dma_pool_create() needs to unlock the mutex in error case. The bug was
introduced in the 3.16 by commit cc6b664aa2 ("mm/dmapool.c: remove
redundant NULL check for dev in dma_pool_create()")/
Signed-off-by: Krzysztof Hałasa <khc@piap.pl>
Cc: stable@vger.kernel.org # v3.16
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The vendor prefix was renamed from "mrvl" to "marvell". Follow this
change in the dts file.
Signed-off-by: Daniel Mack <zonque@gmail.com>
Signed-off-by: Haojian Zhuang <haojian.zhuang@linaro.org>
The vendor prefix was renamed from "mrvl" to "marvell". Follow this
change in the dts file.
Signed-off-by: Daniel Mack <zonque@gmail.com>
Signed-off-by: Haojian Zhuang <haojian.zhuang@linaro.org>
Pull regulator fix from Mark Brown:
"Fix some leaked OF node references in regulator drivers that have been
left over following a fix on a fix to the reference counting"
* tag 'regulator-v3.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:
regulator: remove unnecessary of_node_get() to parent
Pull spi fixes from Mark Brown:
"A few driver specific fixes for v3.17:
- Fix davinci so that GPIO chip selects work with deferred probe of
GPIOs (which could happen in production depending on kernel config)
plus one incremental stylistic fix to that.
- Several fixes for the newly introduced rockchip driver that came up
in wider testing of the device.
- A couple of small things in the sirf driver, one bug that would
stop DMA transfers working and another update to follow the
documented procedure in the datasheet.
- Fix some memory leaks with devm_kzalloc() being used outside of the
device bind path"
* tag 'spi-v3.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: davinci: remove empty function davinci_spi_cleanup
spi: davinci: request cs_gpio's from probe
spi/pl022: Fix error message
spi/rockchip: Mark DMA as optional
spi/rockchip: Don't warn if SPI is busy but disabled
spi/rockchip: Fix the wait_for_idle() timeout
spi: sirf: add fifo reset/start for cmd transfer
spi: sirf: enable RX_IO_DMA_INT interrupt
spi: dw: Don't use devm_kzalloc in master->setup callback
spi: fsl: Don't use devm_kzalloc in master->setup callback
Use the ONE macro instead of REG, and we can simplify proc_cpuset_show().
Signed-off-by: Zefan Li <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Use the ONE macro instead of REG, and we can simplify proc_cgroup_show().
Signed-off-by: Zefan Li <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Instead of using a global work to schedule release agent on removable
cgroups, we change to use a per-cgroup work to do this, which makes
the code much simpler.
v2: use a dedicated work instead of reusing css->destroy_work. (Tejun)
Signed-off-by: Zefan Li <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
James Drew reports another bug whereby the NFS client is now sending
an OPEN_DOWNGRADE in a situation where it should really have sent a
CLOSE: the client is opening the file for O_RDWR, but then trying to
do a downgrade to O_RDONLY, which is not allowed by the NFSv4 spec.
Reported-by: James Drews <drews@engr.wisc.edu>
Link: http://lkml.kernel.org/r/541AD7E5.8020409@engr.wisc.edu
Fixes: aee7af356e (NFSv4: Fix problems with close in the presence...)
Cc: stable@vger.kernel.org # 2.6.33+
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
There is a race between nfs4_state_manager() and
nfs_server_remove_lists() that happens during a nfsv3 mount.
The v3 mount notices there is already a supper block so
nfs_server_remove_lists() called which uses the nfs_client_lock
spin lock to synchronize access to the client list.
At the same time nfs4_state_manager() is running through
the client list looking for work to do, using the same
lock. When nfs4_state_manager() wins the race to the
list, a v3 client pointer is found and not ignored
properly which causes the panic.
Moving some protocol checks before the state checking
avoids the panic.
CC: Stable Tree <stable@vger.kernel.org>
Signed-off-by: Steve Dickson <steved@redhat.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
DRA74x and DRA72x family of processors vary slightly in the number
of CPUs. So, add different instances of PMU for each of these processor
groups. Further, since the interrupts bypass crossbar and are directly
connected to GIC, mark the dts nodes with relevant information.
Tested with perf utility.
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Lucas Weaver <l-weaver@ti.com>
Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Use the more common pr_warn.
Other miscellanea:
o Realign arguments
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
We're moving to the dmaengine API, so let's remove the unused
pieces of the omap legacy DMA code to make sure we don't get
any new users for these:
omap_set_dma_color_mode
omap_set_dma_src_index
omap_set_dma_dest_index
omap_dma_unlink_lch
omap_clear_dma
omap_dma_running
omap_dma_set_prio_lch
omap_set_dma_dst_endian_type
omap_set_dma_src_endian_type
omap_get_dma_index
omap_dma_disable_irq
omap_request_dma_chain
omap_free_dma_chain
omap_dma_chain_a_transfer
omap_start_dma_chain_transfers
omap_stop_dma_chain_transfers
omap_get_dma_chain_index
omap_get_dma_chain_dst_pos
omap_get_dma_chain_src_pos
omap_modify_dma_chain_params
omap_dma_chain_status
Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Tony Lindgren <tony@atomide.com>
In order to handle errata I688, a page of sram was reserved by doing a
static iotable map. Now that we use gen_pool to manage sram, we can
completely remove all of these static mappings and use gen_pool_alloc()
to get the one page of sram space needed to implement errata I688.
omap_bus_sync will be NOP until SRAM initialization happens.
Suggested-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Dave Gerlach <d-gerlach@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Use drivers/misc/sram.c driver to manage SRAM on all DT only
OMAP platforms (am33xx, am43xx, omap4 and omap5) instead of
the existing private plat-omap/sram.c
Address and size related data is removed from mach-omap2/sram.c
and now passed to drivers/misc/sram.c from DT.
Users can hence use general purpose allocator apis instead of
OMAP private ones to manage and use SRAM.
Signed-off-by: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Dave Gerlach <d-gerlach@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
parse_arg() has three possible return values:
-EINVAL if sscanf(), in short, fails;
zero if "count" is zero; and
"count" in all other cases
But "count" will never be zero. See, parse_arg() is called by the
various store functions. And the callchain of these functions starts
with sysfs_kf_write(). And that function checks for a zero "count". So
we can stop checking for a zero "count", drop the "count" argument
entirely, and transform parse_arg() into a function that returns zero on
success or a negative error. That, in turn, allows to make those store
functions just return "count" on success. The net effect is that the
code becomes a bit easier to understand.
A nice side effect is that this GCC warning is silenced too:
drivers/platform/x86/eeepc-laptop.c: In function ‘store_sys_acpi’:
drivers/platform/x86/eeepc-laptop.c:279:10: warning: ‘value’ may be used uninitialized in this function [-Wmaybe-uninitialized]
int rv, value;
Which is, of course, the reason to have a look at parse_arg().
Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Now that we have panel support for DT based booting,
let's make it usable and enable most things as modules.
Note that omap3 boards need also the ads7847 module for
the panel that we're now changing to a loadable module.
And n900 seems to require setting the brightness via
sysfs for acx565akm/brightness after modprobe of
panel_sony_acx565akm and omapfb.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Since many omaps run on battery, we should have the battery
drivers enabled. Let's also enable the reset driver.
Signed-off-by: Tony Lindgren <tony@atomide.com>
cgroup_pidlist_start() holds cgrp->pidlist_mutex and then calls
pidlist_array_load(), and cgroup_pidlist_stop() releases the mutex.
It is wrong that we release the mutex in the failure path in
pidlist_array_load(), because cgroup_pidlist_stop() will be called
no matter if cgroup_pidlist_start() returns errno or not.
Fixes: 4bac00d16a
Cc: <stable@vger.kernel.org> # 3.14+
Signed-off-by: Zefan Li <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Add DSS related pinmux and display data nodes required to support HDMI
and DVI video out on CM-T54.
Signed-off-by: Dmitry Lifshitz <lifshitz@compulab.co.il>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Follow the comment style of mode0_name.modeX_name for pins
which mux mode differs from MUX_MODE0.
Signed-off-by: Dmitry Lifshitz <lifshitz@compulab.co.il>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Follow the comment style of mode0_name.modeX_name for pins
which mux mode differs from MUX_MODE0.
Signed-off-by: Dmitry Lifshitz <lifshitz@compulab.co.il>
Signed-off-by: Tony Lindgren <tony@atomide.com>
With the IO chain reconfigure fixed, we can now enable the PMIC
scripts for LDP.
Note that at least on my es3.0 based LDP, the UART seems to be
flakey after wake-up events from off-idle and hangs but eventually
continues.
Signed-off-by: Tony Lindgren <tony@atomide.com>
As we have support for this in board-rx51-peripherals.c, let's
add it to the .dts files too.
Note that the reset GPIO will eventually go to the driver.
For now let's just pull it down and skip any further reset
in case the bootloader has configured the MAC address so
NFSroot works.
Also note that after 3430-sdp are using proper GPMC timings
we can remove the tests for smsc,lan91c94 in gpmc.c.
Signed-off-by: Tony Lindgren <tony@atomide.com>
There are external pulls on these lines and enabling the
internal pulls can cause issue. This is because the internal
pulls are parallel with the external pulls. So let's clear
the internal I2C pulls.
Signed-off-by: Tony Lindgren <tony@atomide.com>
This is no longer needed as the device specific wake-up event
can now be specified with interrupts-extended property where
the second interrupt is the pinctrl-single register, such as
the UART3 RX pin.
Note that twl4030_omap3.dtsi needs to set WAKEUPENABLE for
off-idle to properly trigger the PMIC scripts. And GPIO pins
still need to set WAKEUPENABLE for wake-up events.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Compared to legacy booting, we don't have wake-up events enabled
for device tree based booting. This means that if deeper idle
states are enabled, the device won't wake up to UART events and
seems like it has hung.
Let's fix that by adding the wake-up interrupt. Note that we
don't need to set the PIN_OFF_WAKEUPENABLE any longer, that's
handled by the wake-up interrupt when the serial driver does
request_irq on it.
Tested with the following on omap3-overo-summit that has the
ES2.1 omap:
#!/bin/bash
uarts=$(find /sys/class/tty/ttyO*/device/power/ -type d)
for uart in $uarts; do
echo 3000 > $uart/autosuspend_delay_ms
done
uarts=$(find /sys/class/tty/ttyO*/power/ -type d)
for uart in $uarts; do
echo enabled > $uart/wakeup
echo auto > $uart/control
done
echo 1 > /sys/kernel/debug/pm_debug/enable_off_mode
# grep -i uart /proc/interrupts
90: 1085 INTC 74 OMAP UART2
338: 5 pinctrl 366 OMAP UART2
# grep ^core_pwrdm /sys/kernel/debug/pm_debug/count
core_pwrdm (ON),OFF:1654,RET:131,INA:39,ON:1825...
Signed-off-by: Tony Lindgren <tony@atomide.com>
Note that we can now use the CONFIG_GENERIC_CPUFREQ_CPU0,
so let's only enable that. Let's use CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND
as suggested by Nishant.
And also let's enable thermal as explained by Nishant Menon:
Many TI SoCs using Highest frequency is not really too nice of an idea for
long periods of time. And not everything is upstream to support things
optimially - example avs class 0, 1.5 ABB consolidation with cpufreq etc..
We definitely need thermal enabled as well for device safety needs.
[tony@atomide.com: updated per Nishant's suggestions]
Acked-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
In sprz318f.pdf "Usage Note 2.7" says that UARTs cannot acknowledge
idle requests in smartidle mode when configured for DMA operations.
This prevents L4 from going idle. So let's use manual idle mode
instead.
Otherwise systems using Sebastian's 8250 patches with DMA will
never enter deeper idle states because of the errata above.
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Commit cc824534d4 ("ARM: OMAP2+: hwmod: Rearm wake-up interrupts
for DT when MUSB is idled") fixed issues with hung UART wake-up
events by calling _reconfigure_io_chain() when MUSB is connected
or disconnected.
As pointed out by Paul Walmsley, we may need to also call
_reconfigure_io_chain() in other cases, so it should be a separate
flag. Let's add HWMOD_RECONFIG_IO_CHAIN as suggested by Paul.
Reviewed-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
We can only determine the final security level when both pairing request
and response have been exchanged. When initiating pairing the starting
target security level is set to MEDIUM unless explicitly specified to be
HIGH, so that we can still perform pairing even if the remote doesn't
have MITM capabilities. However, once we've received the pairing
response we should re-consult the remote and local IO capabilities and
upgrade the target security level if necessary.
Without this patch the resulting Long Term Key will occasionally be
reported to be unauthenticated when it in reality is an authenticated
one.
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Cc: stable@vger.kernel.org
This switches the SAMA5D3 to use the new atmel,sama5d3-pinctrl id that was
added with the drive strength options patch.
Signed-off-by: Marek Roszko <mark.roszko@gmail.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
[nicolas.ferre@atmel.com: second compatible string kept as at91sam9x5]
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Commit 98683650 "Merge branch 'drbd-8.4_ed6' into
for-3.8-drivers-drbd-8.4_ed6" switches to the new augment API, but the
new API requires that the tree is augmented before rb_insert_augmented()
is called, which is missing.
So we add the augment-code to drbd_insert_interval() when it travels the
tree up to down before rb_insert_augmented(). See the example in
include/linux/interval_tree_generic.h or Documentation/rbtree.txt.
drbd_insert_interval() may cancel the insertion when traveling, in this
case, the just added augment-code does nothing before cancel since the
@this node is already in the subtrees in this case.
CC: Michel Lespinasse <walken@google.com>
CC: stable@kernel.org # v3.10+
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Andreas Gruenbacher <agruen@linbit.com>
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
As the SSC integrate in at91sam9g20 support frame sync length
extension, so switch compatible string to support this feature.
Signed-off-by: Bo Shen <voice.shen@atmel.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
As the SSC integrate in at91sam9rl support frame sync length
extension, so switch compatible string to support this feature.
Signed-off-by: Bo Shen <voice.shen@atmel.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
This reverts commit b96de000bc.
This commit is triggering failures to mount by subvolume id in some
configurations. The main problem is how many different ways this
scanning function is used, both for scanning while mounted and
unmounted. A proper cleanup is too big for late rcs.
For now, just revert the commit and we'll put a better fix into a later
merge window.
Signed-off-by: Chris Mason <clm@fb.com>
This reverts commit be0a8887bb.
The original commit f2267089ea
(usb: gadget: composite: dequeue cdev->req before free it in
composite_dev_cleanup) ended up being reverted because it caused
more issues then fixed. We will also revert this counter part
commit so we start clean to properly add that idea back.
Cc: Li Jun <b47624@freescale.com>
Signed-of-by: Felipe Balbi <balbi@ti.com>
The tracepoint of extent map doesn't parse @flag correctly, we set @flag via
set_bit(), so we need to parse it on a bit bias.
Also add the missing flag, EXTENT_FLAG_FS_MAPPING.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <clm@fb.com>
The following commit enhanced the merge_extent_mapping() to reduce
fragment in extent map tree, but it can't handle case which existing
lies before map_start:
51f39 btrfs: Use right extent length when inserting overlap extent map.
[BUG]
When existing extent map's start is before map_start,
the em->len will be minus, which will corrupt the extent map and fail to
insert the new extent map.
This will happen when someone get a large extent map, but when it is
going to insert it into extent map tree, some one has already commit
some write and split the huge extent into small parts.
[REPRODUCER]
It is very easy to tiger using filebench with randomrw personality.
It is about 100% to reproduce when using 8G preallocated file in 60s
randonrw test.
[FIX]
This patch can now handle any existing extent position.
Since it does not directly use existing->start, now it will find the
previous and next extent around map_start.
So the old existing->start < map_start bug will never happen again.
[ENHANCE]
This patch will insert the best fitted extent map into extent map tree,
other than the oldest [map_start, map_start + sectorsize) or the
relatively newer but not perfect [map_start, existing->start).
The patch will first search existing extent that does not intersects with
the desired map range [map_start, map_start + len).
The existing extent will be either before or behind map_start, and based
on the existing extent, we can find out the previous and next extent
around map_start.
So the best fitted extent would be [prev->end, next->start).
For prev or next is not found, em->start would be prev->end and em->end
wold be next->start.
With this patch, the fragment in extent map tree should be reduced much
more than the 51f39 commit and reduce an unneeded extent map tree search.
Reported-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
