From: hch@lst.de (Christoph Hellwig)
Subject: [PATCH 0/3] Provide more fine grained control over multipathing
Date: Wed, 6 Jun 2018 11:51:30 +0200 [thread overview]
Message-ID: <20180606095130.GA10485@lst.de> (raw)
In-Reply-To: <CAL1RGDWvxcz7zAoEa-Ukhmsn=Vuj6O07X666BK+vJOdkA7qwNQ@mail.gmail.com>
On Tue, Jun 05, 2018@03:57:05PM -0700, Roland Dreier wrote:
> That makes sense but I'm not sure it covers everything. Probably the
> most common way to do NVMe/RDMA will be with a single HCA that has
> multiple ports, so there's no sensible CPU locality. On the other
> hand we want to keep both ports to the fabric busy. Setting different
> paths for different queues makes sense, but there may be
> single-threaded applications that want a different policy.
>
> I'm not saying anything very profound, but we have to find the right
> balance between too many and too few knobs.
Agreed. And the philosophy here is to start with a as few knobs
as possible and work from there based on actual use cases.
Single threaded applications will run into issues with general
blk-mq philosophy, so to work around that we'll need to dig deeper
and allow borrowing of other cpu queues if we want to cater for that.
WARNING: multiple messages have this Message-ID (diff)
From: Christoph Hellwig <hch@lst.de>
To: Roland Dreier <roland@purestorage.com>
Cc: Christoph Hellwig <hch@lst.de>, Sagi Grimberg <sagi@grimberg.me>,
Mike Snitzer <snitzer@redhat.com>,
Johannes Thumshirn <jthumshirn@suse.de>,
Keith Busch <keith.busch@intel.com>,
Hannes Reinecke <hare@suse.de>,
Laurence Oberman <loberman@redhat.com>,
Ewan Milne <emilne@redhat.com>,
James Smart <james.smart@broadcom.com>,
Linux Kernel Mailinglist <linux-kernel@vger.kernel.org>,
Linux NVMe Mailinglist <linux-nvme@lists.infradead.org>,
"Martin K . Petersen" <martin.petersen@oracle.com>,
Martin George <marting@netapp.com>,
John Meneghini <John.Meneghini@netapp.com>
Subject: Re: [PATCH 0/3] Provide more fine grained control over multipathing
Date: Wed, 6 Jun 2018 11:51:30 +0200 [thread overview]
Message-ID: <20180606095130.GA10485@lst.de> (raw)
In-Reply-To: <CAL1RGDWvxcz7zAoEa-Ukhmsn=Vuj6O07X666BK+vJOdkA7qwNQ@mail.gmail.com>
On Tue, Jun 05, 2018 at 03:57:05PM -0700, Roland Dreier wrote:
> That makes sense but I'm not sure it covers everything. Probably the
> most common way to do NVMe/RDMA will be with a single HCA that has
> multiple ports, so there's no sensible CPU locality. On the other
> hand we want to keep both ports to the fabric busy. Setting different
> paths for different queues makes sense, but there may be
> single-threaded applications that want a different policy.
>
> I'm not saying anything very profound, but we have to find the right
> balance between too many and too few knobs.
Agreed. And the philosophy here is to start with a as few knobs
as possible and work from there based on actual use cases.
Single threaded applications will run into issues with general
blk-mq philosophy, so to work around that we'll need to dig deeper
and allow borrowing of other cpu queues if we want to cater for that.
next prev parent reply other threads:[~2018-06-06 9:51 UTC|newest]
Thread overview: 104+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-05-25 12:53 [PATCH 0/3] Provide more fine grained control over multipathing Johannes Thumshirn
2018-05-25 12:53 ` Johannes Thumshirn
2018-05-25 12:53 ` [PATCH 1/3] nvme: provide a way to disable nvme mpath per subsystem Johannes Thumshirn
2018-05-25 12:53 ` Johannes Thumshirn
2018-05-25 13:47 ` Mike Snitzer
2018-05-25 13:47 ` Mike Snitzer
2018-05-31 8:17 ` Sagi Grimberg
2018-05-31 8:17 ` Sagi Grimberg
2018-05-25 12:53 ` [PATCH 2/3] nvme multipath: added SUBSYS_ATTR_RW Johannes Thumshirn
2018-05-25 12:53 ` Johannes Thumshirn
2018-05-25 12:53 ` [PATCH 3/3] nvme multipath: add dev_attr_mpath_personality Johannes Thumshirn
2018-05-25 12:53 ` Johannes Thumshirn
2018-05-25 13:05 ` [PATCH 0/3] Provide more fine grained control over multipathing Christoph Hellwig
2018-05-25 13:05 ` Christoph Hellwig
2018-05-25 13:58 ` Mike Snitzer
2018-05-25 13:58 ` Mike Snitzer
2018-05-25 14:12 ` Christoph Hellwig
2018-05-25 14:12 ` Christoph Hellwig
2018-05-25 14:50 ` Mike Snitzer
2018-05-25 14:50 ` Mike Snitzer
2018-05-29 1:19 ` Martin K. Petersen
2018-05-29 1:19 ` Martin K. Petersen
2018-05-29 3:02 ` Mike Snitzer
2018-05-29 3:02 ` Mike Snitzer
2018-05-29 7:18 ` Hannes Reinecke
2018-05-29 7:18 ` Hannes Reinecke
2018-05-29 7:22 ` Johannes Thumshirn
2018-05-29 7:22 ` Johannes Thumshirn
2018-05-29 8:09 ` Christoph Hellwig
2018-05-29 8:09 ` Christoph Hellwig
2018-05-29 9:54 ` Mike Snitzer
2018-05-29 9:54 ` Mike Snitzer
2018-05-29 23:27 ` Mike Snitzer
2018-05-29 23:27 ` Mike Snitzer
2018-05-30 19:05 ` Jens Axboe
2018-05-30 19:05 ` Jens Axboe
2018-05-30 19:59 ` Mike Snitzer
2018-05-30 19:59 ` Mike Snitzer
2018-06-04 6:19 ` Hannes Reinecke
2018-06-04 6:19 ` Hannes Reinecke
2018-06-04 7:18 ` Johannes Thumshirn
2018-06-04 7:18 ` Johannes Thumshirn
2018-06-04 12:59 ` Christoph Hellwig
2018-06-04 12:59 ` Christoph Hellwig
2018-06-04 13:27 ` Mike Snitzer
2018-06-04 13:27 ` Mike Snitzer
2018-05-31 2:42 ` Ming Lei
2018-05-31 2:42 ` Ming Lei
2018-05-30 21:20 ` Sagi Grimberg
2018-05-30 21:20 ` Sagi Grimberg
2018-05-30 22:02 ` Mike Snitzer
2018-05-30 22:02 ` Mike Snitzer
2018-05-31 8:37 ` Sagi Grimberg
2018-05-31 8:37 ` Sagi Grimberg
2018-05-31 12:37 ` Mike Snitzer
2018-05-31 12:37 ` Mike Snitzer
2018-05-31 16:34 ` Christoph Hellwig
2018-05-31 16:34 ` Christoph Hellwig
2018-06-01 4:11 ` Mike Snitzer
2018-06-01 4:11 ` Mike Snitzer
2018-05-31 16:36 ` Christoph Hellwig
2018-05-31 16:36 ` Christoph Hellwig
2018-05-31 16:33 ` Christoph Hellwig
2018-05-31 16:33 ` Christoph Hellwig
2018-05-31 18:17 ` Mike Snitzer
2018-05-31 18:17 ` Mike Snitzer
2018-06-01 2:40 ` Martin K. Petersen
2018-06-01 2:40 ` Martin K. Petersen
2018-06-01 4:24 ` Mike Snitzer
2018-06-01 4:24 ` Mike Snitzer
2018-06-01 14:09 ` Martin K. Petersen
2018-06-01 14:09 ` Martin K. Petersen
2018-06-01 15:21 ` Mike Snitzer
2018-06-01 15:21 ` Mike Snitzer
2018-06-03 11:00 ` Sagi Grimberg
2018-06-03 11:00 ` Sagi Grimberg
2018-06-03 16:06 ` Mike Snitzer
2018-06-03 16:06 ` Mike Snitzer
2018-06-04 11:46 ` Sagi Grimberg
2018-06-04 11:46 ` Sagi Grimberg
2018-06-04 12:48 ` Johannes Thumshirn
2018-06-04 12:48 ` Johannes Thumshirn
2018-05-30 22:44 ` Mike Snitzer
2018-05-30 22:44 ` Mike Snitzer
2018-05-31 8:51 ` Sagi Grimberg
2018-05-31 8:51 ` Sagi Grimberg
2018-05-31 12:41 ` Mike Snitzer
2018-05-31 12:41 ` Mike Snitzer
2018-06-04 21:58 ` Roland Dreier
2018-06-04 21:58 ` Roland Dreier
2018-06-05 4:42 ` Christoph Hellwig
2018-06-05 4:42 ` Christoph Hellwig
2018-06-05 22:57 ` Roland Dreier
2018-06-05 22:57 ` Roland Dreier
2018-06-06 9:51 ` Christoph Hellwig [this message]
2018-06-06 9:51 ` Christoph Hellwig
2018-06-06 9:32 ` Sagi Grimberg
2018-06-06 9:32 ` Sagi Grimberg
2018-06-06 9:50 ` Christoph Hellwig
2018-06-06 9:50 ` Christoph Hellwig
2018-05-25 14:22 ` Johannes Thumshirn
2018-05-25 14:22 ` Johannes Thumshirn
2018-05-25 14:30 ` Christoph Hellwig
2018-05-25 14:30 ` Christoph Hellwig
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=20180606095130.GA10485@lst.de \
--to=hch@lst.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.