Re: [PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-04 Thread Jens Axboe
On 11/04/2016 08:46 AM, Christoph Hellwig wrote: On Thu, Nov 03, 2016 at 10:00:58PM -0600, Jens Axboe wrote: I've applied 1-2 for 4.10, but we probably should look into unifying those parts of sq and mq in general. For instance, it doesn't seem to make a lot of sense why we'd depth limit sq and

Re: [PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-04 Thread Jens Axboe
On 11/04/2016 08:46 AM, Christoph Hellwig wrote: On Thu, Nov 03, 2016 at 10:00:58PM -0600, Jens Axboe wrote: I've applied 1-2 for 4.10, but we probably should look into unifying those parts of sq and mq in general. For instance, it doesn't seem to make a lot of sense why we'd depth limit sq and

Re: [PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-04 Thread Christoph Hellwig
On Thu, Nov 03, 2016 at 10:00:58PM -0600, Jens Axboe wrote: > I've applied 1-2 for 4.10, but we probably should look into unifying > those parts of sq and mq in general. For instance, it doesn't seem to > make a lot of sense why we'd depth limit sq and not mq. I've spent some time looking the the

Re: [PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-04 Thread Christoph Hellwig
On Thu, Nov 03, 2016 at 10:00:58PM -0600, Jens Axboe wrote: > I've applied 1-2 for 4.10, but we probably should look into unifying > those parts of sq and mq in general. For instance, it doesn't seem to > make a lot of sense why we'd depth limit sq and not mq. I've spent some time looking the the

Re: [PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-03 Thread Jens Axboe
On 11/03/2016 06:13 PM, Shaohua Li wrote: On Thu, Nov 03, 2016 at 05:09:54PM -0700, Christoph Hellwig wrote: On Thu, Nov 03, 2016 at 05:03:54PM -0700, Shaohua Li wrote: This is corresponding part for blk-mq. Disk with multiple hardware queues doesn't need this as we only hold 1 request at

Re: [PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-03 Thread Jens Axboe
On 11/03/2016 06:13 PM, Shaohua Li wrote: On Thu, Nov 03, 2016 at 05:09:54PM -0700, Christoph Hellwig wrote: On Thu, Nov 03, 2016 at 05:03:54PM -0700, Shaohua Li wrote: This is corresponding part for blk-mq. Disk with multiple hardware queues doesn't need this as we only hold 1 request at

Re: [PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-03 Thread Shaohua Li
On Thu, Nov 03, 2016 at 05:09:54PM -0700, Christoph Hellwig wrote: > On Thu, Nov 03, 2016 at 05:03:54PM -0700, Shaohua Li wrote: > > This is corresponding part for blk-mq. Disk with multiple hardware > > queues doesn't need this as we only hold 1 request at most. > > Any reason you only do this

Re: [PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-03 Thread Shaohua Li
On Thu, Nov 03, 2016 at 05:09:54PM -0700, Christoph Hellwig wrote: > On Thu, Nov 03, 2016 at 05:03:54PM -0700, Shaohua Li wrote: > > This is corresponding part for blk-mq. Disk with multiple hardware > > queues doesn't need this as we only hold 1 request at most. > > Any reason you only do this

Re: [PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-03 Thread Christoph Hellwig
On Thu, Nov 03, 2016 at 05:03:54PM -0700, Shaohua Li wrote: > This is corresponding part for blk-mq. Disk with multiple hardware > queues doesn't need this as we only hold 1 request at most. Any reason you only do this for the SQ and not the MQ case?

Re: [PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-03 Thread Christoph Hellwig
On Thu, Nov 03, 2016 at 05:03:54PM -0700, Shaohua Li wrote: > This is corresponding part for blk-mq. Disk with multiple hardware > queues doesn't need this as we only hold 1 request at most. Any reason you only do this for the SQ and not the MQ case?

[PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-03 Thread Shaohua Li
This is corresponding part for blk-mq. Disk with multiple hardware queues doesn't need this as we only hold 1 request at most. Signed-off-by: Shaohua Li --- block/blk-mq.c | 7 ++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/block/blk-mq.c b/block/blk-mq.c

[PATCH V2 2/2] blk-mq: immediately dispatch big size request

2016-11-03 Thread Shaohua Li
This is corresponding part for blk-mq. Disk with multiple hardware queues doesn't need this as we only hold 1 request at most. Signed-off-by: Shaohua Li --- block/blk-mq.c | 7 ++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index