| Message ID | 20221105005902.407297-1-longman@redhat.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | blk-cgroup: Optimize blkcg_rstat_flush() | expand |
On 11/4/22 20:58, Waiman Long wrote: > v10: > - Update patch 3 to rename the rstat function to > cgroup_rstat_css_cpu_flush(). > > v9: > - Remove patch "llist: Allow optional sentinel node terminated lockless > list" for now. This will be done as a follow-up patch. > - Add a new lqueued field to blkg_iostat_set to store the status of > whether lnode is in a lockless list. > - Add a new patch 3 to speed up the freeing of blkcg by flushing out > the rstat lockless lists at blkcg offline time. > > v8: > - Update the llist patch to make existing llist functions and macros > work for both NULL and sentinel terminated lockless list as much > as possible and leave only the initialization and removal functions > to have a sentinel terminated llist variants. > > This patch series improves blkcg_rstat_flush() performance by eliminating > unnecessary blkg enumeration and flush operations for those blkg's and > blkg_iostat_set's that haven't been updated since the last flush. > > Waiman Long (3): > blk-cgroup: Return -ENOMEM directly in blkcg_css_alloc() error path > blk-cgroup: Optimize blkcg_rstat_flush() > blk-cgroup: Flush stats at blkgs destruction path > > block/blk-cgroup.c | 103 +++++++++++++++++++++++++++++++++++------ > block/blk-cgroup.h | 10 ++++ > include/linux/cgroup.h | 1 + > kernel/cgroup/rstat.c | 20 ++++++++ > 4 files changed, 119 insertions(+), 15 deletions(-) Jens, do you have any further comment on this patchset? Is it possible to queue it for the next Linux version? Cheers, Longman
On 11/16/22 4:06 PM, Waiman Long wrote: > On 11/4/22 20:58, Waiman Long wrote: >> ? v10: >> ?? - Update patch 3 to rename the rstat function to >> ???? cgroup_rstat_css_cpu_flush(). >> >> ? v9: >> ?? - Remove patch "llist: Allow optional sentinel node terminated lockless >> ???? list" for now. This will be done as a follow-up patch. >> ?? - Add a new lqueued field to blkg_iostat_set to store the status of >> ???? whether lnode is in a lockless list. >> ?? - Add a new patch 3 to speed up the freeing of blkcg by flushing out >> ???? the rstat lockless lists at blkcg offline time. >> >> ? v8: >> ?? - Update the llist patch to make existing llist functions and macros >> ???? work for both NULL and sentinel terminated lockless list as much >> ???? as possible and leave only the initialization and removal functions >> ???? to have a sentinel terminated llist variants. >> >> This patch series improves blkcg_rstat_flush() performance by eliminating >> unnecessary blkg enumeration and flush operations for those blkg's and >> blkg_iostat_set's that haven't been updated since the last flush. >> >> Waiman Long (3): >> ?? blk-cgroup: Return -ENOMEM directly in blkcg_css_alloc() error path >> ?? blk-cgroup: Optimize blkcg_rstat_flush() >> ?? blk-cgroup: Flush stats at blkgs destruction path >> >> ? block/blk-cgroup.c???? | 103 +++++++++++++++++++++++++++++++++++------ >> ? block/blk-cgroup.h???? |? 10 ++++ >> ? include/linux/cgroup.h |?? 1 + >> ? kernel/cgroup/rstat.c? |? 20 ++++++++ >> ? 4 files changed, 119 insertions(+), 15 deletions(-) > > Jens, do you have any further comment on this patchset? Is it possible > to queue it for the next Linux version? Looks good to me, I'll queue it up.
On 11/16/22 18:58, Jens Axboe wrote: > On Fri, 4 Nov 2022 20:58:59 -0400, Waiman Long wrote: >> v10: >> - Update patch 3 to rename the rstat function to >> cgroup_rstat_css_cpu_flush(). >> >> v9: >> - Remove patch "llist: Allow optional sentinel node terminated lockless >> list" for now. This will be done as a follow-up patch. >> - Add a new lqueued field to blkg_iostat_set to store the status of >> whether lnode is in a lockless list. >> - Add a new patch 3 to speed up the freeing of blkcg by flushing out >> the rstat lockless lists at blkcg offline time. >> >> [...] > Applied, thanks! > > [1/3] blk-cgroup: Return -ENOMEM directly in blkcg_css_alloc() error path > commit: b5a9adcbd5dc95d34d1f5fc84eff9af6fc60d284 > [2/3] blk-cgroup: Optimize blkcg_rstat_flush() > commit: 3b8cc6298724021da845f2f9fd7dd4b6829a6817 > [3/3] blk-cgroup: Flush stats at blkgs destruction path > commit: dae590a6c96c799434e0ff8156ef29b88c257e60 > > Best regards, Thanks a lot! Cheers, Longman
v10: - Update patch 3 to rename the rstat function to cgroup_rstat_css_cpu_flush(). v9: - Remove patch "llist: Allow optional sentinel node terminated lockless list" for now. This will be done as a follow-up patch. - Add a new lqueued field to blkg_iostat_set to store the status of whether lnode is in a lockless list. - Add a new patch 3 to speed up the freeing of blkcg by flushing out the rstat lockless lists at blkcg offline time. v8: - Update the llist patch to make existing llist functions and macros work for both NULL and sentinel terminated lockless list as much as possible and leave only the initialization and removal functions to have a sentinel terminated llist variants. This patch series improves blkcg_rstat_flush() performance by eliminating unnecessary blkg enumeration and flush operations for those blkg's and blkg_iostat_set's that haven't been updated since the last flush. Waiman Long (3): blk-cgroup: Return -ENOMEM directly in blkcg_css_alloc() error path blk-cgroup: Optimize blkcg_rstat_flush() blk-cgroup: Flush stats at blkgs destruction path block/blk-cgroup.c | 103 +++++++++++++++++++++++++++++++++++------ block/blk-cgroup.h | 10 ++++ include/linux/cgroup.h | 1 + kernel/cgroup/rstat.c | 20 ++++++++ 4 files changed, 119 insertions(+), 15 deletions(-)