Hi Roman, On Fri, Oct 27, 2017 at 07:55:33PM +0200, Roman Penyaev wrote: > Hi Ming, > > On Fri, Oct 27, 2017 at 7:38 AM, Ming Lei <ming.lei@xxxxxxxxxx> wrote: > > Hello Bart, > > > > On Fri, Oct 27, 2017 at 04:53:18AM +0000, Bart Van Assche wrote: > >> On Fri, 2017-10-27 at 12:43 +0800, Ming Lei wrote: > >> > The 1st patch removes the RESTART for TAG-SHARED because SCSI handles it > >> > by itself, and not necessary to waste CPU to do the expensive RESTART. > >> > And Roman Pen reported that this RESTART cuts half of IOPS in his case. > >> > > >> > The 2nd patch removes the RESTART when .get_budget returns BLK_STS_RESOURCE, > >> > and this RESTART is handled by SCSI's RESTART(scsi_end_request()) too. > >> > >> Hello Ming, > >> > >> There are more block drivers than the SCSI core that share tags. If the > > > > Could you share us what the other in-tree driver which share tags is? > > > > If there are really more, and when all can share similar RESTART mechanism, I > > think that is the time for considering the common RESTART. > > > >> restart mechanism is removed from the blk-mq core, does that mean that all > >> block drivers that share tags will have to follow the example of the SCSI > >> core and implement a restart mechanism themselves? As far as I know there > > > > If there are such drivers, there should have been their own restart mechanism > > which works for long time before blk-mq comes, and more importantly each driver > > has much more knowledge than generic block layer to handle the restart, > > such as SCSI's restart, that means driver's implementation may be more efficient. > > Personally I would be happier if block layer would do restarts for me. My eyes > are bleeding when I see chunks of my own code which do this juggling with hctx, > enqueuing or dequeuing from a percpu lists, etc inside IBNBD driver. The issue is that the current block's RESTART for TAG-SHARED isn't efficient, and never functions as expected. And no one uses that at all actually. Firstly we have handled cross-queue RESTART via BLK_MQ_S_TAG_WAITING already when getting driver tag. Secondly as you see, SCSI handles this kind of RESTART much more efficient and simply via one linked list, I believe your IBNBD can do this in similar way, right? If you can't figure out one easy way to do that, I am happy to provide one patchset[1](which was done via sbitmap weeks ago) to do that efficiently in block layer, but not the current approach, when you upstream your IBNBD patchset. https://github.com/ming1/linux/commits/blk_mq_improve_restart_V1-rc1 Finally you reported that the RESTART for TAG-SHARED has degraded your IO performance by half, and that can't be accepted, IMO. I found the similar performance issue in my test too. > > > Also the RESTART for TAG-SHARED may never function as expected wrt. SCSI. > > E.g. new BLK_MQ_F_NO_RESTARTS flag can help to disable restarts when they > are not needed. This flag is only needed iff there are more than one in-tree drivers which need the RESTART for TAG-SHARED. Also I have mentioned the current RESTART for TAG-SHARED has big performance issue. > > > And more importantly the block's RESTART for TAG-SHARED has caused big performance > > issue for people. > > That's just a bug in code, not a in issue with restarts, which can be fixed > if we put hctx which are needed to be restarted in percpu lists and avoid > long loops and contentions. Again no in-tree driver uses RESTART for TAG-SHARED now, so not necessary to fix anything in the RESTART for TAG-SHARED. If there is driver which need this kind of RESTART in future, one more efficient implementation can be figured out at that time, but not now. As you see, I already have one. Thanks, Ming
