* [PATCH 0/8] loop: cleanup and few improvements
@ 2022-02-14 10:01 Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 1/8] loop: use sysfs_emit() in the sysfs offset show Chaitanya Kulkarni
` (8 more replies)
0 siblings, 9 replies; 12+ messages in thread
From: Chaitanya Kulkarni @ 2022-02-14 10:01 UTC (permalink / raw)
To: linux-block; +Cc: axboe, Chaitanya Kulkarni
Hi,
This has few improvment and cleanups such as using sysfs_emit() for the
sysfs dev attributes and removing variables that are only used once and
a cleanup with fixing declaration.
Below is the test log where 10 loop devices created, each device is
linked to it's own file in ./loopX, formatted with xfs and mounted on
/mnt/loopX. For each device it reads the offset, sizelimit, autoclear,
partscan, and dio attr from sysfs using cat command, then it runs fio
verify job on it.
In summary write-verify fio job seems to work fine :-
write-and-verify: (groupid=0, jobs=1): err= 0: pid=3495: Mon Feb 14 00:43:19 2022
write-and-verify: (groupid=0, jobs=1): err= 0: pid=3960: Mon Feb 14 00:45:17 2022
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4321: Mon Feb 14 00:47:15 2022
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4369: Mon Feb 14 00:49:20 2022
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4443: Mon Feb 14 00:51:25 2022
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4536: Mon Feb 14 00:53:25 2022
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4770: Mon Feb 14 00:55:24 2022
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4933: Mon Feb 14 00:57:31 2022
write-and-verify: (groupid=0, jobs=1): err= 0: pid=5008: Mon Feb 14 00:59:35 2022
write-and-verify: (groupid=0, jobs=1): err= 0: pid=5048: Mon Feb 14 01:01:58 2022
Below is detailed test log.
Chaitanya Kulkarni (8):
loop: use sysfs_emit() in the sysfs offset show
loop: use sysfs_emit() in the sysfs sizelimit show
loop: use sysfs_emit() in the sysfs autoclear show
loop: use sysfs_emit() in the sysfs partscan show
loop: use sysfs_emit() in the sysfs dio show
loop: remove extra variable in lo_fallocate()
loop: remove extra variable in lo_req_flush
loop: allow user to set the queue depth
drivers/block/loop.c | 21 +++++++++++----------
1 file changed, 11 insertions(+), 10 deletions(-)
# git am --skip
Applying: loop: use sysfs_emit() in the sysfs offset show
Applying: loop: use sysfs_emit() in the sysfs sizelimit show
Applying: loop: use sysfs_emit() in the sysfs autoclear show
Applying: loop: use sysfs_emit() in the sysfs partscan show
Applying: loop: use sysfs_emit() in the sysfs dio show
Applying: loop: remove extra variable in lo_fallocate()
Applying: loop: remove extra variable in lo_req_flush
Applying: loop: allow user to set the queue depth
root@dev linux-block (for-next) # ./compile_loop.sh 10
+ FILE=./loop
+ LOOP_MNT=/mnt/loop
+ NN=10
+ unload_loop
++ shuf -i 1-10 -n 10
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop5
umount: /mnt/loop5: no mount point specified.
+ rm -fr ./loop5
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop8
umount: /mnt/loop8: no mount point specified.
+ rm -fr ./loop8
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop1
umount: /mnt/loop1: no mount point specified.
+ rm -fr ./loop1
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop4
umount: /mnt/loop4: no mount point specified.
+ rm -fr ./loop4
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop2
umount: /mnt/loop2: no mount point specified.
+ rm -fr ./loop2
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop3
umount: /mnt/loop3: no mount point specified.
+ rm -fr ./loop3
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop9
umount: /mnt/loop9: no mount point specified.
+ rm -fr ./loop9
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop10
umount: /mnt/loop10: no mount point specified.
+ rm -fr ./loop10
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop6
umount: /mnt/loop6: no mount point specified.
+ rm -fr ./loop6
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop7
umount: /mnt/loop7: no mount point specified.
+ rm -fr ./loop7
+ losetup -D
+ sleep 3
+ rmmod loop
rmmod: ERROR: Module loop is not currently loaded
+ modprobe -r loop
+ lsmod
+ grep loop
++ shuf -i 1-10 -n 10
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop9
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop3
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop10
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop7
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop5
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop4
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop8
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop2
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop6
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop1
+ rm -fr '/mnt/loop*'
+ compile_loop
++ nproc
+ make -j 48 M=drivers/block modules
CC [M] drivers/block/loop.o
MODPOST drivers/block/Module.symvers
CC [M] drivers/block/brd.mod.o
CC [M] drivers/block/loop.mod.o
CC [M] drivers/block/null_blk/null_blk.mod.o
CC [M] drivers/block/pktcdvd.mod.o
CC [M] drivers/block/rbd.mod.o
CC [M] drivers/block/sx8.mod.o
CC [M] drivers/block/virtio_blk.mod.o
CC [M] drivers/block/xen-blkfront.mod.o
CC [M] drivers/block/zram/zram.mod.o
LD [M] drivers/block/rbd.ko
LD [M] drivers/block/xen-blkfront.ko
LD [M] drivers/block/brd.ko
LD [M] drivers/block/null_blk/null_blk.ko
LD [M] drivers/block/zram/zram.ko
LD [M] drivers/block/loop.ko
LD [M] drivers/block/pktcdvd.ko
LD [M] drivers/block/sx8.ko
LD [M] drivers/block/virtio_blk.ko
+ HOST=drivers/block/
++ uname -r
+ HOST_DEST=/lib/modules/5.17.0-rc3blk+/kernel/drivers/block
+ cp drivers/block//loop.ko /lib/modules/5.17.0-rc3blk+/kernel/drivers/block/
+ load_loop
+ insmod drivers/block/loop.ko max_loop=11 hw_queue_depth=32
++ shuf -i 1-10 -n 10
+ for i in `shuf -i 1-$NN -n $NN`
+ mkdir -p /mnt/loop1
+ truncate -s 2048M ./loop1
+ /mnt/data/util-linux/losetup --direct-io=on /dev/loop1 ./loop1
+ /mnt/data/util-linux/losetup
NAME SIZELIMIT OFFSET AUTOCLEAR RO BACK-FILE DIO LOG-SEC
/dev/loop1 0 0 0 0 /mnt/data/linux-block/loop1 1 512
+ sleep 1
+ mkfs.xfs -f /dev/loop1
meta-data=/dev/loop1 isize=512 agcount=4, agsize=131072 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1 bigtime=0
data = bsize=4096 blocks=524288, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Discarding blocks...Done.
+ mount /dev/loop1 /mnt/loop1
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop1/loop/offset : '
cat /sys/block/loop1/loop/offset : + cat /sys/block/loop1/loop/offset
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop1/loop/sizelimit : '
cat /sys/block/loop1/loop/sizelimit : + cat /sys/block/loop1/loop/sizelimit
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop1/loop/autoclear : '
cat /sys/block/loop1/loop/autoclear : + cat /sys/block/loop1/loop/autoclear
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop1/loop/partscan : '
cat /sys/block/loop1/loop/partscan : + cat /sys/block/loop1/loop/partscan
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop1/loop/dio : '
cat /sys/block/loop1/loop/dio : + cat /sys/block/loop1/loop/dio
1
+ for i in `shuf -i 1-$NN -n $NN`
+ mkdir -p /mnt/loop4
+ truncate -s 2048M ./loop4
+ /mnt/data/util-linux/losetup --direct-io=on /dev/loop4 ./loop4
+ /mnt/data/util-linux/losetup
NAME SIZELIMIT OFFSET AUTOCLEAR RO BACK-FILE DIO LOG-SEC
/dev/loop1 0 0 0 0 /mnt/data/linux-block/loop1 1 512
/dev/loop4 0 0 0 0 /mnt/data/linux-block/loop4 1 512
+ sleep 1
+ mkfs.xfs -f /dev/loop4
meta-data=/dev/loop4 isize=512 agcount=4, agsize=131072 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1 bigtime=0
data = bsize=4096 blocks=524288, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Discarding blocks...Done.
+ mount /dev/loop4 /mnt/loop4
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop4/loop/offset : '
cat /sys/block/loop4/loop/offset : + cat /sys/block/loop4/loop/offset
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop4/loop/sizelimit : '
cat /sys/block/loop4/loop/sizelimit : + cat /sys/block/loop4/loop/sizelimit
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop4/loop/autoclear : '
cat /sys/block/loop4/loop/autoclear : + cat /sys/block/loop4/loop/autoclear
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop4/loop/partscan : '
cat /sys/block/loop4/loop/partscan : + cat /sys/block/loop4/loop/partscan
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop4/loop/dio : '
cat /sys/block/loop4/loop/dio : + cat /sys/block/loop4/loop/dio
1
+ for i in `shuf -i 1-$NN -n $NN`
+ mkdir -p /mnt/loop9
+ truncate -s 2048M ./loop9
+ /mnt/data/util-linux/losetup --direct-io=on /dev/loop9 ./loop9
+ /mnt/data/util-linux/losetup
NAME SIZELIMIT OFFSET AUTOCLEAR RO BACK-FILE DIO LOG-SEC
/dev/loop1 0 0 0 0 /mnt/data/linux-block/loop1 1 512
/dev/loop4 0 0 0 0 /mnt/data/linux-block/loop4 1 512
/dev/loop9 0 0 0 0 /mnt/data/linux-block/loop9 1 512
+ sleep 1
+ mkfs.xfs -f /dev/loop9
meta-data=/dev/loop9 isize=512 agcount=4, agsize=131072 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1 bigtime=0
data = bsize=4096 blocks=524288, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Discarding blocks...Done.
+ mount /dev/loop9 /mnt/loop9
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop9/loop/offset : '
cat /sys/block/loop9/loop/offset : + cat /sys/block/loop9/loop/offset
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop9/loop/sizelimit : '
cat /sys/block/loop9/loop/sizelimit : + cat /sys/block/loop9/loop/sizelimit
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop9/loop/autoclear : '
cat /sys/block/loop9/loop/autoclear : + cat /sys/block/loop9/loop/autoclear
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop9/loop/partscan : '
cat /sys/block/loop9/loop/partscan : + cat /sys/block/loop9/loop/partscan
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop9/loop/dio : '
cat /sys/block/loop9/loop/dio : + cat /sys/block/loop9/loop/dio
1
+ for i in `shuf -i 1-$NN -n $NN`
+ mkdir -p /mnt/loop10
+ truncate -s 2048M ./loop10
+ /mnt/data/util-linux/losetup --direct-io=on /dev/loop10 ./loop10
+ /mnt/data/util-linux/losetup
NAME SIZELIMIT OFFSET AUTOCLEAR RO BACK-FILE DIO LOG-SEC
/dev/loop1 0 0 0 0 /mnt/data/linux-block/loop1 1 512
/dev/loop4 0 0 0 0 /mnt/data/linux-block/loop4 1 512
/dev/loop9 0 0 0 0 /mnt/data/linux-block/loop9 1 512
/dev/loop10 0 0 0 0 /mnt/data/linux-block/loop10 1 512
+ sleep 1
+ mkfs.xfs -f /dev/loop10
meta-data=/dev/loop10 isize=512 agcount=4, agsize=131072 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1 bigtime=0
data = bsize=4096 blocks=524288, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Discarding blocks...Done.
+ mount /dev/loop10 /mnt/loop10
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop10/loop/offset : '
cat /sys/block/loop10/loop/offset : + cat /sys/block/loop10/loop/offset
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop10/loop/sizelimit : '
cat /sys/block/loop10/loop/sizelimit : + cat /sys/block/loop10/loop/sizelimit
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop10/loop/autoclear : '
cat /sys/block/loop10/loop/autoclear : + cat /sys/block/loop10/loop/autoclear
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop10/loop/partscan : '
cat /sys/block/loop10/loop/partscan : + cat /sys/block/loop10/loop/partscan
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop10/loop/dio : '
cat /sys/block/loop10/loop/dio : + cat /sys/block/loop10/loop/dio
1
+ for i in `shuf -i 1-$NN -n $NN`
+ mkdir -p /mnt/loop7
+ truncate -s 2048M ./loop7
+ /mnt/data/util-linux/losetup --direct-io=on /dev/loop7 ./loop7
+ /mnt/data/util-linux/losetup
NAME SIZELIMIT OFFSET AUTOCLEAR RO BACK-FILE DIO LOG-SEC
/dev/loop1 0 0 0 0 /mnt/data/linux-block/loop1 1 512
/dev/loop4 0 0 0 0 /mnt/data/linux-block/loop4 1 512
/dev/loop9 0 0 0 0 /mnt/data/linux-block/loop9 1 512
/dev/loop7 0 0 0 0 /mnt/data/linux-block/loop7 1 512
/dev/loop10 0 0 0 0 /mnt/data/linux-block/loop10 1 512
+ sleep 1
+ mkfs.xfs -f /dev/loop7
meta-data=/dev/loop7 isize=512 agcount=4, agsize=131072 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1 bigtime=0
data = bsize=4096 blocks=524288, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Discarding blocks...Done.
+ mount /dev/loop7 /mnt/loop7
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop7/loop/offset : '
cat /sys/block/loop7/loop/offset : + cat /sys/block/loop7/loop/offset
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop7/loop/sizelimit : '
cat /sys/block/loop7/loop/sizelimit : + cat /sys/block/loop7/loop/sizelimit
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop7/loop/autoclear : '
cat /sys/block/loop7/loop/autoclear : + cat /sys/block/loop7/loop/autoclear
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop7/loop/partscan : '
cat /sys/block/loop7/loop/partscan : + cat /sys/block/loop7/loop/partscan
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop7/loop/dio : '
cat /sys/block/loop7/loop/dio : + cat /sys/block/loop7/loop/dio
1
+ for i in `shuf -i 1-$NN -n $NN`
+ mkdir -p /mnt/loop2
+ truncate -s 2048M ./loop2
+ /mnt/data/util-linux/losetup --direct-io=on /dev/loop2 ./loop2
+ /mnt/data/util-linux/losetup
NAME SIZELIMIT OFFSET AUTOCLEAR RO BACK-FILE DIO LOG-SEC
/dev/loop1 0 0 0 0 /mnt/data/linux-block/loop1 1 512
/dev/loop4 0 0 0 0 /mnt/data/linux-block/loop4 1 512
/dev/loop2 0 0 0 0 /mnt/data/linux-block/loop2 1 512
/dev/loop9 0 0 0 0 /mnt/data/linux-block/loop9 1 512
/dev/loop7 0 0 0 0 /mnt/data/linux-block/loop7 1 512
/dev/loop10 0 0 0 0 /mnt/data/linux-block/loop10 1 512
+ sleep 1
+ mkfs.xfs -f /dev/loop2
meta-data=/dev/loop2 isize=512 agcount=4, agsize=131072 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1 bigtime=0
data = bsize=4096 blocks=524288, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Discarding blocks...Done.
+ mount /dev/loop2 /mnt/loop2
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop2/loop/offset : '
cat /sys/block/loop2/loop/offset : + cat /sys/block/loop2/loop/offset
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop2/loop/sizelimit : '
cat /sys/block/loop2/loop/sizelimit : + cat /sys/block/loop2/loop/sizelimit
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop2/loop/autoclear : '
cat /sys/block/loop2/loop/autoclear : + cat /sys/block/loop2/loop/autoclear
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop2/loop/partscan : '
cat /sys/block/loop2/loop/partscan : + cat /sys/block/loop2/loop/partscan
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop2/loop/dio : '
cat /sys/block/loop2/loop/dio : + cat /sys/block/loop2/loop/dio
1
+ for i in `shuf -i 1-$NN -n $NN`
+ mkdir -p /mnt/loop3
+ truncate -s 2048M ./loop3
+ /mnt/data/util-linux/losetup --direct-io=on /dev/loop3 ./loop3
+ /mnt/data/util-linux/losetup
NAME SIZELIMIT OFFSET AUTOCLEAR RO BACK-FILE DIO LOG-SEC
/dev/loop1 0 0 0 0 /mnt/data/linux-block/loop1 1 512
/dev/loop4 0 0 0 0 /mnt/data/linux-block/loop4 1 512
/dev/loop2 0 0 0 0 /mnt/data/linux-block/loop2 1 512
/dev/loop9 0 0 0 0 /mnt/data/linux-block/loop9 1 512
/dev/loop7 0 0 0 0 /mnt/data/linux-block/loop7 1 512
/dev/loop3 0 0 0 0 /mnt/data/linux-block/loop3 1 512
/dev/loop10 0 0 0 0 /mnt/data/linux-block/loop10 1 512
+ sleep 1
+ mkfs.xfs -f /dev/loop3
meta-data=/dev/loop3 isize=512 agcount=4, agsize=131072 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1 bigtime=0
data = bsize=4096 blocks=524288, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Discarding blocks...Done.
+ mount /dev/loop3 /mnt/loop3
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop3/loop/offset : '
cat /sys/block/loop3/loop/offset : + cat /sys/block/loop3/loop/offset
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop3/loop/sizelimit : '
cat /sys/block/loop3/loop/sizelimit : + cat /sys/block/loop3/loop/sizelimit
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop3/loop/autoclear : '
cat /sys/block/loop3/loop/autoclear : + cat /sys/block/loop3/loop/autoclear
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop3/loop/partscan : '
cat /sys/block/loop3/loop/partscan : + cat /sys/block/loop3/loop/partscan
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop3/loop/dio : '
cat /sys/block/loop3/loop/dio : + cat /sys/block/loop3/loop/dio
1
+ for i in `shuf -i 1-$NN -n $NN`
+ mkdir -p /mnt/loop5
+ truncate -s 2048M ./loop5
+ /mnt/data/util-linux/losetup --direct-io=on /dev/loop5 ./loop5
+ /mnt/data/util-linux/losetup
NAME SIZELIMIT OFFSET AUTOCLEAR RO BACK-FILE DIO LOG-SEC
/dev/loop1 0 0 0 0 /mnt/data/linux-block/loop1 1 512
/dev/loop4 0 0 0 0 /mnt/data/linux-block/loop4 1 512
/dev/loop2 0 0 0 0 /mnt/data/linux-block/loop2 1 512
/dev/loop9 0 0 0 0 /mnt/data/linux-block/loop9 1 512
/dev/loop7 0 0 0 0 /mnt/data/linux-block/loop7 1 512
/dev/loop5 0 0 0 0 /mnt/data/linux-block/loop5 1 512
/dev/loop3 0 0 0 0 /mnt/data/linux-block/loop3 1 512
/dev/loop10 0 0 0 0 /mnt/data/linux-block/loop10 1 512
+ sleep 1
+ mkfs.xfs -f /dev/loop5
meta-data=/dev/loop5 isize=512 agcount=4, agsize=131072 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1 bigtime=0
data = bsize=4096 blocks=524288, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Discarding blocks...Done.
+ mount /dev/loop5 /mnt/loop5
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop5/loop/offset : '
cat /sys/block/loop5/loop/offset : + cat /sys/block/loop5/loop/offset
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop5/loop/sizelimit : '
cat /sys/block/loop5/loop/sizelimit : + cat /sys/block/loop5/loop/sizelimit
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop5/loop/autoclear : '
cat /sys/block/loop5/loop/autoclear : + cat /sys/block/loop5/loop/autoclear
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop5/loop/partscan : '
cat /sys/block/loop5/loop/partscan : + cat /sys/block/loop5/loop/partscan
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop5/loop/dio : '
cat /sys/block/loop5/loop/dio : + cat /sys/block/loop5/loop/dio
1
+ for i in `shuf -i 1-$NN -n $NN`
+ mkdir -p /mnt/loop8
+ truncate -s 2048M ./loop8
+ /mnt/data/util-linux/losetup --direct-io=on /dev/loop8 ./loop8
+ /mnt/data/util-linux/losetup
NAME SIZELIMIT OFFSET AUTOCLEAR RO BACK-FILE DIO LOG-SEC
/dev/loop1 0 0 0 0 /mnt/data/linux-block/loop1 1 512
/dev/loop8 0 0 0 0 /mnt/data/linux-block/loop8 1 512
/dev/loop4 0 0 0 0 /mnt/data/linux-block/loop4 1 512
/dev/loop2 0 0 0 0 /mnt/data/linux-block/loop2 1 512
/dev/loop9 0 0 0 0 /mnt/data/linux-block/loop9 1 512
/dev/loop7 0 0 0 0 /mnt/data/linux-block/loop7 1 512
/dev/loop5 0 0 0 0 /mnt/data/linux-block/loop5 1 512
/dev/loop3 0 0 0 0 /mnt/data/linux-block/loop3 1 512
/dev/loop10 0 0 0 0 /mnt/data/linux-block/loop10 1 512
+ sleep 1
+ mkfs.xfs -f /dev/loop8
meta-data=/dev/loop8 isize=512 agcount=4, agsize=131072 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1 bigtime=0
data = bsize=4096 blocks=524288, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Discarding blocks...Done.
+ mount /dev/loop8 /mnt/loop8
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop8/loop/offset : '
cat /sys/block/loop8/loop/offset : + cat /sys/block/loop8/loop/offset
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop8/loop/sizelimit : '
cat /sys/block/loop8/loop/sizelimit : + cat /sys/block/loop8/loop/sizelimit
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop8/loop/autoclear : '
cat /sys/block/loop8/loop/autoclear : + cat /sys/block/loop8/loop/autoclear
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop8/loop/partscan : '
cat /sys/block/loop8/loop/partscan : + cat /sys/block/loop8/loop/partscan
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop8/loop/dio : '
cat /sys/block/loop8/loop/dio : + cat /sys/block/loop8/loop/dio
1
+ for i in `shuf -i 1-$NN -n $NN`
+ mkdir -p /mnt/loop6
+ truncate -s 2048M ./loop6
+ /mnt/data/util-linux/losetup --direct-io=on /dev/loop6 ./loop6
+ /mnt/data/util-linux/losetup
NAME SIZELIMIT OFFSET AUTOCLEAR RO BACK-FILE DIO LOG-SEC
/dev/loop1 0 0 0 0 /mnt/data/linux-block/loop1 1 512
/dev/loop8 0 0 0 0 /mnt/data/linux-block/loop8 1 512
/dev/loop6 0 0 0 0 /mnt/data/linux-block/loop6 1 512
/dev/loop4 0 0 0 0 /mnt/data/linux-block/loop4 1 512
/dev/loop2 0 0 0 0 /mnt/data/linux-block/loop2 1 512
/dev/loop9 0 0 0 0 /mnt/data/linux-block/loop9 1 512
/dev/loop7 0 0 0 0 /mnt/data/linux-block/loop7 1 512
/dev/loop5 0 0 0 0 /mnt/data/linux-block/loop5 1 512
/dev/loop3 0 0 0 0 /mnt/data/linux-block/loop3 1 512
/dev/loop10 0 0 0 0 /mnt/data/linux-block/loop10 1 512
+ sleep 1
+ mkfs.xfs -f /dev/loop6
meta-data=/dev/loop6 isize=512 agcount=4, agsize=131072 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1 bigtime=0
data = bsize=4096 blocks=524288, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Discarding blocks...Done.
+ mount /dev/loop6 /mnt/loop6
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop6/loop/offset : '
cat /sys/block/loop6/loop/offset : + cat /sys/block/loop6/loop/offset
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop6/loop/sizelimit : '
cat /sys/block/loop6/loop/sizelimit : + cat /sys/block/loop6/loop/sizelimit
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop6/loop/autoclear : '
cat /sys/block/loop6/loop/autoclear : + cat /sys/block/loop6/loop/autoclear
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop6/loop/partscan : '
cat /sys/block/loop6/loop/partscan : + cat /sys/block/loop6/loop/partscan
0
+ for attr in offset sizelimit autoclear partscan dio
+ echo -n 'cat /sys/block/loop6/loop/dio : '
cat /sys/block/loop6/loop/dio : + cat /sys/block/loop6/loop/dio
1
+ mount
+ grep loop
/dev/loop1 on /mnt/loop1 type xfs (rw,relatime,seclabel,attr2,inode64,logbufs=8,logbsize=32k,noquota)
/dev/loop4 on /mnt/loop4 type xfs (rw,relatime,seclabel,attr2,inode64,logbufs=8,logbsize=32k,noquota)
/dev/loop9 on /mnt/loop9 type xfs (rw,relatime,seclabel,attr2,inode64,logbufs=8,logbsize=32k,noquota)
/dev/loop10 on /mnt/loop10 type xfs (rw,relatime,seclabel,attr2,inode64,logbufs=8,logbsize=32k,noquota)
/dev/loop7 on /mnt/loop7 type xfs (rw,relatime,seclabel,attr2,inode64,logbufs=8,logbsize=32k,noquota)
/dev/loop2 on /mnt/loop2 type xfs (rw,relatime,seclabel,attr2,inode64,logbufs=8,logbsize=32k,noquota)
/dev/loop3 on /mnt/loop3 type xfs (rw,relatime,seclabel,attr2,inode64,logbufs=8,logbsize=32k,noquota)
/dev/loop5 on /mnt/loop5 type xfs (rw,relatime,seclabel,attr2,inode64,logbufs=8,logbsize=32k,noquota)
/dev/loop8 on /mnt/loop8 type xfs (rw,relatime,seclabel,attr2,inode64,logbufs=8,logbsize=32k,noquota)
/dev/loop6 on /mnt/loop6 type xfs (rw,relatime,seclabel,attr2,inode64,logbufs=8,logbsize=32k,noquota)
+ dmesg -c
[ 0.000000] Linux version 5.17.0-rc3blk+ (root@dev) (gcc (GCC) 11.0.1 20210324 (Red Hat 11.0.1-0), GNU ld version 2.35.1-41.fc34) #52 SMP PREEMPT Mon Feb 14 00:35:51 PST 2022
[ 0.000000] Command line: BOOT_IMAGE=(hd0,msdos1)/vmlinuz-5.17.0-rc3blk+ root=UUID=e5f9bccb-cc5d-4577-8f74-fddb710fae7f ro rootflags=subvol=root rhgb quiet console=ttyS0,115200 kgdboc=ttyS0,115200 nokaslr
[ 88.725708] loop: loading out-of-tree module taints kernel.
[ 88.795243] loop: module loaded
[ 88.886469] loop1: detected capacity change from 0 to 4194304
[ 90.099519] XFS (loop1): Mounting V5 Filesystem
[ 90.137943] XFS (loop1): Ending clean mount
[ 90.142241] xfs filesystem being mounted at /mnt/loop1 supports timestamps until 2038 (0x7fffffff)
[ 90.248857] loop4: detected capacity change from 0 to 4194304
[ 91.445059] XFS (loop4): Mounting V5 Filesystem
[ 91.461389] XFS (loop4): Ending clean mount
[ 91.466664] xfs filesystem being mounted at /mnt/loop4 supports timestamps until 2038 (0x7fffffff)
[ 91.567786] loop9: detected capacity change from 0 to 4194304
[ 92.762834] XFS (loop9): Mounting V5 Filesystem
[ 92.780240] XFS (loop9): Ending clean mount
[ 92.784748] xfs filesystem being mounted at /mnt/loop9 supports timestamps until 2038 (0x7fffffff)
[ 92.883355] loop10: detected capacity change from 0 to 4194304
[ 94.077276] XFS (loop10): Mounting V5 Filesystem
[ 94.095212] XFS (loop10): Ending clean mount
[ 94.101121] xfs filesystem being mounted at /mnt/loop10 supports timestamps until 2038 (0x7fffffff)
[ 94.200828] loop7: detected capacity change from 0 to 4194304
[ 95.387052] XFS (loop7): Mounting V5 Filesystem
[ 95.402344] XFS (loop7): Ending clean mount
[ 95.408219] xfs filesystem being mounted at /mnt/loop7 supports timestamps until 2038 (0x7fffffff)
[ 95.505530] loop2: detected capacity change from 0 to 4194304
[ 96.696228] XFS (loop2): Mounting V5 Filesystem
[ 96.712500] XFS (loop2): Ending clean mount
[ 96.717802] xfs filesystem being mounted at /mnt/loop2 supports timestamps until 2038 (0x7fffffff)
[ 96.815870] loop3: detected capacity change from 0 to 4194304
[ 98.006189] XFS (loop3): Mounting V5 Filesystem
[ 98.022896] XFS (loop3): Ending clean mount
[ 98.028990] xfs filesystem being mounted at /mnt/loop3 supports timestamps until 2038 (0x7fffffff)
[ 98.128314] loop5: detected capacity change from 0 to 4194304
[ 99.317471] XFS (loop5): Mounting V5 Filesystem
[ 99.334259] XFS (loop5): Ending clean mount
[ 99.340014] xfs filesystem being mounted at /mnt/loop5 supports timestamps until 2038 (0x7fffffff)
[ 99.437944] loop8: detected capacity change from 0 to 4194304
[ 100.640549] XFS (loop8): Mounting V5 Filesystem
[ 100.658762] XFS (loop8): Ending clean mount
[ 100.663623] xfs filesystem being mounted at /mnt/loop8 supports timestamps until 2038 (0x7fffffff)
[ 100.768163] loop6: detected capacity change from 0 to 4194304
[ 101.964471] XFS (loop6): Mounting V5 Filesystem
[ 101.985869] XFS (loop6): Ending clean mount
[ 101.991076] xfs filesystem being mounted at /mnt/loop6 supports timestamps until 2038 (0x7fffffff)
+ df -h /mnt/loop1 /mnt/loop10 /mnt/loop2 /mnt/loop3 /mnt/loop4 /mnt/loop5 /mnt/loop6 /mnt/loop7 /mnt/loop8 /mnt/loop9
Filesystem Size Used Avail Use% Mounted on
/dev/loop1 2.0G 47M 2.0G 3% /mnt/loop1
/dev/loop10 2.0G 47M 2.0G 3% /mnt/loop10
/dev/loop2 2.0G 47M 2.0G 3% /mnt/loop2
/dev/loop3 2.0G 47M 2.0G 3% /mnt/loop3
/dev/loop4 2.0G 47M 2.0G 3% /mnt/loop4
/dev/loop5 2.0G 47M 2.0G 3% /mnt/loop5
/dev/loop6 2.0G 47M 2.0G 3% /mnt/loop6
/dev/loop7 2.0G 47M 2.0G 3% /mnt/loop7
/dev/loop8 2.0G 47M 2.0G 3% /mnt/loop8
/dev/loop9 2.0G 47M 2.0G 3% /mnt/loop9
++ shuf -i 1-10 -n 10
+ for i in `shuf -i 1-$NN -n $NN`
+ fallocate -o 0 -l 524288000 /mnt/loop5/testfile
+ fio fio/verify.fio --filename=/mnt/loop5/testfile
write-and-verify: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=16
fio-3.27
Starting 1 process
write-and-verify: Laying out IO file (1 file / 1024MiB)
Jobs: 1 (f=1): [V(1)][81.7%][r=18.4MiB/s][r=4722 IOPS][eta 00m:30s]
write-and-verify: (groupid=0, jobs=1): err= 0: pid=3495: Mon Feb 14 00:43:19 2022
read: IOPS=4612, BW=18.0MiB/s (18.9MB/s)(647MiB/35913msec)
slat (usec): min=7, max=2920, avg=13.50, stdev=23.99
clat (usec): min=269, max=86467, avg=3453.04, stdev=1221.75
lat (usec): min=321, max=86484, avg=3466.65, stdev=1224.18
clat percentiles (usec):
| 1.00th=[ 2409], 5.00th=[ 2474], 10.00th=[ 2540], 20.00th=[ 2671],
| 30.00th=[ 2802], 40.00th=[ 3097], 50.00th=[ 3490], 60.00th=[ 3752],
| 70.00th=[ 3851], 80.00th=[ 3982], 90.00th=[ 4228], 95.00th=[ 4424],
| 99.00th=[ 6587], 99.50th=[ 7242], 99.90th=[ 9765], 99.95th=[11469],
| 99.99th=[46924]
write: IOPS=2671, BW=10.4MiB/s (10.9MB/s)(1024MiB/98116msec); 0 zone resets
slat (usec): min=9, max=136085, avg=32.59, stdev=613.45
clat (usec): min=255, max=344426, avg=5954.50, stdev=13130.50
lat (usec): min=490, max=344452, avg=5987.24, stdev=13189.93
clat percentiles (msec):
| 1.00th=[ 3], 5.00th=[ 3], 10.00th=[ 4], 20.00th=[ 4],
| 30.00th=[ 4], 40.00th=[ 4], 50.00th=[ 4], 60.00th=[ 5],
| 70.00th=[ 6], 80.00th=[ 6], 90.00th=[ 7], 95.00th=[ 11],
| 99.00th=[ 36], 99.50th=[ 93], 99.90th=[ 228], 99.95th=[ 251],
| 99.99th=[ 292]
bw ( KiB/s): min= 240, max=19040, per=99.67%, avg=10652.53, stdev=4662.38, samples=197
iops : min= 60, max= 4760, avg=2663.01, stdev=1165.58, samples=197
lat (usec) : 500=0.01%, 750=0.01%, 1000=0.01%
lat (msec) : 2=0.02%, 4=63.60%, 10=33.10%, 20=1.91%, 50=0.87%
lat (msec) : 100=0.20%, 250=0.25%, 500=0.03%
cpu : usr=1.89%, sys=8.56%, ctx=422844, majf=0, minf=3901
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=100.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.1%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=165662,262144,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=16
Run status group 0 (all jobs):
READ: bw=18.0MiB/s (18.9MB/s), 18.0MiB/s-18.0MiB/s (18.9MB/s-18.9MB/s), io=647MiB (679MB), run=35913-35913msec
WRITE: bw=10.4MiB/s (10.9MB/s), 10.4MiB/s-10.4MiB/s (10.9MB/s-10.9MB/s), io=1024MiB (1074MB), run=98116-98116msec
Disk stats (read/write):
loop5: ios=165665/274948, merge=0/1146, ticks=570916/1710663, in_queue=2291246, util=99.71%
+ for i in `shuf -i 1-$NN -n $NN`
+ fallocate -o 0 -l 524288000 /mnt/loop3/testfile
+ fio fio/verify.fio --filename=/mnt/loop3/testfile
write-and-verify: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=16
fio-3.27
Starting 1 process
write-and-verify: Laying out IO file (1 file / 1024MiB)
Jobs: 1 (f=1): [V(1)][81.4%][r=19.4MiB/s][r=4959 IOPS][eta 00m:27s]
write-and-verify: (groupid=0, jobs=1): err= 0: pid=3960: Mon Feb 14 00:45:17 2022
read: IOPS=4591, BW=17.9MiB/s (18.8MB/s)(647MiB/36059msec)
slat (usec): min=7, max=3764, avg=12.77, stdev=37.77
clat (usec): min=328, max=195361, avg=3469.75, stdev=2029.21
lat (usec): min=380, max=195375, avg=3482.63, stdev=2031.19
clat percentiles (usec):
| 1.00th=[ 2442], 5.00th=[ 2507], 10.00th=[ 2573], 20.00th=[ 2737],
| 30.00th=[ 2900], 40.00th=[ 3097], 50.00th=[ 3458], 60.00th=[ 3720],
| 70.00th=[ 3851], 80.00th=[ 3949], 90.00th=[ 4228], 95.00th=[ 4424],
| 99.00th=[ 6521], 99.50th=[ 7111], 99.90th=[ 9372], 99.95th=[ 9765],
| 99.99th=[17433]
write: IOPS=3197, BW=12.5MiB/s (13.1MB/s)(1024MiB/81996msec); 0 zone resets
slat (usec): min=9, max=36590, avg=26.96, stdev=175.62
clat (usec): min=383, max=140174, avg=4976.36, stdev=4045.11
lat (usec): min=477, max=140202, avg=5003.46, stdev=4051.17
clat percentiles (usec):
| 1.00th=[ 2868], 5.00th=[ 2999], 10.00th=[ 3097], 20.00th=[ 3228],
| 30.00th=[ 3326], 40.00th=[ 3556], 50.00th=[ 3949], 60.00th=[ 4686],
| 70.00th=[ 5211], 80.00th=[ 5604], 90.00th=[ 6325], 95.00th=[ 9241],
| 99.00th=[24511], 99.50th=[31065], 99.90th=[47449], 99.95th=[65799],
| 99.99th=[96994]
bw ( KiB/s): min= 4832, max=18504, per=100.00%, avg=12793.93, stdev=2620.03, samples=164
iops : min= 1208, max= 4626, avg=3198.38, stdev=654.99, samples=164
lat (usec) : 500=0.01%, 750=0.01%, 1000=0.01%
lat (msec) : 2=0.05%, 4=62.80%, 10=34.34%, 20=1.93%, 50=0.81%
lat (msec) : 100=0.05%, 250=0.01%
cpu : usr=2.02%, sys=9.13%, ctx=422417, majf=0, minf=3898
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=100.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.1%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=165574,262144,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=16
Run status group 0 (all jobs):
READ: bw=17.9MiB/s (18.8MB/s), 17.9MiB/s-17.9MiB/s (18.8MB/s-18.8MB/s), io=647MiB (678MB), run=36059-36059msec
WRITE: bw=12.5MiB/s (13.1MB/s), 12.5MiB/s-12.5MiB/s (13.1MB/s-13.1MB/s), io=1024MiB (1074MB), run=81996-81996msec
Disk stats (read/write):
loop3: ios=165516/274739, merge=0/1239, ticks=573166/1354400, in_queue=1935490, util=99.86%
+ for i in `shuf -i 1-$NN -n $NN`
+ fallocate -o 0 -l 524288000 /mnt/loop1/testfile
+ fio fio/verify.fio --filename=/mnt/loop1/testfile
write-and-verify: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=16
fio-3.27
Starting 1 process
write-and-verify: Laying out IO file (1 file / 1024MiB)
Jobs: 1 (f=1): [V(1)][81.2%][r=16.4MiB/s][r=4188 IOPS][eta 00m:27s]
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4321: Mon Feb 14 00:47:15 2022
read: IOPS=4598, BW=18.0MiB/s (18.8MB/s)(648MiB/36048msec)
slat (usec): min=7, max=6302, avg=12.44, stdev=31.83
clat (usec): min=421, max=184955, avg=3464.65, stdev=1920.29
lat (usec): min=450, max=184973, avg=3477.19, stdev=1921.75
clat percentiles (usec):
| 1.00th=[ 2442], 5.00th=[ 2507], 10.00th=[ 2573], 20.00th=[ 2737],
| 30.00th=[ 2868], 40.00th=[ 3163], 50.00th=[ 3523], 60.00th=[ 3752],
| 70.00th=[ 3851], 80.00th=[ 3949], 90.00th=[ 4228], 95.00th=[ 4359],
| 99.00th=[ 6325], 99.50th=[ 6915], 99.90th=[ 8291], 99.95th=[ 9110],
| 99.99th=[15139]
write: IOPS=3213, BW=12.6MiB/s (13.2MB/s)(1024MiB/81578msec); 0 zone resets
slat (usec): min=9, max=103942, avg=26.21, stdev=275.35
clat (usec): min=462, max=202364, avg=4951.53, stdev=4217.15
lat (usec): min=776, max=202401, avg=4977.88, stdev=4237.99
clat percentiles (msec):
| 1.00th=[ 3], 5.00th=[ 4], 10.00th=[ 4], 20.00th=[ 4],
| 30.00th=[ 4], 40.00th=[ 4], 50.00th=[ 4], 60.00th=[ 5],
| 70.00th=[ 6], 80.00th=[ 6], 90.00th=[ 7], 95.00th=[ 10],
| 99.00th=[ 24], 99.50th=[ 30], 99.90th=[ 40], 99.95th=[ 54],
| 99.99th=[ 138]
bw ( KiB/s): min= 4368, max=19240, per=100.00%, avg=12865.96, stdev=2971.61, samples=163
iops : min= 1092, max= 4810, avg=3216.49, stdev=742.90, samples=163
lat (usec) : 500=0.01%, 750=0.01%, 1000=0.01%
lat (msec) : 2=0.01%, 4=64.55%, 10=32.44%, 20=2.12%, 50=0.83%
lat (msec) : 100=0.03%, 250=0.01%
cpu : usr=2.03%, sys=9.06%, ctx=423983, majf=0, minf=3903
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=100.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.1%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=165779,262144,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=16
Run status group 0 (all jobs):
READ: bw=18.0MiB/s (18.8MB/s), 18.0MiB/s-18.0MiB/s (18.8MB/s-18.8MB/s), io=648MiB (679MB), run=36048-36048msec
WRITE: bw=12.6MiB/s (13.2MB/s), 12.6MiB/s-12.6MiB/s (13.2MB/s-13.2MB/s), io=1024MiB (1074MB), run=81578-81578msec
Disk stats (read/write):
loop1: ios=164590/274753, merge=0/972, ticks=569739/1340209, in_queue=1917586, util=99.90%
+ for i in `shuf -i 1-$NN -n $NN`
+ fallocate -o 0 -l 524288000 /mnt/loop7/testfile
+ fio fio/verify.fio --filename=/mnt/loop7/testfile
write-and-verify: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=16
fio-3.27
Starting 1 process
write-and-verify: Laying out IO file (1 file / 1024MiB)
Jobs: 1 (f=1): [V(1)][81.0%][r=18.5MiB/s][r=4743 IOPS][eta 00m:29s]
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4369: Mon Feb 14 00:49:20 2022
read: IOPS=4576, BW=17.9MiB/s (18.7MB/s)(648MiB/36269msec)
slat (usec): min=7, max=3455, avg=12.29, stdev=18.44
clat (usec): min=339, max=183878, avg=3481.44, stdev=1929.01
lat (usec): min=392, max=183901, avg=3493.84, stdev=1930.15
clat percentiles (usec):
| 1.00th=[ 2442], 5.00th=[ 2507], 10.00th=[ 2573], 20.00th=[ 2704],
| 30.00th=[ 2868], 40.00th=[ 3163], 50.00th=[ 3523], 60.00th=[ 3752],
| 70.00th=[ 3851], 80.00th=[ 3982], 90.00th=[ 4228], 95.00th=[ 4424],
| 99.00th=[ 6521], 99.50th=[ 7046], 99.90th=[10159], 99.95th=[11338],
| 99.99th=[14222]
write: IOPS=2964, BW=11.6MiB/s (12.1MB/s)(1024MiB/88415msec); 0 zone resets
slat (usec): min=9, max=22005, avg=26.98, stdev=119.64
clat (usec): min=363, max=331919, avg=5368.06, stdev=8943.04
lat (usec): min=660, max=331946, avg=5395.19, stdev=8945.60
clat percentiles (msec):
| 1.00th=[ 3], 5.00th=[ 4], 10.00th=[ 4], 20.00th=[ 4],
| 30.00th=[ 4], 40.00th=[ 4], 50.00th=[ 4], 60.00th=[ 5],
| 70.00th=[ 6], 80.00th=[ 6], 90.00th=[ 7], 95.00th=[ 11],
| 99.00th=[ 27], 99.50th=[ 33], 99.90th=[ 153], 99.95th=[ 232],
| 99.99th=[ 313]
bw ( KiB/s): min= 200, max=19288, per=99.94%, avg=11853.62, stdev=3752.39, samples=177
iops : min= 50, max= 4822, avg=2963.33, stdev=938.07, samples=177
lat (usec) : 500=0.01%, 750=0.01%, 1000=0.01%
lat (msec) : 2=0.02%, 4=62.68%, 10=33.85%, 20=2.47%, 50=0.82%
lat (msec) : 100=0.05%, 250=0.07%, 500=0.02%
cpu : usr=1.98%, sys=8.76%, ctx=423946, majf=0, minf=3908
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=100.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.1%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=166003,262144,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=16
Run status group 0 (all jobs):
READ: bw=17.9MiB/s (18.7MB/s), 17.9MiB/s-17.9MiB/s (18.7MB/s-18.7MB/s), io=648MiB (680MB), run=36269-36269msec
WRITE: bw=11.6MiB/s (12.1MB/s), 11.6MiB/s-11.6MiB/s (12.1MB/s-12.1MB/s), io=1024MiB (1074MB), run=88415-88415msec
Disk stats (read/write):
loop7: ios=165505/275157, merge=0/445, ticks=575262/1491091, in_queue=2074451, util=99.80%
+ for i in `shuf -i 1-$NN -n $NN`
+ fallocate -o 0 -l 524288000 /mnt/loop2/testfile
+ fio fio/verify.fio --filename=/mnt/loop2/testfile
write-and-verify: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=16
fio-3.27
Starting 1 process
write-and-verify: Laying out IO file (1 file / 1024MiB)
Jobs: 1 (f=1): [V(1)][81.0%][r=19.8MiB/s][r=5069 IOPS][eta 00m:29s]
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4443: Mon Feb 14 00:51:25 2022
read: IOPS=4727, BW=18.5MiB/s (19.4MB/s)(647MiB/35059msec)
slat (usec): min=7, max=4887, avg=12.81, stdev=23.98
clat (usec): min=245, max=132700, avg=3369.45, stdev=1488.12
lat (usec): min=288, max=132731, avg=3382.36, stdev=1489.93
clat percentiles (usec):
| 1.00th=[ 2409], 5.00th=[ 2507], 10.00th=[ 2540], 20.00th=[ 2638],
| 30.00th=[ 2769], 40.00th=[ 2966], 50.00th=[ 3294], 60.00th=[ 3621],
| 70.00th=[ 3785], 80.00th=[ 3916], 90.00th=[ 4228], 95.00th=[ 4359],
| 99.00th=[ 6390], 99.50th=[ 6915], 99.90th=[ 9503], 99.95th=[10552],
| 99.99th=[13829]
write: IOPS=2931, BW=11.5MiB/s (12.0MB/s)(1024MiB/89414msec); 0 zone resets
slat (usec): min=9, max=64714, avg=31.32, stdev=289.47
clat (usec): min=370, max=161699, avg=5424.62, stdev=4208.98
lat (usec): min=524, max=161746, avg=5456.10, stdev=4225.29
clat percentiles (usec):
| 1.00th=[ 3064], 5.00th=[ 3163], 10.00th=[ 3228], 20.00th=[ 3392],
| 30.00th=[ 3621], 40.00th=[ 3851], 50.00th=[ 4178], 60.00th=[ 4621],
| 70.00th=[ 5473], 80.00th=[ 6325], 90.00th=[ 8094], 95.00th=[11469],
| 99.00th=[25560], 99.50th=[31589], 99.90th=[44303], 99.95th=[54789],
| 99.99th=[81265]
bw ( KiB/s): min= 1408, max=19264, per=99.91%, avg=11717.60, stdev=3203.42, samples=179
iops : min= 352, max= 4816, avg=2929.39, stdev=800.85, samples=179
lat (usec) : 250=0.01%, 500=0.01%, 750=0.01%, 1000=0.01%
lat (msec) : 2=0.03%, 4=60.44%, 10=35.67%, 20=2.88%, 50=0.93%
lat (msec) : 100=0.04%, 250=0.01%
cpu : usr=2.12%, sys=9.57%, ctx=423544, majf=0, minf=3902
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=100.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.1%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=165748,262144,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=16
Run status group 0 (all jobs):
READ: bw=18.5MiB/s (19.4MB/s), 18.5MiB/s-18.5MiB/s (19.4MB/s-19.4MB/s), io=647MiB (679MB), run=35059-35059msec
WRITE: bw=11.5MiB/s (12.0MB/s), 11.5MiB/s-11.5MiB/s (12.0MB/s-12.0MB/s), io=1024MiB (1074MB), run=89414-89414msec
Disk stats (read/write):
loop2: ios=165107/275671, merge=0/849, ticks=555353/1492768, in_queue=2056959, util=99.84%
+ for i in `shuf -i 1-$NN -n $NN`
+ fallocate -o 0 -l 524288000 /mnt/loop9/testfile
+ fio fio/verify.fio --filename=/mnt/loop9/testfile
write-and-verify: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=16
fio-3.27
Starting 1 process
write-and-verify: Laying out IO file (1 file / 1024MiB)
Jobs: 1 (f=1): [V(1)][81.6%][r=17.8MiB/s][r=4544 IOPS][eta 00m:27s]
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4536: Mon Feb 14 00:53:25 2022
read: IOPS=4578, BW=17.9MiB/s (18.8MB/s)(648MiB/36203msec)
slat (usec): min=7, max=3656, avg=12.64, stdev=30.11
clat (usec): min=433, max=172133, avg=3479.80, stdev=1835.20
lat (usec): min=461, max=172146, avg=3492.54, stdev=1836.82
clat percentiles (usec):
| 1.00th=[ 2442], 5.00th=[ 2507], 10.00th=[ 2540], 20.00th=[ 2737],
| 30.00th=[ 2868], 40.00th=[ 3195], 50.00th=[ 3556], 60.00th=[ 3785],
| 70.00th=[ 3851], 80.00th=[ 3982], 90.00th=[ 4228], 95.00th=[ 4359],
| 99.00th=[ 6194], 99.50th=[ 6849], 99.90th=[ 9634], 99.95th=[10945],
| 99.99th=[27657]
write: IOPS=3129, BW=12.2MiB/s (12.8MB/s)(1024MiB/83768msec); 0 zone resets
slat (usec): min=9, max=118411, avg=27.16, stdev=269.68
clat (usec): min=399, max=156396, avg=5084.28, stdev=4265.18
lat (usec): min=611, max=156425, avg=5111.58, stdev=4275.66
clat percentiles (msec):
| 1.00th=[ 3], 5.00th=[ 3], 10.00th=[ 4], 20.00th=[ 4],
| 30.00th=[ 4], 40.00th=[ 4], 50.00th=[ 4], 60.00th=[ 5],
| 70.00th=[ 6], 80.00th=[ 6], 90.00th=[ 7], 95.00th=[ 11],
| 99.00th=[ 25], 99.50th=[ 31], 99.90th=[ 48], 99.95th=[ 68],
| 99.99th=[ 127]
bw ( KiB/s): min= 3808, max=17688, per=99.75%, avg=12486.86, stdev=2928.80, samples=168
iops : min= 952, max= 4422, avg=3121.66, stdev=732.17, samples=168
lat (usec) : 500=0.01%, 750=0.01%, 1000=0.01%
lat (msec) : 2=0.06%, 4=62.84%, 10=33.78%, 20=2.44%, 50=0.82%
lat (msec) : 100=0.05%, 250=0.01%
cpu : usr=1.94%, sys=9.19%, ctx=422880, majf=0, minf=3903
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=100.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.1%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=165766,262144,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=16
Run status group 0 (all jobs):
READ: bw=17.9MiB/s (18.8MB/s), 17.9MiB/s-17.9MiB/s (18.8MB/s-18.8MB/s), io=648MiB (679MB), run=36203-36203msec
WRITE: bw=12.2MiB/s (12.8MB/s), 12.2MiB/s-12.2MiB/s (12.8MB/s-12.8MB/s), io=1024MiB (1074MB), run=83768-83768msec
Disk stats (read/write):
loop9: ios=164914/275126, merge=0/801, ticks=573496/1387584, in_queue=1969380, util=99.87%
+ for i in `shuf -i 1-$NN -n $NN`
+ fallocate -o 0 -l 524288000 /mnt/loop8/testfile
+ fio fio/verify.fio --filename=/mnt/loop8/testfile
write-and-verify: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=16
fio-3.27
Starting 1 process
write-and-verify: Laying out IO file (1 file / 1024MiB)
./kernel_compile.sh Jobs: 1 (f=1): [V(1)][75.7%][r=19.5MiB/s][r=4996 IOPS][eta 00m:36s]
Jobs: 1 (f=1): [V(1)][81.5%][r=18.9MiB/s][r=4826 IOPS][eta 00m:27s]
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4770: Mon Feb 14 00:55:24 2022
read: IOPS=4440, BW=17.3MiB/s (18.2MB/s)(647MiB/37270msec)
slat (usec): min=7, max=2362, avg=13.33, stdev=18.79
clat (usec): min=303, max=197336, avg=3587.52, stdev=2050.10
lat (usec): min=355, max=197349, avg=3600.96, stdev=2051.09
clat percentiles (usec):
| 1.00th=[ 2474], 5.00th=[ 2540], 10.00th=[ 2638], 20.00th=[ 2802],
| 30.00th=[ 2999], 40.00th=[ 3392], 50.00th=[ 3720], 60.00th=[ 3851],
| 70.00th=[ 3916], 80.00th=[ 4047], 90.00th=[ 4293], 95.00th=[ 4490],
| 99.00th=[ 6521], 99.50th=[ 7046], 99.90th=[ 9110], 99.95th=[10421],
| 99.99th=[17433]
write: IOPS=3205, BW=12.5MiB/s (13.1MB/s)(1024MiB/81785msec); 0 zone resets
slat (usec): min=9, max=90569, avg=27.29, stdev=290.77
clat (usec): min=360, max=145421, avg=4963.20, stdev=4064.59
lat (usec): min=506, max=145449, avg=4990.62, stdev=4085.24
clat percentiles (usec):
| 1.00th=[ 2835], 5.00th=[ 2966], 10.00th=[ 3032], 20.00th=[ 3163],
| 30.00th=[ 3261], 40.00th=[ 3490], 50.00th=[ 3982], 60.00th=[ 4621],
| 70.00th=[ 5145], 80.00th=[ 5604], 90.00th=[ 6521], 95.00th=[ 9372],
| 99.00th=[25035], 99.50th=[30540], 99.90th=[45351], 99.95th=[58459],
| 99.99th=[88605]
bw ( KiB/s): min= 2632, max=20184, per=99.73%, avg=12787.02, stdev=2904.81, samples=164
iops : min= 658, max= 5046, avg=3196.75, stdev=726.20, samples=164
lat (usec) : 500=0.01%, 750=0.01%, 1000=0.01%
lat (msec) : 2=0.05%, 4=60.54%, 10=36.65%, 20=1.84%, 50=0.86%
lat (msec) : 100=0.04%, 250=0.01%
cpu : usr=1.99%, sys=9.24%, ctx=423283, majf=0, minf=3903
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=100.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.1%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=165505,262144,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=16
Run status group 0 (all jobs):
READ: bw=17.3MiB/s (18.2MB/s), 17.3MiB/s-17.3MiB/s (18.2MB/s-18.2MB/s), io=647MiB (678MB), run=37270-37270msec
WRITE: bw=12.5MiB/s (13.1MB/s), 12.5MiB/s-12.5MiB/s (13.1MB/s-13.1MB/s), io=1024MiB (1074MB), run=81785-81785msec
Disk stats (read/write):
loop8: ios=164724/275455, merge=0/496, ticks=589770/1367679, in_queue=1965080, util=99.91%
+ for i in `shuf -i 1-$NN -n $NN`
+ fallocate -o 0 -l 524288000 /mnt/loop10/testfile
+ fio fio/verify.fio --filename=/mnt/loop10/testfile
write-and-verify: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=16
fio-3.27
Starting 1 process
write-and-verify: Laying out IO file (1 file / 1024MiB)
Jobs: 1 (f=1): [V(1)][81.3%][r=17.8MiB/s][r=4556 IOPS][eta 00m:29s]
write-and-verify: (groupid=0, jobs=1): err= 0: pid=4933: Mon Feb 14 00:57:31 2022
read: IOPS=4707, BW=18.4MiB/s (19.3MB/s)(647MiB/35192msec)
slat (usec): min=7, max=3278, avg=12.42, stdev=17.36
clat (usec): min=188, max=110908, avg=3384.22, stdev=1312.64
lat (usec): min=216, max=110935, avg=3396.74, stdev=1314.44
clat percentiles (usec):
| 1.00th=[ 2376], 5.00th=[ 2442], 10.00th=[ 2474], 20.00th=[ 2573],
| 30.00th=[ 2737], 40.00th=[ 2966], 50.00th=[ 3425], 60.00th=[ 3720],
| 70.00th=[ 3851], 80.00th=[ 3949], 90.00th=[ 4228], 95.00th=[ 4359],
| 99.00th=[ 6390], 99.50th=[ 6915], 99.90th=[ 9765], 99.95th=[10945],
| 99.99th=[13960]
write: IOPS=2875, BW=11.2MiB/s (11.8MB/s)(1024MiB/91151msec); 0 zone resets
slat (usec): min=9, max=188191, avg=31.96, stdev=677.26
clat (usec): min=327, max=498572, avg=5530.16, stdev=11129.13
lat (usec): min=360, max=498617, avg=5562.26, stdev=11210.29
clat percentiles (msec):
| 1.00th=[ 3], 5.00th=[ 3], 10.00th=[ 3], 20.00th=[ 4],
| 30.00th=[ 4], 40.00th=[ 4], 50.00th=[ 4], 60.00th=[ 5],
| 70.00th=[ 6], 80.00th=[ 6], 90.00th=[ 7], 95.00th=[ 11],
| 99.00th=[ 30], 99.50th=[ 48], 99.90th=[ 207], 99.95th=[ 249],
| 99.99th=[ 292]
bw ( KiB/s): min= 208, max=20048, per=99.67%, avg=11466.34, stdev=4330.86, samples=183
iops : min= 52, max= 5012, avg=2866.48, stdev=1082.72, samples=183
lat (usec) : 250=0.01%, 500=0.01%, 750=0.01%, 1000=0.01%
lat (msec) : 2=0.04%, 4=63.80%, 10=32.95%, 20=2.05%, 50=0.86%
lat (msec) : 100=0.11%, 250=0.15%, 500=0.03%
cpu : usr=1.84%, sys=8.64%, ctx=423262, majf=0, minf=3902
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=100.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.1%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=165678,262144,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=16
Run status group 0 (all jobs):
READ: bw=18.4MiB/s (19.3MB/s), 18.4MiB/s-18.4MiB/s (19.3MB/s-19.3MB/s), io=647MiB (679MB), run=35192-35192msec
WRITE: bw=11.2MiB/s (11.8MB/s), 11.2MiB/s-11.2MiB/s (11.8MB/s-11.8MB/s), io=1024MiB (1074MB), run=91151-91151msec
Disk stats (read/write):
loop10: ios=165337/275417, merge=0/558, ticks=558916/1602182, in_queue=2171960, util=99.48%
+ for i in `shuf -i 1-$NN -n $NN`
+ fallocate -o 0 -l 524288000 /mnt/loop6/testfile
+ fio fio/verify.fio --filename=/mnt/loop6/testfile
write-and-verify: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=16
fio-3.27
Starting 1 process
write-and-verify: Laying out IO file (1 file / 1024MiB)
Jobs: 1 (f=1): [V(1)][81.5%][r=17.2MiB/s][r=4407 IOPS][eta 00m:28s]
write-and-verify: (groupid=0, jobs=1): err= 0: pid=5008: Mon Feb 14 00:59:35 2022
read: IOPS=4388, BW=17.1MiB/s (18.0MB/s)(648MiB/37775msec)
slat (usec): min=7, max=3957, avg=13.73, stdev=18.70
clat (usec): min=214, max=30818, avg=3630.04, stdev=887.40
lat (usec): min=246, max=30833, avg=3643.89, stdev=889.77
clat percentiles (usec):
| 1.00th=[ 2474], 5.00th=[ 2573], 10.00th=[ 2671], 20.00th=[ 2835],
| 30.00th=[ 3032], 40.00th=[ 3326], 50.00th=[ 3654], 60.00th=[ 3851],
| 70.00th=[ 3982], 80.00th=[ 4228], 90.00th=[ 4490], 95.00th=[ 4817],
| 99.00th=[ 6521], 99.50th=[ 7242], 99.90th=[ 9634], 99.95th=[10814],
| 99.99th=[26608]
write: IOPS=3058, BW=11.9MiB/s (12.5MB/s)(1024MiB/85702msec); 0 zone resets
slat (usec): min=9, max=64165, avg=28.45, stdev=341.57
clat (usec): min=283, max=292636, avg=5201.04, stdev=7778.65
lat (usec): min=419, max=300151, avg=5229.64, stdev=7823.47
clat percentiles (msec):
| 1.00th=[ 3], 5.00th=[ 3], 10.00th=[ 4], 20.00th=[ 4],
| 30.00th=[ 4], 40.00th=[ 4], 50.00th=[ 4], 60.00th=[ 5],
| 70.00th=[ 5], 80.00th=[ 6], 90.00th=[ 7], 95.00th=[ 11],
| 99.00th=[ 28], 99.50th=[ 35], 99.90th=[ 121], 99.95th=[ 188],
| 99.99th=[ 259]
bw ( KiB/s): min= 368, max=19680, per=99.66%, avg=12193.70, stdev=3835.99, samples=172
iops : min= 92, max= 4920, avg=3048.40, stdev=958.99, samples=172
lat (usec) : 250=0.01%, 500=0.01%, 750=0.01%, 1000=0.01%
lat (msec) : 2=0.05%, 4=61.29%, 10=35.42%, 20=2.16%, 50=0.89%
lat (msec) : 100=0.11%, 250=0.07%, 500=0.01%
cpu : usr=1.93%, sys=8.80%, ctx=423706, majf=0, minf=3901
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=100.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.1%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=165770,262144,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=16
Run status group 0 (all jobs):
READ: bw=17.1MiB/s (18.0MB/s), 17.1MiB/s-17.1MiB/s (18.0MB/s-18.0MB/s), io=648MiB (679MB), run=37775-37775msec
WRITE: bw=11.9MiB/s (12.5MB/s), 11.9MiB/s-11.9MiB/s (12.5MB/s-12.5MB/s), io=1024MiB (1074MB), run=85702-85702msec
Disk stats (read/write):
loop6: ios=165239/275207, merge=0/399, ticks=599041/1461570, in_queue=2069975, util=99.69%
+ for i in `shuf -i 1-$NN -n $NN`
+ fallocate -o 0 -l 524288000 /mnt/loop4/testfile
+ fio fio/verify.fio --filename=/mnt/loop4/testfile
write-and-verify: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=16
fio-3.27
Starting 1 process
write-and-verify: Laying out IO file (1 file / 1024MiB)
Jobs: 1 (f=1): [V(1)][81.1%][r=18.5MiB/s][r=4724 IOPS][eta 00m:33s]
write-and-verify: (groupid=0, jobs=1): err= 0: pid=5048: Mon Feb 14 01:01:58 2022
read: IOPS=4523, BW=17.7MiB/s (18.5MB/s)(647MiB/36630msec)
slat (usec): min=7, max=5014, avg=13.45, stdev=45.87
clat (usec): min=231, max=180178, avg=3521.35, stdev=1920.30
lat (usec): min=263, max=180195, avg=3534.91, stdev=1923.48
clat percentiles (usec):
| 1.00th=[ 2442], 5.00th=[ 2573], 10.00th=[ 2638], 20.00th=[ 2802],
| 30.00th=[ 2933], 40.00th=[ 3130], 50.00th=[ 3425], 60.00th=[ 3687],
| 70.00th=[ 3884], 80.00th=[ 4047], 90.00th=[ 4359], 95.00th=[ 4621],
| 99.00th=[ 6521], 99.50th=[ 7177], 99.90th=[10945], 99.95th=[12256],
| 99.99th=[20055]
write: IOPS=2477, BW=9911KiB/s (10.1MB/s)(1024MiB/105799msec); 0 zone resets
slat (usec): min=9, max=149056, avg=38.32, stdev=830.44
clat (usec): min=576, max=335748, avg=6417.70, stdev=16705.28
lat (usec): min=677, max=335834, avg=6456.18, stdev=16804.79
clat percentiles (msec):
| 1.00th=[ 3], 5.00th=[ 4], 10.00th=[ 4], 20.00th=[ 4],
| 30.00th=[ 4], 40.00th=[ 4], 50.00th=[ 5], 60.00th=[ 5],
| 70.00th=[ 6], 80.00th=[ 6], 90.00th=[ 8], 95.00th=[ 12],
| 99.00th=[ 41], 99.50th=[ 132], 99.90th=[ 271], 99.95th=[ 292],
| 99.99th=[ 330]
bw ( KiB/s): min= 200, max=17667, per=99.85%, avg=9896.22, stdev=4614.79, samples=212
iops : min= 50, max= 4416, avg=2473.96, stdev=1153.68, samples=212
lat (usec) : 250=0.01%, 500=0.01%, 750=0.01%, 1000=0.01%
lat (msec) : 2=0.02%, 4=60.24%, 10=36.23%, 20=2.09%, 50=0.90%
lat (msec) : 100=0.14%, 250=0.29%, 500=0.10%
cpu : usr=1.82%, sys=7.77%, ctx=422823, majf=0, minf=3901
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=100.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.1%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=165702,262144,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=16
Run status group 0 (all jobs):
READ: bw=17.7MiB/s (18.5MB/s), 17.7MiB/s-17.7MiB/s (18.5MB/s-18.5MB/s), io=647MiB (679MB), run=36630-36630msec
WRITE: bw=9911KiB/s (10.1MB/s), 9911KiB/s-9911KiB/s (10.1MB/s-10.1MB/s), io=1024MiB (1074MB), run=105799-105799msec
Disk stats (read/write):
loop4: ios=165422/275347, merge=0/459, ticks=581057/1918398, in_queue=2512842, util=99.54%
+ df -h /mnt/loop1 /mnt/loop10 /mnt/loop2 /mnt/loop3 /mnt/loop4 /mnt/loop5 /mnt/loop6 /mnt/loop7 /mnt/loop8 /mnt/loop9
Filesystem Size Used Avail Use% Mounted on
/dev/loop1 2.0G 1.1G 965M 53% /mnt/loop1
/dev/loop10 2.0G 1.1G 966M 53% /mnt/loop10
/dev/loop2 2.0G 1.1G 966M 53% /mnt/loop2
/dev/loop3 2.0G 1.1G 966M 53% /mnt/loop3
/dev/loop4 2.0G 1.1G 965M 53% /mnt/loop4
/dev/loop5 2.0G 1.1G 966M 53% /mnt/loop5
/dev/loop6 2.0G 1.1G 966M 53% /mnt/loop6
/dev/loop7 2.0G 1.1G 966M 53% /mnt/loop7
/dev/loop8 2.0G 1.1G 965M 53% /mnt/loop8
/dev/loop9 2.0G 1.1G 965M 53% /mnt/loop9
+ unload_loop
++ shuf -i 1-10 -n 10
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop3
+ rm -fr ./loop3
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop10
+ rm -fr ./loop10
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop5
+ rm -fr ./loop5
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop8
+ rm -fr ./loop8
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop9
+ rm -fr ./loop9
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop4
+ rm -fr ./loop4
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop7
+ rm -fr ./loop7
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop2
+ rm -fr ./loop2
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop1
+ rm -fr ./loop1
+ for i in `shuf -i 1-$NN -n $NN`
+ umount /mnt/loop6
+ rm -fr ./loop6
+ losetup -D
+ sleep 3
+ rmmod loop
+ modprobe -r loop
+ lsmod
+ grep loop
++ shuf -i 1-10 -n 10
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop7
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop8
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop1
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop4
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop6
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop10
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop2
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop3
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop5
+ for i in `shuf -i 1-$NN -n $NN`
+ rm -fr ./loop9
+ rm -fr /mnt/loop1 /mnt/loop10 /mnt/loop2 /mnt/loop3 /mnt/loop4 /mnt/loop5 /mnt/loop6 /mnt/loop7 /mnt/loop8 /mnt/loop9
+ dmesg -c
[ 1355.637847] XFS (loop3): Unmounting Filesystem
[ 1355.777851] XFS (loop10): Unmounting Filesystem
[ 1355.916829] XFS (loop5): Unmounting Filesystem
[ 1356.049920] XFS (loop8): Unmounting Filesystem
[ 1356.197833] XFS (loop9): Unmounting Filesystem
[ 1356.332848] XFS (loop4): Unmounting Filesystem
[ 1356.519850] XFS (loop7): Unmounting Filesystem
[ 1356.655913] XFS (loop2): Unmounting Filesystem
[ 1356.781838] XFS (loop1): Unmounting Filesystem
[ 1356.909819] XFS (loop6): Unmounting Filesystem
--
2.29.0
^ permalink raw reply [flat|nested] 12+ messages in thread
* [PATCH 1/8] loop: use sysfs_emit() in the sysfs offset show
2022-02-14 10:01 [PATCH 0/8] loop: cleanup and few improvements Chaitanya Kulkarni
@ 2022-02-14 10:01 ` Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 2/8] loop: use sysfs_emit() in the sysfs sizelimit show Chaitanya Kulkarni
` (7 subsequent siblings)
8 siblings, 0 replies; 12+ messages in thread
From: Chaitanya Kulkarni @ 2022-02-14 10:01 UTC (permalink / raw)
To: linux-block; +Cc: axboe, Chaitanya Kulkarni, Himanshu Madhani
sprintf does not know the PAGE_SIZE maximum of the temporary buffer
used for outputting sysfs content and it's possible to overrun the
PAGE_SIZE buffer length.
Use a generic sysfs_emit function that knows the size of the
temporary buffer and ensures that no overrun is done for offset
attribute.
Signed-off-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
---
drivers/block/loop.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index bdea448d2419..f6ed265d34e7 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -680,7 +680,7 @@ static ssize_t loop_attr_backing_file_show(struct loop_device *lo, char *buf)
static ssize_t loop_attr_offset_show(struct loop_device *lo, char *buf)
{
- return sprintf(buf, "%llu\n", (unsigned long long)lo->lo_offset);
+ return sysfs_emit(buf, "%llu\n", (unsigned long long)lo->lo_offset);
}
static ssize_t loop_attr_sizelimit_show(struct loop_device *lo, char *buf)
--
2.29.0
^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH 2/8] loop: use sysfs_emit() in the sysfs sizelimit show
2022-02-14 10:01 [PATCH 0/8] loop: cleanup and few improvements Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 1/8] loop: use sysfs_emit() in the sysfs offset show Chaitanya Kulkarni
@ 2022-02-14 10:01 ` Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 3/8] loop: use sysfs_emit() in the sysfs autoclear show Chaitanya Kulkarni
` (6 subsequent siblings)
8 siblings, 0 replies; 12+ messages in thread
From: Chaitanya Kulkarni @ 2022-02-14 10:01 UTC (permalink / raw)
To: linux-block; +Cc: axboe, Chaitanya Kulkarni, Himanshu Madhani
sprintf does not know the PAGE_SIZE maximum of the temporary buffer
used for outputting sysfs content and it's possible to overrun the
PAGE_SIZE buffer length.
Use a generic sysfs_emit function that knows the size of the
temporary buffer and ensures that no overrun is done for offset
attribute.
Signed-off-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
---
drivers/block/loop.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index f6ed265d34e7..c7fc790a6390 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -685,7 +685,7 @@ static ssize_t loop_attr_offset_show(struct loop_device *lo, char *buf)
static ssize_t loop_attr_sizelimit_show(struct loop_device *lo, char *buf)
{
- return sprintf(buf, "%llu\n", (unsigned long long)lo->lo_sizelimit);
+ return sysfs_emit(buf, "%llu\n", (unsigned long long)lo->lo_sizelimit);
}
static ssize_t loop_attr_autoclear_show(struct loop_device *lo, char *buf)
--
2.29.0
^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH 3/8] loop: use sysfs_emit() in the sysfs autoclear show
2022-02-14 10:01 [PATCH 0/8] loop: cleanup and few improvements Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 1/8] loop: use sysfs_emit() in the sysfs offset show Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 2/8] loop: use sysfs_emit() in the sysfs sizelimit show Chaitanya Kulkarni
@ 2022-02-14 10:01 ` Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 4/8] loop: use sysfs_emit() in the sysfs partscan show Chaitanya Kulkarni
` (5 subsequent siblings)
8 siblings, 0 replies; 12+ messages in thread
From: Chaitanya Kulkarni @ 2022-02-14 10:01 UTC (permalink / raw)
To: linux-block; +Cc: axboe, Chaitanya Kulkarni, Himanshu Madhani
sprintf does not know the PAGE_SIZE maximum of the temporary buffer
used for outputting sysfs content and it's possible to overrun the
PAGE_SIZE buffer length.
Use a generic sysfs_emit function that knows the size of the
temporary buffer and ensures that no overrun is done for offset
attribute.
Signed-off-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
---
drivers/block/loop.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index c7fc790a6390..c049c1967ec3 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -692,7 +692,7 @@ static ssize_t loop_attr_autoclear_show(struct loop_device *lo, char *buf)
{
int autoclear = (lo->lo_flags & LO_FLAGS_AUTOCLEAR);
- return sprintf(buf, "%s\n", autoclear ? "1" : "0");
+ return sysfs_emit(buf, "%s\n", autoclear ? "1" : "0");
}
static ssize_t loop_attr_partscan_show(struct loop_device *lo, char *buf)
--
2.29.0
^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH 4/8] loop: use sysfs_emit() in the sysfs partscan show
2022-02-14 10:01 [PATCH 0/8] loop: cleanup and few improvements Chaitanya Kulkarni
` (2 preceding siblings ...)
2022-02-14 10:01 ` [PATCH 3/8] loop: use sysfs_emit() in the sysfs autoclear show Chaitanya Kulkarni
@ 2022-02-14 10:01 ` Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 5/8] loop: use sysfs_emit() in the sysfs dio show Chaitanya Kulkarni
` (4 subsequent siblings)
8 siblings, 0 replies; 12+ messages in thread
From: Chaitanya Kulkarni @ 2022-02-14 10:01 UTC (permalink / raw)
To: linux-block; +Cc: axboe, Chaitanya Kulkarni, Himanshu Madhani
sprintf does not know the PAGE_SIZE maximum of the temporary buffer
used for outputting sysfs content and it's possible to overrun the
PAGE_SIZE buffer length.
Use a generic sysfs_emit function that knows the size of the
temporary buffer and ensures that no overrun is done for offset
attribute.
Signed-off-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
---
drivers/block/loop.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index c049c1967ec3..7fa6f68d7e41 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -699,7 +699,7 @@ static ssize_t loop_attr_partscan_show(struct loop_device *lo, char *buf)
{
int partscan = (lo->lo_flags & LO_FLAGS_PARTSCAN);
- return sprintf(buf, "%s\n", partscan ? "1" : "0");
+ return sysfs_emit(buf, "%s\n", partscan ? "1" : "0");
}
static ssize_t loop_attr_dio_show(struct loop_device *lo, char *buf)
--
2.29.0
^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH 5/8] loop: use sysfs_emit() in the sysfs dio show
2022-02-14 10:01 [PATCH 0/8] loop: cleanup and few improvements Chaitanya Kulkarni
` (3 preceding siblings ...)
2022-02-14 10:01 ` [PATCH 4/8] loop: use sysfs_emit() in the sysfs partscan show Chaitanya Kulkarni
@ 2022-02-14 10:01 ` Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 6/8] loop: remove extra variable in lo_fallocate() Chaitanya Kulkarni
` (3 subsequent siblings)
8 siblings, 0 replies; 12+ messages in thread
From: Chaitanya Kulkarni @ 2022-02-14 10:01 UTC (permalink / raw)
To: linux-block; +Cc: axboe, Chaitanya Kulkarni, Himanshu Madhani
sprintf does not know the PAGE_SIZE maximum of the temporary buffer
used for outputting sysfs content and it's possible to overrun the
PAGE_SIZE buffer length.
Use a generic sysfs_emit function that knows the size of the
temporary buffer and ensures that no overrun is done for offset
attribute.
Signed-off-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
---
drivers/block/loop.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 7fa6f68d7e41..a55e5eda1d17 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -706,7 +706,7 @@ static ssize_t loop_attr_dio_show(struct loop_device *lo, char *buf)
{
int dio = (lo->lo_flags & LO_FLAGS_DIRECT_IO);
- return sprintf(buf, "%s\n", dio ? "1" : "0");
+ return sysfs_emit(buf, "%s\n", dio ? "1" : "0");
}
LOOP_ATTR_RO(backing_file);
--
2.29.0
^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH 6/8] loop: remove extra variable in lo_fallocate()
2022-02-14 10:01 [PATCH 0/8] loop: cleanup and few improvements Chaitanya Kulkarni
` (4 preceding siblings ...)
2022-02-14 10:01 ` [PATCH 5/8] loop: use sysfs_emit() in the sysfs dio show Chaitanya Kulkarni
@ 2022-02-14 10:01 ` Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 7/8] loop: remove extra variable in lo_req_flush Chaitanya Kulkarni
` (2 subsequent siblings)
8 siblings, 0 replies; 12+ messages in thread
From: Chaitanya Kulkarni @ 2022-02-14 10:01 UTC (permalink / raw)
To: linux-block; +Cc: axboe, Chaitanya Kulkarni, Himanshu Madhani
The local variable q is used to pass it to the blk_queue_discard(). We
can get away with using lo->lo_queue instead of storing in a local
variable which is not used anywhere else.
No functional change in this patch.
Signed-off-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
---
drivers/block/loop.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index a55e5eda1d17..77c61eaaa6e4 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -308,12 +308,11 @@ static int lo_fallocate(struct loop_device *lo, struct request *rq, loff_t pos,
* a.k.a. discard/zerorange.
*/
struct file *file = lo->lo_backing_file;
- struct request_queue *q = lo->lo_queue;
int ret;
mode |= FALLOC_FL_KEEP_SIZE;
- if (!blk_queue_discard(q)) {
+ if (!blk_queue_discard(lo->lo_queue)) {
ret = -EOPNOTSUPP;
goto out;
}
--
2.29.0
^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH 7/8] loop: remove extra variable in lo_req_flush
2022-02-14 10:01 [PATCH 0/8] loop: cleanup and few improvements Chaitanya Kulkarni
` (5 preceding siblings ...)
2022-02-14 10:01 ` [PATCH 6/8] loop: remove extra variable in lo_fallocate() Chaitanya Kulkarni
@ 2022-02-14 10:01 ` Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 8/8] loop: allow user to set the queue depth Chaitanya Kulkarni
2022-02-14 13:46 ` [PATCH 0/8] loop: cleanup and few improvements Jens Axboe
8 siblings, 0 replies; 12+ messages in thread
From: Chaitanya Kulkarni @ 2022-02-14 10:01 UTC (permalink / raw)
To: linux-block; +Cc: axboe, Chaitanya Kulkarni, Himanshu Madhani
The local variable file is used to pass it to the vfs_fsync(). We can
get away with using lo->lo_backing_file instead of storing in a local
variable which is not used anywhere else.
No functional change in this patch.
Signed-off-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
---
drivers/block/loop.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 77c61eaaa6e4..18b30a56bfc4 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -326,8 +326,7 @@ static int lo_fallocate(struct loop_device *lo, struct request *rq, loff_t pos,
static int lo_req_flush(struct loop_device *lo, struct request *rq)
{
- struct file *file = lo->lo_backing_file;
- int ret = vfs_fsync(file, 0);
+ int ret = vfs_fsync(lo->lo_backing_file, 0);
if (unlikely(ret && ret != -EINVAL))
ret = -EIO;
--
2.29.0
^ permalink raw reply related [flat|nested] 12+ messages in thread
* [PATCH 8/8] loop: allow user to set the queue depth
2022-02-14 10:01 [PATCH 0/8] loop: cleanup and few improvements Chaitanya Kulkarni
` (6 preceding siblings ...)
2022-02-14 10:01 ` [PATCH 7/8] loop: remove extra variable in lo_req_flush Chaitanya Kulkarni
@ 2022-02-14 10:01 ` Chaitanya Kulkarni
2022-02-14 13:46 ` [PATCH 0/8] loop: cleanup and few improvements Jens Axboe
8 siblings, 0 replies; 12+ messages in thread
From: Chaitanya Kulkarni @ 2022-02-14 10:01 UTC (permalink / raw)
To: linux-block; +Cc: axboe, Chaitanya Kulkarni, Himanshu Madhani
Instead of hardcoding queue depth allow user to set the hw queue depth
using module parameter. Set default value to 128 to retain the existing
behavior.
Signed-off-by: Chaitanya Kulkarni <kch@nvidia.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
---
drivers/block/loop.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 18b30a56bfc4..fd2184d63c11 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -1785,6 +1785,9 @@ module_param(max_loop, int, 0444);
MODULE_PARM_DESC(max_loop, "Maximum number of loop devices");
module_param(max_part, int, 0444);
MODULE_PARM_DESC(max_part, "Maximum number of partitions per loop device");
+static int hw_queue_depth = 128;
+module_param_named(hw_queue_depth, hw_queue_depth, int, 0444);
+MODULE_PARM_DESC(hw_queue_depth, "Queue depth for each hardware queue. Default: 128");
MODULE_LICENSE("GPL");
MODULE_ALIAS_BLOCKDEV_MAJOR(LOOP_MAJOR);
@@ -1979,7 +1982,7 @@ static int loop_add(int i)
lo->tag_set.ops = &loop_mq_ops;
lo->tag_set.nr_hw_queues = 1;
- lo->tag_set.queue_depth = 128;
+ lo->tag_set.queue_depth = hw_queue_depth;
lo->tag_set.numa_node = NUMA_NO_NODE;
lo->tag_set.cmd_size = sizeof(struct loop_cmd);
lo->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_STACKING |
--
2.29.0
^ permalink raw reply related [flat|nested] 12+ messages in thread
* Re: [PATCH 0/8] loop: cleanup and few improvements
2022-02-14 10:01 [PATCH 0/8] loop: cleanup and few improvements Chaitanya Kulkarni
` (7 preceding siblings ...)
2022-02-14 10:01 ` [PATCH 8/8] loop: allow user to set the queue depth Chaitanya Kulkarni
@ 2022-02-14 13:46 ` Jens Axboe
8 siblings, 0 replies; 12+ messages in thread
From: Jens Axboe @ 2022-02-14 13:46 UTC (permalink / raw)
To: Chaitanya Kulkarni, linux-block
On 2/14/22 3:01 AM, Chaitanya Kulkarni wrote:
> Hi,
>
> This has few improvment and cleanups such as using sysfs_emit() for the
> sysfs dev attributes and removing variables that are only used once and
> a cleanup with fixing declaration.
>
> Below is the test log where 10 loop devices created, each device is
> linked to it's own file in ./loopX, formatted with xfs and mounted on
> /mnt/loopX. For each device it reads the offset, sizelimit, autoclear,
> partscan, and dio attr from sysfs using cat command, then it runs fio
> verify job on it.
>
> In summary write-verify fio job seems to work fine :-
> write-and-verify: (groupid=0, jobs=1): err= 0: pid=3495: Mon Feb 14 00:43:19 2022
> write-and-verify: (groupid=0, jobs=1): err= 0: pid=3960: Mon Feb 14 00:45:17 2022
> write-and-verify: (groupid=0, jobs=1): err= 0: pid=4321: Mon Feb 14 00:47:15 2022
> write-and-verify: (groupid=0, jobs=1): err= 0: pid=4369: Mon Feb 14 00:49:20 2022
> write-and-verify: (groupid=0, jobs=1): err= 0: pid=4443: Mon Feb 14 00:51:25 2022
> write-and-verify: (groupid=0, jobs=1): err= 0: pid=4536: Mon Feb 14 00:53:25 2022
> write-and-verify: (groupid=0, jobs=1): err= 0: pid=4770: Mon Feb 14 00:55:24 2022
> write-and-verify: (groupid=0, jobs=1): err= 0: pid=4933: Mon Feb 14 00:57:31 2022
> write-and-verify: (groupid=0, jobs=1): err= 0: pid=5008: Mon Feb 14 00:59:35 2022
> write-and-verify: (groupid=0, jobs=1): err= 0: pid=5048: Mon Feb 14 01:01:58 2022
>
> Below is detailed test log.
>
> Chaitanya Kulkarni (8):
> loop: use sysfs_emit() in the sysfs offset show
> loop: use sysfs_emit() in the sysfs sizelimit show
> loop: use sysfs_emit() in the sysfs autoclear show
> loop: use sysfs_emit() in the sysfs partscan show
> loop: use sysfs_emit() in the sysfs dio show
> loop: remove extra variable in lo_fallocate()
> loop: remove extra variable in lo_req_flush
> loop: allow user to set the queue depth
Please collapse patches 1..5.
--
Jens Axboe
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH 8/8] loop: allow user to set the queue depth
2021-09-21 9:21 ` [PATCH 8/8] loop: allow user to set the queue depth Chaitanya Kulkarni
@ 2021-09-21 13:38 ` Himanshu Madhani
0 siblings, 0 replies; 12+ messages in thread
From: Himanshu Madhani @ 2021-09-21 13:38 UTC (permalink / raw)
To: Chaitanya Kulkarni; +Cc: linux-block, Jens Axboe, Chaitanya Kulkarni
> On Sep 21, 2021, at 4:21 AM, Chaitanya Kulkarni <chaitanyak@nvidia.com> wrote:
>
> From: Chaitanya Kulkarni <kch@nvidia.com>
>
> Instead of hardcoding queue depth allow user to set the hw queue depth
> using module parameter. Set default value to 128 to retain the existing
> behavior.
>
> Signed-off-by: Chaitanya Kulkarni <kch@nvidia.com>
> ---
> drivers/block/loop.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/block/loop.c b/drivers/block/loop.c
> index 6478d3b0dd2a..aeba72b5dd2d 100644
> --- a/drivers/block/loop.c
> +++ b/drivers/block/loop.c
> @@ -2096,6 +2096,9 @@ module_param(max_loop, int, 0444);
> MODULE_PARM_DESC(max_loop, "Maximum number of loop devices");
> module_param(max_part, int, 0444);
> MODULE_PARM_DESC(max_part, "Maximum number of partitions per loop device");
> +static int hw_queue_depth = 128;
> +module_param_named(hw_queue_depth, hw_queue_depth, int, 0444);
> +MODULE_PARM_DESC(hw_queue_depth, "Queue depth for each hardware queue. Default: 128");
> MODULE_LICENSE("GPL");
> MODULE_ALIAS_BLOCKDEV_MAJOR(LOOP_MAJOR);
>
> @@ -2328,7 +2331,7 @@ static int loop_add(int i)
> err = -ENOMEM;
> lo->tag_set.ops = &loop_mq_ops;
> lo->tag_set.nr_hw_queues = 1;
> - lo->tag_set.queue_depth = 128;
> + lo->tag_set.queue_depth = hw_queue_depth;
> lo->tag_set.numa_node = NUMA_NO_NODE;
> lo->tag_set.cmd_size = sizeof(struct loop_cmd);
> lo->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_STACKING |
> --
> 2.29.0
>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
--
Himanshu Madhani Oracle Linux Engineering
^ permalink raw reply [flat|nested] 12+ messages in thread
* [PATCH 8/8] loop: allow user to set the queue depth
2021-09-21 9:21 [PATCH 0/8] loop: small clenaup Chaitanya Kulkarni
@ 2021-09-21 9:21 ` Chaitanya Kulkarni
2021-09-21 13:38 ` Himanshu Madhani
0 siblings, 1 reply; 12+ messages in thread
From: Chaitanya Kulkarni @ 2021-09-21 9:21 UTC (permalink / raw)
To: linux-block; +Cc: axboe, Chaitanya Kulkarni
From: Chaitanya Kulkarni <kch@nvidia.com>
Instead of hardcoding queue depth allow user to set the hw queue depth
using module parameter. Set default value to 128 to retain the existing
behavior.
Signed-off-by: Chaitanya Kulkarni <kch@nvidia.com>
---
drivers/block/loop.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 6478d3b0dd2a..aeba72b5dd2d 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -2096,6 +2096,9 @@ module_param(max_loop, int, 0444);
MODULE_PARM_DESC(max_loop, "Maximum number of loop devices");
module_param(max_part, int, 0444);
MODULE_PARM_DESC(max_part, "Maximum number of partitions per loop device");
+static int hw_queue_depth = 128;
+module_param_named(hw_queue_depth, hw_queue_depth, int, 0444);
+MODULE_PARM_DESC(hw_queue_depth, "Queue depth for each hardware queue. Default: 128");
MODULE_LICENSE("GPL");
MODULE_ALIAS_BLOCKDEV_MAJOR(LOOP_MAJOR);
@@ -2328,7 +2331,7 @@ static int loop_add(int i)
err = -ENOMEM;
lo->tag_set.ops = &loop_mq_ops;
lo->tag_set.nr_hw_queues = 1;
- lo->tag_set.queue_depth = 128;
+ lo->tag_set.queue_depth = hw_queue_depth;
lo->tag_set.numa_node = NUMA_NO_NODE;
lo->tag_set.cmd_size = sizeof(struct loop_cmd);
lo->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_STACKING |
--
2.29.0
^ permalink raw reply related [flat|nested] 12+ messages in thread
end of thread, other threads:[~2022-02-14 13:47 UTC | newest]
Thread overview: 12+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-02-14 10:01 [PATCH 0/8] loop: cleanup and few improvements Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 1/8] loop: use sysfs_emit() in the sysfs offset show Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 2/8] loop: use sysfs_emit() in the sysfs sizelimit show Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 3/8] loop: use sysfs_emit() in the sysfs autoclear show Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 4/8] loop: use sysfs_emit() in the sysfs partscan show Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 5/8] loop: use sysfs_emit() in the sysfs dio show Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 6/8] loop: remove extra variable in lo_fallocate() Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 7/8] loop: remove extra variable in lo_req_flush Chaitanya Kulkarni
2022-02-14 10:01 ` [PATCH 8/8] loop: allow user to set the queue depth Chaitanya Kulkarni
2022-02-14 13:46 ` [PATCH 0/8] loop: cleanup and few improvements Jens Axboe
-- strict thread matches above, loose matches on Subject: below --
2021-09-21 9:21 [PATCH 0/8] loop: small clenaup Chaitanya Kulkarni
2021-09-21 9:21 ` [PATCH 8/8] loop: allow user to set the queue depth Chaitanya Kulkarni
2021-09-21 13:38 ` Himanshu Madhani
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.