kernel/include/linux/iomap.h, branch linux-5.17.y

xfs, iomap: limit individual ioend chain lengths in writeback

2022-01-26T17:19:20Z

Trond Myklebust reported soft lockups in XFS IO completion such as this: watchdog: BUG: soft lockup - CPU#12 stuck for 23s! [kworker/12:1:3106] CPU: 12 PID: 3106 Comm: kworker/12:1 Not tainted 4.18.0-305.10.2.el8_4.x86_64 #1 Workqueue: xfs-conv/md127 xfs_end_io [xfs] RIP: 0010:_raw_spin_unlock_irqrestore+0x11/0x20 Call Trace: wake_up_page_bit+0x8a/0x110 iomap_finish_ioend+0xd7/0x1c0 iomap_finish_ioends+0x7f/0xb0 xfs_end_ioend+0x6b/0x100 [xfs] xfs_end_io+0xb9/0xe0 [xfs] process_one_work+0x1a7/0x360 worker_thread+0x1fa/0x390 kthread+0x116/0x130 ret_from_fork+0x35/0x40 Ioends are processed as an atomic completion unit when all the chained bios in the ioend have completed their IO. Logically contiguous ioends can also be merged and completed as a single, larger unit. Both of these things can be problematic as both the bio chains per ioend and the size of the merged ioends processed as a single completion are both unbound. If we have a large sequential dirty region in the page cache, write_cache_pages() will keep feeding us sequential pages and we will keep mapping them into ioends and bios until we get a dirty page at a non-sequential file offset. These large sequential runs can will result in bio and ioend chaining to optimise the io patterns. The pages iunder writeback are pinned within these chains until the submission chaining is broken, allowing the entire chain to be completed. This can result in huge chains being processed in IO completion context. We get deep bio chaining if we have large contiguous physical extents. We will keep adding pages to the current bio until it is full, then we'll chain a new bio to keep adding pages for writeback. Hence we can build bio chains that map millions of pages and tens of gigabytes of RAM if the page cache contains big enough contiguous dirty file regions. This long bio chain pins those pages until the final bio in the chain completes and the ioend can iterate all the chained bios and complete them. OTOH, if we have a physically fragmented file, we end up submitting one ioend per physical fragment that each have a small bio or bio chain attached to them. We do not chain these at IO submission time, but instead we chain them at completion time based on file offset via iomap_ioend_try_merge(). Hence we can end up with unbound ioend chains being built via completion merging. XFS can then do COW remapping or unwritten extent conversion on that merged chain, which involves walking an extent fragment at a time and running a transaction to modify the physical extent information. IOWs, we merge all the discontiguous ioends together into a contiguous file range, only to then process them individually as discontiguous extents. This extent manipulation is computationally expensive and can run in a tight loop, so merging logically contiguous but physically discontigous ioends gains us nothing except for hiding the fact the fact we broke the ioends up into individual physical extents at submission and then need to loop over those individual physical extents at completion. Hence we need to have mechanisms to limit ioend sizes and to break up completion processing of large merged ioend chains: 1. bio chains per ioend need to be bound in length. Pure overwrites go straight to iomap_finish_ioend() in softirq context with the exact bio chain attached to the ioend by submission. Hence the only way to prevent long holdoffs here is to bound ioend submission sizes because we can't reschedule in softirq context. 2. iomap_finish_ioends() has to handle unbound merged ioend chains correctly. This relies on any one call to iomap_finish_ioend() being bound in runtime so that cond_resched() can be issued regularly as the long ioend chain is processed. i.e. this relies on mechanism #1 to limit individual ioend sizes to work correctly. 3. filesystems have to loop over the merged ioends to process physical extent manipulations. This means they can loop internally, and so we break merging at physical extent boundaries so the filesystem can easily insert reschedule points between individual extent manipulations. Signed-off-by: Dave Chinner Reported-and-tested-by: Trond Myklebust Reviewed-by: Darrick J. Wong Signed-off-by: Darrick J. Wong

Merge tag 'libnvdimm-for-5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm

2022-01-12T23:46:11Z

Pull dax and libnvdimm updates from Dan Williams: "The bulk of this is a rework of the dax_operations API after discovering the obstacles it posed to the work-in-progress DAX+reflink support for XFS and other copy-on-write filesystem mechanics. Primarily the need to plumb a block_device through the API to handle partition offsets was a sticking point and Christoph untangled that dependency in addition to other cleanups to make landing the DAX+reflink support easier. The DAX_PMEM_COMPAT option has been around for 4 years and not only are distributions shipping userspace that understand the current configuration API, but some are not even bothering to turn this option on anymore, so it seems a good time to remove it per the deprecation schedule. Recall that this was added after the device-dax subsystem moved from /sys/class/dax to /sys/bus/dax for its sysfs organization. All recent functionality depends on /sys/bus/dax. Some other miscellaneous cleanups and reflink prep patches are included as well. Summary: - Simplify the dax_operations API: - Eliminate bdev_dax_pgoff() in favor of the filesystem maintaining and applying a partition offset to all its DAX iomap operations. - Remove wrappers and device-mapper stacked callbacks for ->copy_from_iter() and ->copy_to_iter() in favor of moving block_device relative offset responsibility to the dax_direct_access() caller. - Remove the need for an @bdev in filesystem-DAX infrastructure - Remove unused uio helpers copy_from_iter_flushcache() and copy_mc_to_iter() as only the non-check_copy_size() versions are used for DAX. - Prepare XFS for the pending (next merge window) DAX+reflink support - Remove deprecated DEV_DAX_PMEM_COMPAT support - Cleanup a straggling misuse of the GUID api" * tag 'libnvdimm-for-5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm: (38 commits) iomap: Fix error handling in iomap_zero_iter() ACPI: NFIT: Import GUID before use dax: remove the copy_from_iter and copy_to_iter methods dax: remove the DAXDEV_F_SYNC flag dax: simplify dax_synchronous and set_dax_synchronous uio: remove copy_from_iter_flushcache() and copy_mc_to_iter() iomap: turn the byte variable in iomap_zero_iter into a ssize_t memremap: remove support for external pgmap refcounts fsdax: don't require CONFIG_BLOCK iomap: build the block based code conditionally dax: fix up some of the block device related ifdefs fsdax: shift partition offset handling into the file systems dax: return the partition offset from fs_dax_get_by_bdev iomap: add a IOMAP_DAX flag xfs: pass the mapping flags to xfs_bmbt_to_iomap xfs: use xfs_direct_write_iomap_ops for DAX zeroing xfs: move dax device handling into xfs_{alloc,free}_buftarg ext4: cleanup the dax handling in ext4_fill_super ext2: cleanup the dax handling in ext2_fill_super fsdax: decouple zeroing from the iomap buffered I/O code ...

iomap,xfs: Convert ->discard_page to ->discard_folio

2021-12-18T05:06:07Z

XFS has the only implementation of ->discard_page today, so convert it to use folios in the same patch as converting the API. Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: Christoph Hellwig Reviewed-by: Darrick J. Wong

iomap: Add iomap_invalidate_folio

2021-12-16T20:49:52Z

Keep iomap_invalidatepage around as a wrapper for use in address_space operations. Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: Christoph Hellwig Reviewed-by: Darrick J. Wong

iomap: add a IOMAP_DAX flag

2021-12-04T16:58:53Z

Add a flag so that the file system can easily detect DAX operations based just on the iomap operation requested instead of looking at inode state using IS_DAX. This will be needed to apply the to be added partition offset only for operations that actually use DAX, but not things like fiemap that are based on the block device. In the long run it should also allow turning the bdev, dax_dev and inline_data into a union. Signed-off-by: Christoph Hellwig Reviewed-by: Dan Williams Reviewed-by: Darrick J. Wong Link: https://lore.kernel.org/r/20211129102203.2243509-25-hch@lst.de Signed-off-by: Dan Williams

Merge tag 'gfs2-v5.15-rc5-mmap-fault' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2

2021-11-02T19:25:03Z

Pull gfs2 mmap + page fault deadlocks fixes from Andreas Gruenbacher: "Functions gfs2_file_read_iter and gfs2_file_write_iter are both accessing the user buffer to write to or read from while holding the inode glock. In the most basic deadlock scenario, that buffer will not be resident and it will be mapped to the same file. Accessing the buffer will trigger a page fault, and gfs2 will deadlock trying to take the same inode glock again while trying to handle that fault. Fix that and similar, more complex scenarios by disabling page faults while accessing user buffers. To make this work, introduce a small amount of new infrastructure and fix some bugs that didn't trigger so far, with page faults enabled" * tag 'gfs2-v5.15-rc5-mmap-fault' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2: gfs2: Fix mmap + page fault deadlocks for direct I/O iov_iter: Introduce nofault flag to disable page faults gup: Introduce FOLL_NOFAULT flag to disable page faults iomap: Add done_before argument to iomap_dio_rw iomap: Support partial direct I/O on user copy failures iomap: Fix iomap_dio_rw return value for user copies gfs2: Fix mmap + page fault deadlocks for buffered I/O gfs2: Eliminate ip->i_gh gfs2: Move the inode glock locking to gfs2_file_buffered_write gfs2: Introduce flag for glock holder auto-demotion gfs2: Clean up function may_grant gfs2: Add wrapper for iomap_file_buffered_write iov_iter: Introduce fault_in_iov_iter_writeable iov_iter: Turn iov_iter_fault_in_readable into fault_in_iov_iter_readable gup: Turn fault_in_pages_{readable,writeable} into fault_in_{readable,writeable} powerpc/kvm: Fix kvm_use_magic_page iov_iter: Fix iov_iter_get_pages{,_alloc} page fault return value

iomap: Add done_before argument to iomap_dio_rw

2021-10-24T13:26:05Z

Add a done_before argument to iomap_dio_rw that indicates how much of the request has already been transferred. When the request succeeds, we report that done_before additional bytes were tranferred. This is useful for finishing a request asynchronously when part of the request has already been completed synchronously. We'll use that to allow iomap_dio_rw to be used with page faults disabled: when a page fault occurs while submitting a request, we synchronously complete the part of the request that has already been submitted. The caller can then take care of the page fault and call iomap_dio_rw again for the rest of the request, passing in the number of bytes already tranferred. Signed-off-by: Andreas Gruenbacher Reviewed-by: Darrick J. Wong

iomap: Support partial direct I/O on user copy failures

2021-10-24T13:26:05Z

In iomap_dio_rw, when iomap_apply returns an -EFAULT error and the IOMAP_DIO_PARTIAL flag is set, complete the request synchronously and return a partial result. This allows the caller to deal with the page fault and retry the remainder of the request. Signed-off-by: Andreas Gruenbacher Reviewed-by: Darrick J. Wong

block: switch polling to be bio based

2021-10-18T12:17:36Z

Replace the blk_poll interface that requires the caller to keep a queue and cookie from the submissions with polling based on the bio. Polling for the bio itself leads to a few advantages: - the cookie construction can made entirely private in blk-mq.c - the caller does not need to remember the request_queue and cookie separately and thus sidesteps their lifetime issues - keeping the device and the cookie inside the bio allows to trivially support polling BIOs remapping by stacking drivers - a lot of code to propagate the cookie back up the submission path can be removed entirely. Signed-off-by: Christoph Hellwig Tested-by: Mark Wunderlich Link: https://lore.kernel.org/r/20211012111226.760968-15-hch@lst.de Signed-off-by: Jens Axboe

block: replace the spin argument to blk_iopoll with a flags argument

2021-10-18T12:17:36Z

Switch the boolean spin argument to blk_poll to passing a set of flags instead. This will allow to control polling behavior in a more fine grained way. Signed-off-by: Christoph Hellwig Tested-by: Mark Wunderlich Link: https://lore.kernel.org/r/20211012111226.760968-10-hch@lst.de [axboe: adapt to changed io_uring iopoll] Signed-off-by: Jens Axboe