From: Daniel Jordan <daniel.m.jordan@oracle.com>
To: Jason Gunthorpe <jgg@nvidia.com>
Cc: Alexander Duyck <alexanderduyck@fb.com>,
Alex Williamson <alex.williamson@redhat.com>,
Andrew Morton <akpm@linux-foundation.org>,
Ben Segall <bsegall@google.com>,
Cornelia Huck <cohuck@redhat.com>,
Dan Williams <dan.j.williams@intel.com>,
Dave Hansen <dave.hansen@linux.intel.com>,
Dietmar Eggemann <dietmar.eggemann@arm.com>,
Herbert Xu <herbert@gondor.apana.org.au>,
Ingo Molnar <mingo@redhat.com>,
Johannes Weiner <hannes@cmpxchg.org>,
Josh Triplett <josh@joshtriplett.org>,
Michal Hocko <mhocko@suse.com>, Nico Pache <npache@redhat.com>,
Pasha Tatashin <pasha.tatashin@soleen.com>,
Peter Zijlstra <peterz@infradead.org>,
Steffen Klassert <steffen.klassert@secunet.com>,
Steve Sistare <steven.sistare@oracle.com>,
Tejun Heo <tj@kernel.org>, Tim Chen <tim.c.chen@linux.intel.com>,
Vincent Guittot <vincent.guittot@linaro.org>,
linux-mm@kvack.org, kvm@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-crypto@vger.kernel.org
Subject: Re: [RFC 00/16] padata, vfio, sched: Multithreaded VFIO page pinning
Date: Tue, 11 Jan 2022 11:20:27 -0500 [thread overview]
Message-ID: <20220111162027.3brb7ga3vgtvv6th@oracle.com> (raw)
In-Reply-To: <20220111001751.GI2328285@nvidia.com>
On Mon, Jan 10, 2022 at 08:17:51PM -0400, Jason Gunthorpe wrote:
> On Mon, Jan 10, 2022 at 05:27:25PM -0500, Daniel Jordan wrote:
>
> > > > Pinning itself, the only thing being optimized, improves 8.5x in that
> > > > experiment, bringing the time from 1.8 seconds to .2 seconds. That's a
> > > > significant savings IMHO
> > >
> > > And here is where I suspect we'd get similar results from folio's
> > > based on the unpin performance uplift we already saw.
> > >
> > > As long as PUP doesn't have to COW its work is largely proportional to
> > > the number of struct pages it processes, so we should be expecting an
> > > upper limit of 512x gains on the PUP alone with foliation.
> > >
> > > This is in line with what we saw with the prior unpin work.
> >
> > "in line with what we saw" Not following. The unpin work had two
> > optimizations, I think, 4.5x and 3.5x which together give 16x. Why is
> > that in line with the potential gains from pup?
>
> It is the same basic issue, doing extra work, dirtying extra memory..
Ok, gotcha.
> I don't know of other users that use such huge memory sizes this would
> matter, besides a VMM..
Right, all the VMMs out there that use vfio.
> > My assumption going into this series was that multithreading VFIO page
> > pinning in the kernel was a viable way forward given the positive
> > feedback I got from the VFIO maintainer last time I posted this, which
> > was admittedly a while ago, and I've since been focused on the other
> > parts of this series rather than what's been happening in the mm lately.
> > Anyway, your arguments are reasonable, so I'll go take a look at some of
> > these optimizations and see where I get.
>
> Well, it is not *unreasonable* it just doesn't seem compelling to me
> yet.
>
> Especially since we are not anywhere close to the limit of single
> threaded performance. Aside from GUP, the whole way we transfer the
> physical pages into the iommu is just begging for optimizations
> eg Matthew's struct phyr needs to be an input and output at the iommu
> layer to make this code really happy.
/nods/ There are other ways forward. As I say, I'll take a look.
prev parent reply other threads:[~2022-01-11 16:21 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-01-06 0:46 [RFC 00/16] padata, vfio, sched: Multithreaded VFIO page pinning Daniel Jordan
2022-01-06 0:46 ` [RFC 01/16] padata: Remove __init from multithreading functions Daniel Jordan
2022-01-06 0:46 ` [RFC 02/16] padata: Return first error from a job Daniel Jordan
2022-01-06 0:46 ` [RFC 03/16] padata: Add undo support Daniel Jordan
2022-01-06 0:46 ` [RFC 04/16] padata: Detect deadlocks between main and helper threads Daniel Jordan
2022-01-06 0:46 ` [RFC 05/16] vfio/type1: Pass mm to vfio_pin_pages_remote() Daniel Jordan
2022-01-06 0:46 ` [RFC 06/16] vfio/type1: Refactor dma map removal Daniel Jordan
2022-01-06 0:46 ` [RFC 07/16] vfio/type1: Parallelize vfio_pin_map_dma() Daniel Jordan
2022-01-06 0:46 ` [RFC 08/16] vfio/type1: Cache locked_vm to ease mmap_lock contention Daniel Jordan
2022-01-06 0:53 ` Jason Gunthorpe
2022-01-06 1:17 ` Daniel Jordan
2022-01-06 12:34 ` Jason Gunthorpe
2022-01-06 21:05 ` Alex Williamson
2022-01-07 0:19 ` Jason Gunthorpe
2022-01-07 3:06 ` Daniel Jordan
2022-01-07 15:18 ` Jason Gunthorpe
2022-01-07 16:39 ` Daniel Jordan
2022-01-06 0:46 ` [RFC 09/16] padata: Use kthreads in do_multithreaded Daniel Jordan
2022-01-06 0:46 ` [RFC 10/16] padata: Helpers should respect main thread's CPU affinity Daniel Jordan
2022-01-06 0:46 ` [RFC 11/16] padata: Cap helpers started to online CPUs Daniel Jordan
2022-01-06 0:46 ` [RFC 12/16] sched, padata: Bound max threads with max_cfs_bandwidth_cpus() Daniel Jordan
2022-01-06 0:46 ` [RFC 13/16] padata: Run helper threads at MAX_NICE Daniel Jordan
2022-01-06 0:46 ` [RFC 14/16] padata: Nice helper threads one by one to prevent starvation Daniel Jordan
2022-01-06 0:46 ` [RFC 15/16] sched/fair: Account kthread runtime debt for CFS bandwidth Daniel Jordan
2022-01-11 11:58 ` Peter Zijlstra
2022-01-11 16:29 ` Daniel Jordan
2022-01-12 20:18 ` Tejun Heo
2022-01-13 21:08 ` Daniel Jordan
2022-01-13 21:11 ` Daniel Jordan
2022-01-14 9:31 ` Peter Zijlstra
2022-01-14 9:40 ` Peter Zijlstra
2022-01-14 16:38 ` Tejun Heo
2022-01-18 17:40 ` Daniel Jordan
2022-01-14 16:30 ` Tejun Heo
2022-01-18 17:32 ` Daniel Jordan
2022-01-06 0:46 ` [RFC 16/16] sched/fair: Consider kthread debt in cputime Daniel Jordan
2022-01-06 1:13 ` [RFC 00/16] padata, vfio, sched: Multithreaded VFIO page pinning Jason Gunthorpe
2022-01-07 3:03 ` Daniel Jordan
2022-01-07 17:12 ` Jason Gunthorpe
2022-01-10 22:27 ` Daniel Jordan
2022-01-11 0:17 ` Jason Gunthorpe
2022-01-11 16:20 ` Daniel Jordan [this message]
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=20220111162027.3brb7ga3vgtvv6th@oracle.com \
--to=daniel.m.jordan@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=alex.williamson@redhat.com \
--cc=alexanderduyck@fb.com \
--cc=bsegall@google.com \
--cc=cohuck@redhat.com \
--cc=dan.j.williams@intel.com \
--cc=dave.hansen@linux.intel.com \
--cc=dietmar.eggemann@arm.com \
--cc=hannes@cmpxchg.org \
--cc=herbert@gondor.apana.org.au \
--cc=jgg@nvidia.com \
--cc=josh@joshtriplett.org \
--cc=kvm@vger.kernel.org \
--cc=linux-crypto@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=mingo@redhat.com \
--cc=npache@redhat.com \
--cc=pasha.tatashin@soleen.com \
--cc=peterz@infradead.org \
--cc=steffen.klassert@secunet.com \
--cc=steven.sistare@oracle.com \
--cc=tim.c.chen@linux.intel.com \
--cc=tj@kernel.org \
--cc=vincent.guittot@linaro.org \
/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).