cpu_die()

WARNING: vmlinux.o(.text+0x226d0):
Section mismatch in reference from the function
platform_cpu_die() to the function .cpuinit.text:omap4_hotplug_cpu()
The function platform_cpu_die() references
the function __cpuinit omap4_hotplug_cpu().
This is often because platform_cpu_die lacks a __cpuinit
annotation or the annotation of omap4_hotplug_cpu is wrong.

Thanks to Russell King for suggesting to use __ref instead of
the initial (and wrong) approach to use __cpuinit.

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: fix section mismatch warning for omap4_hotplug_cpu()
Santosh Shilimkar [Thu, 23 Feb 2012 06:58:29 +0000 (12:28 +0530)]
ARM: OMAP: fix section mismatch warning for omap4_hotplug_cpu()

WARNING: arch/arm/mach-omap2/built-in.o(.text+0x8b80):
Section mismatch in reference from the function omap4_hotplug_cpu() to
the function .cpuinit.text:omap_secondary_startup()
The function omap4_hotplug_cpu() references
the function __cpuinit omap_secondary_startup().
This is often because omap4_hotplug_cpu lacks a __cpuinit
annotation or the annotation of omap_secondary_startup is wrong.

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoMerge branch 'iomap' into cleanup
Tony Lindgren [Tue, 28 Feb 2012 22:27:01 +0000 (14:27 -0800)]
Merge branch 'iomap' into cleanup

13 years agoLinux 3.3-rc5 v3.3-rc5
Linus Torvalds [Sat, 25 Feb 2012 20:18:16 +0000 (12:18 -0800)]
Linux 3.3-rc5

13 years agoMerge tag 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck...
Linus Torvalds [Sat, 25 Feb 2012 20:12:08 +0000 (12:12 -0800)]
Merge tag 'hwmon-for-linus' of git://git./linux/kernel/git/groeck/linux-staging

Couple of minor driver fixes.

* tag 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
  hwmon: (max34440) Fix resetting temperature history
  hwmon: (f75375s) Fix register write order when setting fans to full speed
  hwmon: (ads1015) Fix file leak in probe function
  hwmon: (max6639) Fix PPR register initialization to set both channels
  hwmon: (max6639) Fix FAN_FROM_REG calculation

13 years agoMerge branch 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild
Linus Torvalds [Sat, 25 Feb 2012 20:11:25 +0000 (12:11 -0800)]
Merge branch 'rc-fixes' of git://git./linux/kernel/git/mmarek/kbuild

three kbuild fixes for 3.3:
 - make deb-pkg symlink race fix.
 - make coccicheck fix.
 - Dropping the check for modutils.  This is not a regression, but
   allows the module-init-tools replacement kmod work with the 3.3
   kernel.

* 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild:
  coccicheck: change handling of C={1,2} when M= is set
  builddeb: Don't create files in /tmp with predictable names
  kbuild: do not check for ancient modutils tools

13 years agoautofs: work around unhappy compat problem on x86-64
Ian Kent [Wed, 22 Feb 2012 12:45:44 +0000 (20:45 +0800)]
autofs: work around unhappy compat problem on x86-64

When the autofs protocol version 5 packet type was added in commit
5c0a32fc2cd0 ("autofs4: add new packet type for v5 communications"), it
obvously tried quite hard to be word-size agnostic, and uses explicitly
sized fields that are all correctly aligned.

However, with the final "char name[NAME_MAX+1]" array at the end, the
actual size of the structure ends up being not very well defined:
because the struct isn't marked 'packed', doing a "sizeof()" on it will
align the size of the struct up to the biggest alignment of the members
it has.

And despite all the members being the same, the alignment of them is
different: a "__u64" has 4-byte alignment on x86-32, but native 8-byte
alignment on x86-64.  And while 'NAME_MAX+1' ends up being a nice round
number (256), the name[] array starts out a 4-byte aligned.

End result: the "packed" size of the structure is 300 bytes: 4-byte, but
not 8-byte aligned.

As a result, despite all the fields being in the same place on all
architectures, sizeof() will round up that size to 304 bytes on
architectures that have 8-byte alignment for u64.

Note that this is *not* a problem for 32-bit compat mode on POWER, since
there __u64 is 8-byte aligned even in 32-bit mode.  But on x86, 32-bit
and 64-bit alignment is different for 64-bit entities, and as a result
the structure that has exactly the same layout has different sizes.

So on x86-64, but no other architecture, we will just subtract 4 from
the size of the structure when running in a compat task.  That way we
will write the properly sized packet that user mode expects.

Not pretty.  Sadly, this very subtle, and unnecessary, size difference
has been encoded in user space that wants to read packets of *exactly*
the right size, and will refuse to touch anything else.

Reported-and-tested-by: Thomas Meyer <thomas@m3y3r.de>
Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoMerge tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland...
Linus Torvalds [Sat, 25 Feb 2012 04:03:14 +0000 (20:03 -0800)]
Merge tag 'rdma-for-linus' of git://git./linux/kernel/git/roland/infiniband

One InfiniBand/RDMA regression fix for 3.3:

 - mlx4 SR-IOV changes added static exported functions, which doesn't
   build on powerpc at least.  Fix from Doug Ledford for this.

* tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
  mlx4_core: Exported functions can't be static

13 years agoMerge branch 'for_3.4/dts_updates' of git://git.kernel.org/pub/scm/linux/kernel/git...
Tony Lindgren [Sat, 25 Feb 2012 00:48:27 +0000 (16:48 -0800)]
Merge branch 'for_3.4/dts_updates' of git://git./linux/kernel/git/bcousson/linux-omap-dt into dt

13 years agoMerge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi...
Linus Torvalds [Sat, 25 Feb 2012 00:08:51 +0000 (16:08 -0800)]
Merge tag 'scsi-fixes' of git://git./linux/kernel/git/jejb/scsi-rc-fixes-2.6

SCSI fixes on 20120224:
 "This is a set of assorted bug fixes for power management, mpt2sas,
  ipr, the rdac device handler and quite a big chunk for qla2xxx (plus a
  use after free of scsi_host in scsi_scan.c). "

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6:
  [SCSI] scsi_dh_rdac: Fix for unbalanced reference count
  [SCSI] scsi_pm: Fix bug in the SCSI power management handler
  [SCSI] scsi_scan: Fix 'Poison overwritten' warning caused by using freed 'shost'
  [SCSI] qla2xxx: Update version number to 8.03.07.13-k.
  [SCSI] qla2xxx: Proper detection of firmware abort error code for ISP82xx.
  [SCSI] qla2xxx: Remove resetting memory during device initialization for ISP82xx.
  [SCSI] qla2xxx: Complete mailbox command timedout to avoid initialization failures during next reset cycle.
  [SCSI] qla2xxx: Remove check for null fcport from host reset handler.
  [SCSI] qla2xxx: Correct out of bounds read of ISP2200 mailbox registers.
  [SCSI] qla2xxx: Remove errant clearing of MBX_INTERRUPT flag during CT-IOCB processing.
  [SCSI] qla2xxx: Clear options-flags while issuing stop-firmware mbx command.
  [SCSI] qla2xxx: Add an "is reset active" helper.
  [SCSI] qla2xxx: Add check for null fcport references in qla2xxx_queuecommand.
  [SCSI] qla2xxx: Propagate up abort failures.
  [SCSI] isci: Fix NULL ptr dereference when no firmware is being loaded
  [SCSI] ipr: fix eeh recovery for 64-bit adapters
  [SCSI] mpt2sas: Fix mismatch in mpt2sas_base_hard_reset_handler() mutex lock-unlock

13 years agococcicheck: change handling of C={1,2} when M= is set
Greg Dietsche [Fri, 20 Jan 2012 23:10:35 +0000 (17:10 -0600)]
coccicheck: change handling of C={1,2} when M= is set

This patch reverts a portion of d0bc1fb4 so that coccicheck will
work properly when C=1 or C=2.

Reported-and-tested-by: Brice Goglin <Brice.Goglin@inria.fr>
Signed-off-by: Greg Dietsche <Gregory.Dietsche@cuw.edu>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Michal Marek <mmarek@suse.cz>
13 years agoarm/dts: Add support for TI OMAP3 EVM board
Vaibhav Hiremath [Fri, 16 Dec 2011 07:05:53 +0000 (12:35 +0530)]
arm/dts: Add support for TI OMAP3 EVM board

Add OMAP3 EVM (OMAP3530, AM/DM37x) DTS file to use the omap3.dtsi SoC file,
along with memory node information.

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
13 years agoarm/dts: OMAP4: Update DTS file with new GIC bindings
Benoit Cousson [Fri, 25 Nov 2011 11:11:52 +0000 (12:11 +0100)]
arm/dts: OMAP4: Update DTS file with new GIC bindings

The GIC binding was updated in 3.2 and expects 3 interrupt-cells.
- Update the #interrupt-cells

Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Acked-by: Rob Herring <rob.herring@calxeda.com>
13 years agoarm/dts: OMAP3: Add i2c controller nodes
Benoit Cousson [Tue, 30 Aug 2011 14:50:24 +0000 (16:50 +0200)]
arm/dts: OMAP3: Add i2c controller nodes

Add i2c controllers nodes into the main ocp bus.

Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Acked-by: Rob Herring <rob.herring@calxeda.com>
13 years agoarm/dts: OMAP4: Add i2c controller nodes
Benoit Cousson [Wed, 17 Aug 2011 13:30:03 +0000 (19:00 +0530)]
arm/dts: OMAP4: Add i2c controller nodes

Add i2c controllers nodes into the main ocp bus.

Signed-off-by: Benoit Cousson <b-cousson@ti.com>
Acked-by: Rob Herring <rob.herring@calxeda.com>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
13 years agoARM: OMAP: convert omap_device_build() and callers to __init
Kevin Hilman [Mon, 20 Feb 2012 17:43:30 +0000 (09:43 -0800)]
ARM: OMAP: convert omap_device_build() and callers to __init

Building omap_devices should only be done at init time, and since
omap_device_build() is using early_platform calls which are also
__init, this ensures that omap_device isn't trying to use functions
that disappear.

Signed-off-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: Mark omap_hsmmc_init and omap_mux related functions as __init
Tony Lindgren [Mon, 20 Feb 2012 17:43:30 +0000 (09:43 -0800)]
ARM: OMAP2+: Mark omap_hsmmc_init and omap_mux related functions as __init

Now that omap hsmmc init is split into two functions, it's safe
to mark omap_hsmmc_init and omap_mux related functions to __init.

This basically reverts the following fixes for the case where
TWL was compiled as a module:

a98f77b (ARM: omap: fix section mismatch warning for sdp3430_twl_gpio_setup())
8930b4e (ARM: omap: fix section mismatch warnings in mux.c caused by hsmmc.c)

Additionally it fixes up the remaining section warnings for
all callers of omap_mux functions.

Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoMerge branch 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab...
Linus Torvalds [Fri, 24 Feb 2012 20:32:51 +0000 (12:32 -0800)]
Merge branch 'v4l_for_linus' of git://git./linux/kernel/git/mchehab/linux-media

* 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
  [media] hdpvr: update picture controls to support firmware versions > 0.15
  [media] wl128x: fix build errors when GPIOLIB is not enabled
  [media] hdpvr: fix race conditon during start of streaming
  [media] omap3isp: Fix crash caused by subdevs now having a pointer to devnodes
  [media] imon: don't wedge hardware after early callbacks

13 years agoepoll: ep_unregister_pollwait() can use the freed pwq->whead
Oleg Nesterov [Fri, 24 Feb 2012 19:07:29 +0000 (20:07 +0100)]
epoll: ep_unregister_pollwait() can use the freed pwq->whead

signalfd_cleanup() ensures that ->signalfd_wqh is not used, but
this is not enough. eppoll_entry->whead still points to the memory
we are going to free, ep_unregister_pollwait()->remove_wait_queue()
is obviously unsafe.

Change ep_poll_callback(POLLFREE) to set eppoll_entry->whead = NULL,
change ep_unregister_pollwait() to check pwq->whead != NULL under
rcu_read_lock() before remove_wait_queue(). We add the new helper,
ep_remove_wait_queue(), for this.

This works because sighand_cachep is SLAB_DESTROY_BY_RCU and because
->signalfd_wqh is initialized in sighand_ctor(), not in copy_sighand.
ep_unregister_pollwait()->remove_wait_queue() can play with already
freed and potentially reused ->sighand, but this is fine. This memory
must have the valid ->signalfd_wqh until rcu_read_unlock().

Reported-by: Maxime Bizon <mbizon@freebox.fr>
Cc: <stable@kernel.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoepoll: introduce POLLFREE to flush ->signalfd_wqh before kfree()
Oleg Nesterov [Fri, 24 Feb 2012 19:07:11 +0000 (20:07 +0100)]
epoll: introduce POLLFREE to flush ->signalfd_wqh before kfree()

This patch is intentionally incomplete to simplify the review.
It ignores ep_unregister_pollwait() which plays with the same wqh.
See the next change.

epoll assumes that the EPOLL_CTL_ADD'ed file controls everything
f_op->poll() needs. In particular it assumes that the wait queue
can't go away until eventpoll_release(). This is not true in case
of signalfd, the task which does EPOLL_CTL_ADD uses its ->sighand
which is not connected to the file.

This patch adds the special event, POLLFREE, currently only for
epoll. It expects that init_poll_funcptr()'ed hook should do the
necessary cleanup. Perhaps it should be defined as EPOLLFREE in
eventpoll.

__cleanup_sighand() is changed to do wake_up_poll(POLLFREE) if
->signalfd_wqh is not empty, we add the new signalfd_cleanup()
helper.

ep_poll_callback(POLLFREE) simply does list_del_init(task_list).
This make this poll entry inconsistent, but we don't care. If you
share epoll fd which contains our sigfd with another process you
should blame yourself. signalfd is "really special". I simply do
not know how we can define the "right" semantics if it used with
epoll.

The main problem is, epoll calls signalfd_poll() once to establish
the connection with the wait queue, after that signalfd_poll(NULL)
returns the different/inconsistent results depending on who does
EPOLL_CTL_MOD/signalfd_read/etc. IOW: apart from sigmask, signalfd
has nothing to do with the file, it works with the current thread.

In short: this patch is the hack which tries to fix the symptoms.
It also assumes that nobody can take tasklist_lock under epoll
locks, this seems to be true.

Note:

- we do not have wake_up_all_poll() but wake_up_poll()
  is fine, poll/epoll doesn't use WQ_FLAG_EXCLUSIVE.

- signalfd_cleanup() uses POLLHUP along with POLLFREE,
  we need a couple of simple changes in eventpoll.c to
  make sure it can't be "lost".

Reported-by: Maxime Bizon <mbizon@freebox.fr>
Cc: <stable@kernel.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoARM: OMAP2+: Limit omap_read/write usage to legacy USB drivers
Tony Lindgren [Fri, 24 Feb 2012 18:34:36 +0000 (10:34 -0800)]
ARM: OMAP2+: Limit omap_read/write usage to legacy USB drivers

Drivers should no longer use omap_read/write functions
but instead use ioremap + read/write functions.

As some USB legacy code is still shared between omap1 and
omap2420, let's limit the omap_read/write to plat/usb.h.

Note that the long term fix is to update the drivers to
use ioremap and read/write functions. That can now be
done as a separate patch series that is limited to the
USB drivers.

Also make sure the legacy omap1-keypad.c driver builds
if selected for 2420 based systems.

Cc: Tomi Valkeinen <tomi.valkeinen@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: Remove plat/io.h by splitting it into mach/io.h and mach/hardware.h
Tony Lindgren [Fri, 24 Feb 2012 18:34:35 +0000 (10:34 -0800)]
ARM: OMAP: Remove plat/io.h by splitting it into mach/io.h and mach/hardware.h

This is needed to minimize io.h so the SoC specific io.h
for ARMs can removed.

Note that minimal driver changes for DSS and RNG are needed to
include cpu.h for SoC detection macros.

Cc: Tomi Valkeinen <tomi.valkeinen@ti.com>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: Move most of plat/io.h into local iomap.h
Tony Lindgren [Fri, 24 Feb 2012 18:34:35 +0000 (10:34 -0800)]
ARM: OMAP2+: Move most of plat/io.h into local iomap.h

There's no need to have these defines in plat/io.h.

Note that we now need to ifdef omap_read/write calls
as they will be available for omap1 only.

While at it, clean up the includes to group them like
they typically are grouped.

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP1: Move most of plat/io.h into local iomap.h
Tony Lindgren [Fri, 24 Feb 2012 18:34:34 +0000 (10:34 -0800)]
ARM: OMAP1: Move most of plat/io.h into local iomap.h

There's no need to have these in plat/io.h.

While at it, clean up the includes to group them
like they typically are grouped.

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP1: Move 16xx GPIO system clock to platform init code
Tony Lindgren [Fri, 24 Feb 2012 18:34:33 +0000 (10:34 -0800)]
ARM: OMAP1: Move 16xx GPIO system clock to platform init code

This way we can remove omap_read/write call from the GPIO driver
and remove include to linux/io.h.

Cc: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: Move omap_init_consistent_dma_size() to local common.h
Tony Lindgren [Fri, 24 Feb 2012 18:34:33 +0000 (10:34 -0800)]
ARM: OMAP: Move omap_init_consistent_dma_size() to local common.h

We don't want to keep it in io.h as we want to remove io.h
for omap2+ for the common zImage support.

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: Move SDRC related functions from io.h into local common.h
Tony Lindgren [Fri, 24 Feb 2012 18:34:33 +0000 (10:34 -0800)]
ARM: OMAP2+: Move SDRC related functions from io.h into local common.h

These should be local to omap2/3/4.

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: Drop DISPC L3 firewall code
Tony Lindgren [Fri, 24 Feb 2012 18:34:32 +0000 (10:34 -0800)]
ARM: OMAP2+: Drop DISPC L3 firewall code

This is only needed when using SRAM for framebuffer,
and the support for SRAM framebuffer is about to get
removed.

Otherwise we cannot move most of plat/io.h to be a local
iomap.h for mach-omap2.

Cc: Tomi Valkeinen <tomi.valkeinen@ti.com>
Cc: linux-fbdev@vger.kernel.org
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2xxx: PM: remove obsolete timer disable code in the suspend path
Paul Walmsley [Fri, 24 Feb 2012 18:34:32 +0000 (10:34 -0800)]
ARM: OMAP2xxx: PM: remove obsolete timer disable code in the suspend path

Remove omap_{read,write}l() from the 24xx PM code.  The clocksource
code should now handle what this was supposed to do.

Tested on N800 -- but it's hard to say whether this fixes anything.
OMAP24xx static suspend path is currently broken, and this patch
doesn't change that.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Rob Herring <robherring2@gmail.com>
Acked-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: McSPI: Remove unused flag from struct omap2_mcspi_device_config
Jarkko Nikula [Fri, 24 Feb 2012 18:33:58 +0000 (10:33 -0800)]
ARM: OMAP: McSPI: Remove unused flag from struct omap2_mcspi_device_config

Flag single_channel in struct omap2_mcspi_device_config is not used
by drivers/spi/spi-omap2-mcspi.c so we may remove it from include/plat/mcspi.h
and affected board files.

Signed-off-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux...
Linus Torvalds [Fri, 24 Feb 2012 17:02:53 +0000 (09:02 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/mason/linux-btrfs

Quoth Chris:
 "This is later than I wanted because I got backed up running through
  btrfs bugs from the Oracle QA teams.  But they are all bug fixes that
  we've queued and tested since rc1.

  Nothing in particular stands out, this just reflects bug fixing and QA
  done in parallel by all the btrfs developers.  The most user visible
  of these is:

    Btrfs: clear the extent uptodate bits during parent transid failures

  Because that helps deal with out of date drives (say an iscsi disk
  that has gone away and come back).  The old code wasn't always
  properly retrying the other mirror for this type of failure."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (24 commits)
  Btrfs: fix compiler warnings on 32 bit systems
  Btrfs: increase the global block reserve estimates
  Btrfs: clear the extent uptodate bits during parent transid failures
  Btrfs: add extra sanity checks on the path names in btrfs_mksubvol
  Btrfs: make sure we update latest_bdev
  Btrfs: improve error handling for btrfs_insert_dir_item callers
  Btrfs: be less strict on finding next node in clear_extent_bit
  Btrfs: fix a bug on overcommit stuff
  Btrfs: kick out redundant stuff in convert_extent_bit
  Btrfs: skip states when they does not contain bits to clear
  Btrfs: check return value of lookup_extent_mapping() correctly
  Btrfs: fix deadlock on page lock when doing auto-defragment
  Btrfs: fix return value check of extent_io_ops
  btrfs: honor umask when creating subvol root
  btrfs: silence warning in raid array setup
  btrfs: fix structs where bitfields and spinlock/atomic share 8B word
  btrfs: delalloc for page dirtied out-of-band in fixup worker
  Btrfs: fix memory leak in load_free_space_cache()
  btrfs: don't check DUP chunks twice
  Btrfs: fix trim 0 bytes after a device delete
  ...

13 years agoMerge tag 'for-linus' of git://linux-c6x.org/git/projects/linux-c6x-upstreaming
Linus Torvalds [Fri, 24 Feb 2012 17:01:46 +0000 (09:01 -0800)]
Merge tag 'for-linus' of git://linux-c6x.org/git/projects/linux-c6x-upstreaming

This is the arch/c6x part of commit 7c43185138cf ("Kbuild: Use dtc's -d
(dependency) option") which was dropped because c6x had not yet been
merged at the time.

* tag 'for-linus' of git://linux-c6x.org/git/projects/linux-c6x-upstreaming:
  Kbuild: Use dtc's -d (dependency) option

13 years agoMAINTAINERS: drop me from PA-RISC maintenance
Kyle McMartin [Fri, 24 Feb 2012 15:36:16 +0000 (10:36 -0500)]
MAINTAINERS: drop me from PA-RISC maintenance

I don't even live in the same country as any of my PA-RISC hardware
these days, so the odds of me touching the code are pretty low.
(Also re-order things to ensure jejb gets CC'd since he's been the
primary maintainer for the last few years.)

Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoNOMMU: Don't need to clear vm_mm when deleting a VMA
David Howells [Thu, 23 Feb 2012 13:51:00 +0000 (13:51 +0000)]
NOMMU: Don't need to clear vm_mm when deleting a VMA

Don't clear vm_mm in a deleted VMA as it's unnecessary and might
conceivably break the filesystem or driver VMA close routine.

Reported-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
cc: stable@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoNOMMU: Lock i_mmap_mutex for access to the VMA prio list
David Howells [Thu, 23 Feb 2012 13:50:35 +0000 (13:50 +0000)]
NOMMU: Lock i_mmap_mutex for access to the VMA prio list

Lock i_mmap_mutex for access to the VMA prio list to prevent concurrent
access.  Currently, certain parts of the mmap handling are protected by
the region mutex, but not all.

Reported-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
cc: stable@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoMerge tag 'rmobile-for-linus' of git://github.com/pmundt/linux-sh
Linus Torvalds [Fri, 24 Feb 2012 16:57:22 +0000 (08:57 -0800)]
Merge tag 'rmobile-for-linus' of git://github.com/pmundt/linux-sh

SH/R-Mobile fixes for 3.3-rc5

* tag 'rmobile-for-linus' of git://github.com/pmundt/linux-sh:
  arch/arm/mach-shmobile/board-ag5evm.c: included linux/dma-mapping.h twice
  ARM: mach-shmobile: r8a7779 PFC IPSR4 fix
  ARM: mach-shmobile: sh73a0 PSTR 32-bit access fix
  ARM: mach-shmobile: add GPIO-to-IRQ translation to sh7372
  ARM: mach-shmobile: clock-sh73a0: add DSIxPHY clock support
  arm: fix compile failure in mach-shmobile/board-ag5evm.c
  ARM: mach-shmobile: mackerel: add ak4642 amixer settings on comment
  ARM: mach-shmobile: mackerel: use renesas_usbhs instead of r8a66597_hcd
  ARM: mach-shmobile: simplify MMCIF DMA configuration
  ARM: mach-shmobile: IRQ driven GPIO key support for Kota2
  ARM: mach-shmobile: sh73a0 IRQ sparse alloc fix
  ARM: mach-shmobile: sh73a0 PINT IRQ base fix

13 years agoMerge tag 'sh-for-linus' of git://github.com/pmundt/linux-sh
Linus Torvalds [Fri, 24 Feb 2012 16:56:51 +0000 (08:56 -0800)]
Merge tag 'sh-for-linus' of git://github.com/pmundt/linux-sh

SuperH fixes for 3.3-rc5

* tag 'sh-for-linus' of git://github.com/pmundt/linux-sh:
  sh: Fix sh2a build error for CONFIG_CACHE_WRITETHROUGH
  sh: modify a resource of sh_eth_giga1_resources in board-sh7757lcr
  arch/sh: remove references to cpu_*_map.
  sh: Fix typo in pci-sh7780.c
  sh: add platform_device for SPI1 in setup-sh7757
  sh: modify resource for SPI0 in setup-sh7757
  sh: se7724: fix compile breakage
  sh: clkfwk: bugfix: use clk_reparent() for div6 clocks
  sh: clock-sh7724: fixup sh_fsi clock settings
  sh: sh7757lcr: update to the new MMCIF DMA configuration
  sh: fix the sh_mmcif_plat_data in board-sh7757lcr
  video: pvr2fb: Fix up spurious section mismatch warnings.
  sh: Defer to asm-generic/device.h.

13 years agomm: memcg: Correct unregistring of events attached to the same eventfd
Anton Vorontsov [Fri, 24 Feb 2012 01:14:46 +0000 (05:14 +0400)]
mm: memcg: Correct unregistring of events attached to the same eventfd

There is an issue when memcg unregisters events that were attached to
the same eventfd:

- On the first call mem_cgroup_usage_unregister_event() removes all
  events attached to a given eventfd, and if there were no events left,
  thresholds->primary would become NULL;

- Since there were several events registered, cgroups core will call
  mem_cgroup_usage_unregister_event() again, but now kernel will oops,
  as the function doesn't expect that threshold->primary may be NULL.

That's a good question whether mem_cgroup_usage_unregister_event()
should actually remove all events in one go, but nowadays it can't
do any better as cftype->unregister_event callback doesn't pass
any private event-associated cookie. So, let's fix the issue by
simply checking for threshold->primary.

FWIW, w/o the patch the following oops may be observed:

 BUG: unable to handle kernel NULL pointer dereference at 0000000000000004
 IP: [<ffffffff810be32c>] mem_cgroup_usage_unregister_event+0x9c/0x1f0
 Pid: 574, comm: kworker/0:2 Not tainted 3.3.0-rc4+ #9 Bochs Bochs
 RIP: 0010:[<ffffffff810be32c>]  [<ffffffff810be32c>] mem_cgroup_usage_unregister_event+0x9c/0x1f0
 RSP: 0018:ffff88001d0b9d60  EFLAGS: 00010246
 Process kworker/0:2 (pid: 574, threadinfo ffff88001d0b8000, task ffff88001de91cc0)
 Call Trace:
  [<ffffffff8107092b>] cgroup_event_remove+0x2b/0x60
  [<ffffffff8103db94>] process_one_work+0x174/0x450
  [<ffffffff8103e413>] worker_thread+0x123/0x2d0

Cc: stable <stable@vger.kernel.org>
Signed-off-by: Anton Vorontsov <anton.vorontsov@linaro.org>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Kirill A. Shutemov <kirill@shutemov.name>
Cc: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
13 years agoARM: OMAP2+: Fix multiple randconfig errors with SOC_OMAP and SOC_OMAP_NOOP
Tony Lindgren [Fri, 24 Feb 2012 00:41:53 +0000 (16:41 -0800)]
ARM: OMAP2+: Fix multiple randconfig errors with SOC_OMAP and SOC_OMAP_NOOP

If we don't have ARCH_OMAP2, 3 or 4 selected randconfig will always
fail with multiple errors as the CPU and MACHINE are not set.

Fix this by changing arch/arm/Makefile to build mach-omap2 based on
ARCH_OMAP2PLUS. And let's introduce SOC_OMAP and SOC_OMAP_NOOP that
allow randconfig to generate buildable .config files.

Note that we can also remove few uncecssary ARCH_OMAP2PLUS lines
as they are all within if ARCH_OMAP2PLUS block.

Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP: Fix devexit for dma when CONFIG_HOTPLUG is not set
Tony Lindgren [Thu, 23 Feb 2012 22:58:08 +0000 (14:58 -0800)]
ARM: OMAP: Fix devexit for dma when CONFIG_HOTPLUG is not set

Otherwise we get:

`omap_system_dma_remove' referenced in section `.data' of arch/arm/plat-omap/built-in.o:
defined in discarded section `.devexit.text' of arch/arm/plat-omap/built-in.o

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: Fix devexit for smartreflex when CONFIG_HOTPLUG is not set
Tony Lindgren [Thu, 23 Feb 2012 22:55:40 +0000 (14:55 -0800)]
ARM: OMAP2+: Fix devexit for smartreflex when CONFIG_HOTPLUG is not set

Otherwise we get:

`omap_sr_remove' referenced in section `.data' of arch/arm/mach-omap2/built-in.o:
defined in discarded section `.devexit.text' of arch/arm/mach-omap2/built-in.o

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agohwmon: (max34440) Fix resetting temperature history
Guenter Roeck [Fri, 24 Feb 2012 11:44:34 +0000 (03:44 -0800)]
hwmon: (max34440) Fix resetting temperature history

Temperature history is reset by writing 0x8000 into the peak temperature
register, not 0xffff.

Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Acked-by: Jean Delvare <khali@linux-fr.org>
13 years agoBtrfs: fix compiler warnings on 32 bit systems
Chris Mason [Fri, 24 Feb 2012 15:39:05 +0000 (10:39 -0500)]
Btrfs: fix compiler warnings on 32 bit systems

The enospc tracing code added some interesting uses of
u64 pointer casts.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
13 years agoarch/arm/mach-shmobile/board-ag5evm.c: included linux/dma-mapping.h twice
Danny Kukawka [Thu, 16 Feb 2012 14:46:38 +0000 (15:46 +0100)]
arch/arm/mach-shmobile/board-ag5evm.c: included linux/dma-mapping.h twice

arch/arm/mach-shmobile/board-ag5evm.c: included 'linux/dma-mapping.h'
twice, remove the duplicate.

Signed-off-by: Danny Kukawka <danny.kukawka@bisect.de>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
13 years agoARM: mach-shmobile: r8a7779 PFC IPSR4 fix
Magnus Damm [Mon, 30 Jan 2012 02:25:07 +0000 (11:25 +0900)]
ARM: mach-shmobile: r8a7779 PFC IPSR4 fix

Fix the bit field width information for the IPSR4 register
in the r8a7779 pin function controller (PFC).

Without this fix the Marzen board fails to receive data
over the serial console due to misconfigured pin function
for the RX pin.

Signed-off-by: Magnus Damm <damm@opensource.se>
Tested-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
13 years agoARM: mach-shmobile: sh73a0 PSTR 32-bit access fix
Magnus Damm [Mon, 30 Jan 2012 02:03:49 +0000 (11:03 +0900)]
ARM: mach-shmobile: sh73a0 PSTR 32-bit access fix

Convert the sh73a0 SMP code to use 32-bit PSTR access.

This fixes wakeup from deep sleep for sh73a0 secondary CPUs.

Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
13 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux into rmobile-fixes...
Paul Mundt [Fri, 24 Feb 2012 04:23:23 +0000 (13:23 +0900)]
Merge git://git./linux/kernel/git/torvalds/linux into rmobile-fixes-for-linus

13 years agosh: Fix sh2a build error for CONFIG_CACHE_WRITETHROUGH
Phil Edworthy [Tue, 21 Feb 2012 08:29:57 +0000 (08:29 +0000)]
sh: Fix sh2a build error for CONFIG_CACHE_WRITETHROUGH

Signed-off-by: Phil Edworthy <phil.edworthy@renesas.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
13 years agosh: modify a resource of sh_eth_giga1_resources in board-sh7757lcr
Shimoda, Yoshihiro [Mon, 20 Feb 2012 08:26:50 +0000 (17:26 +0900)]
sh: modify a resource of sh_eth_giga1_resources in board-sh7757lcr

The latest sh_eth driver needs a resource of TSU in the channel 1,
if the controller has TSU registers. So, this patch adds the resource.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
13 years agoarch/sh: remove references to cpu_*_map.
Rusty Russell [Wed, 15 Feb 2012 04:58:04 +0000 (15:28 +1030)]
arch/sh: remove references to cpu_*_map.

This has been obsolescent for a while; time for the final push.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: linux-sh@vger.kernel.org
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
13 years agosh: Fix typo in pci-sh7780.c
Masanari Iida [Sat, 4 Feb 2012 12:40:24 +0000 (21:40 +0900)]
sh: Fix typo in pci-sh7780.c

Correct spelling "erorr" to "error" in
arch/sh/drivers/pci/pci-sh7780.c

Signed-off-by: Masanari Iida <standby24x7@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
13 years agoARM: OMAP2+: Fix zoom LCD backlight if TWL_CORE is not selected
Tony Lindgren [Thu, 23 Feb 2012 22:29:59 +0000 (14:29 -0800)]
ARM: OMAP2+: Fix zoom LCD backlight if TWL_CORE is not selected

Otherwise we get:

arch/arm/mach-omap2/board-zoom-display.c:64: undefined reference to `twl_i2c_read_u8'
arch/arm/mach-omap2/board-zoom-display.c:65: undefined reference to `twl_i2c_read_u8'
arch/arm/mach-omap2/board-zoom-display.c:84: undefined reference to `twl_i2c_write_u8'
arch/arm/mach-omap2/board-zoom-display.c:86: undefined reference to `twl_i2c_write_u8'
arch/arm/mach-omap2/board-zoom-display.c:91: undefined reference to `twl_i2c_write_u8'
arch/arm/mach-omap2/board-zoom-display.c:92: undefined reference to `twl_i2c_write_u8'
arch/arm/mach-omap2/board-zoom-display.c:72: undefined reference to `twl_i2c_write_u8'

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: Fix board_mux section type conflict when OMAP_MUX is not set
Tony Lindgren [Thu, 23 Feb 2012 22:21:28 +0000 (14:21 -0800)]
ARM: OMAP2+: Fix board_mux section type conflict when OMAP_MUX is not set

Otherwise we can get:

arch/arm/mach-omap2/mux.h:249:31: error: board_mux causes a section type conflict

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: Fix OMAP_HDQ_BASE build error
Tony Lindgren [Fri, 24 Feb 2012 00:41:53 +0000 (16:41 -0800)]
ARM: OMAP2+: Fix OMAP_HDQ_BASE build error

If CONFIG_SOC_OMAP3430 is not set and CONFIG_HDQ_MASTER_OMAP
is selected for w1 driver we get the following error:

arch/arm/mach-omap2/devices.c:662:13: error:
'OMAP_HDQ_BASE' undeclared here (not in a function)

Looks like OMAP_HDQ_BASE is valid for all omaps except
2420, so we can remove the ifdef and not register
the device on 2420.

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: Fix Kconfig dependencies for USB_ARCH_HAS_EHCI
Tony Lindgren [Fri, 24 Feb 2012 00:41:52 +0000 (16:41 -0800)]
ARM: OMAP2+: Fix Kconfig dependencies for USB_ARCH_HAS_EHCI

Otherwise we get:

warning: (ARCH_OMAP3 && ARCH_OMAP4) selects USB_ARCH_HAS_EHCI
which has unmet direct dependencies (USB_SUPPORT)

Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoARM: OMAP2+: I2C: always compile I2C reset code, even if I2C driver is not built
Paul Walmsley [Fri, 24 Feb 2012 00:41:52 +0000 (16:41 -0800)]
ARM: OMAP2+: I2C: always compile I2C reset code, even if I2C driver is not built

During kernel init, we reset all IP blocks on the OMAP that we can,
even if there is no driver compiled for that IP block.  Unlike most IP
blocks, the I2C block requires some extra programming for this to
work.  This reset code is incorrectly omitted when the I2C driver is
deselected.  In this circumstance, the build breaks.  Fix by compiling
the I2C reset code unconditionally.

Problem reported by Russell King <linux@arm.linux.org.uk>.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Russell King <linux@arm.linux.org.uk>
Tested-by: Shubhrajyoti <shubhrajyoti@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
13 years agoRestore direct_io / truncate locking API
Anton Altaparmakov [Thu, 23 Feb 2012 23:40:05 +0000 (23:40 +0000)]
Restore direct_io / truncate locking API

With kernel 3.1, Christoph removed i_alloc_sem and replaced it with
calls (namely inode_dio_wait() and inode_dio_done()) which are
EXPORT_SYMBOL_GPL() thus they cannot be used by non-GPL file systems and
further inode_dio_wait() was pushed from notify_change() into the file
system ->setattr() method but no non-GPL file system can make this call.

That means non-GPL file systems cannot exist any more unless they do not
use any VFS functionality related to reading/writing as far as I can
tell or at least as long as they want to implement direct i/o.

Both Linus and Al (and others) have said on LKML that this breakage of
the VFS API should not have happened and that the change was simply
missed as it was not documented in the change logs of the patches that
did those changes.

This patch changes the two function exports in question to be
EXPORT_SYMBOL() thus restoring the VFS API as it used to be - accessible
for all modules.

Christoph, who introduced the two functions and exported them GPL-only
is CC-ed on this patch to give him the opportunity to object to the
symbols being changed in this manner if he did indeed intend them to be
GPL-only and does not want them to become available to all modules.

Signed-off-by: Anton Altaparmakov <anton@tuxera.com>
CC: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>