From 2519369201f36a6b2571bc672c4e48f88c6b68d6 Mon Sep 17 00:00:00 2001 From: Brian Foster Date: Fri, 15 Nov 2024 15:01:53 -0500 Subject: [PATCH 1/3] iomap: reset per-iter state on non-error iter advances iomap_iter_advance() zeroes the processed and mapping fields on every non-error iteration except for the last expected iteration (i.e. return 0 expected to terminate the iteration loop). This appears to be circumstantial as nothing currently relies on these fields after the final iteration. Therefore to better faciliate iomap_iter reuse in subsequent patches, update iomap_iter_advance() to always reset per-iteration state on successful completion. Signed-off-by: Brian Foster Link: https://lore.kernel.org/r/20241115200155.593665-2-bfoster@redhat.com Reviewed-by: Darrick J. Wong Reviewed-by: Christoph Hellwig Signed-off-by: Christian Brauner --- fs/iomap/iter.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/fs/iomap/iter.c b/fs/iomap/iter.c index 79a0614eaab7..3790918646af 100644 --- a/fs/iomap/iter.c +++ b/fs/iomap/iter.c @@ -22,26 +22,25 @@ static inline int iomap_iter_advance(struct iomap_iter *iter) { bool stale = iter->iomap.flags & IOMAP_F_STALE; + int ret = 1; /* handle the previous iteration (if any) */ if (iter->iomap.length) { if (iter->processed < 0) return iter->processed; - if (!iter->processed && !stale) - return 0; if (WARN_ON_ONCE(iter->processed > iomap_length(iter))) return -EIO; iter->pos += iter->processed; iter->len -= iter->processed; - if (!iter->len) - return 0; + if (!iter->len || (!iter->processed && !stale)) + ret = 0; } - /* clear the state for the next iteration */ + /* clear the per iteration state */ iter->processed = 0; memset(&iter->iomap, 0, sizeof(iter->iomap)); memset(&iter->srcmap, 0, sizeof(iter->srcmap)); - return 1; + return ret; } static inline void iomap_iter_done(struct iomap_iter *iter) From 889ac75787cbeb129df7faf917ce7d53a32ea696 Mon Sep 17 00:00:00 2001 From: Brian Foster Date: Fri, 15 Nov 2024 15:01:54 -0500 Subject: [PATCH 2/3] iomap: lift zeroed mapping handling into iomap_zero_range() In preparation for special handling of subranges, lift the zeroed mapping logic from the iterator into the caller. Since this puts the pagecache dirty check and flushing in the same place, streamline the comments a bit as well. Signed-off-by: Brian Foster Link: https://lore.kernel.org/r/20241115200155.593665-3-bfoster@redhat.com Reviewed-by: Darrick J. Wong Signed-off-by: Christian Brauner --- fs/iomap/buffered-io.c | 66 +++++++++++++++--------------------------- 1 file changed, 24 insertions(+), 42 deletions(-) diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index 9ae71b9dafde..cb5aa3cded0e 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -1350,40 +1350,12 @@ static inline int iomap_zero_iter_flush_and_stale(struct iomap_iter *i) return filemap_write_and_wait_range(mapping, i->pos, end); } -static loff_t iomap_zero_iter(struct iomap_iter *iter, bool *did_zero, - bool *range_dirty) +static loff_t iomap_zero_iter(struct iomap_iter *iter, bool *did_zero) { - const struct iomap *srcmap = iomap_iter_srcmap(iter); loff_t pos = iter->pos; loff_t length = iomap_length(iter); loff_t written = 0; - /* - * We must zero subranges of unwritten mappings that might be dirty in - * pagecache from previous writes. We only know whether the entire range - * was clean or not, however, and dirty folios may have been written - * back or reclaimed at any point after mapping lookup. - * - * The easiest way to deal with this is to flush pagecache to trigger - * any pending unwritten conversions and then grab the updated extents - * from the fs. The flush may change the current mapping, so mark it - * stale for the iterator to remap it for the next pass to handle - * properly. - * - * Note that holes are treated the same as unwritten because zero range - * is (ab)used for partial folio zeroing in some cases. Hole backed - * post-eof ranges can be dirtied via mapped write and the flush - * triggers writeback time post-eof zeroing. - */ - if (srcmap->type == IOMAP_HOLE || srcmap->type == IOMAP_UNWRITTEN) { - if (*range_dirty) { - *range_dirty = false; - return iomap_zero_iter_flush_and_stale(iter); - } - /* range is clean and already zeroed, nothing to do */ - return length; - } - do { struct folio *folio; int status; @@ -1435,24 +1407,34 @@ iomap_zero_range(struct inode *inode, loff_t pos, loff_t len, bool *did_zero, bool range_dirty; /* - * Zero range wants to skip pre-zeroed (i.e. unwritten) mappings, but - * pagecache must be flushed to ensure stale data from previous - * buffered writes is not exposed. A flush is only required for certain - * types of mappings, but checking pagecache after mapping lookup is - * racy with writeback and reclaim. + * Zero range can skip mappings that are zero on disk so long as + * pagecache is clean. If pagecache was dirty prior to zero range, the + * mapping converts on writeback completion and so must be zeroed. * - * Therefore, check the entire range first and pass along whether any - * part of it is dirty. If so and an underlying mapping warrants it, - * flush the cache at that point. This trades off the occasional false - * positive (and spurious flush, if the dirty data and mapping don't - * happen to overlap) for simplicity in handling a relatively uncommon - * situation. + * The simplest way to deal with this across a range is to flush + * pagecache and process the updated mappings. To avoid an unconditional + * flush, check pagecache state and only flush if dirty and the fs + * returns a mapping that might convert on writeback. */ range_dirty = filemap_range_needs_writeback(inode->i_mapping, pos, pos + len - 1); + while ((ret = iomap_iter(&iter, ops)) > 0) { + const struct iomap *srcmap = iomap_iter_srcmap(&iter); - while ((ret = iomap_iter(&iter, ops)) > 0) - iter.processed = iomap_zero_iter(&iter, did_zero, &range_dirty); + if (srcmap->type == IOMAP_HOLE || + srcmap->type == IOMAP_UNWRITTEN) { + loff_t proc = iomap_length(&iter); + + if (range_dirty) { + range_dirty = false; + proc = iomap_zero_iter_flush_and_stale(&iter); + } + iter.processed = proc; + continue; + } + + iter.processed = iomap_zero_iter(&iter, did_zero); + } return ret; } EXPORT_SYMBOL_GPL(iomap_zero_range); From fde4c4c3ec1c1590eb09f97f9525fa7dd8df8343 Mon Sep 17 00:00:00 2001 From: Brian Foster Date: Fri, 15 Nov 2024 15:01:55 -0500 Subject: [PATCH 3/3] iomap: elide flush from partial eof zero range iomap zero range flushes pagecache in certain situations to determine which parts of the range might require zeroing if dirty data is present in pagecache. The kernel robot recently reported a regression associated with this flushing in the following stress-ng workload on XFS: stress-ng --timeout 60 --times --verify --metrics --no-rand-seed --metamix 64 This workload involves repeated small, strided, extending writes. On XFS, this produces a pattern of post-eof speculative preallocation, conversion of preallocation from delalloc to unwritten, dirtying pagecache over newly unwritten blocks, and then rinse and repeat from the new EOF. This leads to repetitive flushing of the EOF folio via the zero range call XFS uses for writes that start beyond current EOF. To mitigate this problem, special case EOF block zeroing to prefer zeroing the folio over a flush when the EOF folio is already dirty. To do this, split out and open code handling of an unaligned start offset. This brings most of the performance back by avoiding flushes on zero range calls via write and truncate extension operations. The flush doesn't occur in these situations because the entire range is post-eof and therefore the folio that overlaps EOF is the only one in the range. Signed-off-by: Brian Foster Link: https://lore.kernel.org/r/20241115200155.593665-4-bfoster@redhat.com Reviewed-by: Darrick J. Wong Reviewed-by: Christoph Hellwig Signed-off-by: Christian Brauner --- fs/iomap/buffered-io.c | 28 ++++++++++++++++++++++++---- 1 file changed, 24 insertions(+), 4 deletions(-) diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index cb5aa3cded0e..0708be776740 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -1403,6 +1403,10 @@ iomap_zero_range(struct inode *inode, loff_t pos, loff_t len, bool *did_zero, .len = len, .flags = IOMAP_ZERO, }; + struct address_space *mapping = inode->i_mapping; + unsigned int blocksize = i_blocksize(inode); + unsigned int off = pos & (blocksize - 1); + loff_t plen = min_t(loff_t, len, blocksize - off); int ret; bool range_dirty; @@ -1412,12 +1416,28 @@ iomap_zero_range(struct inode *inode, loff_t pos, loff_t len, bool *did_zero, * mapping converts on writeback completion and so must be zeroed. * * The simplest way to deal with this across a range is to flush - * pagecache and process the updated mappings. To avoid an unconditional - * flush, check pagecache state and only flush if dirty and the fs - * returns a mapping that might convert on writeback. + * pagecache and process the updated mappings. To avoid excessive + * flushing on partial eof zeroing, special case it to zero the + * unaligned start portion if already dirty in pagecache. + */ + if (off && + filemap_range_needs_writeback(mapping, pos, pos + plen - 1)) { + iter.len = plen; + while ((ret = iomap_iter(&iter, ops)) > 0) + iter.processed = iomap_zero_iter(&iter, did_zero); + + iter.len = len - (iter.pos - pos); + if (ret || !iter.len) + return ret; + } + + /* + * To avoid an unconditional flush, check pagecache state and only flush + * if dirty and the fs returns a mapping that might convert on + * writeback. */ range_dirty = filemap_range_needs_writeback(inode->i_mapping, - pos, pos + len - 1); + iter.pos, iter.pos + iter.len - 1); while ((ret = iomap_iter(&iter, ops)) > 0) { const struct iomap *srcmap = iomap_iter_srcmap(&iter);