summaryrefslogtreecommitdiff
path: root/fs
diff options
context:
space:
mode:
authorWang Shilong <wangsl.fnst@cn.fujitsu.com>2014-04-01 14:01:43 +0400
committerChris Mason <clm@fb.com>2014-04-07 20:08:49 +0400
commit3b080b2564287be91605bfd1d5ee985696e61d3c (patch)
treea5a7b6a08703e71047193c41dd3b243ff09c0fea /fs
parent68bb462d42a963169bf7acbe106aae08c17129a5 (diff)
downloadlinux-3b080b2564287be91605bfd1d5ee985696e61d3c.tar.xz
Btrfs: scrub raid56 stripes in the right way
Steps to reproduce: # mkfs.btrfs -f /dev/sda[8-11] -m raid5 -d raid5 # mount /dev/sda8 /mnt # btrfs scrub start -BR /mnt # echo $? <--unverified errors make return value be 3 This is because we don't setup right mapping between physical and logical address for raid56, which makes checksum mismatch. But we will find everthing is fine later when rechecking using btrfs_map_block(). This patch fixed the problem by settuping right mappings and we only verify data stripes' checksums. Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com> Signed-off-by: Chris Mason <clm@fb.com>
Diffstat (limited to 'fs')
-rw-r--r--fs/btrfs/scrub.c108
1 files changed, 89 insertions, 19 deletions
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index db21a1360e13..aee909fc622b 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -2235,6 +2235,47 @@ behind_scrub_pages:
return 0;
}
+/*
+ * Given a physical address, this will calculate it's
+ * logical offset. if this is a parity stripe, it will return
+ * the most left data stripe's logical offset.
+ *
+ * return 0 if it is a data stripe, 1 means parity stripe.
+ */
+static int get_raid56_logic_offset(u64 physical, int num,
+ struct map_lookup *map, u64 *offset)
+{
+ int i;
+ int j = 0;
+ u64 stripe_nr;
+ u64 last_offset;
+ int stripe_index;
+ int rot;
+
+ last_offset = (physical - map->stripes[num].physical) *
+ nr_data_stripes(map);
+ *offset = last_offset;
+ for (i = 0; i < nr_data_stripes(map); i++) {
+ *offset = last_offset + i * map->stripe_len;
+
+ stripe_nr = *offset;
+ do_div(stripe_nr, map->stripe_len);
+ do_div(stripe_nr, nr_data_stripes(map));
+
+ /* Work out the disk rotation on this stripe-set */
+ rot = do_div(stripe_nr, map->num_stripes);
+ /* calculate which stripe this data locates */
+ rot += i;
+ stripe_index = do_div(rot, map->num_stripes);
+ if (stripe_index == num)
+ return 0;
+ if (stripe_index < num)
+ j++;
+ }
+ *offset = last_offset + j * map->stripe_len;
+ return 1;
+}
+
static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
struct map_lookup *map,
struct btrfs_device *scrub_dev,
@@ -2256,6 +2297,7 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
u64 physical;
u64 logical;
u64 logic_end;
+ u64 physical_end;
u64 generation;
int mirror_num;
struct reada_control *reada1;
@@ -2269,16 +2311,10 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
u64 extent_len;
struct btrfs_device *extent_dev;
int extent_mirror_num;
- int stop_loop;
-
- if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
- BTRFS_BLOCK_GROUP_RAID6)) {
- if (num >= nr_data_stripes(map)) {
- return 0;
- }
- }
+ int stop_loop = 0;
nstripes = length;
+ physical = map->stripes[num].physical;
offset = 0;
do_div(nstripes, map->stripe_len);
if (map->type & BTRFS_BLOCK_GROUP_RAID0) {
@@ -2296,6 +2332,11 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
} else if (map->type & BTRFS_BLOCK_GROUP_DUP) {
increment = map->stripe_len;
mirror_num = num % map->num_stripes + 1;
+ } else if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6)) {
+ get_raid56_logic_offset(physical, num, map, &offset);
+ increment = map->stripe_len * nr_data_stripes(map);
+ mirror_num = 1;
} else {
increment = map->stripe_len;
mirror_num = 1;
@@ -2319,7 +2360,15 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
* to not hold off transaction commits
*/
logical = base + offset;
-
+ physical_end = physical + nstripes * map->stripe_len;
+ if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6)) {
+ get_raid56_logic_offset(physical_end, num,
+ map, &logic_end);
+ logic_end += base;
+ } else {
+ logic_end = logical + increment * nstripes;
+ }
wait_event(sctx->list_wait,
atomic_read(&sctx->bios_in_flight) == 0);
scrub_blocked_if_needed(fs_info);
@@ -2328,7 +2377,7 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
key_start.objectid = logical;
key_start.type = BTRFS_EXTENT_ITEM_KEY;
key_start.offset = (u64)0;
- key_end.objectid = base + offset + nstripes * increment;
+ key_end.objectid = logic_end;
key_end.type = BTRFS_METADATA_ITEM_KEY;
key_end.offset = (u64)-1;
reada1 = btrfs_reada_add(root, &key_start, &key_end);
@@ -2338,7 +2387,7 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
key_start.offset = logical;
key_end.objectid = BTRFS_EXTENT_CSUM_OBJECTID;
key_end.type = BTRFS_EXTENT_CSUM_KEY;
- key_end.offset = base + offset + nstripes * increment;
+ key_end.offset = logic_end;
reada2 = btrfs_reada_add(csum_root, &key_start, &key_end);
if (!IS_ERR(reada1))
@@ -2356,11 +2405,17 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
/*
* now find all extents for each stripe and scrub them
*/
- logical = base + offset;
- physical = map->stripes[num].physical;
- logic_end = logical + increment * nstripes;
ret = 0;
- while (logical < logic_end) {
+ while (physical < physical_end) {
+ /* for raid56, we skip parity stripe */
+ if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6)) {
+ ret = get_raid56_logic_offset(physical, num,
+ map, &logical);
+ logical += base;
+ if (ret)
+ goto skip;
+ }
/*
* canceled?
*/
@@ -2504,15 +2559,29 @@ again:
scrub_free_csums(sctx);
if (extent_logical + extent_len <
key.objectid + bytes) {
- logical += increment;
- physical += map->stripe_len;
-
+ if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
+ BTRFS_BLOCK_GROUP_RAID6)) {
+ /*
+ * loop until we find next data stripe
+ * or we have finished all stripes.
+ */
+ do {
+ physical += map->stripe_len;
+ ret = get_raid56_logic_offset(
+ physical, num,
+ map, &logical);
+ logical += base;
+ } while (physical < physical_end && ret);
+ } else {
+ physical += map->stripe_len;
+ logical += increment;
+ }
if (logical < key.objectid + bytes) {
cond_resched();
goto again;
}
- if (logical >= logic_end) {
+ if (physical >= physical_end) {
stop_loop = 1;
break;
}
@@ -2521,6 +2590,7 @@ next:
path->slots[0]++;
}
btrfs_release_path(path);
+skip:
logical += increment;
physical += map->stripe_len;
spin_lock(&sctx->stat_lock);