> On Jan 7, 2025, at 10:12, Darrick J. Wong <djwong@xxxxxxxxxx> wrote: > > On Mon, Jan 06, 2025 at 10:01:04PM +0800, Su Yue wrote: >> mkfs.ocfs2 is using 32bit journal as default. >> For 16T size device support, '-J block64' should be used. >> >> Signed-off-by: Su Yue <glass.su@xxxxxxxx> >> --- >> tests/generic/620 | 3 +++ >> 1 file changed, 3 insertions(+) >> >> diff --git a/tests/generic/620 b/tests/generic/620 >> index 3f1ce45a55fd..60e5a2cacdda 100755 >> --- a/tests/generic/620 >> +++ b/tests/generic/620 >> @@ -41,6 +41,9 @@ sectors=$((2*1024*1024*1024*17)) >> chunk_size=128 >> >> _dmhugedisk_init $sectors $chunk_size >> + >> +[ "$FSTYP" = "ocfs2" ] && MKFS_OPTIONS="$MKFS_OPTIONS -J block64" > > Won't mkfs.ocfs2 turn on block64 on a > 16T volume without prompting? > It won’t. Only error will be printed: ERROR: jbd can only store block numbers in 32 bits. /dev/mapper/huge-test.620 can hold 4563402752 blocks which overflows this limit. If you have a new enough Ocfs2 with JBD2 support, you can try formatting with the "-Jblock64" option to turn on support for this size block device. It’s 2025 now. Maybe it’s time to make mkfs.ocfs2 turn on block64 if device size > 16T. Maintainers may know some stories. Joseph? — Su > --D > >> + >> _mkfs_dev $DMHUGEDISK_DEV >> _mount $DMHUGEDISK_DEV $SCRATCH_MNT || _fail "mount failed for $DMHUGEDISK_DEV $SCRATCH_MNT" >> testfile=$SCRATCH_MNT/testfile-$seq >> -- >> 2.47.1
