ext4: lock block groups when initializing
I noticed when filling a 1T filesystem with 4 threads using the
fs_mark benchmark:
fs_mark -d /mnt/test -D 256 -n 100000 -t 4 -s 20480 -F -S 0
that I occasionally got checksum mismatch errors:
EXT4-fs error (device sdb): ext4_init_inode_bitmap: Checksum bad for group 6935
etc. I'd reliably get 4-5 of them during the run.
It appears that the problem is likely a race to init the bg's
when the uninit_bg feature is enabled.
With the patch below, which adds sb_bgl_locking around initialization,
I was able to complete several runs with no errors or warnings.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 386cb79..1ae5004 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -321,12 +321,15 @@
if (bh_uptodate_or_lock(bh))
return bh;
+ spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group));
if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
ext4_init_block_bitmap(sb, bh, block_group, desc);
set_buffer_uptodate(bh);
unlock_buffer(bh);
+ spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
return bh;
}
+ spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
if (bh_submit_read(bh) < 0) {
put_bh(bh);
ext4_error(sb, __func__,
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 09cdcd5..655e760 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -118,12 +118,15 @@
if (bh_uptodate_or_lock(bh))
return bh;
+ spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group));
if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) {
ext4_init_inode_bitmap(sb, bh, block_group, desc);
set_buffer_uptodate(bh);
unlock_buffer(bh);
+ spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
return bh;
}
+ spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
if (bh_submit_read(bh) < 0) {
put_bh(bh);
ext4_error(sb, __func__,
@@ -735,7 +738,7 @@
/* When marking the block group with
* ~EXT4_BG_INODE_UNINIT we don't want to depend
- * on the value of bg_itable_unsed even though
+ * on the value of bg_itable_unused even though
* mke2fs could have initialized the same for us.
* Instead we calculated the value below
*/
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 8d141a2..4258d32 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -787,13 +787,16 @@
if (bh_uptodate_or_lock(bh[i]))
continue;
+ spin_lock(sb_bgl_lock(EXT4_SB(sb), first_group + i));
if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
ext4_init_block_bitmap(sb, bh[i],
first_group + i, desc);
set_buffer_uptodate(bh[i]);
unlock_buffer(bh[i]);
+ spin_unlock(sb_bgl_lock(EXT4_SB(sb), first_group + i));
continue;
}
+ spin_unlock(sb_bgl_lock(EXT4_SB(sb), first_group + i));
get_bh(bh[i]);
bh[i]->b_end_io = end_buffer_read_sync;
submit_bh(READ, bh[i]);
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 876e1c6..511997e 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1621,6 +1621,7 @@
"(block %llu)!", i, inode_table);
return 0;
}
+ spin_lock(sb_bgl_lock(sbi, i));
if (!ext4_group_desc_csum_verify(sbi, i, gdp)) {
printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: "
"Checksum for group %lu failed (%u!=%u)\n",
@@ -1629,6 +1630,7 @@
if (!(sb->s_flags & MS_RDONLY))
return 0;
}
+ spin_unlock(sb_bgl_lock(sbi, i));
if (!flexbg_flag)
first_block += EXT4_BLOCKS_PER_GROUP(sb);
}