On 10/13/2017 10:22 AM, Ming Lei wrote: > On Fri, Oct 13, 2017 at 10:20:01AM -0600, Jens Axboe wrote: >> On 10/13/2017 10:17 AM, Ming Lei wrote: >>> On Fri, Oct 13, 2017 at 08:44:23AM -0600, Jens Axboe wrote: >>>> On 10/12/2017 06:19 PM, Ming Lei wrote: >>>>> On Thu, Oct 12, 2017 at 12:46:24PM -0600, Jens Axboe wrote: >>>>>> On 10/12/2017 12:37 PM, Ming Lei wrote: >>>>>>> For SCSI devices, there is often per-request-queue depth, which need >>>>>>> to be respected before queuing one request. >>>>>>> >>>>>>> The current blk-mq always dequeues one request first, then calls .queue_rq() >>>>>>> to dispatch the request to lld. One obvious issue of this way is that I/O >>>>>>> merge may not be good, because when the per-request-queue depth can't be >>>>>>> respected, .queue_rq() has to return BLK_STS_RESOURCE, then this request >>>>>>> has to staty in hctx->dispatch list, and never got chance to participate >>>>>>> into I/O merge. >>>>>>> >>>>>>> This patch introduces .get_budget and .put_budget callback in blk_mq_ops, >>>>>>> then we can try to get reserved budget first before dequeuing request. >>>>>>> Once we can't get budget for queueing I/O, we don't need to dequeue request >>>>>>> at all, then I/O merge can get improved a lot. >>>>>> >>>>>> I can't help but think that it would be cleaner to just be able to >>>>>> reinsert the request into the scheduler properly, if we fail to >>>>>> dispatch it. Bart hinted at that earlier as well. >>>>> >>>>> Actually when I start to investigate the issue, the 1st thing I tried >>>>> is to reinsert, but that way is even worse on qla2xxx. >>>>> >>>>> Once request is dequeued, the IO merge chance is decreased a lot. >>>>> With none scheduler, it becomes not possible to merge because >>>>> we only try to merge over the last 8 requests. With mq-deadline, >>>>> when one request is reinserted, another request may be dequeued >>>>> at the same time. >>>> >>>> I don't care too much about 'none'. If perfect merging is crucial for >>>> getting to the performance level you want on the hardware you are using, >>>> you should not be using 'none'. 'none' will work perfectly fine for NVMe >>>> etc style devices, where we are not dependent on merging to the same >>>> extent that we are on other devices. >>> >>> We still have some SCSI device, such as qla2xxx, which is 1:1 multi-queue >>> device, like NVMe, in my test, the big lock of mq-deadline has been >>> an issue for this kind of device, and none actually is better than >>> mq-deadline, even though its merge isn't good. >> >> Kyber should be able to fill that hole, hopefully. > > Yeah, kyber still uses same IO merge with none, :-) Doesn't mean it can't be changed... 'none' has to remain with very low overhead, any extra smarts or logic should be a scheduler thing. -- Jens Axboe
