On Tue, Oct 29, 2019 at 1:17 PM Chengguang Xu <cgxu519@xxxxxxxxxxxx> wrote: > > ---- 在 星期二, 2019-10-29 16:32:32 Amir Goldstein <amir73il@xxxxxxxxx> 撰写 ---- > > On Tue, Oct 29, 2019 at 7:57 AM Chengguang Xu <cgxu519@xxxxxxxxxxxx> wrote: > > > > > > > Can you please send the patch as plain/text. > > Your mailer has sent it with quoted printable encoding and git am > > fails to apply the patch: > > https://lore.kernel.org/fstests/20191029055713.28191-1-cgxu519@xxxxxxxxxxxx/raw > > > > Sorry for that, I'm not clear for the reason, so I send you the patch in attachment first. > FWIW I am using git-send-email... > > > > Make many small holes in 10M test file seems not very > > > helpful for test coverage and it takes too much time > > > on creating test files. So in order to improve test > > > speed we adjust test file size to (10 * iosize) for > > > iosize aligned hole files meanwhile add more test > > > patterns for small random holes and small empty file. > > > > > > Signed-off-by: Chengguang Xu <cgxu519@xxxxxxxxxxxx> > > > --- > > > tests/overlay/066 | 89 +++++++++++++++++++++++++++++++++++------------ > > > 1 file changed, 67 insertions(+), 22 deletions(-) > > > > > > diff --git a/tests/overlay/066 b/tests/overlay/066 > > > index 285a5aff..fb8f5e5c 100755 > > > --- a/tests/overlay/066 > > > +++ b/tests/overlay/066 > > > @@ -40,49 +40,82 @@ _require_scratch > > > # Remove all files from previous tests > > > _scratch_mkfs > > > > > > -# We have totally 14 test files in this test. > > > +# We have totally 16 test files in this test. > > > # The detail as below: > > > -# 1 empty file(10M) + 2^0(K)..2^11(K) hole size files(each 10M) + 1 random hole size file(100M). > > > +# 1 small empty file 4K > > > +# 1 big empty file 4M > > > +# 1 small random hole file 10M > > > +# 1 big random hole file 100M > > > +# > > > +# 12 files with variant iosize aligned holes. > > > +# 2^0(K)..2^11(K) hole size files(file size = 10 * iosize) > > > # > > > # Considering both upper and lower fs will fill zero when copy-up > > > # hole area in the file, this test at least requires double disk > > > # space of the sum of above test files' size. > > > > > > -_require_fs_space $OVL_BASE_SCRATCH_MNT $(((10*1024*13 + 100*1024*1) * 2)) > > > +_require_fs_space $OVL_BASE_SCRATCH_MNT $((((4) + (4096) + (10 * 1024) \ > > > + + (100 * 1024) + (10 * (1 + 2048) * 12 / 2)) * 2)) > > > > > > lowerdir=$OVL_BASE_SCRATCH_MNT/$OVL_LOWER > > > upperdir=$OVL_BASE_SCRATCH_MNT/$OVL_UPPER > > > testfile="copyup_sparse_test" > > > > > > -# Create a completely empty hole file(10M). > > > -file_size=10240 > > > -$XFS_IO_PROG -fc "truncate ${file_size}K" "${lowerdir}/${testfile}_empty_holefile" \ > > > +# Create a small completely empty hole file(4K). > > > +file_size=4 > > > +$XFS_IO_PROG -fc "truncate ${file_size}K" "${lowerdir}/${testfile}_empty_small_holefile" \ > > > + >>$seqres.full > > > + > > > +# Create a big completely empty hole file(4M). > > > +file_size=4096 > > > +$XFS_IO_PROG -fc "truncate ${file_size}K" "${lowerdir}/${testfile}_empty_big_holefile" \ > > > >>$seqres.full > > > > > > -# Create 2^0(K)..2^11(K) hole size test files(each 10M). > > > +# Create 2^0(K)..2^11(K) hole size test files(file size = 10 * iosize). > > > # > > > # The pattern is like below, both hole and data are equal to > > > # iosize except last hole. > > > # > > > # |-- hole --|-- data --| ... |-- data --|-- hole --| > > > > > > -iosize=1 > > > +min_iosize=1 > > > max_iosize=2048 > > > -file_size=10240 > > > -max_pos=`expr $file_size - $max_iosize` > > > +iosize=$min_iosize > > > > > > while [ $iosize -le $max_iosize ]; do > > > + file_size=$(($iosize * 10)) > > > + max_pos=$(($file_size - $iosize)) > > > pos=$iosize > > > $XFS_IO_PROG -fc "truncate ${file_size}K" \ > > > "${lowerdir}/${testfile}_iosize${iosize}K_holefile" >>$seqres.full > > > while [ $pos -lt $max_pos ]; do > > > $XFS_IO_PROG -fc "pwrite ${pos}K ${iosize}K" \ > > > "${lowerdir}/${testfile}_iosize${iosize}K_holefile" >>$seqres.full > > > - pos=`expr $pos + $iosize + $iosize` > > > + pos=$(($pos + $iosize * 2)) > > > done > > > - iosize=`expr $iosize + $iosize` > > > + iosize=$(($iosize * 2)) > > > done > > > > > > +# Create test file with many random small holes(hole size is between 4K and 512K), > > > +# total file size is 10M. > > > + > > > +pos=4 > > > +max_pos=9216 > > > +file_size=10240 > > > +min_hole=4 > > > +max_hole=512 > > > + > > > +$XFS_IO_PROG -fc "truncate ${file_size}K" "${lowerdir}/${testfile}_random_small_holefile" \ > > > + >>$seqres.full > > > + > > > +while [ $pos -le $max_pos ]; do > > > + iosize=$(($RANDOM % ($max_hole - $min_hole) + $min_hole)) > > > + $XFS_IO_PROG -fc "pwrite ${pos}K ${iosize}K" \ > > > + "${lowerdir}/${testfile}_random_small_holefile" >>$seqres.full > > > + pos=$(($pos + $iosize * 2)) > > > +done > > > + > > > + > > > # Create test file with many random holes(hole size is between 1M and 5M), > > > # total file size is 100M. > > > > > > @@ -92,14 +125,14 @@ file_size=102400 > > > min_hole=1024 > > > max_hole=5120 > > > > > > -$XFS_IO_PROG -fc "truncate ${file_size}K" "${lowerdir}/${testfile}_random_holefile" \ > > > +$XFS_IO_PROG -fc "truncate ${file_size}K" "${lowerdir}/${testfile}_random_big_holefile" \ > > > >>$seqres.full > > > > > > while [ $pos -le $max_pos ]; do > > > iosize=$(($RANDOM % ($max_hole - $min_hole) + $min_hole)) > > > $XFS_IO_PROG -fc "pwrite ${pos}K ${iosize}K" \ > > > - "${lowerdir}/${testfile}_random_holefile" >>$seqres.full > > > - pos=`expr $pos + $iosize + $iosize` > > > + "${lowerdir}/${testfile}_random_big_holefile" >>$seqres.full > > > + pos=$(($pos + $iosize * 2)) > > > done > > > > > > _scratch_mount > > > @@ -112,18 +145,30 @@ done > > > echo "Silence is golden" > > > > > > # Check all copy-up files in upper layer. > > > -iosize=1 > > > -while [ $iosize -le 2048 ]; do > > > +min_iosize=1 > > > +max_iosize=2048 > > > > My intention was that you use those "constants" defined above when creating > > the files, not that you re-define them when verifying the files. > > Maybe there is a risk that iosize is changed unexpectedly between creating first test files > and before verifying them in the future. Now I think better solution is just compare all > test files in lower/upper dir by one diff command. > Agreed. diff -qr "${lowerdir}/" "${upperdir}/" Would be enough and provide enough details in 066.bad should the diff fail Thanks, Amir.
