From: Jason Gunthorpe <jgg@ziepe.ca>
To: Christoph Hellwig <hch@infradead.org>
Cc: Martin Oliveira <martin.oliveira@eideticom.com>,
linux-rdma@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>,
Artemy Kovalyov <artemyko@nvidia.com>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Leon Romanovsky <leon@kernel.org>,
Logan Gunthorpe <logang@deltatee.com>,
Michael Guralnik <michaelgur@nvidia.com>,
Mike Marciniszyn <mike.marciniszyn@intel.com>,
Shiraz Saleem <shiraz.saleem@intel.com>,
Tejun Heo <tj@kernel.org>, John Hubbard <jhubbard@nvidia.com>,
Dan Williams <dan.j.williams@intel.com>,
David Sloan <david.sloan@eideticom.com>
Subject: Re: [PATCH v5 3/4] mm/gup: allow FOLL_LONGTERM & FOLL_PCI_P2PDMA
Date: Mon, 12 Aug 2024 20:12:49 -0300 [thread overview]
Message-ID: <20240812231249.GG1985367@ziepe.ca> (raw)
In-Reply-To: <ZrmuGrDaJTZFrKrc@infradead.org>
On Sun, Aug 11, 2024 at 11:39:22PM -0700, Christoph Hellwig wrote:
> On Thu, Aug 08, 2024 at 12:33:39PM -0600, Martin Oliveira wrote:
> > This check existed originally due to concerns that P2PDMA needed to copy
> > fsdax until pgmap refcounts were fixed (see [1]).
> >
> > The P2PDMA infrastructure will only call unmap_mapping_range() when the
> > underlying device is unbound, and immediately after unmapping it waits
> > for the reference of all ZONE_DEVICE pages to be released before
> > continuing. This does not allow for a page to be reused and no user
> > access fault is therefore possible. It does not have the same problem as
> > fsdax.
> >
> > The one minor concern with FOLL_LONGTERM pins is they will block device
> > unbind until userspace releases them all.
>
> This is unfortunately not really minor unless we have a well documented
> way to force this :(
It is not that different from blocking driver unbind while FDs are
open which a lot of places do in various ways?
Jason
next prev parent reply other threads:[~2024-08-12 23:12 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-08-08 18:33 [PATCH v5 0/4] Enable P2PDMA in Userspace RDMA Martin Oliveira
2024-08-08 18:33 ` [PATCH v5 1/4] kernfs: add a WARN_ON_ONCE if ->close is set Martin Oliveira
2024-08-09 5:37 ` Greg Kroah-Hartman
2024-08-09 15:41 ` Martin Oliveira
2024-08-11 8:31 ` Greg Kroah-Hartman
2024-08-12 6:38 ` Christoph Hellwig
2024-08-12 6:38 ` Christoph Hellwig
2024-08-08 18:33 ` [PATCH v5 2/4] kernfs: remove page_mkwrite() from vm_operations_struct Martin Oliveira
2024-08-12 6:38 ` Christoph Hellwig
2024-08-08 18:33 ` [PATCH v5 3/4] mm/gup: allow FOLL_LONGTERM & FOLL_PCI_P2PDMA Martin Oliveira
2024-08-12 6:39 ` Christoph Hellwig
2024-08-12 23:12 ` Jason Gunthorpe [this message]
2024-08-13 5:41 ` Christoph Hellwig
2024-08-13 16:05 ` Jason Gunthorpe
2024-08-13 17:03 ` Dan Williams
2024-08-14 0:02 ` Jason Gunthorpe
2024-08-14 4:26 ` Christoph Hellwig
2024-08-15 16:38 ` Jason Gunthorpe
2024-08-08 18:33 ` [PATCH v5 4/4] RDMA/umem: add support for P2P RDMA Martin Oliveira
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=20240812231249.GG1985367@ziepe.ca \
--to=jgg@ziepe.ca \
--cc=akpm@linux-foundation.org \
--cc=artemyko@nvidia.com \
--cc=dan.j.williams@intel.com \
--cc=david.sloan@eideticom.com \
--cc=gregkh@linuxfoundation.org \
--cc=hch@infradead.org \
--cc=jhubbard@nvidia.com \
--cc=leon@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-rdma@vger.kernel.org \
--cc=logang@deltatee.com \
--cc=martin.oliveira@eideticom.com \
--cc=michaelgur@nvidia.com \
--cc=mike.marciniszyn@intel.com \
--cc=shiraz.saleem@intel.com \
--cc=tj@kernel.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 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.