btrfs: raid56: make error_bitmap update atomic

In the rework of raid56 code, there is very limited concurrency in the
endio context.

Most of the work is done inside the sectors arrays, which different bios
will never touch the same sector.

But there is a concurrency here for error_bitmap. Both read and write
endio functions need to touch them, and we can have multiple write bios
touching the same error bitmap if they all hit some errors.

Here we fix the unprotected bitmap operation by going set_bit() in a
loop.

Since we have a very small ceiling of the sectors (at most 16 sectors),
such set_bit() in a loop should be very acceptable.

Fixes: 2942a50dea74 ("btrfs: raid56: introduce btrfs_raid_bio::error_bitmap")
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>

diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index f35898a78b16b8bc72bcce137ca678bfb35011c3..ff4b1d583788c81b224cc3eb108d8d3c94a5d4ad 100644 (file)
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -1426,12 +1426,20 @@ static void rbio_update_error_bitmap(struct btrfs_raid_bio *rbio, struct bio *bi
        u32 bio_size = 0;
        struct bio_vec *bvec;
        struct bvec_iter_all iter_all;
+       int i;
 
        bio_for_each_segment_all(bvec, bio, iter_all)
                bio_size += bvec->bv_len;
 
-       bitmap_set(rbio->error_bitmap, total_sector_nr,
-                  bio_size >> rbio->bioc->fs_info->sectorsize_bits);
+       /*
+        * Since we can have multiple bios touching the error_bitmap, we cannot
+        * call bitmap_set() without protection.
+        *
+        * Instead use set_bit() for each bit, as set_bit() itself is atomic.
+        */
+       for (i = total_sector_nr; i < total_sector_nr +
+            (bio_size >> rbio->bioc->fs_info->sectorsize_bits); i++)
+               set_bit(i, rbio->error_bitmap);
 }
 
 /* Verify the data sectors at read time. */