On Sun, Nov 06, 2022 at 08:10:31PM +0800, Zorro Lang wrote: > > Thanks Ted, actually I'm going to merge this patch (with g/457 fix by myself) > this week. > > But looks like you missed the change on g/457 (might forgot to commit). Anyway, > I think it's not worth wasting one more week for this small change, I'll help > to change g/457 when I merge this patch. Argh, yeah, sorry, I forgot to do the "git add -u". Anyway, here's the leftover change that was in my git working directory, if it saves you 5 seconds or so. :-) BTW, I noticed there were a number of fsx --replay-ops invocations where we could potentially add the $FSX_AVOID. OTOH, it would probably make those tests completely pointless, so it might be easier just for the test runners to relay on a group-based exclusion in those cases. What do you think? - Ted diff --git a/tests/generic/457 b/tests/generic/457 index da75798f1..ca0f5e622 100755 --- a/tests/generic/457 +++ b/tests/generic/457 @@ -83,7 +83,7 @@ FSX_OPTS="-N $NUM_OPS -d -k -P $SANITY_DIR -i $LOGWRITES_DMDEV" for j in `seq 0 $((NUM_FILES-1))`; do # clone the clone from prev iteration which may have already mutated _cp_reflink $SCRATCH_MNT/testfile$((j-1)) $SCRATCH_MNT/testfile$j - run_check $here/ltp/fsx $FSX_OPTS -S 0 -j $j $SCRATCH_MNT/testfile$j & + run_check $here/ltp/fsx $FSX_OPTS $FSX_AVOID -S 0 -j $j $SCRATCH_MNT/testfile$j & done wait
