| Message ID | 572b5b3f11cc414bd990d1580f8bf287f4797676.1711952124.git.anand.jain@oracle.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | common/btrfs: lookup running processes using pgrep | expand |
在 2024/4/1 16:46, Anand Jain 写道: > Certain helper functions and the testcase btrfs/132 use the following > script to find running processes: > > while ps aux | grep "balance start" | grep -qv grep; do > <> > done > > Instead, using pgrep is more efficient. > > while pgrep -f "btrfs balance start" > /dev/null; do > <> > done > > Signed-off-by: Anand Jain <anand.jain@oracle.com> Looks good to me. Although there are already several test cases utilizing pgrep, I'm not 100% sure if pgrep would exist for all systems. Shouldn't there be some checks first? Thanks, Qu > --- > common/btrfs | 10 +++++----- > tests/btrfs/132 | 2 +- > 2 files changed, 6 insertions(+), 6 deletions(-) > > diff --git a/common/btrfs b/common/btrfs > index 2c086227d8e0..a320b0e41d0d 100644 > --- a/common/btrfs > +++ b/common/btrfs > @@ -327,7 +327,7 @@ _btrfs_kill_stress_balance_pid() > kill $balance_pid &> /dev/null > wait $balance_pid &> /dev/null > # Wait for the balance operation to finish. > - while ps aux | grep "balance start" | grep -qv grep; do > + while pgrep -f "btrfs balance start" > /dev/null; do > sleep 1 > done > } > @@ -381,7 +381,7 @@ _btrfs_kill_stress_scrub_pid() > kill $scrub_pid &> /dev/null > wait $scrub_pid &> /dev/null > # Wait for the scrub operation to finish. > - while ps aux | grep "scrub start" | grep -qv grep; do > + while pgrep -f "btrfs scrub start" > /dev/null; do > sleep 1 > done > } > @@ -415,7 +415,7 @@ _btrfs_kill_stress_defrag_pid() > kill $defrag_pid &> /dev/null > wait $defrag_pid &> /dev/null > # Wait for the defrag operation to finish. > - while ps aux | grep "btrfs filesystem defrag" | grep -qv grep; do > + while pgrep -f "btrfs filesystem defrag" > /dev/null; do > sleep 1 > done > } > @@ -444,7 +444,7 @@ _btrfs_kill_stress_remount_compress_pid() > kill $remount_pid &> /dev/null > wait $remount_pid &> /dev/null > # Wait for the remount loop to finish. > - while ps aux | grep "mount.*${btrfs_mnt}" | grep -qv grep; do > + while pgrep -f "mount.*${btrfs_mnt}" > /dev/null; do > sleep 1 > done > } > @@ -507,7 +507,7 @@ _btrfs_kill_stress_replace_pid() > kill $replace_pid &> /dev/null > wait $replace_pid &> /dev/null > # Wait for the replace operation to finish. > - while ps aux | grep "replace start" | grep -qv grep; do > + while pgrep -f "btrfs replace start" > /dev/null; do > sleep 1 > done > } > diff --git a/tests/btrfs/132 b/tests/btrfs/132 > index f50420f51181..b48395c1884f 100755 > --- a/tests/btrfs/132 > +++ b/tests/btrfs/132 > @@ -70,7 +70,7 @@ kill $pids > wait > > # Wait all writers really exits > -while ps aux | grep "$SCRATCH_MNT" | grep -qv grep; do > +while pgrep -f "$SCRATCH_MNT" > /dev/null; do > sleep 1 > done >
On 4/1/24 14:21, Qu Wenruo wrote: > > > 在 2024/4/1 16:46, Anand Jain 写道: >> Certain helper functions and the testcase btrfs/132 use the following >> script to find running processes: >> >> while ps aux | grep "balance start" | grep -qv grep; do >> <> >> done >> >> Instead, using pgrep is more efficient. >> >> while pgrep -f "btrfs balance start" > /dev/null; do >> <> >> done >> >> Signed-off-by: Anand Jain <anand.jain@oracle.com> > > Looks good to me. > > Although there are already several test cases utilizing pgrep, I'm not > 100% sure if pgrep would exist for all systems. > > Shouldn't there be some checks first? > Actually, I checked on that and noticed that pgrep comes from the same package as ps. So we are fine. $ rpm -ql procps-ng | grep -E "bin/pgrep|bin/ps" /usr/bin/pgrep /usr/bin/ps Thanks, Anand > Thanks, > Qu >> --- >> common/btrfs | 10 +++++----- >> tests/btrfs/132 | 2 +- >> 2 files changed, 6 insertions(+), 6 deletions(-) >> >> diff --git a/common/btrfs b/common/btrfs >> index 2c086227d8e0..a320b0e41d0d 100644 >> --- a/common/btrfs >> +++ b/common/btrfs >> @@ -327,7 +327,7 @@ _btrfs_kill_stress_balance_pid() >> kill $balance_pid &> /dev/null >> wait $balance_pid &> /dev/null >> # Wait for the balance operation to finish. >> - while ps aux | grep "balance start" | grep -qv grep; do >> + while pgrep -f "btrfs balance start" > /dev/null; do >> sleep 1 >> done >> } >> @@ -381,7 +381,7 @@ _btrfs_kill_stress_scrub_pid() >> kill $scrub_pid &> /dev/null >> wait $scrub_pid &> /dev/null >> # Wait for the scrub operation to finish. >> - while ps aux | grep "scrub start" | grep -qv grep; do >> + while pgrep -f "btrfs scrub start" > /dev/null; do >> sleep 1 >> done >> } >> @@ -415,7 +415,7 @@ _btrfs_kill_stress_defrag_pid() >> kill $defrag_pid &> /dev/null >> wait $defrag_pid &> /dev/null >> # Wait for the defrag operation to finish. >> - while ps aux | grep "btrfs filesystem defrag" | grep -qv grep; do >> + while pgrep -f "btrfs filesystem defrag" > /dev/null; do >> sleep 1 >> done >> } >> @@ -444,7 +444,7 @@ _btrfs_kill_stress_remount_compress_pid() >> kill $remount_pid &> /dev/null >> wait $remount_pid &> /dev/null >> # Wait for the remount loop to finish. >> - while ps aux | grep "mount.*${btrfs_mnt}" | grep -qv grep; do >> + while pgrep -f "mount.*${btrfs_mnt}" > /dev/null; do >> sleep 1 >> done >> } >> @@ -507,7 +507,7 @@ _btrfs_kill_stress_replace_pid() >> kill $replace_pid &> /dev/null >> wait $replace_pid &> /dev/null >> # Wait for the replace operation to finish. >> - while ps aux | grep "replace start" | grep -qv grep; do >> + while pgrep -f "btrfs replace start" > /dev/null; do >> sleep 1 >> done >> } >> diff --git a/tests/btrfs/132 b/tests/btrfs/132 >> index f50420f51181..b48395c1884f 100755 >> --- a/tests/btrfs/132 >> +++ b/tests/btrfs/132 >> @@ -70,7 +70,7 @@ kill $pids >> wait >> >> # Wait all writers really exits >> -while ps aux | grep "$SCRATCH_MNT" | grep -qv grep; do >> +while pgrep -f "$SCRATCH_MNT" > /dev/null; do >> sleep 1 >> done >>
在 2024/4/1 18:10, Anand Jain 写道: > > > On 4/1/24 14:21, Qu Wenruo wrote: >> >> >> 在 2024/4/1 16:46, Anand Jain 写道: >>> Certain helper functions and the testcase btrfs/132 use the following >>> script to find running processes: >>> >>> while ps aux | grep "balance start" | grep -qv grep; do >>> <> >>> done >>> >>> Instead, using pgrep is more efficient. >>> >>> while pgrep -f "btrfs balance start" > /dev/null; do >>> <> >>> done >>> >>> Signed-off-by: Anand Jain <anand.jain@oracle.com> >> >> Looks good to me. >> >> Although there are already several test cases utilizing pgrep, I'm not >> 100% sure if pgrep would exist for all systems. >> >> Shouldn't there be some checks first? >> > > > Actually, I checked on that and noticed that pgrep comes from > the same package as ps. So we are fine. > > $ rpm -ql procps-ng | grep -E "bin/pgrep|bin/ps" > /usr/bin/pgrep > /usr/bin/ps > > Thanks, Anand So I guess busybox based system won't be supported anyway for fstests? In that case it looks fine to me. Reviewed-by: Qu Wenruo <wqu@suse.com> Thanks, Qu > > >> Thanks, >> Qu >>> --- >>> common/btrfs | 10 +++++----- >>> tests/btrfs/132 | 2 +- >>> 2 files changed, 6 insertions(+), 6 deletions(-) >>> >>> diff --git a/common/btrfs b/common/btrfs >>> index 2c086227d8e0..a320b0e41d0d 100644 >>> --- a/common/btrfs >>> +++ b/common/btrfs >>> @@ -327,7 +327,7 @@ _btrfs_kill_stress_balance_pid() >>> kill $balance_pid &> /dev/null >>> wait $balance_pid &> /dev/null >>> # Wait for the balance operation to finish. >>> - while ps aux | grep "balance start" | grep -qv grep; do >>> + while pgrep -f "btrfs balance start" > /dev/null; do >>> sleep 1 >>> done >>> } >>> @@ -381,7 +381,7 @@ _btrfs_kill_stress_scrub_pid() >>> kill $scrub_pid &> /dev/null >>> wait $scrub_pid &> /dev/null >>> # Wait for the scrub operation to finish. >>> - while ps aux | grep "scrub start" | grep -qv grep; do >>> + while pgrep -f "btrfs scrub start" > /dev/null; do >>> sleep 1 >>> done >>> } >>> @@ -415,7 +415,7 @@ _btrfs_kill_stress_defrag_pid() >>> kill $defrag_pid &> /dev/null >>> wait $defrag_pid &> /dev/null >>> # Wait for the defrag operation to finish. >>> - while ps aux | grep "btrfs filesystem defrag" | grep -qv >>> grep; do >>> + while pgrep -f "btrfs filesystem defrag" > /dev/null; do >>> sleep 1 >>> done >>> } >>> @@ -444,7 +444,7 @@ _btrfs_kill_stress_remount_compress_pid() >>> kill $remount_pid &> /dev/null >>> wait $remount_pid &> /dev/null >>> # Wait for the remount loop to finish. >>> - while ps aux | grep "mount.*${btrfs_mnt}" | grep -qv grep; do >>> + while pgrep -f "mount.*${btrfs_mnt}" > /dev/null; do >>> sleep 1 >>> done >>> } >>> @@ -507,7 +507,7 @@ _btrfs_kill_stress_replace_pid() >>> kill $replace_pid &> /dev/null >>> wait $replace_pid &> /dev/null >>> # Wait for the replace operation to finish. >>> - while ps aux | grep "replace start" | grep -qv grep; do >>> + while pgrep -f "btrfs replace start" > /dev/null; do >>> sleep 1 >>> done >>> } >>> diff --git a/tests/btrfs/132 b/tests/btrfs/132 >>> index f50420f51181..b48395c1884f 100755 >>> --- a/tests/btrfs/132 >>> +++ b/tests/btrfs/132 >>> @@ -70,7 +70,7 @@ kill $pids >>> wait >>> >>> # Wait all writers really exits >>> -while ps aux | grep "$SCRATCH_MNT" | grep -qv grep; do >>> +while pgrep -f "$SCRATCH_MNT" > /dev/null; do >>> sleep 1 >>> done >>> >
On 4/1/24 15:51, Qu Wenruo wrote: > > > 在 2024/4/1 18:10, Anand Jain 写道: >> >> >> On 4/1/24 14:21, Qu Wenruo wrote: >>> >>> >>> 在 2024/4/1 16:46, Anand Jain 写道: >>>> Certain helper functions and the testcase btrfs/132 use the following >>>> script to find running processes: >>>> >>>> while ps aux | grep "balance start" | grep -qv grep; do >>>> <> >>>> done >>>> >>>> Instead, using pgrep is more efficient. >>>> >>>> while pgrep -f "btrfs balance start" > /dev/null; do >>>> <> >>>> done >>>> >>>> Signed-off-by: Anand Jain <anand.jain@oracle.com> >>> >>> Looks good to me. >>> >>> Although there are already several test cases utilizing pgrep, I'm not >>> 100% sure if pgrep would exist for all systems. >>> >>> Shouldn't there be some checks first? >>> >> >> >> Actually, I checked on that and noticed that pgrep comes from >> the same package as ps. So we are fine. >> >> $ rpm -ql procps-ng | grep -E "bin/pgrep|bin/ps" >> /usr/bin/pgrep >> /usr/bin/ps >> >> Thanks, Anand > > So I guess busybox based system won't be supported anyway for fstests? > There aren't quite a lot of tools required for fstests in BusyBox. pgrep is not new in fstests, as you noticed, so not introducing a new failures, so for now, it should be okay. Thanks, Anand > In that case it looks fine to me. > > Reviewed-by: Qu Wenruo <wqu@suse.com> > > Thanks, > Qu >> >> >>> Thanks, >>> Qu >>>> --- >>>> common/btrfs | 10 +++++----- >>>> tests/btrfs/132 | 2 +- >>>> 2 files changed, 6 insertions(+), 6 deletions(-) >>>> >>>> diff --git a/common/btrfs b/common/btrfs >>>> index 2c086227d8e0..a320b0e41d0d 100644 >>>> --- a/common/btrfs >>>> +++ b/common/btrfs >>>> @@ -327,7 +327,7 @@ _btrfs_kill_stress_balance_pid() >>>> kill $balance_pid &> /dev/null >>>> wait $balance_pid &> /dev/null >>>> # Wait for the balance operation to finish. >>>> - while ps aux | grep "balance start" | grep -qv grep; do >>>> + while pgrep -f "btrfs balance start" > /dev/null; do >>>> sleep 1 >>>> done >>>> } >>>> @@ -381,7 +381,7 @@ _btrfs_kill_stress_scrub_pid() >>>> kill $scrub_pid &> /dev/null >>>> wait $scrub_pid &> /dev/null >>>> # Wait for the scrub operation to finish. >>>> - while ps aux | grep "scrub start" | grep -qv grep; do >>>> + while pgrep -f "btrfs scrub start" > /dev/null; do >>>> sleep 1 >>>> done >>>> } >>>> @@ -415,7 +415,7 @@ _btrfs_kill_stress_defrag_pid() >>>> kill $defrag_pid &> /dev/null >>>> wait $defrag_pid &> /dev/null >>>> # Wait for the defrag operation to finish. >>>> - while ps aux | grep "btrfs filesystem defrag" | grep -qv >>>> grep; do >>>> + while pgrep -f "btrfs filesystem defrag" > /dev/null; do >>>> sleep 1 >>>> done >>>> } >>>> @@ -444,7 +444,7 @@ _btrfs_kill_stress_remount_compress_pid() >>>> kill $remount_pid &> /dev/null >>>> wait $remount_pid &> /dev/null >>>> # Wait for the remount loop to finish. >>>> - while ps aux | grep "mount.*${btrfs_mnt}" | grep -qv grep; do >>>> + while pgrep -f "mount.*${btrfs_mnt}" > /dev/null; do >>>> sleep 1 >>>> done >>>> } >>>> @@ -507,7 +507,7 @@ _btrfs_kill_stress_replace_pid() >>>> kill $replace_pid &> /dev/null >>>> wait $replace_pid &> /dev/null >>>> # Wait for the replace operation to finish. >>>> - while ps aux | grep "replace start" | grep -qv grep; do >>>> + while pgrep -f "btrfs replace start" > /dev/null; do >>>> sleep 1 >>>> done >>>> } >>>> diff --git a/tests/btrfs/132 b/tests/btrfs/132 >>>> index f50420f51181..b48395c1884f 100755 >>>> --- a/tests/btrfs/132 >>>> +++ b/tests/btrfs/132 >>>> @@ -70,7 +70,7 @@ kill $pids >>>> wait >>>> >>>> # Wait all writers really exits >>>> -while ps aux | grep "$SCRATCH_MNT" | grep -qv grep; do >>>> +while pgrep -f "$SCRATCH_MNT" > /dev/null; do >>>> sleep 1 >>>> done >>>> >>
diff --git a/common/btrfs b/common/btrfs index 2c086227d8e0..a320b0e41d0d 100644 --- a/common/btrfs +++ b/common/btrfs @@ -327,7 +327,7 @@ _btrfs_kill_stress_balance_pid() kill $balance_pid &> /dev/null wait $balance_pid &> /dev/null # Wait for the balance operation to finish. - while ps aux | grep "balance start" | grep -qv grep; do + while pgrep -f "btrfs balance start" > /dev/null; do sleep 1 done } @@ -381,7 +381,7 @@ _btrfs_kill_stress_scrub_pid() kill $scrub_pid &> /dev/null wait $scrub_pid &> /dev/null # Wait for the scrub operation to finish. - while ps aux | grep "scrub start" | grep -qv grep; do + while pgrep -f "btrfs scrub start" > /dev/null; do sleep 1 done } @@ -415,7 +415,7 @@ _btrfs_kill_stress_defrag_pid() kill $defrag_pid &> /dev/null wait $defrag_pid &> /dev/null # Wait for the defrag operation to finish. - while ps aux | grep "btrfs filesystem defrag" | grep -qv grep; do + while pgrep -f "btrfs filesystem defrag" > /dev/null; do sleep 1 done } @@ -444,7 +444,7 @@ _btrfs_kill_stress_remount_compress_pid() kill $remount_pid &> /dev/null wait $remount_pid &> /dev/null # Wait for the remount loop to finish. - while ps aux | grep "mount.*${btrfs_mnt}" | grep -qv grep; do + while pgrep -f "mount.*${btrfs_mnt}" > /dev/null; do sleep 1 done } @@ -507,7 +507,7 @@ _btrfs_kill_stress_replace_pid() kill $replace_pid &> /dev/null wait $replace_pid &> /dev/null # Wait for the replace operation to finish. - while ps aux | grep "replace start" | grep -qv grep; do + while pgrep -f "btrfs replace start" > /dev/null; do sleep 1 done } diff --git a/tests/btrfs/132 b/tests/btrfs/132 index f50420f51181..b48395c1884f 100755 --- a/tests/btrfs/132 +++ b/tests/btrfs/132 @@ -70,7 +70,7 @@ kill $pids wait # Wait all writers really exits -while ps aux | grep "$SCRATCH_MNT" | grep -qv grep; do +while pgrep -f "$SCRATCH_MNT" > /dev/null; do sleep 1 done
Certain helper functions and the testcase btrfs/132 use the following script to find running processes: while ps aux | grep "balance start" | grep -qv grep; do <> done Instead, using pgrep is more efficient. while pgrep -f "btrfs balance start" > /dev/null; do <> done Signed-off-by: Anand Jain <anand.jain@oracle.com> --- common/btrfs | 10 +++++----- tests/btrfs/132 | 2 +- 2 files changed, 6 insertions(+), 6 deletions(-)