f2fs: fix panic of IO alignment feature

Since 07173c3ec276 ("block: enable multipage bvecs"), one bio vector
can store multi pages, so that we can not calculate max IO size of
bio as PAGE_SIZE * bio->bi_max_vecs. However IO alignment feature of
f2fs always has that assumption, so finally, it may cause panic during
IO submission as below stack.

 kernel BUG at fs/f2fs/data.c:317!
 RIP: 0010:__submit_merged_bio+0x8b0/0x8c0
 Call Trace:
  f2fs_submit_page_write+0x3cd/0xdd0
  do_write_page+0x15d/0x360
  f2fs_outplace_write_data+0xd7/0x210
  f2fs_do_write_data_page+0x43b/0xf30
  __write_data_page+0xcf6/0x1140
  f2fs_write_cache_pages+0x3ba/0xb40
  f2fs_write_data_pages+0x3dd/0x8b0
  do_writepages+0xbb/0x1e0
  __writeback_single_inode+0xb6/0x800
  writeback_sb_inodes+0x441/0x910
  wb_writeback+0x261/0x650
  wb_workfn+0x1f9/0x7a0
  process_one_work+0x503/0x970
  worker_thread+0x7d/0x820
  kthread+0x1ad/0x210
  ret_from_fork+0x35/0x40

This patch adds one extra condition to check left space in bio while
trying merging page to bio, to avoid panic.

This bug was reported in bugzilla:

https://bugzilla.kernel.org/show_bug.cgi?id=204043

Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 0686306..5bce200 100644 (file)
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -503,6 +503,16 @@ static bool io_is_mergeable(struct f2fs_sb_info *sbi, struct bio *bio,
                                        block_t last_blkaddr,
                                        block_t cur_blkaddr)
 {
+       if (F2FS_IO_ALIGNED(sbi) && (fio->type == DATA || fio->type == NODE)) {
+               unsigned int filled_blocks =
+                               F2FS_BYTES_TO_BLK(bio->bi_iter.bi_size);
+               unsigned int io_size = F2FS_IO_SIZE(sbi);
+               unsigned int left_vecs = bio->bi_max_vecs - bio->bi_vcnt;
+
+               /* IOs in bio is aligned and left space of vectors is not enough */
+               if (!(filled_blocks % io_size) && left_vecs < io_size)
+                       return false;
+       }
        if (!page_is_mergeable(sbi, bio, last_blkaddr, cur_blkaddr))
                return false;
        return io_type_is_mergeable(io, fio);

diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 78a1b87..720f2e6 100644 (file)
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -3202,7 +3202,7 @@ try_onemore:
        if (err)
                goto free_bio_info;
 
-       if (F2FS_IO_SIZE(sbi) > 1) {
+       if (F2FS_IO_ALIGNED(sbi)) {
                sbi->write_io_dummy =
                        mempool_create_page_pool(2 * (F2FS_IO_SIZE(sbi) - 1), 0);
                if (!sbi->write_io_dummy) {

diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h
index 6555990..52af9ac 100644 (file)
--- a/include/linux/f2fs_fs.h
+++ b/include/linux/f2fs_fs.h
@@ -41,6 +41,7 @@
 #define F2FS_IO_SIZE_BYTES(sbi)        (1 << (F2FS_OPTION(sbi).write_io_size_bits + 12)) /* B */
 #define F2FS_IO_SIZE_BITS(sbi) (F2FS_OPTION(sbi).write_io_size_bits) /* power of 2 */
 #define F2FS_IO_SIZE_MASK(sbi) (F2FS_IO_SIZE(sbi) - 1)
+#define F2FS_IO_ALIGNED(sbi)   (F2FS_IO_SIZE(sbi) > 1)
 
 /* This flag is used by node and meta inodes, and by recovery */
 #define GFP_F2FS_ZERO          (GFP_NOFS | __GFP_ZERO)