@@ -64,7 +64,7 @@ _init_flakey
_mount_flakey
# Create the file first.
-$XFS_IO_PROG -f -c "pwrite -S 0xff 0 64K" $SCRATCH_MNT/foo | _filter_xfs_io
+$XFS_IO_PROG -f -c "pwrite -S 0xff 0 256K" $SCRATCH_MNT/foo | _filter_xfs_io
# Now sync the file data to disk using 'sync' and not an fsync. This is because
# in btrfs the first fsync clears the btrfs inode full fsync flag, which must
@@ -80,11 +80,11 @@ sync
# This second msync() used to be a no-op for that btrfs bug (and the first fsync
# didn't log the last 4Kb extent as expected too).
$XFS_IO_PROG \
- -c "mmap -w 0 64K" \
+ -c "mmap -w 0 256K" \
-c "mwrite -S 0xaa 0 4K" \
- -c "mwrite -S 0xbb 60K 4K" \
- -c "msync -s 0K 16K" \
- -c "msync -s 32K 32K" \
+ -c "mwrite -S 0xbb 252K 4K" \
+ -c "msync -s 0K 64K" \
+ -c "msync -s 192K 64K" \
-c "munmap" \
$SCRATCH_MNT/foo | _filter_xfs_io
@@ -1,19 +1,19 @@
QA output created by 325
-wrote 65536/65536 bytes at offset 0
+wrote 262144/262144 bytes at offset 0
XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
File content before crash/reboot:
0000000 aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa
*
0010000 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
*
-0170000 bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb
+0770000 bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb
*
-0200000
+1000000
File content after crash/reboot and fs mount:
0000000 aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa aa
*
0010000 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
*
-0170000 bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb
+0770000 bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb bb
*
-0200000
+1000000
The test case passes 32K as the offset value to msync. This fails on machines with 64K page size. Fix this by creating a larger file and passing offset values which are multiples of 64K. Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com> --- tests/generic/325 | 10 +++++----- tests/generic/325.out | 10 +++++----- 2 files changed, 10 insertions(+), 10 deletions(-)