From: Ming Lei <ming.lei@redhat.com>
To: Kashyap Desai <kashyap.desai@broadcom.com>
Cc: Jens Axboe <axboe@kernel.dk>,
linux-block@vger.kernel.org, linux-scsi@vger.kernel.org,
"Martin K . Petersen" <martin.petersen@oracle.com>,
James Bottomley <James.Bottomley@hansenpartnership.com>,
Bart Van Assche <bvanassche@acm.org>,
Hannes Reinecke <hare@suse.com>,
John Garry <john.garry@huawei.com>,
Don Brace <don.brace@microsemi.com>,
Sathya Prakash Veerichetty <sathya.prakash@broadcom.com>,
Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH 8/9] scsi: megaraid: convert private reply queue to blk-mq hw queue
Date: Sun, 2 Jun 2019 14:42:08 +0800 [thread overview]
Message-ID: <20190602064202.GA2731@ming.t460p> (raw)
In-Reply-To: <7819e1a523b9e8227e3a9d188ee1e083@mail.gmail.com>
Hi Kashyap,
Thanks for your test.
On Sun, Jun 02, 2019 at 03:11:26AM +0530, Kashyap Desai wrote:
> > SCSI's reply qeueue is very similar with blk-mq's hw queue, both
> assigned by
> > IRQ vector, so map te private reply queue into blk-mq's hw queue via
> > .host_tagset.
> >
> > Then the private reply mapping can be removed.
> >
> > Another benefit is that the request/irq lost issue may be solved in
> generic
> > approach because managed IRQ may be shutdown during CPU hotplug.
>
> Ming,
>
> I quickly tested this patch series on MegaRaid Aero controller. Without
> this patch I can get 3.0M IOPS, but once I apply this patch I see only
> 1.2M IOPS (40% Performance drop)
> HBA supports 5089 can_queue.
>
> <perf top> output without patch -
>
> 3.39% [megaraid_sas] [k] complete_cmd_fusion
> 3.36% [kernel] [k] scsi_queue_rq
> 3.26% [kernel] [k] entry_SYSCALL_64
> 2.57% [kernel] [k] syscall_return_via_sysret
> 1.95% [megaraid_sas] [k] megasas_build_and_issue_cmd_fusion
> 1.88% [kernel] [k] _raw_spin_lock_irqsave
> 1.79% [kernel] [k] gup_pmd_range
> 1.73% [kernel] [k] _raw_spin_lock
> 1.68% [kernel] [k] __sched_text_start
> 1.19% [kernel] [k] irq_entries_start
> 1.13% [kernel] [k] scsi_dec_host_busy
> 1.08% [kernel] [k] aio_complete
> 1.07% [kernel] [k] read_tsc
> 1.01% [kernel] [k] blk_mq_get_request
> 0.93% [kernel] [k] __update_load_avg_cfs_rq
> 0.92% [kernel] [k] aio_read_events
> 0.91% [kernel] [k] lookup_ioctx
> 0.91% fio [.] fio_gettime
> 0.87% [kernel] [k] set_next_entity
> 0.87% [megaraid_sas] [k] megasas_build_ldio_fusion
>
> <perf top> output with patch -
>
> 11.30% [kernel] [k] native_queued_spin_lock_slowpath
I guess there must be one global lock required in megaraid submission path,
could you run 'perf record -g -a' to see which lock is and what the stack
trace is?
Thanks,
Ming
next prev parent reply other threads:[~2019-06-02 7:18 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-31 2:27 [PATCH 0/9] blk-mq/scsi: convert private reply queue into blk_mq hw queue Ming Lei
2019-05-31 2:27 ` [PATCH 1/9] blk-mq: allow hw queues to share hostwide tags Ming Lei
2019-05-31 6:07 ` Hannes Reinecke
2019-05-31 15:37 ` Bart Van Assche
2019-06-24 8:44 ` Ming Lei
2019-06-05 14:10 ` John Garry
2019-06-24 8:46 ` Ming Lei
2019-06-24 13:14 ` John Garry
2019-05-31 2:27 ` [PATCH 2/9] block: null_blk: introduce module parameter of 'g_host_tags' Ming Lei
2019-05-31 6:08 ` Hannes Reinecke
2019-05-31 15:39 ` Bart Van Assche
2019-06-24 8:43 ` Ming Lei
2019-06-02 1:56 ` Minwoo Im
2019-05-31 2:27 ` [PATCH 3/9] scsi: Add template flag 'host_tagset' Ming Lei
2019-05-31 6:08 ` Hannes Reinecke
2019-05-31 2:27 ` [PATCH 4/9] scsi_debug: support host tagset Ming Lei
2019-05-31 6:09 ` Hannes Reinecke
2019-06-02 2:03 ` Minwoo Im
2019-06-02 17:01 ` Douglas Gilbert
2019-05-31 2:27 ` [PATCH 5/9] scsi: introduce scsi_cmnd_hctx_index() Ming Lei
2019-05-31 6:10 ` Hannes Reinecke
2019-05-31 2:27 ` [PATCH 6/9] scsi: hpsa: convert private reply queue to blk-mq hw queue Ming Lei
2019-05-31 6:15 ` Hannes Reinecke
2019-05-31 6:30 ` Ming Lei
2019-05-31 6:40 ` Hannes Reinecke
2019-05-31 2:27 ` [PATCH 7/9] scsi: hisi_sas_v3: " Ming Lei
2019-05-31 6:20 ` Hannes Reinecke
2019-05-31 6:34 ` Ming Lei
2019-05-31 6:42 ` Hannes Reinecke
2019-05-31 7:14 ` Ming Lei
2019-05-31 11:38 ` John Garry
2019-06-03 11:00 ` Ming Lei
2019-06-03 13:00 ` John Garry
2019-06-04 13:37 ` Ming Lei
2019-05-31 2:28 ` [PATCH 8/9] scsi: megaraid: " Ming Lei
2019-05-31 6:22 ` Hannes Reinecke
2019-06-01 21:41 ` Kashyap Desai
2019-06-02 6:42 ` Ming Lei [this message]
2019-06-02 7:48 ` Ming Lei
2019-06-02 16:34 ` Kashyap Desai
2019-06-03 3:56 ` Ming Lei
2019-06-03 10:00 ` Kashyap Desai
2019-06-07 9:45 ` Kashyap Desai
2019-05-31 2:28 ` [PATCH 9/9] scsi: mp3sas: " Ming Lei
2019-05-31 6:23 ` Hannes Reinecke
2019-06-06 11:58 ` Sreekanth Reddy
2019-06-04 8:49 ` [PATCH 0/9] blk-mq/scsi: convert private reply queue into blk_mq " John Garry
2019-08-13 8:30 ` John Garry
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=20190602064202.GA2731@ming.t460p \
--to=ming.lei@redhat.com \
--cc=James.Bottomley@hansenpartnership.com \
--cc=axboe@kernel.dk \
--cc=bvanassche@acm.org \
--cc=don.brace@microsemi.com \
--cc=hare@suse.com \
--cc=hch@lst.de \
--cc=john.garry@huawei.com \
--cc=kashyap.desai@broadcom.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=sathya.prakash@broadcom.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).