The kfifo's request_update callback will free the current buffer and allocate a
new one if the size has changed. This will remove any samples that might still
be left in the buffer. If the size has not changed the buffer content is
left untouched though. This is a bit inconsistent and might cause an application
to see data from a previous capture. This patch inserts a call to
kfifo_reset_out() when the size did not change. This makes sure that any pending
samples are removed from the buffer.
Note, due to a different bug the buffer is currently always re-allocated, even
if the size did not change. So this patch will not change the behavior. In the
next patch the bug will be fixed and this patch makes sure that the current
behavior is kept.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
iio:kfifo: Protect against concurrent access from userspace
It is possible for userspace to concurrently access the buffer from multiple
threads or processes. To avoid corruption of the internal state of the buffer we
need to add proper locking. It is possible for multiple processes to try to read
from the buffer concurrently and it is also possible that one process causes a
buffer re-allocation while a different process still access the buffer. Both can
be fixed by protecting the calls to kfifo_to_user() and kfifo_alloc() by the
same mutex. In iio_read_first_n_kfifo() we also use kfifo_recsize() instead of
the buffers bytes_per_datum to avoid a race that can happen if bytes_per_datum
has been changed, but the buffer has not been reallocated yet.
Note that all access to the buffer from within the kernel is already properly
synchronized, so there is no need for extra locking in iio_store_to_kfifo().
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
staging:iio: Allow to build SoC specific drivers when COMPILE_TEST is set
None of the SPEAr, LPC32XX or MXS ADC drivers have a compile time dependency on
their respective platform. So make it possible to build the drivers when
CONFIG_COMPILE_TEST is set. This makes it easier to compile test changes.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Reviewed-by: Marek Vasut <marex@denx.de> Cc: Stefan Roese <sr@denx.de> Cc: Roland Stigge <stigge@antcom.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
The MXS ADC driver uses the stmp_reset_block() which is only provided when the
STMP_DEVICE Kconfig symbol is selected. Hence the driver should select this
symbol. So far this has not been a problem since the driver depends on ARCH_MXS,
which already selects STMP_DEVICE, but will become necessary once we allow the
driver to be built when COMPILE_TEST is selected.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Cc: Marek Vasut <marex@denx.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
The driver is casting from one __iomem pointer to another. Make sure to include
__iomem in the cast, otherwise sparse will complain with the following warning:
drivers/staging/iio/adc/spear_adc.c:321:18: warning: cast removes address space of expression
drivers/staging/iio/adc/spear_adc.c:320:33: warning: incorrect type in assignment (different address spaces)
drivers/staging/iio/adc/spear_adc.c:320:33: expected struct adc_regs_spear3xx [noderef] <asn:2>*adc_base_spear3xx
drivers/staging/iio/adc/spear_adc.c:320:33: got struct adc_regs_spear3xx *<noident>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Cc: Stefan Roese <sr@denx.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Remove the scale_mv variable from the read_raw() callback. Fixes the following
warning:
drivers/staging/iio/adc/spear_adc.c: In function 'spear_read_raw':
drivers/staging/iio/adc/spear_adc.c:149:6: warning: unused variable 'scale_mv'
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Cc: Stefan Roese <sr@denx.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
iio:buffer: Ignore noop requests for iio_update_buffers()
Since the kernel now disables all buffers when a device is unregistered it might
happen that a in-kernel consumer tries to disable that buffer again. So ignore
requests where the buffer already is in the desired state.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
staging:iio:ad7291: Use event spec for threshold hysteresis
Register the event threshold hysteresis attributes by using the new
IIO_EV_INFO_HYSTERESIS event spec type. This allows us to throw away a good
portion of boiler-plate code.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
staging:iio:ad799x: Use event spec for threshold hysteresis
Register the event threshold hysteresis attributes by using the new
IIO_EV_INFO_HYSTERESIS event spec type. This allows us to throw away a good
portion of boiler-plate code.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
For some devices it is possible to configure a hysteresis for threshold (or
similar) events. This patch adds a new hysteresis event info type which allows
for easy creation and read/write handling of the sysfs attribute.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
The event configuration interface of the IIO framework has not been getting the
same attention as other parts. As a result it has not seen the same improvements
as e.g. the channel interface has seen with the introduction of the channel spec
struct. Currently all the event config callbacks take a u64 (the so called event
code) to pass all the different information about for which event the callback
is invoked. The callback function then has to extract the information it is
interested in using some macros with rather long names. Most information encoded
in the event code comes straight from the iio_chan_spec struct the event was
registered for. Since we always have a handle to the channel spec when we call
the event callbacks the first step is to add the channel spec as a parameter to
the event callbacks. The two remaining things encoded in the event code are the
type and direction of the event. Instead of passing them in one parameter, add
one parameter for each of them and remove the eventcode from the event
callbacks. The patch also adds a new iio_event_info parameter to the
{read,write}_event_value callbacks. This makes it possible, similar to the
iio_chan_info_enum for channels, to specify additional properties other than
just the value for an event. Furthermore the new interface will allow to
register shared events. This is e.g. useful if a device allows configuring a
threshold event, but the threshold setting is the same for all channels.
To implement this the patch adds a new iio_event_spec struct which is similar to
the iio_chan_spec struct. It as two field to specify the type and the direction
of the event. Furthermore it has a mask field for each one of the different
iio_shared_by types. These mask fields holds which kind of attributes should be
registered for the event. Creation of the attributes follows the same rules as
the for the channel attributes. E.g. for the separate_mask there will be a
attribute for each channel with this event, for the shared_by_type there will
only be one attribute per channel type. The iio_chan_spec struct gets two new
fields, 'event_spec' and 'num_event_specs', which is used to specify which the
events for this channel. These two fields are going to replace the channel's
event_mask field.
For now both the old and the new event config interface coexist, but over the
few patches all drivers will be converted from the old to the new interface.
Once that is done all code for supporting the old interface will be removed.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
iio: Add a helper to free a list of IIO device attributes
We have the same code to free a IIO device attribute list in multiple place.
This patch adds a new helper function to take care of this and replaces the
custom instances with a call to the helper function. Note that we do not need to
call list_del() for each of the list items since we will never look at any of
the list items nor the list itself again.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
iio:buffer: Add proper locking for iio_update_buffers()
We need to make sure that in-kernel users of iio_update_buffers() do not race
against each other or against unregistration of the device. So we need to take
both the mlock and the info_exist_lock when calling iio_update_buffers() from a
in-kernel consumer.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
iio: Wakeup poll and blocking reads when the device is unregistered
Once the device has been unregistered there won't be any new data no matter how
long a userspace application waits, so we might as well wake them up and let
them know.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
iio: Return -ENODEV for file operations if the device has been unregistered
If the IIO device has been unregistered return -ENODEV for any further file
operations like read() and ioctl(). This avoids userspace being able to grab new
references to the device.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Since the buffer is accessed by userspace we can not just free the buffers
memory once we are done with it in kernel space. There might still be open file
descriptors and userspace still might be accessing the buffer. This patch adds
support for reference counting to the IIO buffers. When a buffer is created and
initialized its initial reference count is set to 1. Instead of freeing the
memory of the buffer the buffer's _free() function will drop that reference
again. But only after the last reference to the buffer has been dropped the
buffer the buffer's memory will be freed. The IIO device will take a reference
to its primary buffer. The patch adds a small helper function for this called
iio_device_attach_buffer() which will get a reference to the buffer and assign
the buffer to the IIO device. This function must be used instead of assigning
the buffer to the device by hand. The reference is only dropped once the IIO
device is freed and we can be sure that there are no more open file handles. A
reference to a buffer will also be taken whenever the buffer is active to avoid
the buffer being freed while data is still being send to it.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Josh Wu [Tue, 8 Oct 2013 03:48:00 +0000 (04:48 +0100)]
iio: at91: ADC start-up time calculation changed since at91sam9x5
Since in at91sam9x5, sama5d3x chip. the start up time calucation is changed.
This patch can choose different start up time calculation formula for different
chips.
Signed-off-by: Josh Wu <josh.wu@atmel.com> Acked-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Aaro Koskinen [Sun, 6 Oct 2013 20:35:15 +0000 (23:35 +0300)]
staging: octeon-ethernet: allow to set IRQ smp_affinity freely
Currently the driver assumes that CPU 0 is handling all the hard IRQs.
This is wrong in Linux SMP systems where user is allowed to assign to
hardware IRQs to any CPU. The driver will stop working if user sets
smp_affinity so that interrupts end up being handled by other than CPU
0. The patch fixes that.
Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Eliminate cvmx_usb_internal_state, just use cvmx_usb_state everywhere.
This also enables to allocate only the needed amount of data for the
USB internal state, instead of always allocating 64 KB.
Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Aaro Koskinen [Sun, 6 Oct 2013 19:22:24 +0000 (22:22 +0300)]
staging: octeon-usb: merge cvmx-usb into octeon-hcd
cvmx-usb module provided Cavium "OS abstraction layer" for USB
functionality. To make this driver a proper Linux driver, we need to
refactor this layer out. By making all the code internal to the HCD
driver makes this task easier.
Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pull SCSI target fixes from Nicholas Bellinger:
"Here are the outstanding target fixes queued up for v3.12-rc4 code.
The highlights include:
- Make vhost/scsi tag percpu_ida_alloc() use GFP_ATOMIC
- Allow sess_cmd_map allocation failure fallback to use vzalloc
- Fix COMPARE_AND_WRITE se_cmd->data_length bug with FILEIO backends
- Fixes for COMPARE_AND_WRITE callback recursive failure OOPs + non
zero scsi_status bug
- Make iscsi-target do acknowledgement tag release from RX context
- Setup iscsi-target with extra (cmdsn_depth / 2) percpu_ida tags
Also included is a iscsi-target patch CC'ed for v3.10+ that avoids
legacy wait_for_task=true release during fast-past StatSN
acknowledgement, and two other SRP target related patches that address
long-standing issues that are CC'ed for v3.3+.
Extra thanks to Thomas Glanzmann for his testing feedback with
COMPARE_AND_WRITE + EXTENDED_COPY VAAI logic"
* git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending:
iscsi-target; Allow an extra tag_num / 2 number of percpu_ida tags
iscsi-target: Perform release of acknowledged tags from RX context
iscsi-target: Only perform wait_for_tasks when performing shutdown
target: Fail on non zero scsi_status in compare_and_write_callback
target: Fix recursive COMPARE_AND_WRITE callback failure
target: Reset data_length for COMPARE_AND_WRITE to NoLB * block_size
ib_srpt: always set response for task management
target: Fall back to vzalloc upon ->sess_cmd_map kzalloc failure
vhost/scsi: Use GFP_ATOMIC with percpu_ida_alloc for obtaining tag
ib_srpt: Destroy cm_id before destroying QP.
target: Fix xop->dbl assignment in target_xcopy_parse_segdesc_02
Linus Torvalds [Sun, 6 Oct 2013 20:35:15 +0000 (13:35 -0700)]
Merge branch 'fixes' of git://git.infradead.org/users/vkoul/slave-dma
Pull slave-dmaengine fixes from Vinod Koul:
"Here is the slave dmanegine fixes. We have the fix for deadlock issue
on imx-dma by Michael and Josh's edma config fix along with author
change"
* 'fixes' of git://git.infradead.org/users/vkoul/slave-dma:
dmaengine: imx-dma: fix callback path in tasklet
dmaengine: imx-dma: fix lockdep issue between irqhandler and tasklet
dmaengine: imx-dma: fix slow path issue in prep_dma_cyclic
dma/Kconfig: Make TI_EDMA select TI_PRIV_EDMA
edma: Update author email address
Merge tag 'iio-for-3.13c' of git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next
Jonathan writes:
Third set of new functionality and cleanups for IIO in the 3.13 cycle.
Driver New Functionality
* MXS - new interrupt driven touch screen support for i.MX23/28. Old
polled implementation dropped.
Driver Cleanups
* Some spi_sync boilerplate dropped by using spi_sync_transfer
* Some switching of drivers to the fractional type for scale reading.
Moves the ugly calculation into one place.
* Fix the documentation for *_voltage_scale which is never been correct
or as implemented in any driver.
* HID sensor hub and children : Open the sensor hub only when someone cares.
* hmc5843 - various minor
Jan Kara [Wed, 2 Oct 2013 14:27:51 +0000 (16:27 +0200)]
lustre: Convert ll_get_user_pages() to use get_user_pages_fast()
CC: Peng Tao <tao.peng@emc.com> Acked-by: Andreas Dilger <andreas.dilger@intel.com> CC: hpdd-discuss@lists.01.org Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Linus Torvalds [Sat, 5 Oct 2013 19:17:24 +0000 (12:17 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs
Pull btrfs fixes from Chris Mason:
"This is a small collection of fixes, including a regression fix from
Liu Bo that solves rare crashes with compression on.
I've merged my for-linus up to 3.12-rc3 because the top commit is only
meant for 3.12. The rest of the fixes are also available in my master
branch on top of my last 3.11 based pull"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
btrfs: Fix crash due to not allocating integrity data for a bioset
Btrfs: fix a use-after-free bug in btrfs_dev_replace_finishing
Btrfs: eliminate races in worker stopping code
Btrfs: fix crash of compressed writes
Btrfs: fix transid verify errors when recovering log tree
Linus Torvalds [Sat, 5 Oct 2013 19:11:40 +0000 (12:11 -0700)]
Merge tag 'gpio-v3.12-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio
Pull GPIO fixes from Linus Walleij:
"Two patches for the OMAP driver, dealing with setting up IRQs properly
on the device tree boot path"
* tag 'gpio-v3.12-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:
gpio/omap: auto-setup a GPIO when used as an IRQ
gpio/omap: maintain GPIO and IRQ usage separately
Linus Torvalds [Sat, 5 Oct 2013 18:54:10 +0000 (11:54 -0700)]
Merge tag 'usb-3.12-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Pull USB fixes from Greg KH:
"Here are none fixes for various USB driver problems. The majority are
gadget/musb fixes, but there are some new device ids in here as well"
* tag 'usb-3.12-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
usb: chipidea: add Intel Clovertrail pci id
usb: gadget: s3c-hsotg: fix can_write limit for non-periodic endpoints
usb: gadget: f_fs: fix error handling
usb: musb: dsps: do not bind to "musb-hdrc"
USB: serial: option: Ignore card reader interface on Huawei E1750
usb: musb: gadget: fix otg active status flag
usb: phy: gpio-vbus: fix deferred probe from __init
usb: gadget: pxa25x_udc: fix deferred probe from __init
usb: musb: fix otg default state
Linus Torvalds [Sat, 5 Oct 2013 18:26:19 +0000 (11:26 -0700)]
Merge tag 'tty-3.12-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Pull tty fixes from Greg KH:
"Here are two tty driver fixes for 3.12-rc4.
One fixes the reported regression in the n_tty code that a number of
people found recently, and the other one fixes an issue with xen
consoles that broke in 3.10"
* tag 'tty-3.12-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
xen/hvc: allow xenboot console to be used again
tty: Fix pty master read() after slave closes
Darrick J. Wong [Fri, 20 Sep 2013 03:37:07 +0000 (20:37 -0700)]
btrfs: Fix crash due to not allocating integrity data for a bioset
When btrfs creates a bioset, we must also allocate the integrity data pool.
Otherwise btrfs will crash when it tries to submit a bio to a checksumming
disk:
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Josef Bacik <jbacik@fusionio.com> Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Linus Torvalds [Sat, 5 Oct 2013 03:50:16 +0000 (20:50 -0700)]
Merge branch 'for-linus' of git://git.samba.org/sfrench/cifs-2.6
Pull CIFS fixes from Steve French:
"Small set of cifs fixes. Most important is Jeff's fix that works
around disconnection problems which can be caused by simultaneous use
of user space tools (starting a long running smbclient backup then
doing a cifs kernel mount) or multiple cifs mounts through a NAT, and
Jim's fix to deal with reexport of cifs share.
I expect to send two more cifs fixes next week (being tested now) -
fixes to address an SMB2 unmount hang when server dies and a fix for
cifs symlink handling of Windows "NFS" symlinks"
* 'for-linus' of git://git.samba.org/sfrench/cifs-2.6:
[CIFS] update cifs.ko version
[CIFS] Remove ext2 flags that have been moved to fs.h
[CIFS] Provide sane values for nlink
cifs: stop trying to use virtual circuits
CIFS: FS-Cache: Uncache unread pages in cifs_readpages() before freeing them
Linus Torvalds [Sat, 5 Oct 2013 03:48:20 +0000 (20:48 -0700)]
Merge tag 'pci-v3.12-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull PCI fix from Bjorn Helgaas:
"We merged what was intended to be an MMCONFIG cleanup, but in fact,
for systems without _CBA (which is almost everything), it broke
extended config space for domain 0 and it broke all config space for
other domains.
This reverts the change"
* tag 'pci-v3.12-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
Revert "x86/PCI: MMCONFIG: Check earlier for MMCONFIG region at address zero"
07f9b61c was intended to be a cleanup that didn't change anything, but in
fact, for systems without _CBA (which is almost everything), it broke
extended config space for domain 0 and all config space for other domains.
Linus Torvalds [Fri, 4 Oct 2013 22:03:42 +0000 (15:03 -0700)]
Merge tag 'pm+acpi-3.12-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull ACPI and power management fixes from Rafael Wysocki:
- The resume part of user space driven hibernation (s2disk) is now
broken after the change that moved the creation of memory bitmaps to
after the freezing of tasks, because I forgot that the resume utility
loaded the image before freezing tasks and needed the bitmaps for
that. The fix adds special handling for that case.
- One of recent commits changed the export of acpi_bus_get_device() to
EXPORT_SYMBOL_GPL(), which was technically correct but broke existing
binary modules using that function including one in particularly
widespread use. Change it back to EXPORT_SYMBOL().
- The intel_pstate driver sometimes fails to disable turbo if its
no_turbo sysfs attribute is set. Fix from Srinivas Pandruvada.
- One of recent cpufreq fixes forgot to update a check in cpufreq-cpu0
which still (incorrectly) treats non-NULL as non-error. Fix from
Philipp Zabel.
- The SPEAr cpufreq driver uses a wrong variable type in one place
preventing it from catching errors returned by one of the functions
called by it. Fix from Sachin Kamat.
* tag 'pm+acpi-3.12-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
ACPI: Use EXPORT_SYMBOL() for acpi_bus_get_device()
intel_pstate: fix no_turbo
cpufreq: cpufreq-cpu0: NULL is a valid regulator, part 2
cpufreq: SPEAr: Fix incorrect variable type
PM / hibernate: Fix user space driven resume regression
Linus Torvalds [Fri, 4 Oct 2013 21:47:22 +0000 (14:47 -0700)]
Merge tag 'xfs-for-linus-v3.12-rc4' of git://oss.sgi.com/xfs/xfs
Pull xfs bugfixes from Ben Myers:
"There are lockdep annotations for project quotas, a fix for dirent
dtype support on v4 filesystems, a fix for a memory leak in recovery,
and a fix for the build error that resulted from it. D'oh"
* tag 'xfs-for-linus-v3.12-rc4' of git://oss.sgi.com/xfs/xfs:
xfs: Use kmem_free() instead of free()
xfs: fix memory leak in xlog_recover_add_to_trans
xfs: dirent dtype presence is dependent on directory magic numbers
xfs: lockdep needs to know about 3 dquot-deep nesting
Ilya Dryomov [Wed, 2 Oct 2013 17:41:01 +0000 (20:41 +0300)]
Btrfs: fix a use-after-free bug in btrfs_dev_replace_finishing
free_device rcu callback, scheduled from btrfs_rm_dev_replace_srcdev,
can be processed before btrfs_scratch_superblock is called, which would
result in a use-after-free on btrfs_device contents. Fix this by
zeroing the superblock before the rcu callback is registered.
Cc: Stefan Behrens <sbehrens@giantdisaster.de> Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Ilya Dryomov [Wed, 2 Oct 2013 16:39:50 +0000 (19:39 +0300)]
Btrfs: eliminate races in worker stopping code
The current implementation of worker threads in Btrfs has races in
worker stopping code, which cause all kinds of panics and lockups when
running btrfs/011 xfstest in a loop. The problem is that
btrfs_stop_workers is unsynchronized with respect to check_idle_worker,
check_busy_worker and __btrfs_start_workers.
E.g., check_idle_worker race flow:
btrfs_stop_workers(): check_idle_worker(aworker):
- grabs the lock
- splices the idle list into the
working list
- removes the first worker from the
working list
- releases the lock to wait for
its kthread's completion
- grabs the lock
- if aworker is on the working list,
moves aworker from the working list
to the idle list
- releases the lock
- grabs the lock
- puts the worker
- removes the second worker from the
working list
......
btrfs_stop_workers returns, aworker is on the idle list
FS is umounted, memory is freed
......
aworker is waken up, fireworks ensue
With this applied, I wasn't able to trigger the problem in 48 hours,
whereas previously I could reliably reproduce at least one of these
races within an hour.
Reported-by: David Sterba <dsterba@suse.cz> Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Liu Bo [Tue, 1 Oct 2013 15:49:49 +0000 (23:49 +0800)]
Btrfs: fix crash of compressed writes
The crash[1] is found by xfstests/generic/208 with "-o compress",
it's not reproduced everytime, but it does panic.
The bug is quite interesting, it's actually introduced by a recent commit
(573aecafca1cf7a974231b759197a1aebcf39c2a,
Btrfs: actually limit the size of delalloc range).
Btrfs implements delay allocation, so during writeback, we
(1) get a page A and lock it
(2) search the state tree for delalloc bytes and lock all pages within the range
(3) process the delalloc range, including find disk space and create
ordered extent and so on.
(4) submit the page A.
It runs well in normal cases, but if we're in a racy case, eg.
buffered compressed writes and aio-dio writes,
sometimes we may fail to lock all pages in the 'delalloc' range,
in which case, we need to fall back to search the state tree again with
a smaller range limit(max_bytes = PAGE_CACHE_SIZE - offset).
The mentioned commit has a side effect, that is, in the fallback case,
we can find delalloc bytes before the index of the page we already have locked,
so we're in the case of (delalloc_end <= *start) and return with (found > 0).
This ends with not locking delalloc pages but making ->writepage still
process them, and the crash happens.
This fixes it by just thinking that we find nothing and returning to caller
as the caller knows how to deal with it properly.
Josef Bacik [Mon, 30 Sep 2013 18:10:43 +0000 (14:10 -0400)]
Btrfs: fix transid verify errors when recovering log tree
If we crash with a log, remount and recover that log, and then crash before we
can commit another transaction we will get transid verify errors on the next
mount. This is because we were not zero'ing out the log when we committed the
transaction after recovery. This is ok as long as we commit another transaction
at some point in the future, but if you abort or something else goes wrong you
can end up in this weird state because the recovery stuff says that the tree log
should have a generation+1 of the super generation, which won't be the case of
the transaction that was started for recovery. Fix this by removing the check
and _always_ zero out the log portion of the super when we commit a transaction.
This fixes the transid verify issues I was seeing with my force errors tests.
Thanks,
Linus Torvalds [Fri, 4 Oct 2013 19:54:11 +0000 (12:54 -0700)]
selinux: remove 'flags' parameter from inode_has_perm
Every single user passes in '0'. I think we had non-zero users back in
some stone age when selinux_inode_permission() was implemented in terms
of inode_has_perm(), but that complicated case got split up into a
totally separate code-path so that we could optimize the much simpler
special cases.
See commit 2e33405785d3 ("SELinux: delay initialization of audit data in
selinux_inode_permission") for example.
projects / linux-beck.git / log