From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Jason Gunthorpe <jgg@nvidia.com>
Cc: Lorenzo Pieralisi <lpieralisi@kernel.org>,
kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
alex.williamson@redhat.com, osamaabb@amazon.com,
linux-pci@vger.kernel.org, Clint Sbisa <csbisa@amazon.com>,
catalin.marinas@arm.com, maz@kernel.org
Subject: Re: VFIO (PCI) and write combine mapping of BARs
Date: Wed, 26 Jul 2023 11:19:18 +1000 [thread overview]
Message-ID: <012cb915ae0ce3a32f1608400cf6389a03f2aa7d.camel@kernel.crashing.org> (raw)
In-Reply-To: <ZL+0h9gvJGTyWKZX@nvidia.com>
On Tue, 2023-07-25 at 08:39 -0300, Jason Gunthorpe wrote:
> On Tue, Jul 25, 2023 at 04:15:39PM +1000, Benjamin Herrenschmidt wrote:
> > > Assuming this is for #2, I think VFIO has fallen into a bit of a trap
> > > by allowing userspace to form the mmap offset. I've seen this happen
> > > in other subsystems too. It seems like a good idea then you realize
> > > you need more stuff in the mmap space and become sad.
> > >
> > > Typically the way out is to covert the mmap offset into a cookie where
> > > userspace issues some ioctl and then the ioctl returns an opaque mmap
> > > offset to use.
> > >
> > > eg in the vfio context you'd do some 'prepare region for mmap' ioctl
> > > where you could specify flags. The kernel would encode the flags in
> > > the cookie and then mmap would do the right thing. Adding more stuff
> > > is done by enhancing the prepare ioctl.
> > >
> > > Legacy mmap offsets are kept working.
> >
> > This indeed what I have in mind. IE. VFIO has legacy regions and add-on
> > regions though the latter is currently only exploited by some drivers
> > that create their own add-on regions. My proposal is to add an ioctl to
> > create them from userspace as "children" of an existing driver-provided
> > region, allowing to set different attributes for mmap.
>
> I wouldn't call it children, you are just getting a different mmap
> cookie for the same region object.
I though they could be subsets but that might be overkill.
> > In the current VFIO the implementation is *entirely* in vfio_pci_core
> > for PCI and entirely in vfio_platform_common.c for platform, so while
> > the same ioctls could be imagined to create sub-regions, it would have
> > to be completely implemented twice unless we do a lot of heavy lifting
> > to move some of that region stuff into common code.
>
> The machinery for managing the mmap cookies should be in common code
Ok. I'll whip up a POC within vfio_pci only intially to test the
concept and to agree on the API, then look at how we can clean all that
up.
Cheers,
Ben.
prev parent reply other threads:[~2023-07-26 1:20 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-14 2:32 VFIO (PCI) and write combine mapping of BARs Benjamin Herrenschmidt
2023-07-14 7:13 ` Lorenzo Pieralisi
2023-07-14 12:37 ` Jason Gunthorpe
2023-07-25 6:15 ` Benjamin Herrenschmidt
2023-07-25 11:39 ` Jason Gunthorpe
2023-07-26 1:19 ` Benjamin Herrenschmidt [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=012cb915ae0ce3a32f1608400cf6389a03f2aa7d.camel@kernel.crashing.org \
--to=benh@kernel.crashing.org \
--cc=alex.williamson@redhat.com \
--cc=catalin.marinas@arm.com \
--cc=csbisa@amazon.com \
--cc=jgg@nvidia.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=lpieralisi@kernel.org \
--cc=maz@kernel.org \
--cc=osamaabb@amazon.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;
as well as URLs for NNTP newsgroup(s).