[BUG]
When running btrfs/215 with 64K page size, 4K sectorsize (subpage RW
support), it fails with the following error:
btrfs/215 [failed, exit status 1]- output mismatch (see ~/xfstests-dev/results//btrfs/215.out.bad)
--- tests/btrfs/215.out 2021-03-19 16:34:26.069634953 +0800
+++ ~/xfstests-dev/results//btrfs/215.out.bad 2021-05-17 16:52:34.743514224 +0800
@@ -1,2 +1,3 @@
QA output created by 215
-Silence is golden
+Errors: 8 expected: 2
+(see ~/xfstests-dev/results//btrfs/215.full for details)
...
(Run 'diff -u ~/xfstests-dev/tests/btrfs/215.out ~/xfstests-dev/results//btrfs/215.out.bad' to see the entire diff)
[CAUSE]
For subpage case, btrfs still tries to read the full page, other than
read just one sector for PAGE_SIZE == sectorsize case.
This means for the 2 sectors corrupted case, since they are in the same
page, all the errors will be reported.
[FIX]
Change the following values:
- filesize
Now it's 8 * pagesize.
- expected error number
Now it's 2 * sectors_per_page or 6 * sectors_per_page.
Signed-off-by: Qu Wenruo <wqu@xxxxxxxx>
---
Note: The btrfs subpage full data RW support is not yet merged into upstream,
but most of its preparation patches are already merged, and it can
already pass most fstests.
The out-of-tree branch can be fetched from my git repo:
https://github.com/adam900710/linux/tree/subpage
Some failures are just false alerts, like btrfs/215 which always
believes sectorsize == PAGE_SIZE for btrfs, thus it's worthy fixing before
bothering testers when the RW support get merged.
---
tests/btrfs/215 | 25 +++++++++++++++++++------
1 file changed, 19 insertions(+), 6 deletions(-)
diff --git a/tests/btrfs/215 b/tests/btrfs/215
index 748287e7..c4959061 100755
--- a/tests/btrfs/215
+++ b/tests/btrfs/215
@@ -46,8 +46,21 @@ _scratch_mkfs > /dev/null
# blobk group
_scratch_mount -o nospace_cache
+pagesize=$(get_page_size)
blocksize=$(_get_block_size $SCRATCH_MNT)
-filesize=$((8*$blocksize))
+
+# For subpage case, since we still do read in full page size, if have 8 corrupted
+# sectors in one page, then even we just try to read one sector of that page,
+# all 8 corruption will be reported.
+# So here we chose the filesize using page size.
+filesize=$((8*$pagesize))
+if [ $blocksize -le $pagesize ]; then
+ sectors_per_page=$(($pagesize / $blocksize))
+else
+ # We don't support multi-page sectorsize yet
+ _notrun "this test doesn't support sectorsize $blocksize with page size $pagesize yet"
+fi
+
uuid=$(findmnt -n -o UUID "$SCRATCH_MNT")
if [ ! -e /sys/fs/btrfs/$uuid/bdi ]; then
@@ -64,24 +77,24 @@ echo "logical = $logical_extent physical=$physical_extent" >> $seqres.full
# corrupt first 4 blocks of file
_scratch_unmount
-$XFS_IO_PROG -d -c "pwrite -S 0xaa -b $blocksize $physical_extent $((4*$blocksize))" $SCRATCH_DEV > /dev/null
+$XFS_IO_PROG -d -c "pwrite -S 0xaa -b $pagesize $physical_extent $((4 * $pagesize))" $SCRATCH_DEV > /dev/null
_scratch_mount
# disable readahead to avoid skewing the counter
echo 0 > /sys/fs/btrfs/$uuid/bdi/read_ahead_kb
-# buffered reads whould result in 2 errors since readahead code always submits
-# at least 1 page worth of IO and it will be counted as an error as well
+# buffered reads whould result in 2 * sectors_per_page errors since readahead code always submits
+# at least 1 page worth of IO and it will be counted as error(s) as well
$XFS_IO_PROG -c "pread -b $filesize 0 $filesize" "$SCRATCH_MNT/foobar" > /dev/null 2>&1
errs=$($BTRFS_UTIL_PROG device stats $SCRATCH_DEV | awk '/corruption_errs/ { print $2 }')
-if [ $errs -ne 2 ]; then
+if [ $errs -ne $((2 * $sectors_per_page)) ]; then
_fail "Errors: $errs expected: 2"
fi
# DIO does check every sector
$XFS_IO_PROG -d -c "pread -b $filesize 0 $filesize" "$SCRATCH_MNT/foobar" > /dev/null 2>&1
errs=$($BTRFS_UTIL_PROG device stats $SCRATCH_DEV | awk '/corruption_errs/ { print $2 }')
-if [ $errs -ne 6 ]; then
+if [ $errs -ne $((6 * $sectors_per_page)) ]; then
_fail "Errors: $errs expected: 6"
fi
--
2.31.1
