Arnd Bergmann [Thu, 21 Mar 2013 16:33:01 +0000 (17:33 +0100)]
Merge tag 'ux500-core-v3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-stericsson into next/soc
From Linus Walleij <linus.walleij@linaro.org>:
Two fixes to the ux500 core machine:
- Enable the 128KB cache on DB8540
- Bump the number of IRQs to the correct value
* tag 'ux500-core-v3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-stericsson:
ARM: mach-ux500: enable 128KB way L2 cache on DB8540
ARM: ux500: bump arch nr of GPIOs
Arnd Bergmann [Thu, 21 Mar 2013 16:29:49 +0000 (17:29 +0100)]
Merge tag 'ux500-defconfig-v3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-stericsson into next/soc
From Linus Walleij <linus.walleij@linaro.org>:
Defconfig updates for ux500:
- Turn on the "heartbeat" LED trigger used by Snowball
- Enable the EXT4_FS and LBDAF by default
* tag 'ux500-defconfig-v3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-stericsson:
ARM: ux500: enable ux500 EXT4_FS and LBDAF support by default
ARM: ux500: Turn on the 'heartbeat' LED trigger
Arnd Bergmann [Thu, 21 Mar 2013 16:25:13 +0000 (17:25 +0100)]
Merge tag 'at91-dt' of git://github.com/at91linux/linux-at91 into next/dt
From Nicolas Ferre <nicolas.ferre@atmel.com>:
DT modifications for at91rm9200 and at91sam9x5 SoCs, mainly around
I2C. Also some cleanup of some unneeded properties and conflicting
nodes.
One more DT-only board based on at91rm9200.
* tag 'at91-dt' of git://github.com/at91linux/linux-at91:
ARM: at91/at91sam9x5cm: add 1-wire chip on CM board
ARM: at91/at91sam9x5ek: i2c1 and i2c2 conflict with macb and lcd
ARM: at91/dt: gpio-keys: remove address-cells and size-cells properties
ARM: at91: add MPA 1600 DT board
ARM: at91: add pinctrl nodes to i2c-gpio on RM92000 DT
ARM: at91: add TWI bindings to RM9200 DT
ARM: at91: dt: at91sam9x5: add i2c-gpio pinctrl
ARM: at91: dt: at91sam9x5: add i2c pinctrl
Richard Genoud [Wed, 20 Mar 2013 11:23:40 +0000 (12:23 +0100)]
ARM: at91/at91sam9x5cm: add 1-wire chip on CM board
This add the 1-wire chip present on the CM board to the DTS.
As the pin is also used by leds, it's disabled by default.
If the board really wants it, it can be enabled in the board DTS.
Signed-off-by: Richard Genoud <richard.genoud@gmail.com> Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Nicolas Ferre [Fri, 15 Mar 2013 10:30:54 +0000 (11:30 +0100)]
ARM: at91/dt: gpio-keys: remove address-cells and size-cells properties
Gpio-keys nodes are not using the "reg" property in their sub nodes.
So, there is no need to define #address-cells and #size-cells properties
in gpio-keys nodes: we remove them.
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Joachim Eastwood <manabian@gmail.com>
[nicolas.ferre@atmel.com: remove unneeded address-cells and size-cells to gpio-keys] Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Shawn Guo [Wed, 13 Mar 2013 02:52:49 +0000 (10:52 +0800)]
ARM: imx: add dependency check for DEBUG_IMX_UART_PORT
While adding i.MX DEBUG_LL selection, commit f8c95fe (ARM: imx: support
DEBUG_LL uart port selection for all i.MX SoCs) leaves Kconfig symbol
DEBUG_IMX_UART_PORT there without any dependency check. This results in
that everyone gets the symbol in their config, which is someting
undesirable. Add "depends on ARCH_MXC" for the symbol to prevent that.
Reported-by: Karl Beldan <karl.beldan@gmail.com> Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
Marek Vasut [Mon, 18 Mar 2013 18:24:02 +0000 (19:24 +0100)]
ARM: video: mxs: Fix mxsfb misconfiguring VDCTRL0
The issue fixed by this patch manifests only then using X11
with mxsfb driver. The X11 will display either shifted image
or otherwise distorted image on the LCD.
The problem is that the X11 tries to reconfigure the framebuffer
and along the way calls fb_ops.fb_set_par() with X11's desired
configuration values. The field of particular interest is
fb_info->var.sync which contains non-standard values if
configured by kernel. These are either FB_SYNC_DATA_ENABLE_HIGH_ACT,
FB_SYNC_DOTCLK_FAILING_ACT or both, depending on the platform
configuration. Both of these values are defined in the
include/linux/mxsfb.h file.
The driver interprets these values and configures the LCD controller
accordingly. Yet X11 only has access to the standard values for this
field defined in include/uapi/linux/fb.h and thus, unlike kernel,
omits these special values. This results in distorted image on the
LCD.
This patch moves these non-standard values into new field of the
mxsfb_platform_data structure so the driver can in turn check this
field instead of the video mode field for these specific portions.
Moreover, this patch prefixes these values with MXSFB_SYNC_ prefix
instead of FB_SYNC_ prefix to prevent confusion of subsequent users.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Fabio Estevam <fabio.estevam@freescale.com> Cc: Linux ARM <linux-arm-kernel@lists.infradead.org> Cc: Linux FBDEV <linux-fbdev@vger.kernel.org> Cc: Lothar Waßmann <LW@karo-electronics.de> Cc: Sascha Hauer <kernel@pengutronix.de> Tested-by: Fabio Estevam <fabio.estevam@freescale.com> Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
Arnd Bergmann [Mon, 18 Mar 2013 16:16:37 +0000 (17:16 +0100)]
Merge branch 'prima2/soc' into next/soc
From Barry Song <Barry.Song@csr.com>:
SiRFatlas6 is much similar with SiRFprimaII with some changes about:
1. deleted l2 cache, sdmmc4, multimedia and so on
2. changed pinmux layout and clock layout
3. DMA access limitation disappears, that makes us not need DMA_ZONE
on atlas6
* prima2/soc:
ARM/dts: prima2: add .dtsi for atlas6 and .dts for atla6-evb board
arm: prima2: add new SiRFatlas6 machine in common board
Barry Song [Mon, 18 Mar 2013 07:04:39 +0000 (15:04 +0800)]
ARM/dts: prima2: add .dtsi for atlas6 and .dts for atla6-evb board
atlas6.dtsi is basically a copy of prima2.dtsi as most components are
compatible with prima2 except that:
1. node of l2 cache is deleted
2. node multimedia engine is deleted
3. node of sata is deleted
4. node of sdmmc4 is deleted
5. powervr is moved to "powervr,sgx510"
6. pinctrl is moved to atlas6 as pinmux layout has big changes in
atlas6
7. clock is moved to atlas6 as clock layout has changes in atlas6
Signed-off-by: Barry Song <Baohua.Song@csr.com> Signed-off-by: Jiansong Chen <Jiansong.Chen@csr.com>
Barry Song [Mon, 18 Mar 2013 07:04:38 +0000 (15:04 +0800)]
arm: prima2: add new SiRFatlas6 machine in common board
SiRFatlas6's machine definition is almost seem with SiRFprimaII
except that prima2 has a 256MB DMA zone.
This patch adds SiRFatlas6 machine in common board files, and
also adds atlas6 arch node in Kconfig.
Signed-off-by: Barry Song <Baohua.Song@csr.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Arnd Bergmann [Mon, 18 Mar 2013 14:46:44 +0000 (15:46 +0100)]
Merge tag 'cleanup_for_v3.10' of git://git.infradead.org/users/jcooper/linux into next/cleanup
From Jason Cooper <jason@lakedaemon.net>:
mvebu cleanup for v3.10
- plat-orion: prep for mvebu-mbus driver
* tag 'cleanup_for_v3.10' of git://git.infradead.org/users/jcooper/linux:
arm: mach-orion5x: use mv_mbus_dram_info() in PCI code
arm: plat-orion: use mv_mbus_dram_info() in PCIe code
arm: plat-orion: only build addr-map.c when needed
Lee Jones [Mon, 21 Jan 2013 13:35:40 +0000 (13:35 +0000)]
ARM: ux500: Turn on the 'heartbeat' LED trigger
The heartbeat LED trigger provides an excellent debugging tool
when hacking on development boards. Here we enable it on all
u8500 based platforms. This will pulse the User LED on the
Snowball low-cost development board only.
Signed-off-by: Lee Jones <lee.jones@linaro.org> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Linus Walleij [Tue, 18 Dec 2012 12:53:09 +0000 (13:53 +0100)]
ARM: ux500: bump arch nr of GPIOs
Set the number of GPIOs for Ux500 to 392. Reasoning:
- Internal pinctrl "Nomadik" SoC: 288 GPIOs
- Then each Ux500 system has one or two GPIO expanders at
maximum 24 GPIOs each:
TC35892 expander: 24 GPIOs
STMPE1601 or 1801 Expander: 24 GPIOs
- Then AB8500/AB8505/AB8540: 56 GPIOs
Sum: maximum 392 GPIOs - no more no less.
Acked-by: Lee Jones <lee.jones@linaro.org> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
David Rientjes [Sun, 17 Mar 2013 22:49:10 +0000 (15:49 -0700)]
perf,x86: fix link failure for non-Intel configs
Commit 1d9d8639c063 ("perf,x86: fix kernel crash with PEBS/BTS after
suspend/resume") introduces a link failure since
perf_restore_debug_store() is only defined for CONFIG_CPU_SUP_INTEL:
arch/x86/power/built-in.o: In function `restore_processor_state':
(.text+0x45c): undefined reference to `perf_restore_debug_store'
Fix it by defining the dummy function appropriately.
Signed-off-by: David Rientjes <rientjes@google.com> Cc: stable@vger.kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Sun, 17 Mar 2013 22:44:43 +0000 (15:44 -0700)]
perf,x86: fix wrmsr_on_cpu() warning on suspend/resume
Commit 1d9d8639c063 ("perf,x86: fix kernel crash with PEBS/BTS after
suspend/resume") fixed a crash when doing PEBS performance profiling
after resuming, but in using init_debug_store_on_cpu() to restore the
DS_AREA mtrr it also resulted in a new WARN_ON() triggering.
init_debug_store_on_cpu() uses "wrmsr_on_cpu()", which in turn uses CPU
cross-calls to do the MSR update. Which is not really valid at the
early resume stage, and the warning is quite reasonable. Now, it all
happens to _work_, for the simple reason that smp_call_function_single()
ends up just doing the call directly on the CPU when the CPU number
matches, but we really should just do the wrmsr() directly instead.
This duplicates the wrmsr() logic, but hopefully we can just remove the
wrmsr_on_cpu() version eventually.
Linus Torvalds [Sun, 17 Mar 2013 18:04:14 +0000 (11:04 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs
Pull btrfs fixes from Chris Mason:
"Eric's rcu barrier patch fixes a long standing problem with our
unmount code hanging on to devices in workqueue helpers. Liu Bo
nailed down a difficult assertion for in-memory extent mappings."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: fix warning of free_extent_map
Btrfs: fix warning when creating snapshots
Btrfs: return as soon as possible when edquot happens
Btrfs: return EIO if we have extent tree corruption
btrfs: use rcu_barrier() to wait for bdev puts at unmount
Btrfs: remove btrfs_try_spin_lock
Btrfs: get better concurrency for snapshot-aware defrag work
arm: mach-orion5x: use mv_mbus_dram_info() in PCI code
The PCI code was directly accessing the orion_mbus_dram_info structure
to get access to a description of the SDRAM chip selects in order to
configure the PCIe -> SDRAM address decoding windows.
However, with the introduction of the mvebu-mbus driver, we are going
to remove this global structure and instead leave only the exported
mv_mbus_dram_info() function to access this description of the SDRAM
chip selects. Therefore, we simply switch to using this API.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: Jason Cooper <jason@lakedaemon.net>
arm: plat-orion: use mv_mbus_dram_info() in PCIe code
The PCIe code was directly accessing the orion_mbus_dram_info
structure to get access to a description of the SDRAM chip selects in
order to configure the PCIe -> SDRAM address decoding
windows.
However, with the introduction of the orion-mbus driver, we are going
to remove this global structure and instead leave only the exported
mv_mbus_dram_info() function to access this description of the SDRAM
chip selects. Therefore, we simply switch to using this API.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: Jason Cooper <jason@lakedaemon.net>
arm: plat-orion: only build addr-map.c when needed
-flagmail-match: MVEBU
X-flagmail-match: KIRKWOOD
X-flagmail-match: DOVE
For now, addr-map.c is needed by all 5 Marvell EBU
sub-architectures. However, we are going to introduce the orion-mbus
driver, which will replace the address decoding code from
addr-map.c. In order to ease the migration process, we will do that
one sub-architecture at a time, which will require us to remove the
compilation of addr-map.c one sub-architecture at a time.
Therefore, we split the unconditional obj-y inclusion of addr-map.c
into 5 conditionals obj-$(CONFIG_...) lines, one per sub-architecture.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: Jason Cooper <jason@lakedaemon.net>
Liu Bo [Fri, 15 Mar 2013 14:46:39 +0000 (08:46 -0600)]
Btrfs: fix warning of free_extent_map
Users report that an extent map's list is still linked when it's actually
going to be freed from cache.
The story is that
a) when we're going to drop an extent map and may split this large one into
smaller ems, and if this large one is flagged as EXTENT_FLAG_LOGGING which means
that it's on the list to be logged, then the smaller ems split from it will also
be flagged as EXTENT_FLAG_LOGGING, and this is _not_ expected.
b) we'll keep ems from unlinking the list and freeing when they are flagged with
EXTENT_FLAG_LOGGING, because the log code holds one reference.
The end result is the warning, but the truth is that we set the flag
EXTENT_FLAG_LOGGING only during fsync.
So clear flag EXTENT_FLAG_LOGGING for extent maps split from a large one.
Reported-by: Johannes Hirte <johannes.hirte@fem.tu-ilmenau.de> Reported-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Linus Torvalds [Sat, 16 Mar 2013 01:06:55 +0000 (18:06 -0700)]
Merge branch 'kbuild' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild
Pull kbuild fix from Michal Marek:
"One fix for for make headers_install/headers_check to not require make
3.81. The requirement has been accidentally introduced in 3.7."
* 'kbuild' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild:
kbuild: fix make headers_check with make 3.80
Linus Torvalds [Sat, 16 Mar 2013 01:05:37 +0000 (18:05 -0700)]
Merge tag 'for-3.9-rc3' of git://openrisc.net/jonas/linux
Pull OpenRISC bug fixes from Jonas Bonn:
- The GPIO descriptor work has exposed how broken the non-GPIOLIB bits
for OpenRISC were. We now require GPIOLIB as this is the preferred
way forward.
- The system.h split introduced a bug in llist.h for arches using
asm-generic/cmpxchg.h directly, which is currently only OpenRISC.
The patch here moves two defines from asm-generic/atomic.h to
asm-generic/cmpxchg.h to make things work as they should.
- The VIRT_TO_BUS selector was added for OpenRISC, but OpenRISC does
not have the virt_to_bus methods, so there's a patch to remove it
again.
* tag 'for-3.9-rc3' of git://openrisc.net/jonas/linux:
openrisc: remove HAVE_VIRT_TO_BUS
asm-generic: move cmpxchg*_local defs to cmpxchg.h
openrisc: require gpiolib
Linus Torvalds [Sat, 16 Mar 2013 01:04:38 +0000 (18:04 -0700)]
Merge tag 'char-misc-3.9-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
Pull char/misc fixes from Greg Kroah-Hartman:
"Here are some tiny fixes for the w1 drivers and the final removal
patch for getting rid of CONFIG_EXPERIMENTAL (all users of it are now
gone from your tree, this just drops the Kconfig item itself.)
All have been in the linux-next tree for a while"
* tag 'char-misc-3.9-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
final removal of CONFIG_EXPERIMENTAL
w1: fix oops when w1_search is called from netlink connector
w1-gpio: fix unused variable warning
w1-gpio: remove erroneous __exit and __exit_p()
ARM: w1-gpio: fix erroneous gpio requests
Linus Torvalds [Sat, 16 Mar 2013 00:35:49 +0000 (17:35 -0700)]
Merge tag 'sound-3.9' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"A collection of small fixes, as expected for the middle rc:
- A couple of fixes for potential NULL dereferences and out-of-range
array accesses revealed by static code parsers
- A fix for the wrong error handling detected by trinity
- A regression fix for missing audio on some MacBooks
- CA0132 DSP loader fixes
- Fix for EAPD control of IDT codecs on machines w/o speaker
- Fix a regression in the HD-audio widget list parser code
- Workaround for the NuForce UDH-100 USB audio"
* tag 'sound-3.9' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: hda - Fix missing EAPD/GPIO setup for Cirrus codecs
sound: sequencer: cap array index in seq_chn_common_event()
ALSA: hda/ca0132 - Remove extra setting of dsp_state.
ALSA: hda/ca0132 - Check download state of DSP.
ALSA: hda/ca0132 - Check if dspload_image succeeded.
ALSA: hda - Disable IDT eapd_switch if there are no internal speakers
ALSA: hda - Fix snd_hda_get_num_raw_conns() to return a correct value
ALSA: usb-audio: add a workaround for the NuForce UDH-100
ALSA: asihpi - fix potential NULL pointer dereference
ALSA: seq: Fix missing error handling in snd_seq_timer_open()
Linus Torvalds [Sat, 16 Mar 2013 00:35:03 +0000 (17:35 -0700)]
Merge branch 'fixes-for-3.9' of git://git.linaro.org/people/mszyprowski/linux-dma-mapping
Pull DMA-mapping fix from Marek Szyprowski:
"An important fix for all ARM architectures which use ZONE_DMA.
Without it dma_alloc_* calls with GFP_ATOMIC flag might have allocated
buffers outsize DMA zone."
* 'fixes-for-3.9' of git://git.linaro.org/people/mszyprowski/linux-dma-mapping:
ARM: DMA-mapping: add missing GFP_DMA flag for atomic buffer allocation
Linus Torvalds [Sat, 16 Mar 2013 00:34:01 +0000 (17:34 -0700)]
Merge tag 'mfd-fixes-3.9-1' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-fixes
Pull MFD fixes from Samuel Ortiz:
"This is the first batch of MFD fixes for 3.9.
With this one we have:
- An ab8500 build failure fix.
- An ab8500 device tree parsing fix.
- A fix for twl4030_madc remove routine to work properly (when
built-in).
- A fix for properly registering palmas interrupt handler.
- A fix for omap-usb init routine to actually write into the
hostconfig register.
- A couple of warning fixes for ab8500-gpadc and tps65912"
* tag 'mfd-fixes-3.9-1' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-fixes:
mfd: twl4030-madc: Remove __exit_p annotation
mfd: ab8500: Kill "reg" property from binding
mfd: ab8500-gpadc: Complain if we fail to enable vtvout LDO
mfd: wm831x: Don't forward declare enum wm831x_auxadc
mfd: twl4030-audio: Fix argument type for twl4030_audio_disable_resource()
mfd: tps65912: Declare and use tps65912_irq_exit()
mfd: palmas: Provide irq flags through DT/platform data
mfd: Make AB8500_CORE select POWER_SUPPLY to fix build error
mfd: omap-usb-host: Actually update hostconfig
Linus Torvalds [Sat, 16 Mar 2013 00:33:13 +0000 (17:33 -0700)]
Merge tag 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging
Pull hwmon fixes from Guenter Roeck:
"Bug fixes for pmbus, ltc2978, and lineage-pem drivers
Added specific maintainer for some hwmon drivers"
* tag 'hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
hwmon: (pmbus/ltc2978) Fix temperature reporting
hwmon: (pmbus) Fix krealloc() misuse in pmbus_add_attribute()
hwmon: (lineage-pem) Add missing terminating entry for pem_[input|fan]_attributes
MAINTAINERS: Add maintainer for MAX6697, INA209, and INA2XX drivers
Arnd Bergmann [Fri, 15 Mar 2013 22:14:44 +0000 (23:14 +0100)]
Merge tag 'at91-soc' of git://github.com/at91linux/linux-at91 into next/cleanup
From Nicolas Ferre <nicolas.ferre@atmel.com>:
Simplification of Kconfig and splitting of the non-DT part.
Common pattern for DT board file naming.
* tag 'at91-soc' of git://github.com/at91linux/linux-at91:
ARM: at91: renamme rm9200 dt file
ARM: at91: rename board-dt to more specific name board-dt-sam9
ARM: at91: move non DT Kconfig to Kconfig.non_dt
Arnd Bergmann [Fri, 15 Mar 2013 22:08:09 +0000 (23:08 +0100)]
Merge tag 'at91-driversLCD' of git://github.com/at91linux/linux-at91 into next/drivers
From Nicolas Ferre <nicolas.ferre@atmel.com>:
Some Atmel framebuffer driver enhancements with modification
of configuration data in ARM/AT91 and AVR32/AP7 trees.
A merge of these modifications seems easier through arm-soc
git tree nowadays.
* tag 'at91-driversLCD' of git://github.com/at91linux/linux-at91:
ARM: at91/avr32/atmel_lcdfb: add platform device-id table
atmel_lcdfb: move lcdcon2 register access to compute_hozval
ARM: at91/avr32/atmel_lcdfb: add bus-clock entry
ARM: at91: fix LCD-wiring mode
atmel_lcdfb: fix 16-bpp modes on older SOCs
Arnd Bergmann [Fri, 15 Mar 2013 22:02:39 +0000 (23:02 +0100)]
Merge tag 'mvebu_dt_for_v3.10' of git://git.infradead.org/users/jcooper/linux into next/dt
From Jason Cooper <jason@lakedaemon.net>:
mvebu DT changes for v3.10
- mvebu: gpio expander for the Mirabox
- kirkwood: thermal sensor
- dove: thermal sensor
* tag 'mvebu_dt_for_v3.10' of git://git.infradead.org/users/jcooper/linux:
Dove: Thermal: Add DT node and enable in defconfig
ARM: Kirkwood: Add support thermal sensor for 88F6282 and 88F6283
arm: mvebu: enable gpio expander over i2c on Mirabox platform
Arnd Bergmann [Fri, 15 Mar 2013 21:57:45 +0000 (22:57 +0100)]
Merge tag 'clksrc-cleanup-for-3.10' of git://sources.calxeda.com/kernel/linux into next/cleanup
From Rob Herring <robherring2@gmail.com>:
- Add device_node ptr to clocksource init functions
- Add CLKSRC_OF support to twd_smp timer
* tag 'clksrc-cleanup-for-3.10' of git://sources.calxeda.com/kernel/linux:
ARM: smp_twd: convert to use CLKSRC_OF init
clocksource: tegra20: use the device_node pointer passed to init
clocksource: pass DT node pointer to init functions
clocksource: add empty version of clocksource_of_init
Arnd Bergmann [Fri, 15 Mar 2013 21:51:54 +0000 (22:51 +0100)]
Merge branch 'next/pinctrl-exynos' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung into next/drivers
From Kukjin Kim <kgene.kim@samsung.com>:
Here is support pinctrl-exynos5250 and that already got ack from Linus Walleij.
* 'next/pinctrl-exynos' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung:
ARM: EXYNOS: skip wakeup interrupt registration for exynos5250 if pinctrl is enabled
gpio: samsung: skip gpiolib registration if pinctrl support is enabled for exynos5250
pinctrl: exynos: add exynos5250 SoC specific data
Arnd Bergmann [Fri, 15 Mar 2013 21:46:51 +0000 (22:46 +0100)]
Merge branch 'next/irq-s3c24xx' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung into next/drivers
From Kukjin Kim <kgene.kim@samsung.com>:
Here is finish the irq rework for s3c2412, s3c2440 and s3c2442 into the new
structure and eint0 to 3 on the s3c2412.
* 'next/irq-s3c24xx' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung:
gpio: samsung: fixes build warning with s3c2410_defconfig
ARM: S3C24XX: handle s3c2412 eints using new infrastructure
ARM: S3C24XX: add soc_is_s3c2412 option
ARM: S3C24XX: include first 4 bits of the eint register in irq mapping
ARM: S3C24XX: transform s3c2412 irqs into new structure
ARM: S3C24XX: modify s3c2412 irq init to initialize all irqs
ARM: S3C24XX: move s3c2412 irq init to common code
ARM: S3C24XX: use samsung_sync_wakemask in s3c2412 pm
ARM: S3C24XX: transform s3c2440 irqs into new structure
ARM: S3C24XX: transform s3c2442 irqs into new structure
ARM: S3C24XX: integrate s3c2440 irqs into common init
ARM: S3C24XX: move s3c2440 irqs to common irq code
ARM: S3C24XX: create dedicated irq init functions for s3c2440 and s3c2442
ARM: S3C24XX: move s3c244x irq init to common irq code
Arnd Bergmann [Fri, 15 Mar 2013 21:45:05 +0000 (22:45 +0100)]
Merge branch 'next/timer-samsung' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung into next/drivers
From Kukjin Kim <kgene.kim@samsung.com>:
Here is re-work samsung-time which was s5p-time to support
clocksource/clockevent API for s3c and s5pc100 timer driver. And this enables to
support high resolution timer and tickles mode on them.
Note, this depends on previous pull request (cleanup-s3c) because of touching
while s3c stuff.
* 'next/timer-samsung' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung:
ARM: SAMSUNG: Remove unused plat-samsung/time.c
ARM: S5PC100: Add samsung-time support for s5pc100
ARM: S3C64XX: Add samsung-time support for s3c64xx
ARM: S3C24XX: Add samsung-time support for s3c24xx
ARM: SAMSUNG: Rename s5p-time to samsung-time
ARM: S3C24XX: cleanup the included soc init functions in common.h
ARM: S3C24XX: move plat-samsung/s3c24XX headers to local common.h
ARM: S3C24XX: remove plat/irq.h in plat-samsung
ARM: S3C24XX: plat/common-smdk.h local
mmc: s3cmci: moved mach/regs-sdi.h into s3cmci device driver
Arnd Bergmann [Fri, 15 Mar 2013 21:41:38 +0000 (22:41 +0100)]
Merge branch 'next/cleanup-s5p' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung into next/cleanup
From Kukjin Kim <kgene.kim@samsung.com>:
Two small cleanups for the s5p platforms.
* 'next/cleanup-s5p' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung:
ARM: S5PV210: remove useless ifdef in common.h
ARM: S5PC100: remove useless ifdef in common.h
Arnd Bergmann [Fri, 15 Mar 2013 21:39:46 +0000 (22:39 +0100)]
Merge branch 'next/cleanup-s3c' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung into next/cleanup
From Kukjin Kim <kgene.kim@samsung.com>:
Cleanups for the Samsung S3C24XX platforms.
* 'next/cleanup-s3c' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung:
ARM: S3C24XX: cleanup the included soc init functions in common.h
ARM: S3C24XX: move plat-samsung/s3c24XX headers to local common.h
ARM: S3C24XX: remove plat/irq.h in plat-samsung
ARM: S3C24XX: plat/common-smdk.h local
mmc: s3cmci: moved mach/regs-sdi.h into s3cmci device driver
Arnd Bergmann [Fri, 15 Mar 2013 21:31:37 +0000 (22:31 +0100)]
Merge branch 'spear/multiplatform' into next/multiplatform
The spear platform has been cleaned up a lot by Viresh
Kumar, and is relatively easy to convert to multiplatform,
getting us one more step closer to having all ARMv7
platforms included.
I originally did this before 3.8 but merging it never
worked out so far, mostly because of conflicts against
patches in the DMA tree, but also because I did not push very
hard for these.
* spear/multiplatform:
ARM: spear: enable spear13xx in multi_v7_defconfig
ARM: spear: fix build error in restart.c
ARM: spear: use multiplatform configuration options.
ARM: spear: rename duplicate pl080_plat_data
ARM: spear: make clock driver independent of headers
ARM: spear: move generic.h and pl080.h into private dir
ARM: spear: move all files to mach-spear
ARM: spear: move spear.h and misc_regs.h into plat-spear
ARM: spear: merge Kconfig files
ARM: spear: make spear3xx/6xx mach/spear.h files identical
ARM: spear: move identical headers to plat-spear/include/mach
ARM: spear: unify mach/generic.h and mach/irqs.h
ARM: spear13xx: make mach/dma.h local
Arnd Bergmann [Fri, 15 Mar 2013 21:29:38 +0000 (22:29 +0100)]
Merge branch 'cns3xxx/multiplatform' into next/multiplatform
I've looked at all the platforms recently to see what their
state is. cns3xxx seems quite clean but not very actively
maintained. Since it is really easy to convert to multiplatform,
that's what I did here.
* cns3xxx/multiplatform:
ARM: cns3xxx: initial DT support
ARM: cns3xxx: enable multiplatform support
ARM: cns3xxx: move debug_ll code to include/debug/
ARM: cns3xxx: enable sparse IRQ support
ARM: cns3xxx: make mach header files local
Arnd Bergmann [Fri, 15 Mar 2013 21:25:45 +0000 (22:25 +0100)]
Merge tag 'nomadik-multiplatform' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-nomadik into next/multiplatform
From Linus Walleij <linus.walleij@linaro.org>:
Multiplatform support for the Nomadik
- Get rid of the last header files in the <mach/*> namespace
- Move the debug macro to the common place
- Make the necessary Kconfig fixes and move the platform Kconfig
fragment down to the mach directory.
- Include necessary defconfig update to get the platform going.
* tag 'nomadik-multiplatform' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-nomadik:
ARM: nomadik: delete remnant include files
ARM: nomadik: convert to multiplatform
ARM: nomadik: move debugmacro to debug includes
ARM: nomadik: delete IRQ header
Arnd Bergmann [Fri, 15 Mar 2013 21:17:17 +0000 (22:17 +0100)]
Merge tag 'ux500-dt-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-stericsson into next/fixes-non-critical
From Linus Walleij <linus.walleij@linaro.org>:
Ux500 Device Tree fixes
All are necessary to make a proper DT boot on the v3.9
series:
- IRQ edges.
- Register defines.
- FSMC clock.
- Ethernet clk fixup (one patch to drivers/net ACKed by
David Miller)
- Proper IOS and regulator voltages for MMCI.
- AB8500 GPIOs restored after they were fixed in the v3.9
merge window from the pinctrl tree.
- There is also a minor cleanup in the platform code, but it
is strongly connected to the other patches.
* tag 'ux500-dt-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-stericsson:
ARM: ux500: allow Snowball access to the AB8500 GPIO pins
ARM: ux500: enable AB8500 GPIO for HREF
ARM: ux500: Remove traces of the ios_handler from platform code
ARM: ux500: Use the GPIO regulator framework for SDI0's 'en' and 'vsel'
ARM: ux500: Setup correct settling time for the MMCI regulator
ARM: ux500: Use the correct name when supplying a GPIO enable pin
ARM: ux500: Specify which IOS regulator to use for MMCI
ARM: ux500: Specify the ux5x0 MMCI regulator's on/off GPIO as high-enable
ARM: ux500: Set correct MMCI regulator voltages in the ux5x0 Device Tree
mmc: mmci: Move ios_handler functionality into the driver
net/smsc911x: Provide common clock functionality
clk: ux500: Provide an alias for the SMSC911x Ethernet chip
clk: ux500: Ensure the FMSC clock is obtainable
ARM: ux500: Provide a means to obtain the SMSC9115 clock when DT is enabled
ARM: ux500: Include the PRCMU's Secure Registers in DB8500's DT
ARM: ux500: Change IRQ from low-to-high edge triggered to high-to-low
Stephane Eranian [Fri, 15 Mar 2013 13:26:07 +0000 (14:26 +0100)]
perf,x86: fix kernel crash with PEBS/BTS after suspend/resume
This patch fixes a kernel crash when using precise sampling (PEBS)
after a suspend/resume. Turns out the CPU notifier code is not invoked
on CPU0 (BP). Therefore, the DS_AREA (used by PEBS) is not restored properly
by the kernel and keeps it power-on/resume value of 0 causing any PEBS
measurement to crash when running on CPU0.
The workaround is to add a hook in the actual resume code to restore
the DS Area MSR value. It is invoked for all CPUS. So for all but CPU0,
the DS_AREA will be restored twice but this is harmless.
Arnd Bergmann [Thu, 14 Mar 2013 20:25:32 +0000 (21:25 +0100)]
ARM: kill Hynix h720x platform
The platform was merged about 10 years ago, and has seen few updates
for most of the time since. The people that merged the code seem
no longer interested in it either, so let's remove it now.
Signed-off-by: Arnd Bergmann <arnd@arndb.de> Acked-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: Robert Schwebel <r.schwebel@pengutronix.de> Acked-by: Sascha Hauer <s.hauer@pengutronix.de>
Sascha Hauer [Tue, 26 Feb 2013 09:55:18 +0000 (10:55 +0100)]
ARM: i.MX35: enable MAX clock
The i.MX35 has two bits per clock gate which are decoded as follows:
0b00 -> clock off
0b01 -> clock is on in run mode, off in wait/doze
0b10 -> clock is on in run/wait mode, off in doze
0b11 -> clock is always on
The reset value for the MAX clock is 0b10.
The MAX clock is needed by the SoC, yet unused in the Kernel, so the
common clock framework will disable it during late init time. It will
only disable clocks though which it detects as being turned on. This
detection is made depending on the lower bit of the gate. If the reset
value has been altered by the bootloader to 0b11 the clock framework
will detect the clock as turned on, yet unused, hence it will turn it
off and the system locks up.
This patch turns the MAX clock on unconditionally making the Kernel
independent of the bootloader.
Takashi Iwai [Fri, 15 Mar 2013 13:23:32 +0000 (14:23 +0100)]
ALSA: hda - Fix missing EAPD/GPIO setup for Cirrus codecs
During the transition to the generic parser, the hook to the codec
specific automute function was forgotten. This resulted in the silent
output on some MacBooks.
Haojian Zhuang [Fri, 15 Mar 2013 08:27:53 +0000 (16:27 +0800)]
ARM: mmp: add platform_device head file in gplugd
arch/arm/mach-mmp/gplugd.c: In function ‘gplugd_init’:
arch/arm/mach-mmp/gplugd.c:188:2: error: implicit declaration of
function ‘platform_device_register’
[-Werror=implicit-function-declaration]
cc1: some warnings being treated as errors
make[1]: *** [arch/arm/mach-mmp/gplugd.o] Error 1
make: *** [arch/arm/mach-mmp] Error 2
So append platform_device.h to resolve build issue.
Dan Carpenter [Fri, 15 Mar 2013 06:14:22 +0000 (09:14 +0300)]
sound: sequencer: cap array index in seq_chn_common_event()
"chn" here is a number between 0 and 255, but ->chn_info[] only has
16 elements so there is a potential write beyond the end of the
array.
If the seq_mode isn't SEQ_2 then we let the individual drivers
(either opl3.c or midi_synth.c) handle it. Those functions all
do a bounds check on "chn" so I haven't changed anything here.
The opl3.c driver has up to 18 channels and not 16.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Arnd Bergmann [Thu, 14 Mar 2013 21:56:38 +0000 (22:56 +0100)]
mfd: twl4030-madc: Remove __exit_p annotation
4740f73fe5 "mfd: remove use of __devexit" removed the __devexit annotation
on the twl4030_madc_remove function, but left an __exit_p() present on the
pointer to this function. Using __exit_p was as wrong with the devexit in
place as it is now, but now we get a gcc warning about an unused function.
In order for the twl4030_madc_remove to work correctly in built-in code, we
have to remove the __exit_p.
Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Dylan Reid [Fri, 15 Mar 2013 00:27:45 +0000 (17:27 -0700)]
ALSA: hda/ca0132 - Check download state of DSP.
Instead of using the dspload_is_loaded() function, check the dsp_state
that is kept in the spec. The dspload_is_loaded() function returns
true if the DSP transfer was never started. This false-positive leads
to multiple second delays when ca0132_setup_efaults() times out on
each write.
Dylan Reid [Fri, 15 Mar 2013 00:27:44 +0000 (17:27 -0700)]
ALSA: hda/ca0132 - Check if dspload_image succeeded.
If dspload_image() fails, it was ignored and dspload_wait_loaded() was
still called. dsp_loaded should never be set to true in this case,
skip it. The check in dspload_wait_loaded() return true if the DSP is
loaded or if it never started.
The vm_flags introduced in 6d7825b10dbe ("mm/fremap.c: fix oops on error
path") is supposed to avoid a compiler warning about unitialized
vm_flags without changing the generated code.
However I am concerned that this is going to be very brittle, and fail
with some compiler versions. The failure could be either of:
- compiler could actually load vma->vm_flags before checking for the
!vma condition, thus reintroducing the oops
- compiler could optimize out the !vma check, since the pointer just got
dereferenced shortly before (so the compiler knows it can't be NULL!)
I propose reversing this part of the change and initializing vm_flags to 0
just to avoid the bogus uninitialized use warning.
Signed-off-by: Michel Lespinasse <walken@google.com> Cc: Tommi Rantala <tt.rantala@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Arnd Bergmann [Thu, 14 Mar 2013 23:00:50 +0000 (00:00 +0100)]
Merge at91 lcdfb bug fixes into fixes
These are part of a longer series that has been submitted some time
ago for the frame buffer tree, but it was never accepted there.
The first two of the five patches are bug fixes, so let's merge
this through arm-soc to get a working 3.9 kernel for at91.
Linus Torvalds [Thu, 14 Mar 2013 21:53:07 +0000 (14:53 -0700)]
Merge branch 'rcu/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu
Pull fix for hlist_entry_safe() regression from Paul McKenney:
"This contains a single commit that fixes a regression in
hlist_entry_safe(). This macro references its argument twice, which
can cause NULL-pointer errors. This commit applies a gcc statement
expression, creating a temporary variable to avoid the double
reference. This has been posted to LKML at
https://lkml.org/lkml/2013/3/9/75.
Kudos to CAI Qian, whose testing uncovered this, to Eric Dumazet, who
spotted root cause, and to Li Zefan, who tested this commit."
* 'rcu/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu:
list: Fix double fetch of pointer in hlist_entry_safe()
Arnd Bergmann [Wed, 13 Feb 2013 16:11:09 +0000 (17:11 +0100)]
input/joystick: use get_cycles on ARM
ARM normally has an accurate clock source, so
we can theoretically use analog joysticks more
accurately and at the same time avoid the
build warning
#warning Precise timer not defined for this architecture.
from the joystick driver.
Now, why anybody would use that driver no ARM I have no
idea, but Ben Dooks enabled it in the s3c2410_defconfig
along with a bunch of other drivers, even though that
platform has neither ISA nor PCI support. It still
seems to be the right thing to fix this quirk.
Arnd Bergmann [Thu, 14 Feb 2013 22:11:33 +0000 (23:11 +0100)]
[media] s5p-fimc: fix s5pv210 build
56bc911 "[media] s5p-fimc: Redefine platform data structure for fimc-is"
changed the bus_type member of struct fimc_source_info treewide, but
got one instance wrong in mach-s5pv210, which was evidently not
even build tested.
This adds the missing change to get s5pv210_defconfig to build again.
Applies on the Mauro's media tree.
Arnd Bergmann [Thu, 14 Mar 2013 21:27:32 +0000 (22:27 +0100)]
ARM: cns3xxx: initial DT support
This adds very minimal support for booting cns3xxx using a device
tree. It should support the same devices that cns3420vb provides
but gets them from the DT. All devices that don't have their own
binding are probed through auxdata. This is completely untested
and likely incomplete.
Booting through ATAGS is made optional, so it can be turned off
by anybody who has a DTB file.
list: Fix double fetch of pointer in hlist_entry_safe()
The current version of hlist_entry_safe() fetches the pointer twice,
once to test for NULL and the other to compute the offset back to the
enclosing structure. This is OK for normal lock-based use because in
that case, the pointer cannot change. However, when the pointer is
protected by RCU (as in "rcu_dereference(p)"), then the pointer can
change at any time. This use case can result in the following sequence
of events:
1. CPU 0 invokes hlist_entry_safe(), fetches the RCU-protected
pointer as sees that it is non-NULL.
2. CPU 1 invokes hlist_del_rcu(), deleting the entry that CPU 0
just fetched a pointer to. Because this is the last entry
in the list, the pointer fetched by CPU 0 is now NULL.
3. CPU 0 refetches the pointer, obtains NULL, and then gets a
NULL-pointer crash.
This commit therefore applies gcc's "({ })" statement expression to
create a temporary variable so that the specified pointer is fetched
only once, avoiding the above sequence of events. Please note that
it is the caller's responsibility to use rcu_dereference() as needed.
This allows RCU-protected uses to work correctly without imposing
any additional overhead on the non-RCU case.
Many thanks to Eric Dumazet for spotting root cause!
Reported-by: CAI Qian <caiqian@redhat.com> Reported-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Tested-by: Li Zefan <lizefan@huawei.com>
Arnd Bergmann [Fri, 4 Jan 2013 13:48:44 +0000 (13:48 +0000)]
ARM: gemini: remove platform support
The gemini platform has not seen any updates from its
maintainer since 2010, and has no defconfig file, which
has lead to bitrot in a number of places:
* System restart has been broken for a long time despite
repeated attempts from Russell to get the obvious
fix at least acknowledged by someone who can test it.
* The platform still uses mach/gpio.h, which has been
deprecated for some time, and since Kconfig does not
select NEED_MACH_GPIO_H, it actually failed to build
because of this.
* A change to asm/io.h introduced a number of compiler
warnings for gemini specific code that confuses
physical and virtual addresses.
This removes gemini support entirely under the assumption
that nobody uses it on mainline kernels. Of course, if
there are still active users out there, they should let
us know so we can put support back in and get it better
tested.
Arnd Bergmann [Fri, 4 Jan 2013 13:38:03 +0000 (13:38 +0000)]
ARM: gemini: get platform to build again
There is no defconfig file for gemini, which has lead to a lot
of bitrot. This makes the broken board files, the gpio implementation
and the reset logic work again, and fixes the build warnings
that got introduced with the changes to the readl/writel prototypes.
Linus Torvalds [Thu, 14 Mar 2013 19:11:28 +0000 (12:11 -0700)]
Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs
Pull ext2, ext3, reiserfs, quota fixes from Jan Kara:
"A fix for regression in ext2, and a format string issue in ext3. The
rest isn't too serious."
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
ext2: Fix BUG_ON in evict() on inode deletion
reiserfs: Use kstrdup instead of kmalloc/strcpy
ext3: Fix format string issues
quota: add missing use of dq_data_lock in __dquot_initialize
Liu Bo [Wed, 13 Mar 2013 13:43:03 +0000 (07:43 -0600)]
Btrfs: fix warning when creating snapshots
Creating snapshot passes extent_root to commit its transaction,
but it can lead to the warning of checking root for quota in
the __btrfs_end_transaction() when someone else is committing
the current transaction. Since we've recorded the needed root
in trans_handle, just use it to get rid of the warning.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Wang Shilong [Wed, 6 Mar 2013 11:51:47 +0000 (11:51 +0000)]
Btrfs: return as soon as possible when edquot happens
If one of qgroup fails to reserve firstly, we should return immediately,
it is unnecessary to continue check.
Signed-off-by: Wang Shilong <wangsl-fnst@cn.fujitsu.com> Signed-off-by: Josef Bacik <jbacik@fusionio.com> Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Josef Bacik [Fri, 8 Mar 2013 20:41:02 +0000 (15:41 -0500)]
Btrfs: return EIO if we have extent tree corruption
The callers of lookup_inline_extent_info all handle getting an error back
properly, so return an error if we have corruption instead of being a jerk and
panicing. Still WARN_ON() since this is kind of crucial and I've been seeing it
a bit too much recently for my taste, I think we're doing something wrong
somewhere. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com> Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Eric Sandeen [Sat, 9 Mar 2013 15:18:39 +0000 (15:18 +0000)]
btrfs: use rcu_barrier() to wait for bdev puts at unmount
Doing this would reliably fail with -EBUSY for me:
# mount /dev/sdb2 /mnt/scratch; umount /mnt/scratch; mkfs.btrfs -f /dev/sdb2
...
unable to open /dev/sdb2: Device or resource busy
because mkfs.btrfs tries to open the device O_EXCL, and somebody still has it.
Using systemtap to track bdev gets & puts shows a kworker thread doing a
blkdev put after mkfs attempts a get; this is left over from the unmount
path:
so unmount might complete before __free_device fires & does its blkdev_put.
Adding an rcu_barrier() to btrfs_close_devices() causes unmount to wait
until all blkdev_put()s are done, and the device is truly free once
unmount completes.
Cc: stable@vger.kernel.org Signed-off-by: Eric Sandeen <sandeen@redhat.com> Signed-off-by: Josef Bacik <jbacik@fusionio.com> Signed-off-by: Chris Mason <chris.mason@fusionio.com>