CVE-2026-89772

For large folios or subpage sectorsize, it is possible to submit a bio which does not cover the whole folio. When this happens, we will have a bio in flight for a folio that we have not called folio clear dirty for io() on. If a task with an existing mmap-ed PTE writes (without faulting..) in this window, it can result in corruptions. If the write arrives while the checksumming or writing itself is underway, this can result in an invalid checksum and later corruption reports on read. If the write arrives after checksumming/writing is done but before the last sector dirty is cleared, then the write is present in page cache but doesn't affect the dirty tracking and will be lost when the folio is fully finished being submitted and the dirty bit is cleared. This results in losing the write even if fsync() is called.

For zoned submissions which are done in batch separate from the main extent writepage() loop, we also risk csum violations for those submissions. Zoned writes are clamped to max zone append size and are not aligned with folios, so a submission can span two folios. The first folio being processed in extent write cache pages() will call extent write locked range() which will submit the partial range of the next folio, while the rest of that folio could still be dirty. So clearing dirty on the submitted sectors doesn't call folio clear dirty for io() and we have the same issue. Since extent write cache pages() skips these batch submitted folios (they are already marked for writeback from submission by the preceding folio), we must add the extra write protection in lock delalloc folios().

For inline extents this will subtly risk losing writes that happen after/while we copy the inline extent but before we clear dirty on the folio.

For folios spanning EOF, mmap could tamper with the zeroed bytes past EOF and cause them to be persisted where future faults would improperly see them instead of zeros.

Finally, for compressed extents, we risk modifying the folios while we work on compressing them which will result in corrupted compressed data. Specifically, in run delalloc compressed() we queue up work to do compress file range() in BTRFS COMPRESSION CHUNK SIZE (512K) chunks which will call btrfs folio clamp clear dirty() on the range. For non-subpage, this will always clear the whole folio, safely. For subpage, we risk a partial clear here as well. In particular, imagine a 2M folio broken up into 512K chunks of work which might start compression work on one chunk before all the chunks compress file range() workers have gotten far enough to finish clearing all the dirty bitmaps of the folio and getting to folio clear dirty for io(). Large folios on the edges of submission ranges are similarly at risk to be only partly cleared. This particular gap was introduced by a second patch in the same series: commit a4ef54dbb576 ("btrfs: make extent range clear dirty for io() to handle sector size < page size cases")