From: Johannes Thumshirn <jthumshirn@suse.de>
To: Sagi Grimberg <sagi@grimberg.me>
Cc: Jens Axboe <axboe@kernel.dk>,
Christoph Hellwig <hch@infradead.org>,
Linux-scsi@vger.kernel.org, linux-nvme@lists.infradead.org,
linux-block@vger.kernel.org, Keith Busch <keith.busch@intel.com>,
"lsf-pc@lists.linux-foundation.org"
<lsf-pc@lists.linux-foundation.org>
Subject: Re: [LSF/MM TOPIC][LSF/MM ATTEND] NAPI polling for block drivers
Date: Wed, 18 Jan 2017 14:51:56 +0100 [thread overview]
Message-ID: <20170118135156.GG3514@linux-x5ow.site> (raw)
In-Reply-To: <6df6bf6a-7cd3-1700-2b0a-e140325ebf47@grimberg.me>
On Tue, Jan 17, 2017 at 06:38:43PM +0200, Sagi Grimberg wrote:
>
> >Just for the record, all tests you've run are with the upper irq_poll_budget of
> >256 [1]?
>
> Yes, but that's the point, I never ever reach this budget because
> I'm only processing 1-2 completions per interrupt.
>
> >We (Hannes and me) recently stumbed accross this when trying to poll for more
> >than 256 queue entries in the drivers we've been testing.
s/stumbed/stumbled/
>
> What do you mean by stumbed? irq-poll should be agnostic to the fact
> that drivers can poll more than their given budget?
So what you say is you saw a consomed == 1 [1] most of the time?
[1] from http://git.infradead.org/nvme.git/commitdiff/eed5a9d925c59e43980047059fde29e3aa0b7836
--
Johannes Thumshirn Storage
jthumshirn@suse.de +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 N�rnberg
GF: Felix Imend�rffer, Jane Smithard, Graham Norton
HRB 21284 (AG N�rnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850
WARNING: multiple messages have this Message-ID (diff)
From: jthumshirn@suse.de (Johannes Thumshirn)
Subject: [LSF/MM TOPIC][LSF/MM ATTEND] NAPI polling for block drivers
Date: Wed, 18 Jan 2017 14:51:56 +0100 [thread overview]
Message-ID: <20170118135156.GG3514@linux-x5ow.site> (raw)
In-Reply-To: <6df6bf6a-7cd3-1700-2b0a-e140325ebf47@grimberg.me>
On Tue, Jan 17, 2017@06:38:43PM +0200, Sagi Grimberg wrote:
>
> >Just for the record, all tests you've run are with the upper irq_poll_budget of
> >256 [1]?
>
> Yes, but that's the point, I never ever reach this budget because
> I'm only processing 1-2 completions per interrupt.
>
> >We (Hannes and me) recently stumbed accross this when trying to poll for more
> >than 256 queue entries in the drivers we've been testing.
s/stumbed/stumbled/
>
> What do you mean by stumbed? irq-poll should be agnostic to the fact
> that drivers can poll more than their given budget?
So what you say is you saw a consomed == 1 [1] most of the time?
[1] from http://git.infradead.org/nvme.git/commitdiff/eed5a9d925c59e43980047059fde29e3aa0b7836
--
Johannes Thumshirn Storage
jthumshirn at suse.de +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 N?rnberg
GF: Felix Imend?rffer, Jane Smithard, Graham Norton
HRB 21284 (AG N?rnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850
WARNING: multiple messages have this Message-ID (diff)
From: Johannes Thumshirn <jthumshirn@suse.de>
To: Sagi Grimberg <sagi@grimberg.me>
Cc: Jens Axboe <axboe@kernel.dk>,
Christoph Hellwig <hch@infradead.org>,
Linux-scsi@vger.kernel.org, linux-nvme@lists.infradead.org,
linux-block@vger.kernel.org, Keith Busch <keith.busch@intel.com>,
"lsf-pc@lists.linux-foundation.org"
<lsf-pc@lists.linux-foundation.org>
Subject: Re: [LSF/MM TOPIC][LSF/MM ATTEND] NAPI polling for block drivers
Date: Wed, 18 Jan 2017 14:51:56 +0100 [thread overview]
Message-ID: <20170118135156.GG3514@linux-x5ow.site> (raw)
In-Reply-To: <6df6bf6a-7cd3-1700-2b0a-e140325ebf47@grimberg.me>
On Tue, Jan 17, 2017 at 06:38:43PM +0200, Sagi Grimberg wrote:
>
> >Just for the record, all tests you've run are with the upper irq_poll_budget of
> >256 [1]?
>
> Yes, but that's the point, I never ever reach this budget because
> I'm only processing 1-2 completions per interrupt.
>
> >We (Hannes and me) recently stumbed accross this when trying to poll for more
> >than 256 queue entries in the drivers we've been testing.
s/stumbed/stumbled/
>
> What do you mean by stumbed? irq-poll should be agnostic to the fact
> that drivers can poll more than their given budget?
So what you say is you saw a consomed == 1 [1] most of the time?
[1] from http://git.infradead.org/nvme.git/commitdiff/eed5a9d925c59e43980047059fde29e3aa0b7836
--
Johannes Thumshirn Storage
jthumshirn@suse.de +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850
next prev parent reply other threads:[~2017-01-18 13:51 UTC|newest]
Thread overview: 120+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-01-11 13:43 [LSF/MM TOPIC][LSF/MM ATTEND] NAPI polling for block drivers Johannes Thumshirn
2017-01-11 13:43 ` Johannes Thumshirn
2017-01-11 13:43 ` Johannes Thumshirn
2017-01-11 13:46 ` Hannes Reinecke
2017-01-11 13:46 ` Hannes Reinecke
2017-01-11 13:46 ` Hannes Reinecke
2017-01-11 15:07 ` Jens Axboe
2017-01-11 15:07 ` Jens Axboe
2017-01-11 15:13 ` Jens Axboe
2017-01-11 15:13 ` Jens Axboe
2017-01-12 8:23 ` Sagi Grimberg
2017-01-12 8:23 ` Sagi Grimberg
2017-01-12 10:02 ` Johannes Thumshirn
2017-01-12 10:02 ` Johannes Thumshirn
2017-01-12 10:02 ` Johannes Thumshirn
2017-01-12 11:44 ` Sagi Grimberg
2017-01-12 11:44 ` Sagi Grimberg
2017-01-12 12:53 ` Johannes Thumshirn
2017-01-12 12:53 ` Johannes Thumshirn
2017-01-12 12:53 ` Johannes Thumshirn
2017-01-12 14:41 ` [Lsf-pc] " Sagi Grimberg
2017-01-12 14:41 ` Sagi Grimberg
2017-01-12 18:59 ` Johannes Thumshirn
2017-01-12 18:59 ` Johannes Thumshirn
2017-01-12 18:59 ` Johannes Thumshirn
2017-01-17 15:38 ` Sagi Grimberg
2017-01-17 15:38 ` Sagi Grimberg
2017-01-17 15:45 ` Sagi Grimberg
2017-01-17 15:45 ` Sagi Grimberg
2017-01-20 12:22 ` Johannes Thumshirn
2017-01-20 12:22 ` Johannes Thumshirn
2017-01-20 12:22 ` Johannes Thumshirn
2017-01-17 16:15 ` Sagi Grimberg
2017-01-17 16:15 ` Sagi Grimberg
2017-01-17 16:27 ` Johannes Thumshirn
2017-01-17 16:27 ` Johannes Thumshirn
2017-01-17 16:27 ` Johannes Thumshirn
2017-01-17 16:38 ` Sagi Grimberg
2017-01-17 16:38 ` Sagi Grimberg
2017-01-18 13:51 ` Johannes Thumshirn [this message]
2017-01-18 13:51 ` Johannes Thumshirn
2017-01-18 13:51 ` Johannes Thumshirn
2017-01-18 14:27 ` Sagi Grimberg
2017-01-18 14:27 ` Sagi Grimberg
2017-01-18 14:36 ` Andrey Kuzmin
2017-01-18 14:36 ` Andrey Kuzmin
2017-01-18 14:40 ` Sagi Grimberg
2017-01-18 14:40 ` Sagi Grimberg
2017-01-18 15:35 ` Andrey Kuzmin
2017-01-18 15:35 ` Andrey Kuzmin
2017-01-18 14:58 ` Johannes Thumshirn
2017-01-18 14:58 ` Johannes Thumshirn
2017-01-18 14:58 ` Johannes Thumshirn
2017-01-18 15:14 ` Sagi Grimberg
2017-01-18 15:14 ` Sagi Grimberg
2017-01-18 15:16 ` Johannes Thumshirn
2017-01-18 15:16 ` Johannes Thumshirn
2017-01-18 15:16 ` Johannes Thumshirn
2017-01-18 15:39 ` Hannes Reinecke
2017-01-18 15:39 ` Hannes Reinecke
2017-01-18 15:39 ` Hannes Reinecke
2017-01-19 8:12 ` Sagi Grimberg
2017-01-19 8:12 ` Sagi Grimberg
2017-01-19 8:23 ` Sagi Grimberg
2017-01-19 8:23 ` Sagi Grimberg
2017-01-19 9:18 ` Johannes Thumshirn
2017-01-19 9:18 ` Johannes Thumshirn
2017-01-19 9:18 ` Johannes Thumshirn
2017-01-19 9:13 ` Johannes Thumshirn
2017-01-19 9:13 ` Johannes Thumshirn
2017-01-19 9:13 ` Johannes Thumshirn
2017-01-17 16:44 ` Andrey Kuzmin
2017-01-17 16:50 ` Sagi Grimberg
2017-01-17 16:50 ` Sagi Grimberg
2017-01-18 14:02 ` Hannes Reinecke
2017-01-18 14:02 ` Hannes Reinecke
2017-01-20 0:13 ` Jens Axboe
2017-01-20 0:13 ` Jens Axboe
2017-01-13 15:56 ` Johannes Thumshirn
2017-01-13 15:56 ` Johannes Thumshirn
2017-01-13 15:56 ` Johannes Thumshirn
2017-01-11 15:16 ` Hannes Reinecke
2017-01-11 15:16 ` Hannes Reinecke
2017-01-11 15:16 ` Hannes Reinecke
2017-01-12 4:36 ` Stephen Bates
2017-01-12 4:44 ` Jens Axboe
2017-01-12 4:44 ` Jens Axboe
2017-01-12 4:56 ` Stephen Bates
2017-01-12 4:56 ` Stephen Bates
2017-01-19 10:57 ` Ming Lei
2017-01-19 10:57 ` Ming Lei
2017-01-19 11:03 ` Hannes Reinecke
2017-01-19 11:03 ` Hannes Reinecke
2017-01-11 16:08 ` Bart Van Assche
2017-01-11 16:08 ` Bart Van Assche
2017-01-11 16:08 ` Bart Van Assche
2017-01-11 16:12 ` hch
2017-01-11 16:12 ` hch
2017-01-11 16:15 ` Jens Axboe
2017-01-11 16:15 ` Jens Axboe
2017-01-11 16:22 ` Hannes Reinecke
2017-01-11 16:22 ` Hannes Reinecke
2017-01-11 16:22 ` Hannes Reinecke
2017-01-11 16:26 ` Bart Van Assche
2017-01-11 16:26 ` Bart Van Assche
2017-01-11 16:26 ` Bart Van Assche
2017-01-11 16:45 ` Hannes Reinecke
2017-01-11 16:45 ` Hannes Reinecke
2017-01-11 16:45 ` Hannes Reinecke
2017-01-12 8:52 ` sagi grimberg
2017-01-12 8:52 ` sagi grimberg
2017-01-11 16:14 ` Johannes Thumshirn
2017-01-11 16:14 ` Johannes Thumshirn
2017-01-11 16:14 ` Johannes Thumshirn
2017-01-12 8:41 ` Sagi Grimberg
2017-01-12 8:41 ` Sagi Grimberg
2017-01-12 8:41 ` Sagi Grimberg
2017-01-12 19:13 ` Bart Van Assche
2017-01-12 19:13 ` Bart Van Assche
2017-01-12 19:13 ` Bart Van Assche
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=20170118135156.GG3514@linux-x5ow.site \
--to=jthumshirn@suse.de \
--cc=Linux-scsi@vger.kernel.org \
--cc=axboe@kernel.dk \
--cc=hch@infradead.org \
--cc=keith.busch@intel.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=lsf-pc@lists.linux-foundation.org \
--cc=sagi@grimberg.me \
/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.