From: Jason Gunthorpe <jgg@nvidia.com>
To: "Cédric Le Goater" <clg@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
linux-mm@kvack.org, Peter Xu <peterx@redhat.com>,
Lorenzo Stoakes <ljs@kernel.org>,
David Hildenbrand <david@kernel.org>,
Alex Williamson <alex.williamson@redhat.com>,
Zi Yan <ziy@nvidia.com>,
stable@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] mm/huge_memory: let special huge VMAs bypass the THP policy check
Date: Wed, 5 Aug 2026 09:15:24 -0300 [thread overview]
Message-ID: <20260805121524.GM27883@nvidia.com> (raw)
In-Reply-To: <20260805055544.1568534-1-clg@redhat.com>
On Wed, Aug 05, 2026 at 07:55:40AM +0200, Cédric Le Goater wrote:
> The result: each 256 GB BAR is mapped at 4 KiB granularity -- 67 million
> page faults per GPU instead of a few thousand PMD/PUD faults. On hosts
> with 8 GPUs (2 TB of BAR space), this causes VM boot times to degrade
> severely, with 99.98% of CPU time spent in the VFIO BAR mapping path.
Broadly upstream we expect people to use iommufd and dmabuf for these
configurations to avoid all this nonsense overhead. :\
Jason
next prev parent reply other threads:[~2026-08-05 12:15 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-05 5:55 [PATCH] mm/huge_memory: let special huge VMAs bypass the THP policy check Cédric Le Goater
2026-08-05 10:41 ` Lorenzo Stoakes (ARM)
2026-08-05 16:21 ` Cédric Le Goater
2026-08-05 16:26 ` Lorenzo Stoakes (ARM)
2026-08-05 16:29 ` Cédric Le Goater
2026-08-06 1:44 ` Matthew Wilcox
2026-08-06 6:29 ` Lorenzo Stoakes (ARM)
2026-08-06 16:19 ` David Hildenbrand (Arm)
2026-08-06 16:34 ` Cédric Le Goater
2026-08-06 16:45 ` David Hildenbrand (Arm)
2026-08-05 12:15 ` Jason Gunthorpe [this message]
2026-08-05 16:29 ` Lorenzo Stoakes (ARM)
2026-08-05 16:52 ` Jason Gunthorpe
2026-08-05 16:54 ` Lorenzo Stoakes (ARM)
2026-08-06 14:26 ` David Hildenbrand (Arm)
2026-08-06 14:28 ` Lorenzo Stoakes (ARM)
2026-08-06 14:42 ` Cédric Le Goater
2026-08-06 14:47 ` David Hildenbrand (Arm)
2026-08-06 14:53 ` Cédric Le Goater
2026-08-06 14:59 ` David Hildenbrand (Arm)
2026-08-06 16:45 ` Jason Gunthorpe
2026-08-06 16:47 ` David Hildenbrand (Arm)
2026-08-06 16:50 ` AI slop (was Re: [PATCH] mm/huge_memory: let special huge VMAs bypass the THP) " Lorenzo Stoakes (ARM)
2026-08-06 17:05 ` John Hubbard
2026-08-06 17:08 ` Zi Yan
2026-08-06 17:40 ` Gregory Price
2026-08-06 17:42 ` Zi Yan
2026-08-06 17:42 ` Lorenzo Stoakes (ARM)
2026-08-06 19:28 ` Gregory Price
2026-08-06 19:33 ` Lorenzo Stoakes (ARM)
2026-08-06 17:41 ` Lorenzo Stoakes (ARM)
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=20260805121524.GM27883@nvidia.com \
--to=jgg@nvidia.com \
--cc=akpm@linux-foundation.org \
--cc=alex.williamson@redhat.com \
--cc=clg@redhat.com \
--cc=david@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=ljs@kernel.org \
--cc=peterx@redhat.com \
--cc=stable@vger.kernel.org \
--cc=ziy@nvidia.com \
/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