[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <782e3b9b-2aae-e013-bedc-3c35cb0f7edd@wdc.com>
Date: Tue, 17 Oct 2017 11:09:15 -0700
From: Bart Van Assche <bart.vanassche@....com>
To: Hannes Reinecke <hare@...e.de>, Ming Lei <ming.lei@...hat.com>
Cc: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
"hch@...radead.org" <hch@...radead.org>,
"tom81094@...il.com" <tom81094@...il.com>,
"himanshu.madhani@...ium.com" <himanshu.madhani@...ium.com>,
"paolo.valente@...aro.org" <paolo.valente@...aro.org>,
"axboe@...nel.dk" <axboe@...nel.dk>,
"linux-scsi@...r.kernel.org" <linux-scsi@...r.kernel.org>,
"oleksandr@...alenko.name" <oleksandr@...alenko.name>,
"john.garry@...wei.com" <john.garry@...wei.com>,
"osandov@...com" <osandov@...com>,
"loberman@...hat.com" <loberman@...hat.com>,
"james.smart@...adcom.com" <james.smart@...adcom.com>
Subject: Re: [PATCH V7 4/6] blk-mq: introduce .get_budget and .put_budget in
blk_mq_ops
On 10/16/17 23:38, Hannes Reinecke wrote:
> Again, these are just some pre-defined values to avoid I/O starvation
> when having several LUNs. _if_ we can guarantee I/O fairness between
> several (hundreds!) devices all sharing the same tagspace we wouldn't
> need these variables.
I think we already have two mechanisms for guaranteeing I/O fairness:
* Scsi_Host.starved_list for both the legacy SCSI core and scsi-mq.
* For scsi-mq, the active_queues counter in the blk-mq core. See also
hctx_may_queue().
Bart.
Powered by blists - more mailing lists