On Wed, Feb 5, 2020 at 2:02 AM Darrick J. Wong <darrick.wong@xxxxxxxxxx> wrote: > > From: Darrick J. Wong <darrick.wong@xxxxxxxxxx> > > If xfs_io's utimes command cannot interpret the arguments that are given > to it, it will print out "Bad value for [am]time". Detect when this > happens and drop the file out of the test entirely. > > This is particularly noticeable on 32-bit platforms and the largest > timestamp seconds supported by the filesystem is INT_MAX. In this case, > the maximum value we can cram into tv_sec is INT_MAX, and there is no > way to actually test setting a timestamp of INT_MAX + 1 to test the > clamping. > > Signed-off-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx> > --- > tests/generic/402 | 11 ++++++++++- > 1 file changed, 10 insertions(+), 1 deletion(-) > > > diff --git a/tests/generic/402 b/tests/generic/402 > index 2a34d127..32988866 100755 > --- a/tests/generic/402 > +++ b/tests/generic/402 > @@ -63,10 +63,19 @@ run_test_individual() > # check if the time needs update > if [ $update_time -eq 1 ]; then > echo "Updating file: $file to timestamp $timestamp" >> $seqres.full > - $XFS_IO_PROG -f -c "utimes $timestamp 0 $timestamp 0" $file > + $XFS_IO_PROG -f -c "utimes $timestamp 0 $timestamp 0" $file >> $tmp.utimes 2>&1 Maybe use > instead of >> to be safe. Also I would feel more comfortable if we special case the 0 timestamp against being skipped, to be safe that we don't have a silent regression in xfs_io or something causing all files to be skipped. Otherwise: Reviewed-by: Amir Goldstein <amir73il@xxxxxxxxx> > + cat $tmp.utimes >> $seqres.full > + if grep -q "Bad value" "$tmp.utimes"; then > + rm -f $file $tmp.utimes > + return > + fi > + cat $tmp.utimes > + rm $tmp.utimes > if [ $? -ne 0 ]; then > echo "Failed to update times on $file" | tee -a $seqres.full > fi > + else > + test -f $file || return > fi > > tsclamp=$((timestamp<tsmin?tsmin:timestamp>tsmax?tsmax:timestamp)) >
