| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Pull iomap updates from Darrick Wong:
"We've got some big changes for this release -- I'm very happy to be
landing willy's work to enable large folios for the page cache for
general read and write IOs when the fs can make contiguous space
allocations, and Ritesh's work to track sub-folio dirty state to
eliminate the write amplification problems inherent in using large
folios.
As a bonus, io_uring can now process write completions in the caller's
context instead of bouncing through a workqueue, which should reduce
io latency dramatically. IOWs, XFS should see a nice performance bump
for both IO paths.
Summary:
- Make large writes to the page cache fill sparse parts of the cache
with large folios, then use large memcpy calls for the large folio.
- Track the per-block dirty state of each large folio so that a
buffered write to a single byte on a large folio does not result in
a (potentially) multi-megabyte writeback IO.
- Allow some directio completions to be performed in the initiating
task's context instead of punting through a workqueue. This will
reduce latency for some io_uring requests"
* tag 'iomap-6.6-merge-3' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: (26 commits)
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()
iomap: Add per-block dirty state tracking to improve performance
iomap: Allocate ifs in ->write_begin() early
iomap: Refactor iomap_write_delalloc_punch() function out
iomap: Use iomap_punch_t typedef
iomap: Fix possible overflow condition in iomap_write_delalloc_scan
iomap: Add some uptodate state handling helpers for ifs state bitmap
iomap: Drop ifs argument from iomap_set_range_uptodate()
iomap: Rename iomap_page to iomap_folio_state and others
iomap: Copy larger chunks from userspace
iomap: Create large folios in the buffered write path
filemap: Allow __filemap_get_folio to allocate large folios
filemap: Add fgf_t typedef
...
|
|
Commit 16d7fd3cfa72 ("zonefs: use iomap for synchronous direct writes")
changes zonefs code from a self-built zone append BIO to using iomap for
synchronous direct writes. This change relies on iomap submit BIO
callback to change the write BIO built by iomap to a zone append BIO.
However, this change overlooked the fact that a write BIO may be very
large as it is split when issued. The change from a regular write to a
zone append operation for the built BIO can result in a block layer
warning as zone append BIO are not allowed to be split.
WARNING: CPU: 18 PID: 202210 at block/bio.c:1644 bio_split+0x288/0x350
Call Trace:
? __warn+0xc9/0x2b0
? bio_split+0x288/0x350
? report_bug+0x2e6/0x390
? handle_bug+0x41/0x80
? exc_invalid_op+0x13/0x40
? asm_exc_invalid_op+0x16/0x20
? bio_split+0x288/0x350
bio_split_rw+0x4bc/0x810
? __pfx_bio_split_rw+0x10/0x10
? lockdep_unlock+0xf2/0x250
__bio_split_to_limits+0x1d8/0x900
blk_mq_submit_bio+0x1cf/0x18a0
? __pfx_iov_iter_extract_pages+0x10/0x10
? __pfx_blk_mq_submit_bio+0x10/0x10
? find_held_lock+0x2d/0x110
? lock_release+0x362/0x620
? mark_held_locks+0x9e/0xe0
__submit_bio+0x1ea/0x290
? __pfx___submit_bio+0x10/0x10
? seqcount_lockdep_reader_access.constprop.0+0x82/0x90
submit_bio_noacct_nocheck+0x675/0xa20
? __pfx_bio_iov_iter_get_pages+0x10/0x10
? __pfx_submit_bio_noacct_nocheck+0x10/0x10
iomap_dio_bio_iter+0x624/0x1280
__iomap_dio_rw+0xa22/0x18a0
? lock_is_held_type+0xe3/0x140
? __pfx___iomap_dio_rw+0x10/0x10
? lock_release+0x362/0x620
? zonefs_file_write_iter+0x74c/0xc80 [zonefs]
? down_write+0x13d/0x1e0
iomap_dio_rw+0xe/0x40
zonefs_file_write_iter+0x5ea/0xc80 [zonefs]
do_iter_readv_writev+0x18b/0x2c0
? __pfx_do_iter_readv_writev+0x10/0x10
? inode_security+0x54/0xf0
do_iter_write+0x13b/0x7c0
? lock_is_held_type+0xe3/0x140
vfs_writev+0x185/0x550
? __pfx_vfs_writev+0x10/0x10
? __handle_mm_fault+0x9bd/0x1c90
? find_held_lock+0x2d/0x110
? lock_release+0x362/0x620
? find_held_lock+0x2d/0x110
? lock_release+0x362/0x620
? __up_read+0x1ea/0x720
? do_pwritev+0x136/0x1f0
do_pwritev+0x136/0x1f0
? __pfx_do_pwritev+0x10/0x10
? syscall_enter_from_user_mode+0x22/0x90
? lockdep_hardirqs_on+0x7d/0x100
do_syscall_64+0x58/0x80
This error depends on the hardware used, specifically on the max zone
append bytes and max_[hw_]sectors limits. Tests using AMD Epyc machines
that have low limits did not reveal this issue while runs on Intel Xeon
machines with larger limits trigger it.
Manually splitting the zone append BIO using bio_split_rw() can solve
this issue but also requires issuing the fragment BIOs synchronously
with submit_bio_wait(), to avoid potential reordering of the zone append
BIO fragments, which would lead to data corruption. That is, this
solution is not better than using regular write BIOs which are subject
to serialization using zone write locking at the IO scheduler level.
Given this, fix the issue by removing zone append support and using
regular write BIOs for synchronous direct writes. This allows preseving
the use of iomap and having identical synchronous and asynchronous
sequential file write path. Zone append support will be reintroduced
later through io_uring commands to ensure that the needed special
handling is done correctly.
Reported-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Fixes: 16d7fd3cfa72 ("zonefs: use iomap for synchronous direct writes")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Tested-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
|
|
When filesystem blocksize is less than folio size (either with
mapping_large_folio_support() or with blocksize < pagesize) and when the
folio is uptodate in pagecache, then even a byte write can cause
an entire folio to be written to disk during writeback. This happens
because we currently don't have a mechanism to track per-block dirty
state within struct iomap_folio_state. We currently only track uptodate
state.
This patch implements support for tracking per-block dirty state in
iomap_folio_state->state bitmap. This should help improve the filesystem
write performance and help reduce write amplification.
Performance testing of below fio workload reveals ~16x performance
improvement using nvme with XFS (4k blocksize) on Power (64K pagesize)
FIO reported write bw scores improved from around ~28 MBps to ~452 MBps.
1. <test_randwrite.fio>
[global]
ioengine=psync
rw=randwrite
overwrite=1
pre_read=1
direct=0
bs=4k
size=1G
dir=./
numjobs=8
fdatasync=1
runtime=60
iodepth=64
group_reporting=1
[fio-run]
2. Also our internal performance team reported that this patch improves
their database workload performance by around ~83% (with XFS on Power)
Reported-by: Aravinda Herle <araherle@in.ibm.com>
Reported-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull mm updates from Andrew Morton:
- Yosry Ahmed brought back some cgroup v1 stats in OOM logs
- Yosry has also eliminated cgroup's atomic rstat flushing
- Nhat Pham adds the new cachestat() syscall. It provides userspace
with the ability to query pagecache status - a similar concept to
mincore() but more powerful and with improved usability
- Mel Gorman provides more optimizations for compaction, reducing the
prevalence of page rescanning
- Lorenzo Stoakes has done some maintanance work on the
get_user_pages() interface
- Liam Howlett continues with cleanups and maintenance work to the
maple tree code. Peng Zhang also does some work on maple tree
- Johannes Weiner has done some cleanup work on the compaction code
- David Hildenbrand has contributed additional selftests for
get_user_pages()
- Thomas Gleixner has contributed some maintenance and optimization
work for the vmalloc code
- Baolin Wang has provided some compaction cleanups,
- SeongJae Park continues maintenance work on the DAMON code
- Huang Ying has done some maintenance on the swap code's usage of
device refcounting
- Christoph Hellwig has some cleanups for the filemap/directio code
- Ryan Roberts provides two patch series which yield some
rationalization of the kernel's access to pte entries - use the
provided APIs rather than open-coding accesses
- Lorenzo Stoakes has some fixes to the interaction between pagecache
and directio access to file mappings
- John Hubbard has a series of fixes to the MM selftesting code
- ZhangPeng continues the folio conversion campaign
- Hugh Dickins has been working on the pagetable handling code, mainly
with a view to reducing the load on the mmap_lock
- Catalin Marinas has reduced the arm64 kmalloc() minimum alignment
from 128 to 8
- Domenico Cerasuolo has improved the zswap reclaim mechanism by
reorganizing the LRU management
- Matthew Wilcox provides some fixups to make gfs2 work better with the
buffer_head code
- Vishal Moola also has done some folio conversion work
- Matthew Wilcox has removed the remnants of the pagevec code - their
functionality is migrated over to struct folio_batch
* tag 'mm-stable-2023-06-24-19-15' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (380 commits)
mm/hugetlb: remove hugetlb_set_page_subpool()
mm: nommu: correct the range of mmap_sem_read_lock in task_mem()
hugetlb: revert use of page_cache_next_miss()
Revert "page cache: fix page_cache_next/prev_miss off by one"
mm/vmscan: fix root proactive reclaim unthrottling unbalanced node
mm: memcg: rename and document global_reclaim()
mm: kill [add|del]_page_to_lru_list()
mm: compaction: convert to use a folio in isolate_migratepages_block()
mm: zswap: fix double invalidate with exclusive loads
mm: remove unnecessary pagevec includes
mm: remove references to pagevec
mm: rename invalidate_mapping_pagevec to mapping_try_invalidate
mm: remove struct pagevec
net: convert sunrpc from pagevec to folio_batch
i915: convert i915_gpu_error to use a folio_batch
pagevec: rename fbatch_count()
mm: remove check_move_unevictable_pages()
drm: convert drm_gem_put_pages() to use a folio_batch
i915: convert shmem_sg_free_table() to use a folio_batch
scatterlist: add sg_set_folio()
...
|
|
Pull splice updates from Jens Axboe:
"This kills off ITER_PIPE to avoid a race between truncate,
iov_iter_revert() on the pipe and an as-yet incomplete DMA to a bio
with unpinned/unref'ed pages from an O_DIRECT splice read. This causes
memory corruption.
Instead, we either use (a) filemap_splice_read(), which invokes the
buffered file reading code and splices from the pagecache into the
pipe; (b) copy_splice_read(), which bulk-allocates a buffer, reads
into it and then pushes the filled pages into the pipe; or (c) handle
it in filesystem-specific code.
Summary:
- Rename direct_splice_read() to copy_splice_read()
- Simplify the calculations for the number of pages to be reclaimed
in copy_splice_read()
- Turn do_splice_to() into a helper, vfs_splice_read(), so that it
can be used by overlayfs and coda to perform the checks on the
lower fs
- Make vfs_splice_read() jump to copy_splice_read() to handle
direct-I/O and DAX
- Provide shmem with its own splice_read to handle non-existent pages
in the pagecache. We don't want a ->read_folio() as we don't want
to populate holes, but filemap_get_pages() requires it
- Provide overlayfs with its own splice_read to call down to a lower
layer as overlayfs doesn't provide ->read_folio()
- Provide coda with its own splice_read to call down to a lower layer
as coda doesn't provide ->read_folio()
- Direct ->splice_read to copy_splice_read() in tty, procfs, kernfs
and random files as they just copy to the output buffer and don't
splice pages
- Provide wrappers for afs, ceph, ecryptfs, ext4, f2fs, nfs, ntfs3,
ocfs2, orangefs, xfs and zonefs to do locking and/or revalidation
- Make cifs use filemap_splice_read()
- Replace pointers to generic_file_splice_read() with pointers to
filemap_splice_read() as DIO and DAX are handled in the caller;
filesystems can still provide their own alternate ->splice_read()
op
- Remove generic_file_splice_read()
- Remove ITER_PIPE and its paraphernalia as generic_file_splice_read
was the only user"
* tag 'for-6.5/splice-2023-06-23' of git://git.kernel.dk/linux: (31 commits)
splice: kdoc for filemap_splice_read() and copy_splice_read()
iov_iter: Kill ITER_PIPE
splice: Remove generic_file_splice_read()
splice: Use filemap_splice_read() instead of generic_file_splice_read()
cifs: Use filemap_splice_read()
trace: Convert trace/seq to use copy_splice_read()
zonefs: Provide a splice-read wrapper
xfs: Provide a splice-read wrapper
orangefs: Provide a splice-read wrapper
ocfs2: Provide a splice-read wrapper
ntfs3: Provide a splice-read wrapper
nfs: Provide a splice-read wrapper
f2fs: Provide a splice-read wrapper
ext4: Provide a splice-read wrapper
ecryptfs: Provide a splice-read wrapper
ceph: Provide a splice-read wrapper
afs: Provide a splice-read wrapper
9p: Add splice_read wrapper
net: Make sock_splice_read() use copy_splice_read() by default
tty, proc, kernfs, random: Use copy_splice_read()
...
|
|
Since commit a2ad63daa88b ("VFS: add FMODE_CAN_ODIRECT file flag") file
systems can just set the FMODE_CAN_ODIRECT flag at open time instead of
wiring up a dummy direct_IO method to indicate support for direct I/O.
Do that for zonefs so that noop_direct_IO can eventually be removed.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
|
|
Remove the function zonefs_file_dio_append() that is used to manually
issue REQ_OP_ZONE_APPEND BIOs for processing synchronous direct writes
and use iomap instead.
To preserve the use of zone append operations for synchronous writes,
different struct iomap_dio_ops are defined. For synchronous direct
writes using zone append, zonefs_zone_append_dio_ops is introduced.
The submit_bio operation of this structure is defined as the function
zonefs_file_zone_append_dio_submit_io() which is used to change the BIO
opreation for synchronous direct IO writes to REQ_OP_ZONE_APPEND.
In order to preserve the write location check on completion of zone
append BIOs, the end_io operation is also defined using the function
zonefs_file_zone_append_dio_bio_end_io(). This check now relies on the
zonefs_zone_append_bio structure, allocated together with zone append
BIOs with a dedicated BIO set. This structure include the target inode
of a zone append BIO as well as the target append offset location for
the zone append operation. This is used to perform a check against
bio->bi_iter.bi_sector when the BIO completes, without needing to use
the zone information z_wpoffset field, thus removing the need for
taking the inode truncate mutex.
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
|
|
All callers of iomap_file_buffered_write need to updated ki_pos, move it
into common code.
Link: https://lkml.kernel.org/r/20230601145904.1385409-8-hch@lst.de
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Andreas Gruenbacher <agruenba@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Acked-by: Damien Le Moal <dlemoal@kernel.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Anna Schumaker <anna@kernel.org>
Cc: Chao Yu <chao@kernel.org>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Ilya Dryomov <idryomov@gmail.com>
Cc: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Miklos Szeredi <miklos@szeredi.hu>
Cc: Miklos Szeredi <mszeredi@redhat.com>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: Trond Myklebust <trond.myklebust@hammerspace.com>
Cc: Xiubo Li <xiubli@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Provide a splice_read wrapper for zonefs. This does some checks before
proceeding and locks the inode across the call to filemap_splice_read() and
a size check in case of truncation. Splicing from direct I/O is handled by
the caller.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Christoph Hellwig <hch@lst.de>
cc: Al Viro <viro@zeniv.linux.org.uk>
cc: Jens Axboe <axboe@kernel.dk>
cc: Darrick J. Wong <djwong@kernel.org>
cc: linux-xfs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-block@vger.kernel.org
cc: linux-mm@kvack.org
Acked-by: Damien Le Moal <dlemoal@kernel.org>
Link: https://lore.kernel.org/r/20230522135018.2742245-26-dhowells@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
The call to invalidate_inode_pages2_range() in __iomap_dio_rw() may
fail, in which case -ENOTBLK is returned and this error code is
propagated back to user space trhough iomap_dio_rw() ->
zonefs_file_dio_write() return chain. This error code is fairly obscure
and may confuse the user. Avoid this and be consistent with the behavior
of zonefs_file_dio_append() for similar invalidate_inode_pages2_range()
errors by returning -EBUSY to user space when iomap_dio_rw() returns
-ENOTBLK.
Suggested-by: Christoph Hellwig <hch@infradead.org>
Fixes: 8dcc1a9d90c1 ("fs: New zonefs file system")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
|
|
When a direct append write is executed, the append offset may correspond
to the last page of a sequential file inode which might have been cached
already by buffered reads, page faults with mmap-read or non-direct
readahead. To ensure that the on-disk and cached data is consistant for
such last cached page, make sure to always invalidate it in
zonefs_file_dio_append(). If the invalidation fails, return -EBUSY to
userspace to differentiate from IO errors.
This invalidation will always be a no-op when the FS block size (device
zone write granularity) is equal to the page size (e.g. 4K).
Reported-by: Hans Holmberg <Hans.Holmberg@wdc.com>
Fixes: 02ef12a663c7 ("zonefs: use REQ_OP_ZONE_APPEND for sync DIO")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Tested-by: Hans Holmberg <hans.holmberg@wdc.com>
|
|
Since the expected write location in a sequential file is always at the
end of the file (append write), when an invalid write append location is
detected in zonefs_file_dio_append(), print the invalid written location
instead of the expected write location.
Fixes: a608da3bd730 ("zonefs: Detect append writes at invalid locations")
Cc: stable@vger.kernel.org
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
|
|
In zonefs_file_dio_append(), initialize the variable size to 0 to
prevent compilation and static code analizers warning such as:
New smatch warnings:
fs/zonefs/file.c:441 zonefs_file_dio_append() error: uninitialized
symbol 'size'.
The warning is a false positive as size is never actually used
uninitialized.
No functional change.
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <error27@gmail.com>
Link: https://lore.kernel.org/r/202303191227.GL8Dprbi-lkp@intel.com/
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
|
|
In preparation for adding dynamic inode allocation, separate an inode
zone information from the zonefs inode structure. The new data structure
zonefs_zone is introduced to store in memory information about a zone
that must be kept throughout the lifetime of the device mount.
Linking between a zone file inode and its zone information is done by
setting the inode i_private field to point to a struct zonefs_zone.
Using the i_private pointer avoids the need for adding a pointer in
struct zonefs_inode_info. Beside the vfs inode, this structure is
reduced to a mutex and a write open counter.
One struct zonefs_zone is created per file inode on mount. These
structures are organized in an array using the new struct
zonefs_zone_group data structure to represent zone groups. The
zonefs_zone arrays are indexed per file number (the index of a struct
zonefs_zone in its array directly gives the file number/name for that
zone file inode).
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
|
|
Instead of using the i_ztype field in struct zonefs_inode_info to
indicate the zone type of an inode, introduce the new inode flag
ZONEFS_ZONE_CNV to be set in the i_flags field of struct
zonefs_inode_info to identify conventional zones. If this flag is not
set, the zone of an inode is considered to be a sequential zone.
The helpers zonefs_zone_is_cnv(), zonefs_zone_is_seq(),
zonefs_inode_is_cnv() and zonefs_inode_is_seq() are introduced to
simplify testing the zone type of a struct zonefs_inode_info and of a
struct inode.
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
|
|
Move all code related to zone file operations from super.c to the new
file.c file. Inode and zone management code remains in super.c.
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
|
