| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Pull iomap updates from Christian Brauner:
"This contains some minor work for the iomap subsystem:
- Add documentation on the design of iomap and how to port to it
- Optimize iomap_read_folio()
- Bring back the change to iomap_write_end() to no increase i_size.
This is accompanied by a change to xfs to reserve blocks for
truncating large realtime inodes to avoid exposing stale data when
iomap_write_end() stops increasing i_size"
* tag 'vfs-6.11.iomap' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
iomap: don't increase i_size in iomap_write_end()
xfs: reserve blocks for truncating large realtime inode
Documentation: the design of iomap and how to port
iomap: Optimize iomap_read_folio
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Pull PG_error removal updates from Christian Brauner:
"This contains work to remove almost all remaining users of PG_error
from filesystems and filesystem helper libraries. An additional patch
will be coming in via the jfs tree which tests the PG_error bit.
Afterwards nothing will be testing it anymore and it's safe to remove
all places which set or clear the PG_error bit.
The goal is to fully remove PG_error by the next merge window"
* tag 'vfs-6.11.pg_error' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
buffer: Remove calls to set and clear the folio error flag
iomap: Remove calls to set and clear folio error flag
vboxsf: Convert vboxsf_read_folio() to use a folio
ufs: Remove call to set the folio error flag
romfs: Convert romfs_read_folio() to use a folio
reiserfs: Remove call to folio_set_error()
orangefs: Remove calls to set/clear the error flag
nfs: Remove calls to folio_set_error
jffs2: Remove calls to set/clear the folio error flag
hostfs: Convert hostfs_read_folio() to use a folio
isofs: Convert rock_ridge_symlink_read_folio to use a folio
hpfs: Convert hpfs_symlink_read_folio to use a folio
efs: Convert efs_symlink_read_folio to use a folio
cramfs: Convert cramfs_read_folio to use a folio
coda: Convert coda_symlink_filler() to use folio_end_read()
befs: Convert befs_symlink_read_folio() to use folio_end_read()
|
|
This reverts commit '0841ea4a3b41 ("iomap: keep on increasing i_size in
iomap_write_end()")'.
After xfs could zero out the tail blocks aligned to the allocation
unitsize and convert the tail blocks to unwritten for realtime inode on
truncate down, it couldn't expose any stale data when unaligned truncate
down realtime inodes, so we could keep on keeping i_size for
IOMAP_UNSHARE and IOMAP_ZERO in iomap_write_end().
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://lore.kernel.org/r/20240618142112.1315279-3-yi.zhang@huaweicloud.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
iomap_readpage_iter() handles "uptodate blocks" and "not uptodate blocks"
within a folio separately. This makes iomap_read_folio() to call into
->iomap_begin() to request for extent mapping even though it might already
have an extent which is not fully processed.
This happens when we either have a large folio or with bs < ps. In these
cases we can have sub blocks which can be uptodate (say for e.g. due to
previous writes). With iomap_read_folio_iter(), this is handled more
efficiently by not calling ->iomap_begin() call until all the sub blocks
with the current folio are processed.
iomap_read_folio_iter() handles multiple sub blocks within a given
folio but it's implementation logic is similar to how
iomap_readahead_iter() handles multiple folios within a single mapped
extent. Both of them iterate over a given range of folio/mapped extent
and call iomap_readpage_iter() for reading.
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Link: https://lore.kernel.org/r/92ae9f3333c9a7e66214568d08f45664261c899c.1715067055.git.ritesh.list@gmail.com
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
cc: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
If the extent spans the block that contains i_size, we need to handle
both halves separately so that we properly zero data in the page cache
for blocks that are entirely outside of i_size. But this is needed only
when i_size is within the current folio under processing.
"orig_pos + length > isize" can be true for all folios if the mapped
extent length is greater than the folio size. That is making plen to
break for every folio instead of only the last folio.
So use orig_plen for checking if "orig_pos + orig_plen > isize".
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Link: https://lore.kernel.org/r/a32e5f9a4fcfdb99077300c4020ed7ae61d6e0f9.1715067055.git.ritesh.list@gmail.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
cc: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Commit '943bc0882ceb ("iomap: don't increase i_size if it's not a write
operation")' breaks xfs with realtime device on generic/561, the problem
is when unaligned truncate down a xfs realtime inode with rtextsize > 1
fs block, xfs only zero out the EOF block but doesn't zero out the tail
blocks that aligned to rtextsize, so if we don't increase i_size in
iomap_write_end(), it could expose stale data after we do an append
write beyond the aligned EOF block.
xfs should zero out the tail blocks when truncate down, but before we
finish that, let's fix the issue by just revert the changes in
iomap_write_end().
Fixes: 943bc0882ceb ("iomap: don't increase i_size if it's not a write operation")
Reported-by: Chandan Babu R <chandanbabu@kernel.org>
Link: https://lore.kernel.org/linux-xfs/0b92a215-9d9b-3788-4504-a520778953c2@huaweicloud.com
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://lore.kernel.org/r/20240603112222.2109341-1-yi.zhang@huaweicloud.com
Tested-by: Chandan Babu R <chandanbabu@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
The folio error flag is not checked anywhere, so we can remove the calls
to set and clear it.
Cc: Christian Brauner <brauner@kernel.org>
Cc: linux-xfs@vger.kernel.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Link: https://lore.kernel.org/r/20240530202110.2653630-16-willy@infradead.org
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Since commit (5d8edfb900d5 "iomap: Copy larger chunks from userspace"),
iomap will try to copy in larger chunks than PAGE_SIZE. However, if the
mapping doesn't support large folio, only one page of maximum 4KB will
be created and 4KB data will be writen to pagecache each time. Then,
next 4KB will be handled in next iteration. This will cause potential
write performance problem.
If chunk is 2MB, total 512 pages need to be handled finally. During this
period, fault_in_iov_iter_readable() is called to check iov_iter readable
validity. Since only 4KB will be handled each time, below address space
will be checked over and over again:
start end
-
buf, buf+2MB
buf+4KB, buf+2MB
buf+8KB, buf+2MB
...
buf+2044KB buf+2MB
Obviously the checking size is wrong since only 4KB will be handled each
time. So this will get a correct chunk to let iomap work well in non-large
folio case.
With this change, the write speed will be stable. Tested on ARM64 device.
Before:
- dd if=/dev/zero of=/dev/sda bs=400K count=10485 (334 MB/s)
- dd if=/dev/zero of=/dev/sda bs=800K count=5242 (278 MB/s)
- dd if=/dev/zero of=/dev/sda bs=1600K count=2621 (204 MB/s)
- dd if=/dev/zero of=/dev/sda bs=2200K count=1906 (170 MB/s)
- dd if=/dev/zero of=/dev/sda bs=3000K count=1398 (150 MB/s)
- dd if=/dev/zero of=/dev/sda bs=4500K count=932 (139 MB/s)
After:
- dd if=/dev/zero of=/dev/sda bs=400K count=10485 (339 MB/s)
- dd if=/dev/zero of=/dev/sda bs=800K count=5242 (330 MB/s)
- dd if=/dev/zero of=/dev/sda bs=1600K count=2621 (332 MB/s)
- dd if=/dev/zero of=/dev/sda bs=2200K count=1906 (333 MB/s)
- dd if=/dev/zero of=/dev/sda bs=3000K count=1398 (333 MB/s)
- dd if=/dev/zero of=/dev/sda bs=4500K count=932 (333 MB/s)
Fixes: 5d8edfb900d5 ("iomap: Copy larger chunks from userspace")
Cc: stable@vger.kernel.org
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Xu Yang <xu.yang_2@nxp.com>
Link: https://lore.kernel.org/r/20240521114939.2541461-2-xu.yang_2@nxp.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild
Pull Kbuild updates from Masahiro Yamada:
- Avoid 'constexpr', which is a keyword in C23
- Allow 'dtbs_check' and 'dt_compatible_check' run independently of
'dt_binding_check'
- Fix weak references to avoid GOT entries in position-independent code
generation
- Convert the last use of 'optional' property in arch/sh/Kconfig
- Remove support for the 'optional' property in Kconfig
- Remove support for Clang's ThinLTO caching, which does not work with
the .incbin directive
- Change the semantics of $(src) so it always points to the source
directory, which fixes Makefile inconsistencies between upstream and
downstream
- Fix 'make tar-pkg' for RISC-V to produce a consistent package
- Provide reasonable default coverage for objtool, sanitizers, and
profilers
- Remove redundant OBJECT_FILES_NON_STANDARD, KASAN_SANITIZE, etc.
- Remove the last use of tristate choice in drivers/rapidio/Kconfig
- Various cleanups and fixes in Kconfig
* tag 'kbuild-v6.10' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild: (46 commits)
kconfig: use sym_get_choice_menu() in sym_check_prop()
rapidio: remove choice for enumeration
kconfig: lxdialog: remove initialization with A_NORMAL
kconfig: m/nconf: merge two item_add_str() calls
kconfig: m/nconf: remove dead code to display value of bool choice
kconfig: m/nconf: remove dead code to display children of choice members
kconfig: gconf: show checkbox for choice correctly
kbuild: use GCOV_PROFILE and KCSAN_SANITIZE in scripts/Makefile.modfinal
Makefile: remove redundant tool coverage variables
kbuild: provide reasonable defaults for tool coverage
modules: Drop the .export_symbol section from the final modules
kconfig: use menu_list_for_each_sym() in sym_check_choice_deps()
kconfig: use sym_get_choice_menu() in conf_write_defconfig()
kconfig: add sym_get_choice_menu() helper
kconfig: turn defaults and additional prompt for choice members into error
kconfig: turn missing prompt for choice members into error
kconfig: turn conf_choice() into void function
kconfig: use linked list in sym_set_changed()
kconfig: gconf: use MENU_CHANGED instead of SYMBOL_CHANGED
kconfig: gconf: remove debug code
...
|
|
Kbuild conventionally uses $(obj)/ for generated files, and $(src)/ for
checked-in source files. It is merely a convention without any functional
difference. In fact, $(obj) and $(src) are exactly the same, as defined
in scripts/Makefile.build:
src := $(obj)
When the kernel is built in a separate output directory, $(src) does
not accurately reflect the source directory location. While Kbuild
resolves this discrepancy by specifying VPATH=$(srctree) to search for
source files, it does not cover all cases. For example, when adding a
header search path for local headers, -I$(srctree)/$(src) is typically
passed to the compiler.
This introduces inconsistency between upstream and downstream Makefiles
because $(src) is used instead of $(srctree)/$(src) for the latter.
To address this inconsistency, this commit changes the semantics of
$(src) so that it always points to the directory in the source tree.
Going forward, the variables used in Makefiles will have the following
meanings:
$(obj) - directory in the object tree
$(src) - directory in the source tree (changed by this commit)
$(objtree) - the top of the kernel object tree
$(srctree) - the top of the kernel source tree
Consequently, $(srctree)/$(src) in upstream Makefiles need to be replaced
with $(src).
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>
|
|
Since iomap_write_end() can never return a partial write length, the
comparison between written, copied and bytes becomes useless, just
merge them with the unwritten branch.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://lore.kernel.org/r/20240320110548.2200662-10-yi.zhang@huaweicloud.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
For now, we can make sure iomap_write_end() always return 0 or copied
bytes, so instead of return written bytes, convert to return a boolean
to indicate the copied bytes have been written to the pagecache.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://lore.kernel.org/r/20240320110548.2200662-9-yi.zhang@huaweicloud.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
In iomap_write_iter(), the status variable used to receive the return
value from iomap_write_end() is confusing, replace it with a new written
variable to represent the written bytes in each cycle, no logic changes.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://lore.kernel.org/r/20240320110548.2200662-8-yi.zhang@huaweicloud.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Increase i_size in iomap_zero_range() and iomap_unshare_iter() is not
needed, the caller should handle it. Especially, when truncate partial
block, we should not increase i_size beyond the new EOF here. It doesn't
affect xfs and gfs2 now because they set the new file size after zero
out, it doesn't matter that a transient increase in i_size, but it will
affect ext4 because it set file size before truncate. So move the i_size
updating logic to iomap_write_iter().
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://lore.kernel.org/r/20240320110548.2200662-7-yi.zhang@huaweicloud.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Unsharing and zeroing can only happen within EOF, so there is never a
need to perform posteof pagecache truncation if write begin fails, also
partial write could never theoretically happened from iomap_write_end(),
so remove both of them.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://lore.kernel.org/r/20240320110548.2200662-6-yi.zhang@huaweicloud.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
This removes one indirect function call per folio, and adds type safety
by not casting through a void pointer.
Based on a patch by Matthew Wilcox.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20240412061614.1511629-1-hch@lst.de
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
gitolite.kernel.org:pub/scm/linux/kernel/git/vfs/vfs
Pull write hint fix from Christian Brauner:
UFS devices are widely used in mobile applications, e.g. in smartphones.
UFS vendors need data lifetime information to achieve good performance.
Providing data lifetime information to UFS devices can result in up to
40% lower write amplification. Hence this patch series that restores the
bi_write_hint member in struct bio. After this patch series has been
merged, patches that implement data lifetime support in the SCSI disk
(sd) driver will be sent to the Linux kernel SCSI maintainer.
The following changes are included in this patch series:
- Improvements for the F_GET_RW_HINT and F_SET_RW_HINT fcntls.
- Move enum rw_hint into a new header file.
- Support F_SET_RW_HINT for block devices to make it easy to test data
lifetime support.
- Restore the bio.bi_write_hint member and restore support in the VFS
layer and also in the block layer for data lifetime information.
The shell script that has been used to test the patch series combined
with the SCSI patches is available at the end of this cover letter.
* tag 'vfs-6.9.rw_hint' of gitolite.kernel.org:pub/scm/linux/kernel/git/vfs/vfs:
block, fs: Restore the per-bio/request data lifetime fields
fs: Propagate write hints to the struct block_device inode
fs: Move enum rw_hint into a new header file
fs: Split fcntl_rw_hint()
fs: Verify write lifetime constants at compile time
fs: Fix rw_hint validation
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
processed: The number of bytes processed by the body in the
most recent iteration, or a negative errno. 0 causes the iteration to
stop.
The processed is useful to check when the loop breaks.
Signed-off-by: Kassey Li <quic_yingangl@quicinc.com>
Link: https://lore.kernel.org/r/20240219021138.3481763-1-quic_yingangl@quicinc.com
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Since commit fd07e0aa23c4 ("iomap: map multiple blocks at a time"), we
could map multi-blocks once a time, and the dirty_len indicates the
expected map length, map_len won't large than it. The pos and dirty_len
means the dirty range that should be mapped to write, add them into
trace_iomap_writepage_map() could be more useful for debug.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://lore.kernel.org/r/20240220115759.3445025-1-yi.zhang@huaweicloud.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Restore support for passing data lifetime information from filesystems to
block drivers. This patch reverts commit b179c98f7697 ("block: Remove
request.write_hint") and commit c75e707fe1aa ("block: remove the
per-bio/request write hint").
This patch does not modify the size of struct bio because the new
bi_write_hint member fills a hole in struct bio. pahole reports the
following for struct bio on an x86_64 system with this patch applied:
/* size: 112, cachelines: 2, members: 20 */
/* sum members: 110, holes: 1, sum holes: 2 */
/* last cacheline: 48 bytes */
Reviewed-by: Kanchan Joshi <joshi.k@samsung.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20240202203926.2478590-7-bvanassche@acm.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Let the file system know how much dirty data exists at the passed
in offset. This allows file systems to allocate the right amount
of space that actually is written back if they can't eagerly
convert (e.g. because they don't support unwritten extents).
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-15-hch@lst.de
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
The ->map_blocks interface returns a valid range for writeback, but we
still call back into it for every block, which is a bit inefficient.
Change iomap_writepage_map to use the valid range in the map until the
end of the folio or the dirty range inside the folio instead of calling
back into every block.
Note that the range is not used over folio boundaries as we need to be
able to check the mapping sequence count under the folio lock.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-14-hch@lst.de
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Currently the writeback code delays submitting fill ioends until we
reach the end of the folio. The reason for that is that otherwise
the end I/O handler could clear the writeback bit before we've even
finished submitting all I/O for the folio.
Add a bias to ifs->write_bytes_pending while we are submitting I/O
for a folio so that it never reaches zero until all I/O is completed
to prevent the early writeback bit clearing, and remove the now
superfluous submit_list.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-13-hch@lst.de
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Split the loop body that calls into the file system to map a block and
add it to the ioend into a separate helper to prefer for refactoring of
the surrounding code.
Note that this was the only place in iomap_writepage_map that could
return an error, so include the call to ->discard_folio into the new
helper as that will help to avoid code duplication in the future.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-12-hch@lst.de
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Instead of clling mapping_set_error once per folio, only do that once
per bio, and consolidate all the writeback error handling code in
iomap_finish_ioend.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-11-hch@lst.de
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Back in the days when a single bio could only be filled to the hardware
limits, and we scheduled a work item for each bio completion, chaining
multiple bios for a single ioend made a lot of sense to reduce the number
of completions. But these days bios can be filled until we reach the
number of vectors or total size limit, which means we can always fit at
least 1 megabyte worth of data in the worst case, but usually a lot more
due to large folios. The only thing bio chaining is buying us now is
to reduce the size of the allocation from an ioend with an embedded bio
into a plain bio, which is a 52 bytes differences on 64-bit systems.
This is not worth the added complexity, so remove the bio chaining and
only use the bio embedded into the ioend. This will help to simplify
further changes to the iomap writeback code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-10-hch@lst.de
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
The calculation in iomap_sector is pretty trivial and most of the time
iomap_add_to_ioend only callers either iomap_can_add_to_ioend or
iomap_alloc_ioend from a single invocation.
Calculate the sector in the two lower level functions and stop passing it
from iomap_add_to_ioend and update the iomap_alloc_ioend argument passing
order to match that of iomap_add_to_ioend.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-9-hch@lst.de
Reviewed-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Switch to the same argument order as iomap_writepage_map and remove the
ifs argument that can be trivially recalculated.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-8-hch@lst.de
Reviewed-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Move the tracepoint and the iomap check from iomap_do_writepage into
iomap_writepage_map. This keeps all logic in one places, and leaves
iomap_do_writepage just as the wrapper for the callback conventions of
write_cache_pages, which will go away when that is converted to an
iterator.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-7-hch@lst.de
Reviewed-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Most of iomap_do_writepage is dedidcated to handling a folio crossing or
beyond i_size. Split this is into a separate helper and update the
commens to deal with folios instead of pages and make them more readable.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-6-hch@lst.de
Reviewed-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
The iomap writepage implementation has been removed in commit
478af190cb6c ("iomap: remove iomap_writepage") and this code is now only
called through ->writepages which never happens from memory reclaim.
Nove the check from iomap_do_writepage to iomap_writepages so that is
only called once per ->writepage invocation.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-5-hch@lst.de
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
The io_folios member in struct iomap_ioend counts the number of folios
added to an ioend. It is only used at submission time and can thus be
moved to iomap_writepage_ctx instead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-4-hch@lst.de
Reviewed-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
iomap_writepage_map aready warns about inline data, but then just ignores
it. Treat it as an error and return -EIO.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-3-hch@lst.de
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
write_cache_pages always clear the page dirty bit before calling into the
file systems, and leaves folios with a writeback failure without the
dirty bit after return. We also clear the per-block writeback bits for
writeback failures unless no I/O has submitted, which will leave the
folio in an inconsistent state where it doesn't have the folio dirty,
but one or more per-block dirty bits. This seems to be due the place
where the iomap_clear_range_dirty call was inserted into the existing
not very clearly structured code when adding per-block dirty bit support
and not actually intentional. Switch to always clearing the dirty on
writeback failure.
Fixes: 4ce02c679722 ("iomap: Add per-block dirty state tracking to improve performance")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20231207072710.176093-2-hch@lst.de
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
The iomap code was limited to PAGE_SIZE bytes; generalise it to cover
an arbitrary-sized folio, and move it to be a common helper.
[akpm@linux-foundation.org: fix folio_fill_tail(), per Andreas Gruenbacher]
Link: https://lkml.kernel.org/r/20231107212643.3490372-3-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Andreas Gruenbacher <agruenba@redhat.com>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Darrick J. Wong <djwong@kernel.org>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull MM updates from Andrew Morton:
"Many singleton patches against the MM code. The patch series which are
included in this merge do the following:
- Kemeng Shi has contributed some compation maintenance work in the
series 'Fixes and cleanups to compaction'
- Joel Fernandes has a patchset ('Optimize mremap during mutual
alignment within PMD') which fixes an obscure issue with mremap()'s
pagetable handling during a subsequent exec(), based upon an
implementation which Linus suggested
- More DAMON/DAMOS maintenance and feature work from SeongJae Park i
the following patch series:
mm/damon: misc fixups for documents, comments and its tracepoint
mm/damon: add a tracepoint for damos apply target regions
mm/damon: provide pseudo-moving sum based access rate
mm/damon: implement DAMOS apply intervals
mm/damon/core-test: Fix memory leaks in core-test
mm/damon/sysfs-schemes: Do DAMOS tried regions update for only one apply interval
- In the series 'Do not try to access unaccepted memory' Adrian
Hunter provides some fixups for the recently-added 'unaccepted
memory' feature. To increase the feature's checking coverage. 'Plug
a few gaps where RAM is exposed without checking if it is
unaccepted memory'
- In the series 'cleanups for lockless slab shrink' Qi Zheng has done
some maintenance work which is preparation for the lockless slab
shrinking code
- Qi Zheng has redone the earlier (and reverted) attempt to make slab
shrinking lockless in the series 'use refcount+RCU method to
implement lockless slab shrink'
- David Hildenbrand contributes some maintenance work for the rmap
code in the series 'Anon rmap cleanups'
- Kefeng Wang does more folio conversions and some maintenance work
in the migration code. Series 'mm: migrate: more folio conversion
and unification'
- Matthew Wilcox has fixed an issue in the buffer_head code which was
causing long stalls under some heavy memory/IO loads. Some cleanups
were added on the way. Series 'Add and use bdev_getblk()'
- In the series 'Use nth_page() in place of direct struct page
manipulation' Zi Yan has fixed a potential issue with the direct
manipulation of hugetlb page frames
- In the series 'mm: hugetlb: Skip initialization of gigantic tail
struct pages if freed by HVO' has improved our handling of gigantic
pages in the hugetlb vmmemmep optimizaton code. This provides
significant boot time improvements when significant amounts of
gigantic pages are in use
- Matthew Wilcox has sent the series 'Small hugetlb cleanups' - code
rationalization and folio conversions in the hugetlb code
- Yin Fengwei has improved mlock()'s handling of large folios in the
series 'support large folio for mlock'
- In the series 'Expose swapcache stat for memcg v1' Liu Shixin has
added statistics for memcg v1 users which are available (and
useful) under memcg v2
- Florent Revest has enhanced the MDWE (Memory-Deny-Write-Executable)
prctl so that userspace may direct the kernel to not automatically
propagate the denial to child processes. The series is named 'MDWE
without inheritance'
- Kefeng Wang has provided the series 'mm: convert numa balancing
functions to use a folio' which does what it says
- In the series 'mm/ksm: add fork-exec support for prctl' Stefan
Roesch makes is possible for a process to propagate KSM treatment
across exec()
- Huang Ying has enhanced memory tiering's calculation of memory
distances. This is used to permit the dax/kmem driver to use 'high
bandwidth memory' in addition to Optane Data Center Persistent
Memory Modules (DCPMM). The series is named 'memory tiering:
calculate abstract distance based on ACPI HMAT'
- In the series 'Smart scanning mode for KSM' Stefan Roesch has
optimized KSM by teaching it to retain and use some historical
information from previous scans
- Yosry Ahmed has fixed some inconsistencies in memcg statistics in
the series 'mm: memcg: fix tracking of pending stats updates
values'
- In the series 'Implement IOCTL to get and optionally clear info
about PTEs' Peter Xu has added an ioctl to /proc/<pid>/pagemap
which permits us to atomically read-then-clear page softdirty
state. This is mainly used by CRIU
- Hugh Dickins contributed the series 'shmem,tmpfs: general
maintenance', a bunch of relatively minor maintenance tweaks to
this code
- Matthew Wilcox has increased the use of the VMA lock over
file-backed page faults in the series 'Handle more faults under the
VMA lock'. Some rationalizations of the fault path became possible
as a result
- In the series 'mm/rmap: convert page_move_anon_rmap() to
folio_move_anon_rmap()' David Hildenbrand has implemented some
cleanups and folio conversions
- In the series 'various improvements to the GUP interface' Lorenzo
Stoakes has simplified and improved the GUP interface with an eye
to providing groundwork for future improvements
- Andrey Konovalov has sent along the series 'kasan: assorted fixes
and improvements' which does those things
- Some page allocator maintenance work from Kemeng Shi in the series
'Two minor cleanups to break_down_buddy_pages'
- In thes series 'New selftest for mm' Breno Leitao has developed
another MM self test which tickles a race we had between madvise()
and page faults
- In the series 'Add folio_end_read' Matthew Wilcox provides cleanups
and an optimization to the core pagecache code
- Nhat Pham has added memcg accounting for hugetlb memory in the
series 'hugetlb memcg accounting'
- Cleanups and rationalizations to the pagemap code from Lorenzo
Stoakes, in the series 'Abstract vma_merge() and split_vma()'
- Audra Mitchell has fixed issues in the procfs page_owner code's new
timestamping feature which was causing some misbehaviours. In the
series 'Fix page_owner's use of free timestamps'
- Lorenzo Stoakes has fixed the handling of new mappings of sealed
files in the series 'permit write-sealed memfd read-only shared
mappings'
- Mike Kravetz has optimized the hugetlb vmemmap optimization in the
series 'Batch hugetlb vmemmap modification operations'
- Some buffer_head folio conversions and cleanups from Matthew Wilcox
in the series 'Finish the create_empty_buffers() transition'
- As a page allocator performance optimization Huang Ying has added
automatic tuning to the allocator's per-cpu-pages feature, in the
series 'mm: PCP high auto-tuning'
- Roman Gushchin has contributed the patchset 'mm: improve
performance of accounted kernel memory allocations' which improves
their performance by ~30% as measured by a micro-benchmark
- folio conversions from Kefeng Wang in the series 'mm: convert page
cpupid functions to folios'
- Some kmemleak fixups in Liu Shixin's series 'Some bugfix about
kmemleak'
- Qi Zheng has improved our handling of memoryless nodes by keeping
them off the allocation fallback list. This is done in the series
'handle memoryless nodes more appropriately'
- khugepaged conversions from Vishal Moola in the series 'Some
khugepaged folio conversions'"
[ bcachefs conflicts with the dynamically allocated shrinkers have been
resolved as per Stephen Rothwell in
https://lore.kernel.org/all/20230913093553.4290421e@canb.auug.org.au/
with help from Qi Zheng.
The clone3 test filtering conflict was half-arsed by yours truly ]
* tag 'mm-stable-2023-11-01-14-33' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (406 commits)
mm/damon/sysfs: update monitoring target regions for online input commit
mm/damon/sysfs: remove requested targets when online-commit inputs
selftests: add a sanity check for zswap
Documentation: maple_tree: fix word spelling error
mm/vmalloc: fix the unchecked dereference warning in vread_iter()
zswap: export compression failure stats
Documentation: ubsan: drop "the" from article title
mempolicy: migration attempt to match interleave nodes
mempolicy: mmap_lock is not needed while migrating folios
mempolicy: alloc_pages_mpol() for NUMA policy without vma
mm: add page_rmappable_folio() wrapper
mempolicy: remove confusing MPOL_MF_LAZY dead code
mempolicy: mpol_shared_policy_init() without pseudo-vma
mempolicy trivia: use pgoff_t in shared mempolicy tree
mempolicy trivia: slightly more consistent naming
mempolicy trivia: delete those ancient pr_debug()s
mempolicy: fix migrate_pages(2) syscall return nr_failed
kernfs: drop shared NUMA mempolicy hooks
hugetlbfs: drop shared NUMA mempolicy pretence
mm/damon/sysfs-test: add a unit test for damon_sysfs_set_targets()
...
|
|
Starting with commit 5d8edfb900d5 ("iomap: Copy larger chunks from
userspace"), iomap_write_iter() can get into endless loop. This can
be reproduced with LTP writev07 which uses partially valid iovecs:
struct iovec wr_iovec[] = {
{ buffer, 64 },
{ bad_addr, 64 },
{ buffer + 64, 64 },
{ buffer + 64 * 2, 64 },
};
commit bc1bb416bbb9 ("generic_perform_write()/iomap_write_actor():
saner logics for short copy") previously introduced the logic, which
made short copy retry in next iteration with amount of "bytes" it
managed to copy:
if (unlikely(status == 0)) {
/*
* A short copy made iomap_write_end() reject the
* thing entirely. Might be memory poisoning
* halfway through, might be a race with munmap,
* might be severe memory pressure.
*/
if (copied)
bytes = copied;
However, since 5d8edfb900d5 "bytes" is no longer carried into next
iteration, because it is now always initialized at the beginning of
the loop. And for iov_iter_count < PAGE_SIZE, "bytes" ends up with
same value as previous iteration, making the loop retry same copy
over and over, which leads to writev07 testcase hanging.
Make next iteration retry with amount of bytes we managed to copy.
Fixes: 5d8edfb900d5 ("iomap: Copy larger chunks from userspace")
Signed-off-by: Jan Stancek <jstancek@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
Combine the setting of the uptodate flag with the clearing of the locked
flag.
Link: https://lkml.kernel.org/r/20231004165317.1061855-7-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Alexander Gordeev <agordeev@linux.ibm.com>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Christian Borntraeger <borntraeger@linux.ibm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Heiko Carstens <hca@linux.ibm.com>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Richard Henderson <richard.henderson@linaro.org>
Cc: Sven Schnelle <svens@linux.ibm.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Perform one atomic operation (acquiring the spinlock) instead of two
(spinlock & atomic_sub) per read completion.
Link: https://lkml.kernel.org/r/20231004165317.1061855-3-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Alexander Gordeev <agordeev@linux.ibm.com>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Christian Borntraeger <borntraeger@linux.ibm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Heiko Carstens <hca@linux.ibm.com>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Richard Henderson <richard.henderson@linaro.org>
Cc: Sven Schnelle <svens@linux.ibm.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "Add folio_end_read", v2.
The core of this patchset is the new folio_end_read() call which
filesystems can use when finishing a page cache read instead of separate
calls to mark the folio uptodate and unlock it. As an illustration of its
use, I converted ext4, iomap & mpage; more can be converted.
I think that's useful by itself, but the interesting optimisation is that
we can implement that with a single XOR instruction that sets the uptodate
bit, clears the lock bit, tests the waiter bit and provides a write memory
barrier. That removes one memory barrier and one atomic instruction from
each page read, which seems worth doing. That's in patch 15.
The last two patches could be a separate series, but basically we can do
the same thing with the writeback flag that we do with the unlock flag;
clear it and test the waiters bit at the same time.
This patch (of 17):
This is really preparation for the next patch, but it lets us call
folio_mark_uptodate() in just one place instead of two.
Link: https://lkml.kernel.org/r/20231004165317.1061855-1-willy@infradead.org
Link: https://lkml.kernel.org/r/20231004165317.1061855-2-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Richard Henderson <richard.henderson@linaro.org>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Heiko Carstens <hca@linux.ibm.com>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Alexander Gordeev <agordeev@linux.ibm.com>
Cc: Christian Borntraeger <borntraeger@linux.ibm.com>
Cc: Sven Schnelle <svens@linux.ibm.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Fix a misspelling of "preceding".
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Bill O'Donnell <bodonnel@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
Convert iomap_unshare_iter to create large folios if possible, since the
write and zeroing paths already do that. I think this got missed in the
conversion of the write paths that landed in 6.6-rc1.
Cc: ritesh.list@gmail.com, willy@infradead.org
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
|
|
Prior to commit a01b8f225248e, we would always read in the contents of a
!uptodate folio prior to writing userspace data into the folio,
allocated a folio state object, etc. Ritesh introduced an optimization
that skips all of that if the write would cover the entire folio.
Unfortunately, the optimization misses the unshare case, where we always
have to read in the folio contents since there isn't a data buffer
supplied by userspace. This can result in stale kernel memory exposure
if userspace issues a FALLOC_FL_UNSHARE_RANGE call on part of a shared
file that isn't already cached.
This was caught by observing fstests regressions in the "unshare around"
mechanism that is used for unaligned writes to a reflinked realtime
volume when the realtime extent size is larger than 1FSB, though I think
it applies to any shared file.
Cc: ritesh.list@gmail.com, willy@infradead.org
Fixes: a01b8f225248e ("iomap: Allocate ifs in ->write_begin() early")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
|
|
Pull block updates from Jens Axboe:
"Pretty quiet round for this release. This contains:
- Add support for zoned storage to ublk (Andreas, Ming)
- Series improving performance for drivers that mark themselves as
needing a blocking context for issue (Bart)
- Cleanup the flush logic (Chengming)
- sed opal keyring support (Greg)
- Fixes and improvements to the integrity support (Jinyoung)
- Add some exports for bcachefs that we can hopefully delete again in
the future (Kent)
- deadline throttling fix (Zhiguo)
- Series allowing building the kernel without buffer_head support
(Christoph)
- Sanitize the bio page adding flow (Christoph)
- Write back cache fixes (Christoph)
- MD updates via Song:
- Fix perf regression for raid0 large sequential writes (Jan)
- Fix split bio iostat for raid0 (David)
- Various raid1 fixes (Heinz, Xueshi)
- raid6test build fixes (WANG)
- Deprecate bitmap file support (Christoph)
- Fix deadlock with md sync thread (Yu)
- Refactor md io accounting (Yu)
- Various non-urgent fixes (Li, Yu, Jack)
- Various fixes and cleanups (Arnd, Azeem, Chengming, Damien, Li,
Ming, Nitesh, Ruan, Tejun, Thomas, Xu)"
* tag 'for-6.6/block-2023-08-28' of git://git.kernel.dk/linux: (113 commits)
block: use strscpy() to instead of strncpy()
block: sed-opal: keyring support for SED keys
block: sed-opal: Implement IOC_OPAL_REVERT_LSP
block: sed-opal: Implement IOC_OPAL_DISCOVERY
blk-mq: prealloc tags when increase tagset nr_hw_queues
blk-mq: delete redundant tagset map update when fallback
blk-mq: fix tags leak when shrink nr_hw_queues
ublk: zoned: support REQ_OP_ZONE_RESET_ALL
md: raid0: account for split bio in iostat accounting
md/raid0: Fix performance regression for large sequential writes
md/raid0: Factor out helper for mapping and submitting a bio
md raid1: allow writebehind to work on any leg device set WriteMostly
md/raid1: hold the barrier until handle_read_error() finishes
md/raid1: free the r1bio before waiting for blocked rdev
md/raid1: call free_r1bio() before allow_barrier() in raid_end_bio_io()
blk-cgroup: Fix NULL deref caused by blkg_policy_data being installed before init
drivers/rnbd: restore sysfs interface to rnbd-client
md/raid5-cache: fix null-ptr-deref for r5l_flush_stripe_to_raid()
raid6: test: only check for Altivec if building on powerpc hosts
raid6: test: make sure all intermediate and artifact files are .gitignored
...
|
|
block_page_mkwrite_return is neither block nor mkwrite specific, and
should not be under CONFIG_BLOCK. Move it to mm.h and rename it to
vmf_fs_error.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Link: https://lore.kernel.org/r/20230801172201.1923299-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
iomap-6.6-mergeA
Improve iomap/xfs async dio write performance
iomap always punts async dio write completions to a workqueue, which has
a cost in terms of efficiency (now you need an unrelated worker to
process it) and latency (now you're bouncing a completion through an
async worker, which is a classic slowdown scenario).
io_uring handles IRQ completions via task_work, and for writes that
don't need to do extra IO at completion time, we can safely complete
them inline from that. This patchset adds IOCB_DIO_CALLER_COMP, which an
IO issuer can set to inform the completion side that any extra work that
needs doing for that completion can be punted to a safe task context.
The iomap dio completion will happen in hard/soft irq context, and we
need a saner context to process these completions. IOCB_DIO_CALLER_COMP
is added, which can be set in a struct kiocb->ki_flags by the issuer. If
the completion side of the iocb handling understands this flag, it can
choose to set a kiocb->dio_complete() handler and just call ki_complete
from IRQ context. The issuer must then ensure that this callback is
processed from a task. io_uring punts IRQ completions to task_work
already, so it's trivial wire it up to run more of the completion before
posting a CQE. This is good for up to a 37% improvement in
throughput/latency for low queue depth IO, patch 5 has the details.
If we need to do real work at completion time, iomap will clear the
IOMAP_DIO_CALLER_COMP flag.
This work came about when Andres tested low queue depth dio writes for
postgres and compared it to doing sync dio writes, showing that the
async processing slows us down a lot.
* tag 'xfs-async-dio.6-2023-08-01' of git://git.kernel.dk/linux:
iomap: support IOCB_DIO_CALLER_COMP
io_uring/rw: add write support for IOCB_DIO_CALLER_COMP
fs: add IOCB flags related to passing back dio completions
iomap: add IOMAP_DIO_INLINE_COMP
iomap: only set iocb->private for polled bio
iomap: treat a write through cache the same as FUA
iomap: use an unsigned type for IOMAP_DIO_* defines
iomap: cleanup up iomap_dio_bio_end_io()
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
If IOCB_DIO_CALLER_COMP is set, utilize that to set kiocb->dio_complete
handler and data for that callback. Rather than punt the completion to a
workqueue, we pass back the handler and data to the issuer and will get
a callback from a safe task context.
Using the following fio job to randomly dio write 4k blocks at
queue depths of 1..16:
fio --name=dio-write --filename=/data1/file --time_based=1 \
--runtime=10 --bs=4096 --rw=randwrite --norandommap --buffered=0 \
--cpus_allowed=4 --ioengine=io_uring --iodepth=$depth
shows the following results before and after this patch:
Stock Patched Diff
=======================================
QD1 155K 162K + 4.5%
QD2 290K 313K + 7.9%
QD4 533K 597K +12.0%
QD8 604K 827K +36.9%
QD16 615K 845K +37.4%
which shows nice wins all around. If we factored in per-IOP efficiency,
the wins look even nicer. This becomes apparent as queue depth rises,
as the offloaded workqueue completions runs out of steam.
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Rather than gate whether or not we need to punt a dio completion to a
workqueue on whether the IO is a write or not, add an explicit flag for
it. For now we treat them the same, reads always set the flags and async
writes do not.
No functional changes in this patch.
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
iocb->private is only used for polled IO, where the completer will
find the bio to poll through that field.
Assign it when we're submitting a polled bio, and get rid of the
dio->poll_bio indirection.
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Whether we have a write back cache and are using FUA or don't have
a write back cache at all is the same situation. Treat them the same.
This makes the IOMAP_DIO_WRITE_FUA name a bit misleading, as we have
two cases that provide stable writes:
1) Volatile write cache with FUA writes
2) Normal write without a volatile write cache
Rename that flag to IOMAP_DIO_STABLE_WRITE to make that clearer, and
update some of the FUA comments as well.
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
