Message ID | 1433760101-25540-1-git-send-email-eguan@redhat.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
On Mon, 8 Jun 2015, Eryu Guan wrote: > Date: Mon, 8 Jun 2015 18:41:41 +0800 > From: Eryu Guan <eguan@redhat.com> > To: fstests@vger.kernel.org > Cc: lczerner@redhat.com, Eryu Guan <eguan@redhat.com> > Subject: [PATCH] generic: concurrent IO test with mixed IO types > > Test concurrent buffered I/O, DIO, AIO, mmap I/O and splice I/O on the > same files. Hi Eryu, Great! Thanks for turning it into a test. Couple of comments below. > > Signed-off-by: Eryu Guan <eguan@redhat.com> > --- > > This fio job file has been proven to be potent, it triggers WARNINGs on ext4 > and xfs with 4.1-rc6 kernel. > > ext4: WARNING: at fs/ext4/inode.c:1328 > xfs: WARNING: CPU: 7 PID: 3090 at fs/xfs/xfs_file.c:726 xfs_file_dio_aio_write+0x176/0x2a8 [xfs]() > > The ext4 issue should be fixed by Lukas's patch > ext4: fix reservation release on invalidatepage for delalloc fs > > And it ever paniced kernel in mm code and hung xfs. > > I reduced the numjobs and iodepth to reduce the test time(~25s on my test host) > and scale them by $LOAD_FACTOR. And it still could trigger the warning on ext4 > and xfs with reduced workload. > > tests/generic/090 | 121 ++++++++++++++++++++++++++++++++++++++++++++++++++ > tests/generic/090.out | 2 + > tests/generic/group | 1 + > 3 files changed, 124 insertions(+) > create mode 100755 tests/generic/090 > create mode 100644 tests/generic/090.out > > diff --git a/tests/generic/090 b/tests/generic/090 > new file mode 100755 > index 0000000..748e935 > --- /dev/null > +++ b/tests/generic/090 > @@ -0,0 +1,121 @@ > +#! /bin/bash > +# FS QA Test generic/090 > +# > +# Concurrent mixed I/O (buffer I/O, aiodio, mmap, splice) on the same files > +# > +#----------------------------------------------------------------------- > +# Copyright (c) 2015 Red Hat Inc. All Rights Reserved. > +# > +# This program is free software; you can redistribute it and/or > +# modify it under the terms of the GNU General Public License as > +# published by the Free Software Foundation. > +# > +# This program is distributed in the hope that it would be useful, > +# but WITHOUT ANY WARRANTY; without even the implied warranty of > +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > +# GNU General Public License for more details. > +# > +# You should have received a copy of the GNU General Public License > +# along with this program; if not, write the Free Software Foundation, > +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA > +#----------------------------------------------------------------------- > +# > + > +seq=`basename $0` > +seqres=$RESULT_DIR/$seq > +echo "QA output created by $seq" > + > +here=`pwd` > +tmp=/tmp/$$ Not a big deal, but using mktemp woud probably be better. > +fio_config=$tmp.fio > +status=1 # failure is the default! > +trap "_cleanup; exit \$status" 0 1 2 3 15 > + > +_cleanup() > +{ > + cd / > + rm -f $tmp.* > +} > + > +# get standard environment, filters and checks > +. ./common/rc > +. ./common/filter > + > +# real QA test starts here > +_supported_fs generic > +_supported_os Linux > +_require_scratch > + > +iodepth=$((16 * LOAD_FACTOR)) > +iodepth_batch=$((8 * LOAD_FACTOR)) > +numjobs=$((5 * LOAD_FACTOR)) > +cat >$fio_config <<EOF > +[global] > +bs=8k > +iodepth=$iodepth > +iodepth_batch=$iodepth_batch > +randrepeat=1 > +size=1m > +directory=$SCRATCH_MNT > +numjobs=$numjobs > +[job1] > +ioengine=sync > +bs=1k > +direct=1 > +rw=randread > +filename=file1:file2 > +[job2] > +ioengine=libaio > +rw=randwrite > +direct=1 > +filename=file1:file2 > +[job3] > +bs=1k > +ioengine=posixaio > +rw=randwrite > +direct=1 > +filename=file1:file2 > +[job4] > +ioengine=splice > +direct=1 > +rw=randwrite > +filename=file1:file2 > +[job5] > +bs=1k > +ioengine=sync > +rw=randread > +filename=file1:file2 > +[job6] > +ioengine=posixaio > +rw=randwrite > +filename=file1:file2 > +[job7] > +ioengine=splice > +rw=randwrite > +filename=file1:file2 > +[job8] > +ioengine=mmap > +rw=randwrite > +bs=1k > +filename=file1:file2 > +[job9] > +ioengine=mmap > +rw=randwrite > +direct=1 > +filename=file1:file2 > +EOF > +# with ioengine=mmap and direct=1, fio requires bs to be at least pagesize, > +# which is a fio built-in var. > +echo 'bs=$pagesize' >> $fio_config Can't this be part of the the fio script creation above ? Thanks! -Lukas > + > +rm -f $seqres.full > + > +_require_fio $fio_config > + > +echo "Silence is golden" > +$FIO_PROG $fio_config >>$seqres.full 2>&1 > + > +# all done, expect no hang no oops no fs corruption, > +# _check_dmesg and _check_filesystems will do the check work for us > +status=0 > +exit > diff --git a/tests/generic/090.out b/tests/generic/090.out > new file mode 100644 > index 0000000..2b5100d > --- /dev/null > +++ b/tests/generic/090.out > @@ -0,0 +1,2 @@ > +QA output created by 090 > +Silence is golden > diff --git a/tests/generic/group b/tests/generic/group > index 0c8964c..2e534a5 100644 > --- a/tests/generic/group > +++ b/tests/generic/group > @@ -92,6 +92,7 @@ > 087 perms auto quick > 088 perms auto quick > 089 metadata auto > +090 auto rw stress > 091 rw auto quick > 092 auto quick prealloc > 093 attr cap udf auto > -- To unsubscribe from this list: send the line "unsubscribe fstests" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Mon, Jun 08, 2015 at 01:02:48PM +0200, Lukáš Czerner wrote: > On Mon, 8 Jun 2015, Eryu Guan wrote: > > > Date: Mon, 8 Jun 2015 18:41:41 +0800 > > From: Eryu Guan <eguan@redhat.com> > > To: fstests@vger.kernel.org > > Cc: lczerner@redhat.com, Eryu Guan <eguan@redhat.com> > > Subject: [PATCH] generic: concurrent IO test with mixed IO types > > > > Test concurrent buffered I/O, DIO, AIO, mmap I/O and splice I/O on the > > same files. > > Hi Eryu, > > Great! Thanks for turning it into a test. Couple of comments below. > > > > > Signed-off-by: Eryu Guan <eguan@redhat.com> > > --- > > > > This fio job file has been proven to be potent, it triggers WARNINGs on ext4 > > and xfs with 4.1-rc6 kernel. > > > > ext4: WARNING: at fs/ext4/inode.c:1328 > > xfs: WARNING: CPU: 7 PID: 3090 at fs/xfs/xfs_file.c:726 xfs_file_dio_aio_write+0x176/0x2a8 [xfs]() > > > > The ext4 issue should be fixed by Lukas's patch > > ext4: fix reservation release on invalidatepage for delalloc fs > > > > And it ever paniced kernel in mm code and hung xfs. > > > > I reduced the numjobs and iodepth to reduce the test time(~25s on my test host) > > and scale them by $LOAD_FACTOR. And it still could trigger the warning on ext4 > > and xfs with reduced workload. > > > > tests/generic/090 | 121 ++++++++++++++++++++++++++++++++++++++++++++++++++ > > tests/generic/090.out | 2 + > > tests/generic/group | 1 + > > 3 files changed, 124 insertions(+) > > create mode 100755 tests/generic/090 > > create mode 100644 tests/generic/090.out > > > > diff --git a/tests/generic/090 b/tests/generic/090 > > new file mode 100755 > > index 0000000..748e935 > > --- /dev/null > > +++ b/tests/generic/090 > > @@ -0,0 +1,121 @@ > > +#! /bin/bash > > +# FS QA Test generic/090 > > +# > > +# Concurrent mixed I/O (buffer I/O, aiodio, mmap, splice) on the same files > > +# > > +#----------------------------------------------------------------------- > > +# Copyright (c) 2015 Red Hat Inc. All Rights Reserved. > > +# > > +# This program is free software; you can redistribute it and/or > > +# modify it under the terms of the GNU General Public License as > > +# published by the Free Software Foundation. > > +# > > +# This program is distributed in the hope that it would be useful, > > +# but WITHOUT ANY WARRANTY; without even the implied warranty of > > +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > > +# GNU General Public License for more details. > > +# > > +# You should have received a copy of the GNU General Public License > > +# along with this program; if not, write the Free Software Foundation, > > +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA > > +#----------------------------------------------------------------------- > > +# > > + > > +seq=`basename $0` > > +seqres=$RESULT_DIR/$seq > > +echo "QA output created by $seq" > > + > > +here=`pwd` > > +tmp=/tmp/$$ > > Not a big deal, but using mktemp woud probably be better. Will do in v2. > > > +fio_config=$tmp.fio > > +status=1 # failure is the default! > > +trap "_cleanup; exit \$status" 0 1 2 3 15 > > + > > +_cleanup() > > +{ > > + cd / > > + rm -f $tmp.* > > +} > > + > > +# get standard environment, filters and checks > > +. ./common/rc > > +. ./common/filter > > + > > +# real QA test starts here > > +_supported_fs generic > > +_supported_os Linux > > +_require_scratch > > + > > +iodepth=$((16 * LOAD_FACTOR)) > > +iodepth_batch=$((8 * LOAD_FACTOR)) > > +numjobs=$((5 * LOAD_FACTOR)) > > +cat >$fio_config <<EOF > > +[global] > > +bs=8k > > +iodepth=$iodepth > > +iodepth_batch=$iodepth_batch > > +randrepeat=1 > > +size=1m > > +directory=$SCRATCH_MNT > > +numjobs=$numjobs > > +[job1] > > +ioengine=sync > > +bs=1k > > +direct=1 > > +rw=randread > > +filename=file1:file2 > > +[job2] > > +ioengine=libaio > > +rw=randwrite > > +direct=1 > > +filename=file1:file2 > > +[job3] > > +bs=1k > > +ioengine=posixaio > > +rw=randwrite > > +direct=1 > > +filename=file1:file2 > > +[job4] > > +ioengine=splice > > +direct=1 > > +rw=randwrite > > +filename=file1:file2 > > +[job5] > > +bs=1k > > +ioengine=sync > > +rw=randread > > +filename=file1:file2 > > +[job6] > > +ioengine=posixaio > > +rw=randwrite > > +filename=file1:file2 > > +[job7] > > +ioengine=splice > > +rw=randwrite > > +filename=file1:file2 > > +[job8] > > +ioengine=mmap > > +rw=randwrite > > +bs=1k > > +filename=file1:file2 > > +[job9] > > +ioengine=mmap > > +rw=randwrite > > +direct=1 > > +filename=file1:file2 > > +EOF > > +# with ioengine=mmap and direct=1, fio requires bs to be at least pagesize, > > +# which is a fio built-in var. > > +echo 'bs=$pagesize' >> $fio_config > > Can't this be part of the the fio script creation above ? The 'here-document' replaces every var to its value, e.g. $numjobs => 10 (if LOAD_FACTOR is 1), so we don't have $pagesize set, and it'll be empty in the fio config file, that's not what we want. "bs=$pagesize" is literally what we want in the config file, $pagesize is a special var for fio, it will be replaced by real pagesize internally by fio. cat >file <<'EOF' text EOF This kind of 'here-document' ('' around EOF) won't do any variable expansion, but we need $numjobs to be expanded. So they can't be merged together. Thanks, Eryu > > Thanks! > -Lukas > > > + > > +rm -f $seqres.full > > + > > +_require_fio $fio_config > > + > > +echo "Silence is golden" > > +$FIO_PROG $fio_config >>$seqres.full 2>&1 > > + > > +# all done, expect no hang no oops no fs corruption, > > +# _check_dmesg and _check_filesystems will do the check work for us > > +status=0 > > +exit > > diff --git a/tests/generic/090.out b/tests/generic/090.out > > new file mode 100644 > > index 0000000..2b5100d > > --- /dev/null > > +++ b/tests/generic/090.out > > @@ -0,0 +1,2 @@ > > +QA output created by 090 > > +Silence is golden > > diff --git a/tests/generic/group b/tests/generic/group > > index 0c8964c..2e534a5 100644 > > --- a/tests/generic/group > > +++ b/tests/generic/group > > @@ -92,6 +92,7 @@ > > 087 perms auto quick > > 088 perms auto quick > > 089 metadata auto > > +090 auto rw stress > > 091 rw auto quick > > 092 auto quick prealloc > > 093 attr cap udf auto > > -- To unsubscribe from this list: send the line "unsubscribe fstests" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Mon, 8 Jun 2015, Eryu Guan wrote: > Date: Mon, 8 Jun 2015 19:59:47 +0800 > From: Eryu Guan <eguan@redhat.com> > To: Lukáš Czerner <lczerner@redhat.com> > Cc: fstests@vger.kernel.org > Subject: Re: [PATCH] generic: concurrent IO test with mixed IO types > > On Mon, Jun 08, 2015 at 01:02:48PM +0200, Lukáš Czerner wrote: > > On Mon, 8 Jun 2015, Eryu Guan wrote: > > > > > Date: Mon, 8 Jun 2015 18:41:41 +0800 > > > From: Eryu Guan <eguan@redhat.com> > > > To: fstests@vger.kernel.org > > > Cc: lczerner@redhat.com, Eryu Guan <eguan@redhat.com> > > > Subject: [PATCH] generic: concurrent IO test with mixed IO types > > > > > > Test concurrent buffered I/O, DIO, AIO, mmap I/O and splice I/O on the > > > same files. > > > > Hi Eryu, > > > > Great! Thanks for turning it into a test. Couple of comments below. > > > > > > > > Signed-off-by: Eryu Guan <eguan@redhat.com> > > > --- > > > > > > This fio job file has been proven to be potent, it triggers WARNINGs on ext4 > > > and xfs with 4.1-rc6 kernel. > > > > > > ext4: WARNING: at fs/ext4/inode.c:1328 > > > xfs: WARNING: CPU: 7 PID: 3090 at fs/xfs/xfs_file.c:726 xfs_file_dio_aio_write+0x176/0x2a8 [xfs]() > > > > > > The ext4 issue should be fixed by Lukas's patch > > > ext4: fix reservation release on invalidatepage for delalloc fs > > > > > > And it ever paniced kernel in mm code and hung xfs. > > > > > > I reduced the numjobs and iodepth to reduce the test time(~25s on my test host) > > > and scale them by $LOAD_FACTOR. And it still could trigger the warning on ext4 > > > and xfs with reduced workload. > > > > > > tests/generic/090 | 121 ++++++++++++++++++++++++++++++++++++++++++++++++++ > > > tests/generic/090.out | 2 + > > > tests/generic/group | 1 + > > > 3 files changed, 124 insertions(+) > > > create mode 100755 tests/generic/090 > > > create mode 100644 tests/generic/090.out > > > > > > diff --git a/tests/generic/090 b/tests/generic/090 > > > new file mode 100755 > > > index 0000000..748e935 > > > --- /dev/null > > > +++ b/tests/generic/090 > > > @@ -0,0 +1,121 @@ > > > +#! /bin/bash > > > +# FS QA Test generic/090 > > > +# > > > +# Concurrent mixed I/O (buffer I/O, aiodio, mmap, splice) on the same files > > > +# > > > +#----------------------------------------------------------------------- > > > +# Copyright (c) 2015 Red Hat Inc. All Rights Reserved. > > > +# > > > +# This program is free software; you can redistribute it and/or > > > +# modify it under the terms of the GNU General Public License as > > > +# published by the Free Software Foundation. > > > +# > > > +# This program is distributed in the hope that it would be useful, > > > +# but WITHOUT ANY WARRANTY; without even the implied warranty of > > > +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > > > +# GNU General Public License for more details. > > > +# > > > +# You should have received a copy of the GNU General Public License > > > +# along with this program; if not, write the Free Software Foundation, > > > +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA > > > +#----------------------------------------------------------------------- > > > +# > > > + > > > +seq=`basename $0` > > > +seqres=$RESULT_DIR/$seq > > > +echo "QA output created by $seq" > > > + > > > +here=`pwd` > > > +tmp=/tmp/$$ > > > > Not a big deal, but using mktemp woud probably be better. > > Will do in v2. > > > > > > +fio_config=$tmp.fio > > > +status=1 # failure is the default! > > > +trap "_cleanup; exit \$status" 0 1 2 3 15 > > > + > > > +_cleanup() > > > +{ > > > + cd / > > > + rm -f $tmp.* > > > +} > > > + > > > +# get standard environment, filters and checks > > > +. ./common/rc > > > +. ./common/filter > > > + > > > +# real QA test starts here > > > +_supported_fs generic > > > +_supported_os Linux > > > +_require_scratch > > > + > > > +iodepth=$((16 * LOAD_FACTOR)) > > > +iodepth_batch=$((8 * LOAD_FACTOR)) > > > +numjobs=$((5 * LOAD_FACTOR)) > > > +cat >$fio_config <<EOF > > > +[global] > > > +bs=8k > > > +iodepth=$iodepth > > > +iodepth_batch=$iodepth_batch > > > +randrepeat=1 > > > +size=1m > > > +directory=$SCRATCH_MNT > > > +numjobs=$numjobs > > > +[job1] > > > +ioengine=sync > > > +bs=1k > > > +direct=1 > > > +rw=randread > > > +filename=file1:file2 > > > +[job2] > > > +ioengine=libaio > > > +rw=randwrite > > > +direct=1 > > > +filename=file1:file2 > > > +[job3] > > > +bs=1k > > > +ioengine=posixaio > > > +rw=randwrite > > > +direct=1 > > > +filename=file1:file2 > > > +[job4] > > > +ioengine=splice > > > +direct=1 > > > +rw=randwrite > > > +filename=file1:file2 > > > +[job5] > > > +bs=1k > > > +ioengine=sync > > > +rw=randread > > > +filename=file1:file2 > > > +[job6] > > > +ioengine=posixaio > > > +rw=randwrite > > > +filename=file1:file2 > > > +[job7] > > > +ioengine=splice > > > +rw=randwrite > > > +filename=file1:file2 > > > +[job8] > > > +ioengine=mmap > > > +rw=randwrite > > > +bs=1k > > > +filename=file1:file2 > > > +[job9] > > > +ioengine=mmap > > > +rw=randwrite > > > +direct=1 > > > +filename=file1:file2 > > > +EOF > > > +# with ioengine=mmap and direct=1, fio requires bs to be at least pagesize, > > > +# which is a fio built-in var. > > > +echo 'bs=$pagesize' >> $fio_config > > > > Can't this be part of the the fio script creation above ? > > The 'here-document' replaces every var to its value, e.g. $numjobs => 10 > (if LOAD_FACTOR is 1), so we don't have $pagesize set, and it'll be > empty in the fio config file, that's not what we want. > > "bs=$pagesize" is literally what we want in the config file, $pagesize > is a special var for fio, it will be replaced by real pagesize > internally by fio. > > cat >file <<'EOF' > text > EOF > > This kind of 'here-document' ('' around EOF) won't do any variable > expansion, but we need $numjobs to be expanded. So they can't be merged > together. > > Thanks, > Eryu Ah right :) I missed the quotes, sorry. Thanks! -Lukas > > > > > Thanks! > > -Lukas > > > > > + > > > +rm -f $seqres.full > > > + > > > +_require_fio $fio_config > > > + > > > +echo "Silence is golden" > > > +$FIO_PROG $fio_config >>$seqres.full 2>&1 > > > + > > > +# all done, expect no hang no oops no fs corruption, > > > +# _check_dmesg and _check_filesystems will do the check work for us > > > +status=0 > > > +exit > > > diff --git a/tests/generic/090.out b/tests/generic/090.out > > > new file mode 100644 > > > index 0000000..2b5100d > > > --- /dev/null > > > +++ b/tests/generic/090.out > > > @@ -0,0 +1,2 @@ > > > +QA output created by 090 > > > +Silence is golden > > > diff --git a/tests/generic/group b/tests/generic/group > > > index 0c8964c..2e534a5 100644 > > > --- a/tests/generic/group > > > +++ b/tests/generic/group > > > @@ -92,6 +92,7 @@ > > > 087 perms auto quick > > > 088 perms auto quick > > > 089 metadata auto > > > +090 auto rw stress > > > 091 rw auto quick > > > 092 auto quick prealloc > > > 093 attr cap udf auto > > > >
On Mon, Jun 08, 2015 at 01:02:48PM +0200, Lukáš Czerner wrote: > On Mon, 8 Jun 2015, Eryu Guan wrote: > > > Date: Mon, 8 Jun 2015 18:41:41 +0800 From: Eryu Guan > > <eguan@redhat.com> To: fstests@vger.kernel.org Cc: > > lczerner@redhat.com, Eryu Guan <eguan@redhat.com> Subject: > > [PATCH] generic: concurrent IO test with mixed IO types > > > > Test concurrent buffered I/O, DIO, AIO, mmap I/O and splice I/O > > on the same files. > > Hi Eryu, > > Great! Thanks for turning it into a test. Couple of comments > below. .... > > + +seq=`basename $0` +seqres=$RESULT_DIR/$seq +echo "QA output > > created by $seq" + +here=`pwd` +tmp=/tmp/$$ > > Not a big deal, but using mktemp woud probably be better. Not better at all - this reduces the easy of debugging problems. e.g. if a test is hung, I know where it's temporary output files are just by looking at the process listing and grabbing the pid of the test process being run. IOWs, we *want* deterministic temporary filenames here - obscuring the temporary file names that tests use doesn't gain us anything. Yes, I see that some btrfs tests have used mktemp and a couple of generic tests use $TEST_DIR/$$ - they should be reverted to /tmp/$$ so they are consistent with the majority of tests and the new test template.... Cheers, Dave.
diff --git a/tests/generic/090 b/tests/generic/090 new file mode 100755 index 0000000..748e935 --- /dev/null +++ b/tests/generic/090 @@ -0,0 +1,121 @@ +#! /bin/bash +# FS QA Test generic/090 +# +# Concurrent mixed I/O (buffer I/O, aiodio, mmap, splice) on the same files +# +#----------------------------------------------------------------------- +# Copyright (c) 2015 Red Hat Inc. All Rights Reserved. +# +# This program is free software; you can redistribute it and/or +# modify it under the terms of the GNU General Public License as +# published by the Free Software Foundation. +# +# This program is distributed in the hope that it would be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write the Free Software Foundation, +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA +#----------------------------------------------------------------------- +# + +seq=`basename $0` +seqres=$RESULT_DIR/$seq +echo "QA output created by $seq" + +here=`pwd` +tmp=/tmp/$$ +fio_config=$tmp.fio +status=1 # failure is the default! +trap "_cleanup; exit \$status" 0 1 2 3 15 + +_cleanup() +{ + cd / + rm -f $tmp.* +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter + +# real QA test starts here +_supported_fs generic +_supported_os Linux +_require_scratch + +iodepth=$((16 * LOAD_FACTOR)) +iodepth_batch=$((8 * LOAD_FACTOR)) +numjobs=$((5 * LOAD_FACTOR)) +cat >$fio_config <<EOF +[global] +bs=8k +iodepth=$iodepth +iodepth_batch=$iodepth_batch +randrepeat=1 +size=1m +directory=$SCRATCH_MNT +numjobs=$numjobs +[job1] +ioengine=sync +bs=1k +direct=1 +rw=randread +filename=file1:file2 +[job2] +ioengine=libaio +rw=randwrite +direct=1 +filename=file1:file2 +[job3] +bs=1k +ioengine=posixaio +rw=randwrite +direct=1 +filename=file1:file2 +[job4] +ioengine=splice +direct=1 +rw=randwrite +filename=file1:file2 +[job5] +bs=1k +ioengine=sync +rw=randread +filename=file1:file2 +[job6] +ioengine=posixaio +rw=randwrite +filename=file1:file2 +[job7] +ioengine=splice +rw=randwrite +filename=file1:file2 +[job8] +ioengine=mmap +rw=randwrite +bs=1k +filename=file1:file2 +[job9] +ioengine=mmap +rw=randwrite +direct=1 +filename=file1:file2 +EOF +# with ioengine=mmap and direct=1, fio requires bs to be at least pagesize, +# which is a fio built-in var. +echo 'bs=$pagesize' >> $fio_config + +rm -f $seqres.full + +_require_fio $fio_config + +echo "Silence is golden" +$FIO_PROG $fio_config >>$seqres.full 2>&1 + +# all done, expect no hang no oops no fs corruption, +# _check_dmesg and _check_filesystems will do the check work for us +status=0 +exit diff --git a/tests/generic/090.out b/tests/generic/090.out new file mode 100644 index 0000000..2b5100d --- /dev/null +++ b/tests/generic/090.out @@ -0,0 +1,2 @@ +QA output created by 090 +Silence is golden diff --git a/tests/generic/group b/tests/generic/group index 0c8964c..2e534a5 100644 --- a/tests/generic/group +++ b/tests/generic/group @@ -92,6 +92,7 @@ 087 perms auto quick 088 perms auto quick 089 metadata auto +090 auto rw stress 091 rw auto quick 092 auto quick prealloc 093 attr cap udf auto
Test concurrent buffered I/O, DIO, AIO, mmap I/O and splice I/O on the same files. Signed-off-by: Eryu Guan <eguan@redhat.com> --- This fio job file has been proven to be potent, it triggers WARNINGs on ext4 and xfs with 4.1-rc6 kernel. ext4: WARNING: at fs/ext4/inode.c:1328 xfs: WARNING: CPU: 7 PID: 3090 at fs/xfs/xfs_file.c:726 xfs_file_dio_aio_write+0x176/0x2a8 [xfs]() The ext4 issue should be fixed by Lukas's patch ext4: fix reservation release on invalidatepage for delalloc fs And it ever paniced kernel in mm code and hung xfs. I reduced the numjobs and iodepth to reduce the test time(~25s on my test host) and scale them by $LOAD_FACTOR. And it still could trigger the warning on ext4 and xfs with reduced workload. tests/generic/090 | 121 ++++++++++++++++++++++++++++++++++++++++++++++++++ tests/generic/090.out | 2 + tests/generic/group | 1 + 3 files changed, 124 insertions(+) create mode 100755 tests/generic/090 create mode 100644 tests/generic/090.out