Commit eade5404 authored by Sheng Yong's avatar Sheng Yong Committed by Gao Xiang
Browse files

erofs: set fileio bio failed in short read case 



For file-backed mount, IO requests are handled by vfs_iocb_iter_read().
However, it can be interrupted by SIGKILL, returning the number of
bytes actually copied. Unused folios in bio are unexpectedly marked
as uptodate.

  vfs_read
    filemap_read
      filemap_get_pages
        filemap_readahead
          erofs_fileio_readahead
            erofs_fileio_rq_submit
              vfs_iocb_iter_read
                filemap_read
                  filemap_get_pages  <= detect signal
              erofs_fileio_ki_complete  <= set all folios uptodate

This patch addresses this by setting short read bio with an error
directly.

Fixes: bc804a8d ("erofs: handle end of filesystem properly for file-backed mounts")
Reported-by: default avatarchenguanyou <chenguanyou@xiaomi.com>
Signed-off-by: default avatarYunlei He <heyunlei@xiaomi.com>
Signed-off-by: default avatarSheng Yong <shengyong1@xiaomi.com>
Reviewed-by: default avatarGao Xiang <hsiangkao@linux.alibaba.com>
Reviewed-by: default avatarChao Yu <chao@kernel.org>
Signed-off-by: default avatarGao Xiang <hsiangkao@linux.alibaba.com>
parent 1f318b96

fs/erofs/fileio.c

+2 −4
Original line number Diff line number Diff line
@@ -25,10 +25,8 @@ static void erofs_fileio_ki_complete(struct kiocb *iocb, long ret) 
			container_of(iocb, struct erofs_fileio_rq, iocb);

	struct folio_iter fi;



	if (ret >= 0 && ret != rq->bio.bi_iter.bi_size) {

		bio_advance(&rq->bio, ret);

		zero_fill_bio(&rq->bio);

	}

	if (ret >= 0 && ret != rq->bio.bi_iter.bi_size)

		ret = -EIO;

	if (!rq->bio.bi_end_io) {

		bio_for_each_folio_all(fi, &rq->bio) {

			DBG_BUGON(folio_test_uptodate(fi.folio));