From: Keith Busch <kbusch@kernel.org>
To: Christoph Hellwig <hch@infradead.org>
Cc: Keith Busch <keith.busch@gmail.com>, Jens Axboe <axboe@kernel.dk>,
"Busch, Keith" <keith.busch@intel.com>,
Bart Van Assche <bvanassche@acm.org>,
"linux-nvme@lists.infradead.org" <linux-nvme@lists.infradead.org>,
Ming Lei <ming.lei@redhat.com>,
"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
Jianchao Wang <jianchao.w.wang@oracle.com>,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [PATCH] blk-mq: Wait for for hctx requests on CPU unplug
Date: Mon, 8 Apr 2019 09:23:02 -0600 [thread overview]
Message-ID: <20190408152302.GE32498@localhost.localdomain> (raw)
In-Reply-To: <20190407075123.GA22003@infradead.org>
On Sun, Apr 07, 2019 at 12:51:23AM -0700, Christoph Hellwig wrote:
> On Fri, Apr 05, 2019 at 05:36:32PM -0600, Keith Busch wrote:
> > On Fri, Apr 5, 2019 at 5:04 PM Jens Axboe <axboe@kernel.dk> wrote:
> > > Looking at current peak testing, I've got around 1.2% in queue enter
> > > and exit. It's definitely not free, hence my question. Probably safe
> > > to assume that we'll double that cycle counter, per IO.
> >
> > Okay, that's not negligible at all. I don't know of a faster reference
> > than the percpu_ref, but that much overhead would have to rule out
> > having a per hctx counter.
>
> Can we just replace queue_enter/exit with the per-hctx reference
> entirely?
I don't think that we can readily do that. We still need to protect a
request_queue access prior to selecting the hctx.
WARNING: multiple messages have this Message-ID (diff)
From: kbusch@kernel.org (Keith Busch)
Subject: [PATCH] blk-mq: Wait for for hctx requests on CPU unplug
Date: Mon, 8 Apr 2019 09:23:02 -0600 [thread overview]
Message-ID: <20190408152302.GE32498@localhost.localdomain> (raw)
In-Reply-To: <20190407075123.GA22003@infradead.org>
On Sun, Apr 07, 2019@12:51:23AM -0700, Christoph Hellwig wrote:
> On Fri, Apr 05, 2019@05:36:32PM -0600, Keith Busch wrote:
> > On Fri, Apr 5, 2019@5:04 PM Jens Axboe <axboe@kernel.dk> wrote:
> > > Looking at current peak testing, I've got around 1.2% in queue enter
> > > and exit. It's definitely not free, hence my question. Probably safe
> > > to assume that we'll double that cycle counter, per IO.
> >
> > Okay, that's not negligible at all. I don't know of a faster reference
> > than the percpu_ref, but that much overhead would have to rule out
> > having a per hctx counter.
>
> Can we just replace queue_enter/exit with the per-hctx reference
> entirely?
I don't think that we can readily do that. We still need to protect a
request_queue access prior to selecting the hctx.
next prev parent reply other threads:[~2019-04-08 15:21 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-05 21:59 [PATCH] blk-mq: Wait for for hctx requests on CPU unplug Keith Busch
2019-04-05 21:59 ` Keith Busch
2019-04-05 22:23 ` Jens Axboe
2019-04-05 22:23 ` Jens Axboe
2019-04-05 22:37 ` Keith Busch
2019-04-05 22:37 ` Keith Busch
2019-04-05 23:04 ` Jens Axboe
2019-04-05 23:04 ` Jens Axboe
2019-04-05 23:36 ` Keith Busch
2019-04-05 23:36 ` Keith Busch
2019-04-06 9:44 ` Dongli Zhang
2019-04-06 9:44 ` Dongli Zhang
2019-04-06 21:27 ` Ming Lei
2019-04-06 21:27 ` Ming Lei
2019-04-07 13:55 ` Dongli Zhang
2019-04-07 13:55 ` Dongli Zhang
2019-04-08 9:49 ` Ming Lei
2019-04-08 9:49 ` Ming Lei
2019-04-08 15:36 ` Keith Busch
2019-04-08 15:36 ` Keith Busch
2019-04-08 15:21 ` Keith Busch
2019-04-08 15:21 ` Keith Busch
2019-04-07 7:51 ` Christoph Hellwig
2019-04-07 7:51 ` Christoph Hellwig
2019-04-08 15:23 ` Keith Busch [this message]
2019-04-08 15:23 ` Keith Busch
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=20190408152302.GE32498@localhost.localdomain \
--to=kbusch@kernel.org \
--cc=axboe@kernel.dk \
--cc=bvanassche@acm.org \
--cc=hch@infradead.org \
--cc=jianchao.w.wang@oracle.com \
--cc=keith.busch@gmail.com \
--cc=keith.busch@intel.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=ming.lei@redhat.com \
--cc=tglx@linutronix.de \
/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.