From: Andreas Dilger <adilger@dilger.ca>
To: Jan Kara <jack@suse.cz>
Cc: Ted Tso <tytso@mit.edu>, linux-ext4@vger.kernel.org
Subject: Re: [PATCH 6/7] tests: Add test to excercise indexed directories with metadata_csum
Date: Tue, 18 Feb 2020 13:34:25 -0700 [thread overview]
Message-ID: <8A45678F-C17D-417C-9BF0-9D123C846C58@dilger.ca> (raw)
In-Reply-To: <20200213101602.29096-7-jack@suse.cz>
[-- Attachment #1: Type: text/plain, Size: 5363 bytes --]
On Feb 13, 2020, at 3:16 AM, Jan Kara <jack@suse.cz> wrote:
>
> Indexed directories have somewhat different format when metadata_csum is
> enabled. Add test to excercise linking in indexed directories and e2fsck
> rehash code in this case.
>
> Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Andreas Dilger <adilger@dilger.ca>
> ---
> tests/f_large_dir_csum/expect | 32 ++++++++++++++
> tests/f_large_dir_csum/is_slow_test | 0
> tests/f_large_dir_csum/name | 1 +
> tests/f_large_dir_csum/script | 84 +++++++++++++++++++++++++++++++++++++
> 4 files changed, 117 insertions(+)
> create mode 100644 tests/f_large_dir_csum/expect
> create mode 100644 tests/f_large_dir_csum/is_slow_test
> create mode 100644 tests/f_large_dir_csum/name
> create mode 100644 tests/f_large_dir_csum/script
>
> diff --git a/tests/f_large_dir_csum/expect b/tests/f_large_dir_csum/expect
> new file mode 100644
> index 000000000000..aa9f33f1d25d
> --- /dev/null
> +++ b/tests/f_large_dir_csum/expect
> @@ -0,0 +1,32 @@
> +Creating filesystem with 31002 1k blocks and 64 inodes
> +Superblock backups stored on blocks:
> + 8193, 24577
> +
> +Allocating group tables: \b\b\bdone
> +Writing inode tables: \b\b\bdone
> +Writing superblocks and filesystem accounting information: \b\b\bdone
> +
> +Pass 1: Checking inodes, blocks, and sizes
> +Pass 2: Checking directory structure
> +Pass 3: Checking directory connectivity
> +Pass 3A: Optimizing directories
> +Pass 4: Checking reference counts
> +Inode 13 ref count is 1, should be 5. Fix? yes
> +
> +Pass 5: Checking group summary information
> +
> +test.img: ***** FILE SYSTEM WAS MODIFIED *****
> +test.img: 13/64 files (0.0% non-contiguous), 766/31002 blocks
> +Exit status is 1
> +Pass 1: Checking inodes, blocks, and sizes
> +Pass 2: Checking directory structure
> +Pass 3: Checking directory connectivity
> +Pass 3A: Optimizing directories
> +Pass 4: Checking reference counts
> +Inode 13 ref count is 5, should be 46504. Fix? yes
> +
> +Pass 5: Checking group summary information
> +
> +test.img: ***** FILE SYSTEM WAS MODIFIED *****
> +test.img: 13/64 files (0.0% non-contiguous), 16390/31002 blocks
> +Exit status is 1
> diff --git a/tests/f_large_dir_csum/is_slow_test b/tests/f_large_dir_csum/is_slow_test
> new file mode 100644
> index 000000000000..e69de29bb2d1
> diff --git a/tests/f_large_dir_csum/name b/tests/f_large_dir_csum/name
> new file mode 100644
> index 000000000000..2b37c8c21f79
> --- /dev/null
> +++ b/tests/f_large_dir_csum/name
> @@ -0,0 +1 @@
> +optimize 3 level htree directories with metadata checksums
> diff --git a/tests/f_large_dir_csum/script b/tests/f_large_dir_csum/script
> new file mode 100644
> index 000000000000..286a965d5e6a
> --- /dev/null
> +++ b/tests/f_large_dir_csum/script
> @@ -0,0 +1,84 @@
> +OUT=$test_name.log
> +EXP=$test_dir/expect
> +E2FSCK=../e2fsck/e2fsck
> +
> +NAMELEN=255
> +DIRENT_SZ=8
> +BLOCKSZ=1024
> +INODESZ=128
> +CSUM_SZ=8
> +CSUM_TAIL_SZ=12
> +DIRENT_PER_LEAF=$(((BLOCKSZ - CSUM_TAIL_SZ) / (NAMELEN + DIRENT_SZ)))
> +HEADER=32
> +INDEX_SZ=8
> +INDEX_L1=$(((BLOCKSZ - HEADER - CSUM_SZ) / INDEX_SZ))
> +INDEX_L2=$(((BLOCKSZ - DIRENT_SZ - CSUM_SZ) / INDEX_SZ))
> +DIRBLK=$((3 + INDEX_L1 * INDEX_L2))
> +ENTRIES=$((DIRBLK * DIRENT_PER_LEAF))
> +# directory leaf blocks - get twice as much because the leaves won't be full
> +# and there are also other filesystem blocks.
> +FSIZE=$((DIRBLK * 2))
> +
> +$MKE2FS -b 1024 -O extents,64bit,large_dir,uninit_bg,metadata_csum -N 50 \
> + -I $INODESZ -F $TMPFILE $FSIZE > $OUT.new 2>&1
> +RC=$?
> +if [ $RC -eq 0 ]; then
> +{
> + # First some initial fs setup to create indexed dir
> + echo "mkdir /foo"
> + echo "cd /foo"
> + touch $TMPFILE.tmp
> + echo "write $TMPFILE.tmp foofile"
> + i=0
> + while test $i -lt $DIRENT_PER_LEAF ; do
> + printf "ln foofile f%0254u\n" $i
> + i=$((i + 1));
> + done
> + echo "expand ./"
> + printf "ln foofile f%0254u\n" $i
> +} | $DEBUGFS -w $TMPFILE > /dev/null 2>> $OUT.new
> + RC=$?
> + # e2fsck should optimize the dir to become indexed
> + $E2FSCK -yfD $TMPFILE >> $OUT.new 2>&1
> + status=$?
> + echo Exit status is $status >> $OUT.new
> +fi
> +
> +if [ $RC -eq 0 ]; then
> +{
> + START=$SECONDS
> + i=$(($DIRENT_PER_LEAF+1))
> + last=$i
> + echo "cd /foo"
> + while test $i -lt $ENTRIES ; do
> + ELAPSED=$((SECONDS - START))
> + if test $((i % 5000)) -eq 0 -a $ELAPSED -gt 10; then
> + RATE=$(((i - last) / ELAPSED))
> + echo "$test_name: $i/$ENTRIES links, ${ELAPSED}s @ $RATE/s" >&2
> + START=$SECONDS
> + last=$i
> + fi
> + printf "ln foofile f%0254u\n" $i
> + i=$((i + 1))
> + done
> +} | $DEBUGFS -w $TMPFILE > /dev/null 2>> $OUT.new
> + RC=$?
> +fi
> +
> +if [ $RC -eq 0 ]; then
> + $E2FSCK -yfD $TMPFILE >> $OUT.new 2>&1
> + status=$?
> + echo Exit status is $status >> $OUT.new
> + sed -f $cmd_dir/filter.sed -e "s;$TMPFILE;test.img;" $OUT.new > $OUT
> + rm -f $OUT.new
> +
> + cmp -s $OUT $EXP
> + RC=$?
> +fi
> +if [ $RC -eq 0 ]; then
> + echo "$test_name: $test_description: ok"
> + touch $test_name.ok
> +else
> + echo "$test_name: $test_description: failed"
> + diff -u $EXP $OUT > $test_name.failed
> +fi
> --
> 2.16.4
>
Cheers, Andreas
[-- Attachment #2: Message signed with OpenPGP --]
[-- Type: application/pgp-signature, Size: 873 bytes --]
next prev parent reply other threads:[~2020-02-18 20:34 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-13 10:15 [PATCH 0/7 v2] e2fsprogs: Better handling of indexed directories Jan Kara
2020-02-13 10:15 ` [PATCH 1/7] e2fsck: Clarify overflow link count error message Jan Kara
2020-02-14 19:27 ` Andreas Dilger
2020-03-07 18:52 ` Theodore Y. Ts'o
2020-02-13 10:15 ` [PATCH 2/7] e2fsck: Fix indexed dir rehash failure with metadata_csum enabled Jan Kara
2020-02-14 19:28 ` Andreas Dilger
2020-03-07 23:17 ` Theodore Y. Ts'o
2020-03-16 9:30 ` Jan Kara
2020-02-13 10:15 ` [PATCH 3/7] ext2fs: Update allocation info earlier in ext2fs_mkdir() and ext2fs_symlink() Jan Kara
2020-02-14 19:37 ` Andreas Dilger
2020-03-08 0:02 ` Theodore Y. Ts'o
2020-03-08 2:20 ` Theodore Y. Ts'o
2020-03-15 16:15 ` Theodore Y. Ts'o
2020-03-16 9:32 ` Jan Kara
2020-02-13 10:15 ` [PATCH 4/7] ext2fs: Implement dir entry creation in htree directories Jan Kara
2020-03-15 16:43 ` Theodore Y. Ts'o
2020-02-13 10:16 ` [PATCH 5/7] tests: Modify f_large_dir test to excercise indexed dir handling Jan Kara
2020-02-18 20:29 ` Andreas Dilger
2020-03-15 16:43 ` Theodore Y. Ts'o
2020-02-13 10:16 ` [PATCH 6/7] tests: Add test to excercise indexed directories with metadata_csum Jan Kara
2020-02-18 20:34 ` Andreas Dilger [this message]
2020-02-13 10:16 ` [PATCH 7/7] tune2fs: Update dir checksums when clearing dir_index feature Jan Kara
2020-02-18 20:50 ` Andreas Dilger
2020-02-19 10:23 ` Jan Kara
2020-03-15 17:15 ` Theodore Y. Ts'o
2020-03-16 0:11 ` Andreas Dilger
2020-03-16 9:27 ` Jan Kara
2020-03-26 14:27 ` Jan Kara
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=8A45678F-C17D-417C-9BF0-9D123C846C58@dilger.ca \
--to=adilger@dilger.ca \
--cc=jack@suse.cz \
--cc=linux-ext4@vger.kernel.org \
--cc=tytso@mit.edu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).