From: Don Brace <don.brace@microsemi.com>
To: Ming Lei <ming.lei@redhat.com>, Jens Axboe <axboe@kernel.dk>,
"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
"Christoph Hellwig" <hch@infradead.org>,
Mike Snitzer <snitzer@redhat.com>
Cc: "linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
Hannes Reinecke <hare@suse.de>, Arun Easi <arun.easi@cavium.com>,
Omar Sandoval <osandov@fb.com>,
"Martin K . Petersen" <martin.petersen@oracle.com>,
"James Bottomley" <james.bottomley@hansenpartnership.com>,
Christoph Hellwig <hch@lst.de>,
Kashyap Desai <kashyap.desai@broadcom.com>,
Peter Rivera <peter.rivera@broadcom.com>,
Paolo Bonzini <pbonzini@redhat.com>,
"Laurence Oberman" <loberman@redhat.com>
Subject: RE: [PATCH V2 4/8] block: null_blk: introduce module parameter of 'g_global_tags'
Date: Mon, 5 Feb 2018 20:26:33 +0000 [thread overview]
Message-ID: <892b0510b3254668a1a1e5c5fb2f84c2@microsemi.com> (raw)
In-Reply-To: <20180205152035.15016-5-ming.lei@redhat.com>
> This patch introduces the parameter of 'g_global_tags' so that we can
> test this feature by null_blk easiy.
>=20
> Not see obvious performance drop with global_tags when the whole hw
> depth is kept as same:
>=20
> 1) no 'global_tags', each hw queue depth is 1, and 4 hw queues
> modprobe null_blk queue_mode=3D2 nr_devices=3D4 shared_tags=3D1 global_ta=
gs=3D0
> submit_queues=3D4 hw_queue_depth=3D1
>=20
> 2) 'global_tags', global hw queue depth is 4, and 4 hw queues
> modprobe null_blk queue_mode=3D2 nr_devices=3D4 shared_tags=3D1 global_ta=
gs=3D1
> submit_queues=3D4 hw_queue_depth=3D4
>=20
> 3) fio test done in above two settings:
> fio --bs=3D4k --size=3D512G --rw=3Drandread --norandommap --direct=3D=
1 --
> ioengine=3Dlibaio --iodepth=3D4 --runtime=3D$RUNTIME --group_reporting=3D=
1 --
> name=3Dnullb0 --filename=3D/dev/nullb0 --name=3Dnullb1 --filename=3D/dev/=
nullb1 --
> name=3Dnullb2 --filename=3D/dev/nullb2 --name=3Dnullb3 --filename=3D/dev/=
nullb3
>=20
> 1M IOPS can be reached in both above tests which is done in one VM.
>=20
I am getting 1.1M IOPS for both cases.
Tested-by: Don Brace <don.brace@microsemi.com>
WARNING: multiple messages have this Message-ID (diff)
From: Don Brace <don.brace@microsemi.com>
To: Ming Lei <ming.lei@redhat.com>, Jens Axboe <axboe@kernel.dk>,
"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
Christoph Hellwig <hch@infradead.org>,
Mike Snitzer <snitzer@redhat.com>
Cc: "linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
Hannes Reinecke <hare@suse.de>, Arun Easi <arun.easi@cavium.com>,
Omar Sandoval <osandov@fb.com>,
"Martin K . Petersen" <martin.petersen@oracle.com>,
James Bottomley <james.bottomley@hansenpartnership.com>,
Christoph Hellwig <hch@lst.de>,
Kashyap Desai <kashyap.desai@broadcom.com>,
Peter Rivera <peter.rivera@broadcom.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Laurence Oberman <loberman@redhat.com>
Subject: RE: [PATCH V2 4/8] block: null_blk: introduce module parameter of 'g_global_tags'
Date: Mon, 5 Feb 2018 20:26:33 +0000 [thread overview]
Message-ID: <892b0510b3254668a1a1e5c5fb2f84c2@microsemi.com> (raw)
In-Reply-To: <20180205152035.15016-5-ming.lei@redhat.com>
> This patch introduces the parameter of 'g_global_tags' so that we can
> test this feature by null_blk easiy.
>
> Not see obvious performance drop with global_tags when the whole hw
> depth is kept as same:
>
> 1) no 'global_tags', each hw queue depth is 1, and 4 hw queues
> modprobe null_blk queue_mode=2 nr_devices=4 shared_tags=1 global_tags=0
> submit_queues=4 hw_queue_depth=1
>
> 2) 'global_tags', global hw queue depth is 4, and 4 hw queues
> modprobe null_blk queue_mode=2 nr_devices=4 shared_tags=1 global_tags=1
> submit_queues=4 hw_queue_depth=4
>
> 3) fio test done in above two settings:
> fio --bs=4k --size=512G --rw=randread --norandommap --direct=1 --
> ioengine=libaio --iodepth=4 --runtime=$RUNTIME --group_reporting=1 --
> name=nullb0 --filename=/dev/nullb0 --name=nullb1 --filename=/dev/nullb1 --
> name=nullb2 --filename=/dev/nullb2 --name=nullb3 --filename=/dev/nullb3
>
> 1M IOPS can be reached in both above tests which is done in one VM.
>
I am getting 1.1M IOPS for both cases.
Tested-by: Don Brace <don.brace@microsemi.com>
next prev parent reply other threads:[~2018-02-05 20:26 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-02-05 15:20 [PATCH V2 0/8] blk-mq/scsi-mq: support global tags & introduce force_blk_mq Ming Lei
2018-02-05 15:20 ` [PATCH V2 1/8] blk-mq: tags: define several fields of tags as pointer Ming Lei
2018-02-06 21:41 ` Omar Sandoval
2018-02-05 15:20 ` [PATCH V2 2/8] blk-mq: introduce BLK_MQ_F_GLOBAL_TAGS Ming Lei
2018-02-06 20:33 ` Omar Sandoval
2018-02-07 0:44 ` Ming Lei
2018-02-06 23:18 ` Jens Axboe
2018-02-07 0:43 ` Ming Lei
2018-02-07 16:09 ` Bart Van Assche
2018-02-07 16:59 ` Bart Van Assche
2018-02-07 16:59 ` Bart Van Assche
2018-02-08 15:25 ` Bart Van Assche
2018-02-08 15:25 ` Bart Van Assche
2018-02-05 15:20 ` [PATCH V2 3/8] scsi: Add template flag 'host_tagset' Ming Lei
2018-02-05 15:20 ` [PATCH V2 4/8] block: null_blk: introduce module parameter of 'g_global_tags' Ming Lei
2018-02-05 20:26 ` Don Brace [this message]
2018-02-05 20:26 ` Don Brace
2018-02-06 21:43 ` Omar Sandoval
2018-02-05 15:20 ` [PATCH V2 5/8] scsi: introduce force_blk_mq Ming Lei
2018-02-06 20:20 ` Omar Sandoval
2018-02-07 0:46 ` Ming Lei
2018-02-05 15:20 ` [PATCH V2 6/8] scsi: virtio_scsi: fix IO hang by irq vector automatic affinity Ming Lei
2018-02-05 15:56 ` Paolo Bonzini
2018-02-05 15:20 ` [PATCH V2 7/8] scsi: hpsa: call hpsa_hba_inquiry() after adding host Ming Lei
2018-02-05 18:55 ` Don Brace
2018-02-05 18:55 ` Don Brace
2018-02-06 8:32 ` Hannes Reinecke
2018-02-05 15:20 ` [PATCH V2 8/8] scsi: hpsa: use blk_mq to solve irq affinity issue Ming Lei
2018-02-05 15:58 ` Laurence Oberman
2018-02-05 15:58 ` Laurence Oberman
2018-02-05 16:07 ` Don Brace
2018-02-05 16:07 ` Don Brace
2018-02-05 18:54 ` Don Brace
2018-02-05 18:54 ` Don Brace
2018-02-06 2:18 ` chenxiang (M)
2018-02-06 2:18 ` chenxiang (M)
2018-02-06 8:23 ` Ming Lei
2018-02-06 8:39 ` Hannes Reinecke
2018-02-06 9:51 ` Ming Lei
2018-02-06 23:15 ` [PATCH V2 0/8] blk-mq/scsi-mq: support global tags & introduce force_blk_mq Jens Axboe
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=892b0510b3254668a1a1e5c5fb2f84c2@microsemi.com \
--to=don.brace@microsemi.com \
--cc=arun.easi@cavium.com \
--cc=axboe@kernel.dk \
--cc=hare@suse.de \
--cc=hch@infradead.org \
--cc=hch@lst.de \
--cc=james.bottomley@hansenpartnership.com \
--cc=kashyap.desai@broadcom.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=loberman@redhat.com \
--cc=martin.petersen@oracle.com \
--cc=ming.lei@redhat.com \
--cc=osandov@fb.com \
--cc=pbonzini@redhat.com \
--cc=peter.rivera@broadcom.com \
--cc=snitzer@redhat.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.