From: Lorenzo Stoakes <ljs@kernel.org>
To: Ryan Roberts <ryan.roberts@arm.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>,
Matthew Wilcox <willy@infradead.org>,
Peter Xu <peterx@redhat.com>, Alex Williamson <alex@shazbot.org>,
Anthony Pighin <anthony.pighin@nokia.com>,
linux-kernel@vger.kernel.org,
Kefeng Wang <wangkefeng.wang@huawei.com>,
kvm@vger.kernel.org, linux-mm@kvack.org,
"Liam R. Howlett" <liam@infradead.org>
Subject: Re: [PATCH] vfio: Request THP-aligned mmap for device fds
Date: Mon, 22 Jun 2026 16:48:19 +0100 [thread overview]
Message-ID: <ajlYwMYswMO1LUXk@lucifer> (raw)
In-Reply-To: <160f8cdf-0d26-43f3-b487-4181e6374782@arm.com>
On Fri, Jun 19, 2026 at 04:04:18PM +0100, Ryan Roberts wrote:
> On 18/06/2026 15:55, Lorenzo Stoakes wrote:
> > +cc Ryan for contPMD
> >
> >>
> >>> And maybe there's a CONTPMD architecture we should also consider?
> >>
> >> ARM HW supports "CONTPMD" but I suppose it is not implemented..
> >
> > Maybe Ryan has thoughts?
>
> Sorry not quite sure what you're asking so will give a generic answer:
>
> Arm arch does indeed support CONTPMD at level 2 in the same way that it supports
> CONTPTE at level 3, as follows:
>
> +-----------+-----------+-----------+-----------+-----------+
> | Page Size | CONTPTE | PMD | CONTPMD | PUD |
> +-----------+-----------+-----------+-----------+-----------+
> | 4KB | 64KB | 2MB | 32MB | 1GB |
> | 16KB | 2MB | 32MB | 1GB | |
> | 64KB | 2MB | 512MB | 16GB | |
> +-----------+-----------+-----------+-----------+-----------+
>
> For HugeTLB, Linux/arm64 support all of the listed sizes today. For !HugeTLB
> (i.e. THP, large file filos) Linux/arm64 supports CONTPTE and PMD only. We could
> (and probably would) extend to support CONTPMD once THP PUDs are supported.
>
> Not sure if that answers the question?
Thanks, that's really useful!
I think Jason's concern was that we'd have to account for contpmd here, but
that'd only be w.r.t. THP rather than hugetlb, so we're safe for now I think!
I guess this would be a new >PMD sized mTHP size which... I really hope we have
the THP code in better shape once we come to that :)
>
> Thanks,
> Ryan
>
> >
> >>
> >> Jason
> >
> > Thanks, Lorenzo
>
Cheers, Lorenzo
prev parent reply other threads:[~2026-06-22 15:48 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-06-16 18:01 [PATCH] vfio: Request THP-aligned mmap for device fds Anthony Pighin
2026-06-16 22:30 ` Alex Williamson
2026-06-17 14:21 ` Peter Xu
2026-06-17 18:34 ` Matthew Wilcox
2026-06-17 19:29 ` Jason Gunthorpe
2026-06-18 14:55 ` Lorenzo Stoakes
2026-06-18 15:04 ` Matthew Wilcox
2026-06-18 15:30 ` Jason Gunthorpe
2026-06-18 15:56 ` Lorenzo Stoakes
2026-06-19 13:16 ` Jason Gunthorpe
2026-06-22 16:54 ` Lorenzo Stoakes
2026-06-18 15:28 ` Jason Gunthorpe
2026-06-19 16:11 ` Matthew Wilcox
2026-06-19 17:07 ` Jason Gunthorpe
2026-06-22 15:42 ` Lorenzo Stoakes
2026-06-22 15:44 ` Lorenzo Stoakes
2026-06-19 15:04 ` Ryan Roberts
2026-06-22 15:48 ` Lorenzo Stoakes [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=ajlYwMYswMO1LUXk@lucifer \
--to=ljs@kernel.org \
--cc=alex@shazbot.org \
--cc=anthony.pighin@nokia.com \
--cc=jgg@ziepe.ca \
--cc=kvm@vger.kernel.org \
--cc=liam@infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=peterx@redhat.com \
--cc=ryan.roberts@arm.com \
--cc=wangkefeng.wang@huawei.com \
--cc=willy@infradead.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