From: Zorro Lang <zlang@redhat.com>
To: Yang Xu <xuyang2018.jy@cn.fujitsu.com>
Cc: fstests@vger.kernel.org, linux-xfs@vger.kernel.org
Subject: Re: [PATCH v3] xfs/191: update mkfs.xfs input results
Date: Thu, 26 Mar 2020 21:21:30 +0800 [thread overview]
Message-ID: <20200326132130.GO14282@dhcp-12-102.nay.redhat.com> (raw)
In-Reply-To: <44cc5352-b9b6-e927-0146-321026bb11cd@cn.fujitsu.com>
On Thu, Mar 26, 2020 at 05:11:37PM +0800, Yang Xu wrote:
>
> on 2020/03/26 13:39, Zorro Lang wrote:
> > On Wed, Jun 19, 2019 at 03:39:23PM +0800, Yang Xu wrote:
> > > Currently, on 5.2.0-rc4+ kernel, when I run xfs/191 with upstream
> > > xfsprogs, I get the following errors because mkfs.xfs binary has
> > > changed a lot.
> > >
> > > -------------------------
> > > pass -n size=2b /dev/sda11
> > > pass -d agsize=8192b /dev/sda11
> > > pass -d agsize=65536s /dev/sda11
> > > pass -d su=0,sw=64 /dev/sda11
> > > pass -d su=4096s,sw=64 /dev/sda11
> > > pass -d su=4096b,sw=64 /dev/sda11
> > > pass -l su=10b /dev/sda11
> > > fail -n log=15 /dev/sda11
> > > fail -r size=65536,rtdev=$fsimg /dev/sda11
> > > fail -r rtdev=$fsimg /dev/sda11
> > > fail -i log=10 /dev/sda11
> > > --------------------------
> > >
> > > "pass -d su=0,sw=64 /dev/sda11", expect fail, this behavior has been
> > > fixed by commit 16adcb88(mkfs: more sunit/swidth sanity checking).
> > >
> > > "fail -n log=15 /dev/sda11" "fail -i log=10 /dev/sda11", expect pass,
> > > this option has been removed since commit 2cf637c(mkfs: remove
> > > logarithm based CLI option).
> > >
> > > "fail -r size=65536,rtdev=$fsimg /dev/sda11" "fail -r rtdev=$fsimg
> > > /dev/sda11" works well if we disable reflink, fail if we enable
> > > reflink. It fails because reflink was not supported in realtime
> > > devices since commit bfa66ec.
> > >
> > > "b" or "s" suffix without specifying their size has been supported
> > > since xfsprogs v4.15.0-rc1.
> > >
> > > I change the expected result for compatibility with current xfsprogs
> > > and add rtdev test with reflink.
> > >
> > > Signed-off-by: Yang Xu <xuyang2018.jy@cn.fujitsu.com>
> > > ---
> >
> > I'd suggest to use a loop device to replace SCRATCH_DEV, to avoid some failures
> > caused by different stripe alignment of SCRATCH_DEV (refer to xfs/513). Or we
> > need to think about the stripe alignment in this case.
> I also met stripe size problem with log section in[1], but it failed on
> mount step. Can you give me a mkfs example and let me understand?
Sure.
# mkfs.xfs -f /dev/mapper/xxx-xfstest
meta-data=/dev/mapper/xxx-xfstest isize=512 agcount=16, agsize=409600 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1
data = bsize=4096 blocks=6553600, imaxpct=25
= sunit=64 swidth=64 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=5184, version=2
= sectsz=512 sunit=64 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
# ./check xfs/191-input-validation
FSTYP -- xfs (non-debug)
PLATFORM -- Linux/x86_64
MKFS_OPTIONS -- -f -bsize=4096 /dev/mapper/xxx-xfscratch
MOUNT_OPTIONS -- -o context=system_u:object_r:root_t:s0 /dev/mapper/xxx-xfscratch /mnt/scratch
xfs/191-input-validation 11s ... - output mismatch (see /home/xfstests-zlang/results//xfs/191-input-validation.out.bad)
--- tests/xfs/191-input-validation.out 2018-10-24 02:06:10.616609603 -0400
+++ /home/xfstests-zlang/results//xfs/191-input-validation.out.bad 2020-03-26 08:58:47.269671087 -0400
@@ -1,2 +1,11 @@
QA output created by 191-input-validation
silence is golden
+fail -d agsize=32m /dev/mapper/xxx-xfscratch
+fail -d agsize=32M /dev/mapper/xxx-xfscratch
+fail -d agsize=33554432 /dev/mapper/xxx-xfscratch
+fail -b size=4096 -d agsize=8192b /dev/mapper/xxx-xfscratch
+fail -d agsize=8192b /dev/mapper/xxx-xfscratch
...
(Run 'diff -u /home/xfstests-zlang/tests/xfs/191-input-validation.out /home/xfstests-zlang/results//xfs/191-input-validation.out.bad' to see the entire diff)
Ran: xfs/191-input-validation
Failures: xfs/191-input-validation
Failed 1 of 1 tests
# mkfs.xfs -f -d agsize=32m /dev/mapper/xxx-xfscratch
Due to stripe alignment, the internal log size (8192) is too large.
Must fit within an allocation group.
Usage: mkfs.xfs
...
...
All these failures gone after I turn to use /dev/loop0 to be SCRATCH_DEV.
# ./check xfs/191-input-validation
FSTYP -- xfs (non-debug)
PLATFORM -- Linux/x86_64
MKFS_OPTIONS -- -f -bsize=4096 /dev/loop0
MOUNT_OPTIONS -- -o context=system_u:object_r:root_t:s0 /dev/loop0 /mnt/scratch
xfs/191-input-validation 11s ... 10s
Ran: xfs/191-input-validation
Passed all 1 tests
Thanks,
Zorro
>
> [1]https://patchwork.kernel.org/patch/11393385/
>
> Best Regards
> Yang Xu
> >
> > Thanks,
> > Zorro
> >
> > > tests/xfs/191-input-validation | 39 ++++++++++++++++++++++------------
> > > 1 file changed, 25 insertions(+), 14 deletions(-)
> > >
> > > diff --git a/tests/xfs/191-input-validation b/tests/xfs/191-input-validation
> > > index b6658015..9f8de500 100755
> > > --- a/tests/xfs/191-input-validation
> > > +++ b/tests/xfs/191-input-validation
> > > @@ -31,11 +31,10 @@ _cleanup()
> > > # Modify as appropriate.
> > > _supported_fs xfs
> > > _supported_os Linux
> > > -_require_scratch
> > > +_require_scratch_nocheck
> > > _require_xfs_mkfs_validation
> > > -
> > > rm -f $seqres.full
> > > echo silence is golden
> > > @@ -112,10 +111,11 @@ do_mkfs_fail -b size=2b $SCRATCH_DEV
> > > do_mkfs_fail -b size=nfi $SCRATCH_DEV
> > > do_mkfs_fail -b size=4096nfi $SCRATCH_DEV
> > > do_mkfs_fail -n size=2s $SCRATCH_DEV
> > > -do_mkfs_fail -n size=2b $SCRATCH_DEV
> > > do_mkfs_fail -n size=nfi $SCRATCH_DEV
> > > do_mkfs_fail -n size=4096nfi $SCRATCH_DEV
> > > +do_mkfs_pass -n size=2b $SCRATCH_DEV
> > > +
> > > # bad label length
> > > do_mkfs_fail -L thisiswaytoolong $SCRATCH_DEV
> > > @@ -129,6 +129,8 @@ do_mkfs_pass -d agsize=32M $SCRATCH_DEV
> > > do_mkfs_pass -d agsize=1g $SCRATCH_DEV
> > > do_mkfs_pass -d agsize=$((32 * 1024 * 1024)) $SCRATCH_DEV
> > > do_mkfs_pass -b size=4096 -d agsize=8192b $SCRATCH_DEV
> > > +do_mkfs_pass -d agsize=8192b $SCRATCH_DEV
> > > +do_mkfs_pass -d agsize=65536s $SCRATCH_DEV
> > > do_mkfs_pass -d sectsize=512,agsize=65536s $SCRATCH_DEV
> > > do_mkfs_pass -s size=512 -d agsize=65536s $SCRATCH_DEV
> > > do_mkfs_pass -d noalign $SCRATCH_DEV
> > > @@ -136,7 +138,10 @@ do_mkfs_pass -d sunit=0,swidth=0 $SCRATCH_DEV
> > > do_mkfs_pass -d sunit=8,swidth=8 $SCRATCH_DEV
> > > do_mkfs_pass -d sunit=8,swidth=64 $SCRATCH_DEV
> > > do_mkfs_pass -d su=0,sw=0 $SCRATCH_DEV
> > > +do_mkfs_pass -d su=0,sw=64 $SCRATCH_DEV
> > > do_mkfs_pass -d su=4096,sw=1 $SCRATCH_DEV
> > > +do_mkfs_pass -d su=4096s,sw=64 $SCRATCH_DEV
> > > +do_mkfs_pass -d su=4096b,sw=64 $SCRATCH_DEV
> > > do_mkfs_pass -d su=4k,sw=1 $SCRATCH_DEV
> > > do_mkfs_pass -d su=4K,sw=8 $SCRATCH_DEV
> > > do_mkfs_pass -b size=4096 -d su=1b,sw=8 $SCRATCH_DEV
> > > @@ -147,8 +152,6 @@ do_mkfs_pass -s size=512 -d su=8s,sw=8 $SCRATCH_DEV
> > > do_mkfs_fail -d size=${fssize}b $SCRATCH_DEV
> > > do_mkfs_fail -d size=${fssize}s $SCRATCH_DEV
> > > do_mkfs_fail -d size=${fssize}yerk $SCRATCH_DEV
> > > -do_mkfs_fail -d agsize=8192b $SCRATCH_DEV
> > > -do_mkfs_fail -d agsize=65536s $SCRATCH_DEV
> > > do_mkfs_fail -d agsize=32Mbsdfsdo $SCRATCH_DEV
> > > do_mkfs_fail -d agsize=1GB $SCRATCH_DEV
> > > do_mkfs_fail -d agcount=1k $SCRATCH_DEV
> > > @@ -159,13 +162,10 @@ do_mkfs_fail -d sunit=64,swidth=0 $SCRATCH_DEV
> > > do_mkfs_fail -d sunit=64,swidth=64,noalign $SCRATCH_DEV
> > > do_mkfs_fail -d sunit=64k,swidth=64 $SCRATCH_DEV
> > > do_mkfs_fail -d sunit=64,swidth=64m $SCRATCH_DEV
> > > -do_mkfs_fail -d su=0,sw=64 $SCRATCH_DEV
> > > do_mkfs_fail -d su=4096,sw=0 $SCRATCH_DEV
> > > do_mkfs_fail -d su=4097,sw=1 $SCRATCH_DEV
> > > do_mkfs_fail -d su=4096,sw=64,noalign $SCRATCH_DEV
> > > do_mkfs_fail -d su=4096,sw=64s $SCRATCH_DEV
> > > -do_mkfs_fail -d su=4096s,sw=64 $SCRATCH_DEV
> > > -do_mkfs_fail -d su=4096b,sw=64 $SCRATCH_DEV
> > > do_mkfs_fail -d su=4096garabge,sw=64 $SCRATCH_DEV
> > > do_mkfs_fail -d su=4096,sw=64,sunit=64,swidth=64 $SCRATCH_DEV
> > > do_mkfs_fail -d sectsize=10,agsize=65536s $SCRATCH_DEV
> > > @@ -206,6 +206,7 @@ do_mkfs_pass -l sunit=64 $SCRATCH_DEV
> > > do_mkfs_pass -l sunit=64 -d sunit=8,swidth=8 $SCRATCH_DEV
> > > do_mkfs_pass -l sunit=8 $SCRATCH_DEV
> > > do_mkfs_pass -l su=$((4096*10)) $SCRATCH_DEV
> > > +do_mkfs_pass -l su=10b $SCRATCH_DEV
> > > do_mkfs_pass -b size=4096 -l su=10b $SCRATCH_DEV
> > > do_mkfs_pass -l sectsize=512,su=$((4096*10)) $SCRATCH_DEV
> > > do_mkfs_pass -l internal $SCRATCH_DEV
> > > @@ -228,7 +229,6 @@ do_mkfs_fail -l agnum=32 $SCRATCH_DEV
> > > do_mkfs_fail -l sunit=0 $SCRATCH_DEV
> > > do_mkfs_fail -l sunit=63 $SCRATCH_DEV
> > > do_mkfs_fail -l su=1 $SCRATCH_DEV
> > > -do_mkfs_fail -l su=10b $SCRATCH_DEV
> > > do_mkfs_fail -l su=10s $SCRATCH_DEV
> > > do_mkfs_fail -l su=$((4096*10+1)) $SCRATCH_DEV
> > > do_mkfs_fail -l sectsize=10,agsize=65536s $SCRATCH_DEV
> > > @@ -246,7 +246,6 @@ do_mkfs_fail -l version=0 $SCRATCH_DEV
> > > # naming section, should pass
> > > do_mkfs_pass -n size=65536 $SCRATCH_DEV
> > > -do_mkfs_pass -n log=15 $SCRATCH_DEV
> > > do_mkfs_pass -n version=2 $SCRATCH_DEV
> > > do_mkfs_pass -n version=ci $SCRATCH_DEV
> > > do_mkfs_pass -n ftype=0 -m crc=0 $SCRATCH_DEV
> > > @@ -257,6 +256,7 @@ do_mkfs_fail -n version=1 $SCRATCH_DEV
> > > do_mkfs_fail -n version=cid $SCRATCH_DEV
> > > do_mkfs_fail -n ftype=4 $SCRATCH_DEV
> > > do_mkfs_fail -n ftype=0 $SCRATCH_DEV
> > > +do_mkfs_fail -n log=15 $SCRATCH_DEV
> > > reset_fsimg
> > > @@ -273,14 +273,24 @@ do_mkfs_fail -m crc=0,finobt=1 $SCRATCH_DEV
> > > do_mkfs_fail -m crc=1 -n ftype=0 $SCRATCH_DEV
> > > +# realtime section, results depend on reflink
> > > +_scratch_mkfs_xfs_supported -m reflink=0 >/dev/null 2>&1
> > > +if [ $? -eq 0 ]; then
> > > + do_mkfs_pass -m reflink=0 -r rtdev=$fsimg $SCRATCH_DEV
> > > + do_mkfs_pass -m reflink=0 -r size=65536,rtdev=$fsimg $SCRATCH_DEV
> > > + do_mkfs_fail -m reflink=1 -r rtdev=$fsimg $SCRATCH_DEV
> > > + do_mkfs_fail -m reflink=1 -r size=65536,rtdev=$fsimg $SCRATCH_DEV
> > > +else
> > > + do_mkfs_pass -r rtdev=$fsimg $SCRATCH_DEV
> > > + do_mkfs_pass -r size=65536,rtdev=$fsimg $SCRATCH_DEV
> > > +fi
> > > +
> > > +
> > > # realtime section, should pass
> > > -do_mkfs_pass -r rtdev=$fsimg $SCRATCH_DEV
> > > do_mkfs_pass -r extsize=4k $SCRATCH_DEV
> > > do_mkfs_pass -r extsize=1G $SCRATCH_DEV
> > > -do_mkfs_pass -r size=65536,rtdev=$fsimg $SCRATCH_DEV
> > > do_mkfs_pass -r noalign $SCRATCH_DEV
> > > -
> > > # realtime section, should fail
> > > do_mkfs_fail -r rtdev=$SCRATCH_DEV
> > > do_mkfs_fail -r extsize=256 $SCRATCH_DEV
> > > @@ -293,7 +303,6 @@ do_mkfs_fail -r size=65536 $SCRATCH_DEV
> > > do_mkfs_pass -i size=256 -m crc=0 $SCRATCH_DEV
> > > do_mkfs_pass -i size=512 $SCRATCH_DEV
> > > do_mkfs_pass -i size=2048 $SCRATCH_DEV
> > > -do_mkfs_pass -i log=10 $SCRATCH_DEV
> > > do_mkfs_pass -i perblock=2 $SCRATCH_DEV
> > > do_mkfs_pass -i maxpct=10 $SCRATCH_DEV
> > > do_mkfs_pass -i maxpct=100 $SCRATCH_DEV
> > > @@ -317,6 +326,8 @@ do_mkfs_fail -i align=2 $SCRATCH_DEV
> > > do_mkfs_fail -i sparse -m crc=0 $SCRATCH_DEV
> > > do_mkfs_fail -i align=0 -m crc=1 $SCRATCH_DEV
> > > do_mkfs_fail -i attr=1 -m crc=1 $SCRATCH_DEV
> > > +do_mkfs_fail -i log=10 $SCRATCH_DEV
> > > +
> > > status=0
> > > exit
> > > --
> > > 2.18.1
> > >
> > >
> > >
> >
> >
> >
>
>
next prev parent reply other threads:[~2020-03-26 13:10 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1560414701-2590-1-git-send-email-xuyang2018.jy@cn.fujitsu.com>
2019-06-16 14:39 ` [PATCH v2] xfs/191: update mkfs.xfs input results Eryu Guan
2019-06-19 7:39 ` [PATCH v3] " Yang Xu
2020-03-25 12:37 ` Christoph Hellwig
2020-03-26 5:39 ` Zorro Lang
2020-03-26 9:11 ` Yang Xu
2020-03-26 13:21 ` Zorro Lang [this message]
2020-03-27 1:22 ` Yang Xu
2019-06-19 7:58 ` [PATCH v2] " Yang Xu
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=20200326132130.GO14282@dhcp-12-102.nay.redhat.com \
--to=zlang@redhat.com \
--cc=fstests@vger.kernel.org \
--cc=linux-xfs@vger.kernel.org \
--cc=xuyang2018.jy@cn.fujitsu.com \
/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).