From: Bart Van Assche <Bart.VanAssche@wdc.com>
To: "ming.lei@redhat.com" <ming.lei@redhat.com>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
"hch@infradead.org" <hch@infradead.org>,
"himanshu.madhani@cavium.com" <himanshu.madhani@cavium.com>,
"tom81094@gmail.com" <tom81094@gmail.com>,
"paolo.valente@linaro.org" <paolo.valente@linaro.org>,
"axboe@kernel.dk" <axboe@kernel.dk>,
"linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
"oleksandr@natalenko.name" <oleksandr@natalenko.name>,
"john.garry@huawei.com" <john.garry@huawei.com>,
"osandov@fb.com" <osandov@fb.com>,
"loberman@redhat.com" <loberman@redhat.com>,
"james.smart@broadcom.com" <james.smart@broadcom.com>
Subject: Re: [PATCH V7 4/6] blk-mq: introduce .get_budget and .put_budget in blk_mq_ops
Date: Fri, 13 Oct 2017 17:47:43 +0000 [thread overview]
Message-ID: <1507916862.13345.27.camel@wdc.com> (raw)
In-Reply-To: <20171013172950.GA32319@ming.t460p>
T24gU2F0LCAyMDE3LTEwLTE0IGF0IDAxOjI5ICswODAwLCBNaW5nIExlaSB3cm90ZToNCj4gLT5j
YW5fcXVldWUgaXMgc2l6ZSBvZiB0aGUgd2hvbGUgdGFnIHNwYWNlIHNoYXJlZCBieSBhbGwgTFVO
cywgbG9va3MgaXQgaXNuJ3QNCj4gcmVhc29uYWJsZSB0byBpbmNyZWFzZSBjbWRfcGVyX2x1biB0
byAuY2FuX3F1ZXVlLg0KDQpTb3JyeSBidXQgSSBkaXNhZ3JlZS4gU2V0dGluZyBjbWRfcGVyX2x1
biB0byBhIHZhbHVlIGxvd2VyIHRoYW4gY2FuX3F1ZXVlDQp3aWxsIHJlc3VsdCBpbiBzdWJvcHRp
bWFsIHBlcmZvcm1hbmNlIGlmIHRoZXJlIGlzIG9ubHkgYSBzaW5nbGUgTFVOIHBlciBTQ1NJDQpo
b3N0LiBJZiB0aGVyZSBhcmUgbXVsdGlwbGUgTFVOcyBwZXIgU0NTSSBob3N0IHRoZW4gdGhlIGJs
ay1tcSBjb3JlIHRyYWNrcw0KdGhlIG51bWJlciBvZiBhY3RpdmUgTFVOcyB0aHJvdWdoIHRoZSBi
bGtfbXFfdGFncy5hY3RpdmVfcXVldWVzIHZhcmlhYmxlLg0KU2VlIGFsc28gaGN0eF9tYXlfcXVl
dWUoKS4gVGhlIGNvbW1lbnQgYWJvdmUgdGhhdCBmdW5jdGlvbiBpcyBhcyBmb2xsb3dzOg0KDQov
Kg0KICogRm9yIHNoYXJlZCB0YWcgdXNlcnMsIHdlIHRyYWNrIHRoZSBudW1iZXIgb2YgY3VycmVu
dGx5IGFjdGl2ZSB1c2Vycw0KICogYW5kIGF0dGVtcHQgdG8gcHJvdmlkZSBhIGZhaXIgc2hhcmUg
b2YgdGhlIHRhZyBkZXB0aCBmb3IgZWFjaCBvZiB0aGVtLg0KICovDQoNCkJUVywgdGhlIGliX3Ny
cCBpbml0aWF0b3IgZHJpdmVyIHNldHMgY21kX3Blcl9sdW4gdG8gY2FuX3F1ZXVlIGFuZCBpcyBh
YmxlDQp0byBhY2hpZXZlIG1vcmUgdGhhbiBvbmUgbWlsbGlvbiBJT1BTIGV2ZW4gaW4gdGVzdHMg
d2l0aCBtdWx0aXBsZSBMVU5zIHBlcg0KU0NTSSBob3N0Lg0KDQpCYXJ0Lg==
WARNING: multiple messages have this Message-ID (diff)
From: Bart Van Assche <Bart.VanAssche@wdc.com>
To: "ming.lei@redhat.com" <ming.lei@redhat.com>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
"hch@infradead.org" <hch@infradead.org>,
"himanshu.madhani@cavium.com" <himanshu.madhani@cavium.com>,
"tom81094@gmail.com" <tom81094@gmail.com>,
"paolo.valente@linaro.org" <paolo.valente@linaro.org>,
"axboe@kernel.dk" <axboe@kernel.dk>,
"linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
"oleksandr@natalenko.name" <oleksandr@natalenko.name>,
"john.garry@huawei.com" <john.garry@huawei.com>,
"osandov@fb.com" <osandov@fb.com>,
"loberman@redhat.com" <loberman@redhat.com>,
"james.smart@broadcom.com" <james.smart@broadcom.com>
Subject: Re: [PATCH V7 4/6] blk-mq: introduce .get_budget and .put_budget in blk_mq_ops
Date: Fri, 13 Oct 2017 17:47:43 +0000 [thread overview]
Message-ID: <1507916862.13345.27.camel@wdc.com> (raw)
In-Reply-To: <20171013172950.GA32319@ming.t460p>
On Sat, 2017-10-14 at 01:29 +0800, Ming Lei wrote:
> ->can_queue is size of the whole tag space shared by all LUNs, looks it isn't
> reasonable to increase cmd_per_lun to .can_queue.
Sorry but I disagree. Setting cmd_per_lun to a value lower than can_queue
will result in suboptimal performance if there is only a single LUN per SCSI
host. If there are multiple LUNs per SCSI host then the blk-mq core tracks
the number of active LUNs through the blk_mq_tags.active_queues variable.
See also hctx_may_queue(). The comment above that function is as follows:
/*
* For shared tag users, we track the number of currently active users
* and attempt to provide a fair share of the tag depth for each of them.
*/
BTW, the ib_srp initiator driver sets cmd_per_lun to can_queue and is able
to achieve more than one million IOPS even in tests with multiple LUNs per
SCSI host.
Bart.
next prev parent reply other threads:[~2017-10-13 17:47 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-12 18:36 [PATCH V7 0/6] blk-mq-sched: improve sequential I/O performance Ming Lei
2017-10-12 18:36 ` [PATCH V7 1/6] blk-mq-sched: fix scheduler bad performance Ming Lei
2017-10-12 18:37 ` [PATCH V7 2/6] blk-mq-sched: move actual dispatching into one helper Ming Lei
2017-10-12 18:37 ` [PATCH V7 3/6] sbitmap: introduce __sbitmap_for_each_set() Ming Lei
2017-10-12 18:37 ` [PATCH V7 4/6] blk-mq: introduce .get_budget and .put_budget in blk_mq_ops Ming Lei
2017-10-12 18:46 ` Jens Axboe
2017-10-13 0:19 ` Ming Lei
2017-10-13 14:44 ` Jens Axboe
2017-10-13 16:07 ` Ming Lei
2017-10-13 16:19 ` Jens Axboe
2017-10-13 16:21 ` Ming Lei
2017-10-13 16:28 ` Jens Axboe
2017-10-13 16:31 ` Bart Van Assche
2017-10-13 16:31 ` Bart Van Assche
2017-10-13 16:33 ` Jens Axboe
2017-10-13 16:45 ` Ming Lei
2017-10-13 17:08 ` Bart Van Assche
2017-10-13 17:08 ` Bart Van Assche
2017-10-13 17:29 ` Ming Lei
2017-10-13 17:47 ` Bart Van Assche [this message]
2017-10-13 17:47 ` Bart Van Assche
2017-10-16 11:30 ` Hannes Reinecke
2017-10-16 16:06 ` Bart Van Assche
2017-10-16 16:06 ` Bart Van Assche
2017-10-17 1:29 ` Ming Lei
2017-10-17 6:38 ` Hannes Reinecke
2017-10-17 9:36 ` Ming Lei
2017-10-17 18:09 ` Bart Van Assche
2017-10-13 16:17 ` Ming Lei
2017-10-13 16:20 ` Jens Axboe
2017-10-13 16:22 ` Ming Lei
2017-10-13 16:28 ` Jens Axboe
2017-10-12 18:37 ` [PATCH V7 5/6] blk-mq-sched: improve dispatching from sw queue Ming Lei
2017-10-12 18:48 ` Bart Van Assche
2017-10-13 0:20 ` Ming Lei
2017-10-12 18:37 ` [PATCH V7 6/6] SCSI: implement .get_budget and .put_budget for blk-mq Ming Lei
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1507916862.13345.27.camel@wdc.com \
--to=bart.vanassche@wdc.com \
--cc=axboe@kernel.dk \
--cc=hch@infradead.org \
--cc=himanshu.madhani@cavium.com \
--cc=james.smart@broadcom.com \
--cc=john.garry@huawei.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=loberman@redhat.com \
--cc=ming.lei@redhat.com \
--cc=oleksandr@natalenko.name \
--cc=osandov@fb.com \
--cc=paolo.valente@linaro.org \
--cc=tom81094@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.