Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-06-04 Thread Jens Axboe
On 2014-06-04 20:33, Shaohua Li wrote: On Wed, Jun 04, 2014 at 08:16:32PM -0600, Jens Axboe wrote: On 2014-06-04 20:09, Shaohua Li wrote: On Wed, Jun 04, 2014 at 02:08:46PM -0600, Jens Axboe wrote: On 06/04/2014 05:29 AM, Matias Bjørling wrote: It's in blk_io_account_start part_round_stat

Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-06-04 Thread Shaohua Li
On Wed, Jun 04, 2014 at 08:16:32PM -0600, Jens Axboe wrote: > On 2014-06-04 20:09, Shaohua Li wrote: > >On Wed, Jun 04, 2014 at 02:08:46PM -0600, Jens Axboe wrote: > >>On 06/04/2014 05:29 AM, Matias Bjørling wrote: > >>>It's in > >>> > >>>blk_io_account_start > >>> part_round_stats > >>> part

Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-06-04 Thread Jens Axboe
On 2014-06-04 20:09, Shaohua Li wrote: On Wed, Jun 04, 2014 at 02:08:46PM -0600, Jens Axboe wrote: On 06/04/2014 05:29 AM, Matias Bjørling wrote: It's in blk_io_account_start part_round_stats part_round_state_single part_in_flight I like the granularity idea. And similarly fr

Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-06-04 Thread Shaohua Li
On Wed, Jun 04, 2014 at 02:08:46PM -0600, Jens Axboe wrote: > On 06/04/2014 05:29 AM, Matias Bjørling wrote: > > It's in > > > > blk_io_account_start > > part_round_stats > > part_round_state_single > > part_in_flight > > > > I like the granularity idea. > > And similarly from blk_io

Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-06-04 Thread Jens Axboe
On 06/04/2014 05:29 AM, Matias Bjørling wrote: > It's in > > blk_io_account_start > part_round_stats > part_round_state_single > part_in_flight > > I like the granularity idea. And similarly from blk_io_account_done() - which makes it even worse, since it at both ends of the IO chain

Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-06-04 Thread Jens Axboe
On 2014-06-04 04:39, Shaohua Li wrote: On Fri, May 30, 2014 at 07:49:52AM -0600, Jens Axboe wrote: On 2014-05-30 06:11, Shaohua Li wrote: On Fri, May 09, 2014 at 10:41:27AM -0600, Jens Axboe wrote: On 05/09/2014 08:12 AM, Jens Axboe wrote: On 05/09/2014 03:17 AM, Matias Bjørling wrote: With

Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-06-04 Thread Matias Bjørling
It's in blk_io_account_start part_round_stats part_round_state_single part_in_flight I like the granularity idea. Thanks, Matias On Wed, Jun 4, 2014 at 12:39 PM, Shaohua Li wrote: > On Fri, May 30, 2014 at 07:49:52AM -0600, Jens Axboe wrote: >> On 2014-05-30 06:11, Shaohua Li wrote

Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-06-04 Thread Shaohua Li
On Fri, May 30, 2014 at 07:49:52AM -0600, Jens Axboe wrote: > On 2014-05-30 06:11, Shaohua Li wrote: > >On Fri, May 09, 2014 at 10:41:27AM -0600, Jens Axboe wrote: > >>On 05/09/2014 08:12 AM, Jens Axboe wrote: > >>>On 05/09/2014 03:17 AM, Matias Bjørling wrote: > With multi-million IOPS and mul

Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-05-30 Thread Jens Axboe
On 2014-05-30 06:11, Shaohua Li wrote: On Fri, May 09, 2014 at 10:41:27AM -0600, Jens Axboe wrote: On 05/09/2014 08:12 AM, Jens Axboe wrote: On 05/09/2014 03:17 AM, Matias Bjørling wrote: With multi-million IOPS and multi-node workloads, the atomic_t in_flight tracking becomes a bottleneck. Ch

Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-05-30 Thread Shaohua Li
On Fri, May 09, 2014 at 10:41:27AM -0600, Jens Axboe wrote: > On 05/09/2014 08:12 AM, Jens Axboe wrote: > > On 05/09/2014 03:17 AM, Matias Bjørling wrote: > >> With multi-million IOPS and multi-node workloads, the atomic_t in_flight > >> tracking becomes a bottleneck. Change the in-flight accountin

Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-05-09 Thread Jens Axboe
On 05/09/2014 08:12 AM, Jens Axboe wrote: > On 05/09/2014 03:17 AM, Matias Bjørling wrote: >> With multi-million IOPS and multi-node workloads, the atomic_t in_flight >> tracking becomes a bottleneck. Change the in-flight accounting to per-cpu >> counters to elevate. > > The part stats are a pain

Re: [PATCH] block: per-cpu counters for in-flight IO accounting

2014-05-09 Thread Jens Axboe
On 05/09/2014 03:17 AM, Matias Bjørling wrote: > With multi-million IOPS and multi-node workloads, the atomic_t in_flight > tracking becomes a bottleneck. Change the in-flight accounting to per-cpu > counters to elevate. The part stats are a pain in the butt, I've tried to come up with a great fix

[PATCH] block: per-cpu counters for in-flight IO accounting

2014-05-09 Thread Matias Bjørling
With multi-million IOPS and multi-node workloads, the atomic_t in_flight tracking becomes a bottleneck. Change the in-flight accounting to per-cpu counters to elevate. Signed-off-by: Matias Bjørling --- block/partition-generic.c | 4 ++-- include/linux/genhd.h | 25 ++---