When BLKZEROOUT ioctl and data read race, the data read leaves stale
page cache. To avoid the stale page cache, hold invalidate_lock of the
block device file mapping. The stale page cache is observed when
blktests test case block/009 is modified to call "blkdiscard -z" command
and repeated hundreds of times.
This patch can be applied back to the stable kernel version v5.15.y.
Rework is required for older stable kernels.
Fixes: 22dd6d3566 ("block: invalidate the page cache when issuing BLKZEROOUT")
Signed-off-by: Shin'ichiro Kawasaki <[email protected]>
Cc: [email protected] # v5.15
Reviewed-by: Jan Kara <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
When BLKDISCARD ioctl and data read race, the data read leaves stale
page cache. To avoid the stale page cache, hold invalidate_lock of the
block device file mapping. The stale page cache is observed when
blktests test case block/009 is repeated hundreds of times.
This patch can be applied back to the stable kernel version v5.15.y
with slight patch edit. Rework is required for older stable kernels.
Fixes: 351499a172 ("block: Invalidate cache on discard v2")
Signed-off-by: Shin'ichiro Kawasaki <[email protected]>
Cc: [email protected] # v5.15
Reviewed-by: Jan Kara <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
The switch to go through blkdev_get_by_dev means we now ignore the
return value from bdev_disk_changed in __blkdev_get. Add a manual
check to restore the old semantics.
Fixes: 4601b4b130 ("block: reopen the device in blkdev_reread_part")
Reported-by: Karel Zak <[email protected]>
Signed-off-by: Christoph Hellwig <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>
Historically the BLKRRPART ioctls called into the now defunct ->revalidate
method, which caused the sd driver to check if any media is present.
When the ->revalidate method was removed this revalidation was lost,
leading to lots of I/O errors when using the eject command. Fix this by
reopening the device to rescan the partitions, and thus calling the
revalidation logic in the sd driver.
Fixes: 471bd0af54 ("sd: use bdev_check_media_change")
Reported--by: Tom Seewald <[email protected]>
Signed-off-by: Christoph Hellwig <[email protected]>
Tested-by: Tom Seewald <[email protected]>
Reviewed-by: Ming Lei <[email protected]>
Reviewed-by: Minwoo Im <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Move the policy field to struct block_device and rename it to the
more descriptive bd_read_only. Also turn the field into a bool as it
is used as such.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Switch the block device lookup interfaces to directly work with a dev_t
so that struct block_device references are only acquired by the
blkdev_get variants (and the blk-cgroup special case). This means that
we now don't need an extra reference in the inode and can generally
simplify handling of struct block_device to keep the lookups contained
in the core block layer code.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Acked-by: Tejun Heo <[email protected]>
Acked-by: Coly Li <[email protected]> [bcache]
Signed-off-by: Jens Axboe <[email protected]>
Now that all drivers that want to hook into setting or clearing the
read-only flag use the set_read_only method, this code can be removed.
Signed-off-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Add a new method to allow for driver-specific processing when setting or
clearing the block device read-only state. This allows to replace the
cumbersome and error-prone override of the whole ioctl implementation.
Signed-off-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
BLKFLSBUF is entirely contained in the block core, and there is no
good reason to give the driver a hook into processing it.
Signed-off-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Add a littler helper to make the somewhat arcane bd_contains checks a
little more obvious.
Signed-off-by: Christoph Hellwig <[email protected]>
Acked-by: Ulf Hansson <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Discarding blocks and buffers under a mounted filesystem is hardly
anything admin wants to do. Usually it will confuse the filesystem and
sometimes the loss of buffer_head state (including b_private field) can
even cause crashes like:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000008
PGD 0 P4D 0
Oops: 0002 [#1] SMP PTI
CPU: 4 PID: 203778 Comm: jbd2/dm-3-8 Kdump: loaded Tainted: G O --------- - - 4.18.0-147.5.0.5.h126.eulerosv2r9.x86_64 #1
Hardware name: Huawei RH2288H V3/BC11HGSA0, BIOS 1.57 08/11/2015
RIP: 0010:jbd2_journal_grab_journal_head+0x1b/0x40 [jbd2]
...
Call Trace:
__jbd2_journal_insert_checkpoint+0x23/0x70 [jbd2]
jbd2_journal_commit_transaction+0x155f/0x1b60 [jbd2]
kjournald2+0xbd/0x270 [jbd2]
So if we don't have block device open with O_EXCL already, claim the
block device while we truncate buffer cache. This makes sure any
exclusive block device user (such as filesystem) cannot operate on the
device while we are discarding buffer cache.
Reported-by: Ye Bin <[email protected]>
Signed-off-by: Jan Kara <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
[axboe: fix !CONFIG_BLOCK error in truncate_bdev_range()]
Signed-off-by: Jens Axboe <[email protected]>
Split each sub-command out into a separate helper, and move those helpers
to block/partitions/core.c instead of having a lot of partition
manipulation logic open coded in block/ioctl.c.
Signed-off-by: Christoph Hellwig <[email protected]
Signed-off-by: Jens Axboe <[email protected]>
Having separate implementations of blkdev_ioctl() often leads to these
getting out of sync, despite the comment at the top.
Since most of the ioctl commands are compatible, and we try very hard
not to add any new incompatible ones, move all the common bits into a
shared function and leave only the ones that are historically different
in separate functions for native/compat mode.
To deal with the compat_ptr() conversion, pass both the integer
argument and the pointer argument into the new blkdev_common_ioctl()
and make sure to always use the correct one of these.
blkdev_ioctl() is now only kept as a separate exported interfact
for drivers/char/raw.c, which lacks a compat_ioctl variant.
We should probably either move raw.c to staging if there are no
more users, or export blkdev_compat_ioctl() as well.
Reviewed-by: Ben Hutchings <[email protected]>
Signed-off-by: Arnd Bergmann <[email protected]>
There is no need to go through a compat_alloc_user_space()
copy any more, just wrap the function in a small helper that
works the same way for native and compat mode.
Reviewed-by: Ben Hutchings <[email protected]>
Signed-off-by: Arnd Bergmann <[email protected]>
Having both in the same file allows a number of simplifications
to the compat path, and makes it more likely that changes to
the native path get applied to the compat version as well.
Reviewed-by: Ben Hutchings <[email protected]>
Signed-off-by: Arnd Bergmann <[email protected]>
A lot of block drivers need only a trivial .compat_ioctl callback.
Add a helper function that can be set as the callback pointer
to only convert the argument using the compat_ptr() conversion
and otherwise assume all input and output data is compatible,
or handled using in_compat_syscall() checks.
This mirrors the compat_ptr_ioctl() helper function used in
character devices.
Reviewed-by: Ben Hutchings <[email protected]>
Signed-off-by: Arnd Bergmann <[email protected]>
Simplify the arguments to blkdev_nr_zones by passing a gendisk instead
of the block_device and capacity. This also removes the need for
__blkdev_nr_zones as all callers are outside the fast path and can
deal with the additional branch.
Signed-off-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
In general drivers should never mess with partition tables directly.
Unfortunately s390 and loop do for somewhat historic reasons, but they
can use bdev_disk_changed directly instead when we export it as they
satisfy the sanity checks we have in __blkdev_reread_part.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Stefan Haberland <[email protected]> [dasd]
Reviewed-by: Jan Kara <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
We still have to set the capacity to 0 if invalidating or call
revalidate_disk if not even if the disk has no partitions. Fix
that by merging rescan_partitions into bdev_disk_changed and just
stubbing out blk_add_partitions and blk_drop_partitions for
non-partitioned devices.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
A lot of the logic in invalidate_partitions and rescan_partitions is
shared. Merge the two functions to simplify things. There is a small
behavior change in that we now send the kevent change notice also if we
were not invalidating but no partitions were found, which seems like
the right thing to do.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Introduce three new ioctl commands BLKOPENZONE, BLKCLOSEZONE and
BLKFINISHZONE to allow applications to control the condition of zones
on a zoned block device through the execution of the REQ_OP_ZONE_OPEN,
REQ_OP_ZONE_CLOSE and REQ_OP_ZONE_FINISH operations.
Contains contributions from Matias Bjorling, Hans Holmberg,
Dmitry Fomichev, Keith Busch, Damien Le Moal and Christoph Hellwig.
Reviewed-by: Javier González <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Ajay Joshi <[email protected]>
Signed-off-by: Matias Bjorling <[email protected]>
Signed-off-by: Hans Holmberg <[email protected]>
Signed-off-by: Dmitry Fomichev <[email protected]>
Signed-off-by: Keith Busch <[email protected]>
Signed-off-by: Damien Le Moal <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Various block layer files do not have any licensing information at all.
Add SPDX tags for the default kernel GPLv2 license to those.
Reviewed-by: Chaitanya Kulkarni <[email protected]>
Signed-off-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Get a zoned block device total number of zones. The device can be a
partition of the whole device. The number of zones is always 0 for
regular block devices.
Reviewed-by: Hannes Reinecke <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Damien Le Moal <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Get a zoned block device zone size in number of 512 B sectors.
The zone size is always 0 for regular block devices.
Reviewed-by: Hannes Reinecke <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Damien Le Moal <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
The 'lend' parameter of truncate_inode_pages_range is required to be
inclusive, so follow the rule.
This patch fixes one memory corruption triggered by discard.
Cc: <[email protected]>
Cc: Dmitry Monakhov <[email protected]>
Fixes: 351499a172 ("block: Invalidate cache on discard v2")
Reviewed-by: Bart Van Assche <[email protected]>
Signed-off-by: Ming Lei <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Check for CAP_SYS_ADMIN before calling into the driver, similar to
blkdev_flushbuf(). This is safer and can spare a check in the driver.
(Currently BLKROSET is overridden by md and rbd, rbd is missing the
check. md has the check, but it covers a lot more than BLKROSET.)
Acked-by: Al Viro <[email protected]>
Signed-off-by: Ilya Dryomov <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
It is reasonable drop page cache on discard, otherwise that pages may
be written by writeback second later, so thin provision devices will
not be happy. This seems to be a security leak in case of secure discard case.
Also add check for queue_discard flag on early stage.
Signed-off-by: Dmitry Monakhov <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Now that we use the proper REQ_OP_WRITE_ZEROES operation everywhere we can
kill this hack.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Martin K. Petersen <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Turn the existing discard flag into a new BLKDEV_ZERO_UNMAP flag with
similar semantics, but without referring to diѕcard.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Martin K. Petersen <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
blk_get_backing_dev_info() is now a simple dereference. Remove that
function and simplify some code around that.
Signed-off-by: Jan Kara <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
This was entirely automated, using the script by Al:
PATT='^[[:blank:]]*#[[:blank:]]*include[[:blank:]]*<asm/uaccess.h>'
sed -i -e "s!$PATT!#include <linux/uaccess.h>!" \
$(git grep -l "$PATT"|grep -v ^include/linux/uaccess.h)
to do the replacement at the end of the merge window.
Requested-by: Al Viro <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Partitions that are not aligned to the blocksize of a device may cause
invalid I/O requests because the blocklayer cares only about alignment
within the partition when building requests on partitions.
device
|--------4096--------|--------4096--------|--------4096--------|
partition offset 512byte
|-512-|--------4096--------|--------4096--------|--------4096--------|
When reading/writing one 4k block of the partition this maps to
reading/writing with an offset of 512 byte of the device leading to
unaligned requests for the device which in turn may cause unexpected
behavior of the device driver.
For DASD devices we have to translate the block number into a cylinder,
head, record format. The unaligned requests lead to wrong calculation
and therefore to misdirected I/O. In a "good" case this leads to I/O
errors because the underlying hardware detects the wrong addressing.
In a worst case scenario this might destroy data on the device.
To prevent partitions that are not aligned to the physical blocksize
of a device check for the alignment in the blkpg_ioctl.
Signed-off-by: Stefan Haberland <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Adds the new BLKREPORTZONE and BLKRESETZONE ioctls for respectively
obtaining the zone configuration of a zoned block device and resetting
the write pointer of sequential zones of a zoned block device.
The BLKREPORTZONE ioctl maps directly to a single call of the function
blkdev_report_zones. The zone information result is passed as an array
of struct blk_zone identical to the structure used internally for
processing the REQ_OP_ZONE_REPORT operation. The BLKRESETZONE ioctl
maps to a single call of the blkdev_reset_zones function.
Signed-off-by: Shaun Tancheff <[email protected]>
Signed-off-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Reviewed-by: Martin K. Petersen <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Patch series "fallocate for block devices", v11.
This is a patchset to fix page cache coherency with BLKZEROOUT and
implement fallocate for block devices.
The first patch is a fix to the existing BLKZEROOUT ioctl to invalidate
the page cache if the zeroing command to the underlying device succeeds.
Without this patch we still have the pagecache coherence bug that's been
in the kernel forever.
The second patch changes the internal block device functions to reject
attempts to discard or zeroout that are not aligned to the logical block
size. Previously, we only checked that the start/len parameters were
512-byte aligned, which caused kernel BUG_ONs for unaligned IOs to 4k-LBA
devices.
The third patch creates an fallocate handler for block devices, wires up
the FALLOC_FL_PUNCH_HOLE flag to zeroing-discard, and connects
FALLOC_FL_ZERO_RANGE to write-same so that we can have a consistent
fallocate interface between files and block devices. It also allows the
combination of PUNCH_HOLE and NO_HIDE_STALE to invoke non-zeroing discard.
Test cases for the new block device fallocate are now in xfstests as
generic/349-351.
This patch (of 3):
Invalidate the page cache (as a regular O_DIRECT write would do) to avoid
returning stale cache contents at a later time.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Darrick J. Wong <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Reviewed-by: Martin K. Petersen <[email protected]>
Reviewed-by: Bart Van Assche <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Cc: Theodore Ts'o <[email protected]>
Cc: Mike Snitzer <[email protected]>
Cc: Brian Foster <[email protected]>
Cc: Jens Axboe <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Pull misc DAX updates from Vishal Verma:
"DAX error handling for 4.7
- Until now, dax has been disabled if media errors were found on any
device. This enables the use of DAX in the presence of these
errors by making all sector-aligned zeroing go through the driver.
- The driver (already) has the ability to clear errors on writes that
are sent through the block layer using 'DSMs' defined in ACPI 6.1.
Other misc changes:
- When mounting DAX filesystems, check to make sure the partition is
page aligned. This is a requirement for DAX, and previously, we
allowed such unaligned mounts to succeed, but subsequent
reads/writes would fail.
- Misc/cleanup fixes from Jan that remove unused code from DAX
related to zeroing, writeback, and some size checks"
* tag 'dax-misc-for-4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
dax: fix a comment in dax_zero_page_range and dax_truncate_page
dax: for truncate/hole-punch, do zeroing through the driver if possible
dax: export a low-level __dax_zero_page_range helper
dax: use sb_issue_zerout instead of calling dax_clear_sectors
dax: enable dax in the presence of known media errors (badblocks)
dax: fallback from pmd to pte on error
block: Update blkdev_dax_capable() for consistency
xfs: Add alignment check for DAX mount
ext2: Add alignment check for DAX mount
ext4: Add alignment check for DAX mount
block: Add bdev_dax_supported() for dax mount checks
block: Add vfs_msg() interface
dax: Remove redundant inode size checks
dax: Remove pointless writeback from dax_do_io()
dax: Remove zeroing from dax_io()
dax: Remove dead zeroing code from fault handlers
ext2: Avoid DAX zeroing to corrupt data
ext2: Fix block zeroing in ext2_get_blocks() for DAX
dax: Remove complete_unwritten argument
DAX: move RADIX_DAX_ definitions to dax.c
Shin'ichiro Kawasaki