جزئیات آسیب‌پذیری

In the Linux kernel, the following vulnerability has been resolved: iomap: fix out-of-bounds bitmap_set() with zero-length range ifs_set_range_dirty() and ifs_set_range_uptodate() compute last_blk as (off + len - 1) >> i_blkbits. When off is 0 and len is 0, the unsigned subtraction underflows to SIZE_MAX, producing a huge last_blk and nr_blks value that causes bitmap_set() to write far beyond the ifs->state allocation. Regarding ifs_set_range_uptodate(), it is temporarily safe because len cannot be passed in as 0. However, for ifs_set_range_dirty() this is reachable from __iomap_write_end(): when copy_folio_from_iter_atomic() returns 0 (e.g. user buffer fault) and the folio is already uptodate, the guard at the top of __iomap_write_end() does not trigger because !folio_test_uptodate() is false, and iomap_set_range_dirty() is called with copied == 0. Add a !len guard to both functions before the computation, so that a zero-length range is a no-op.

• https://git.kernel.org/stable/c/48829622212f6b8f49155889aecc818ba28ba680
• https://git.kernel.org/stable/c/fb4fad9105c88b1d82f1b3c39e3b6abea8249af6
• https://git.kernel.org/stable/c/7037e7bdcd26f46c080b8ce307dee5cb471c4b7c
• https://git.kernel.org/stable/c/c5b6a48a8a716a7730e39af1cad083dc4ec955ce
• https://git.kernel.org/stable/c/9c7d8f7c8994c790fca501dc45ce66e7356cbe05