From: Jason Gunthorpe <jgg@nvidia.com>
To: Alexey Kardashevskiy <aik@ozlabs.ru>
Cc: Max Gurtovoy <mgurtovoy@nvidia.com>,
alex.williamson@redhat.com, cohuck@redhat.com,
kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
liranl@nvidia.com, oren@nvidia.com, tzahio@nvidia.com,
leonro@nvidia.com, yarong@nvidia.com, aviadye@nvidia.com,
shahafs@nvidia.com, artemp@nvidia.com, kwankhede@nvidia.com,
ACurrid@nvidia.com, cjia@nvidia.com, yishaih@nvidia.com,
mjrosato@linux.ibm.com, hch@lst.de
Subject: Re: [PATCH 8/9] vfio/pci: export nvlink2 support into vendor vfio_pci drivers
Date: Thu, 11 Mar 2021 13:01:44 -0400 [thread overview]
Message-ID: <20210311170144.GM2356281@nvidia.com> (raw)
In-Reply-To: <73c99da0-6624-7aa2-2857-ef68092c0d07@ozlabs.ru>
On Thu, Mar 11, 2021 at 06:54:09PM +1100, Alexey Kardashevskiy wrote:
> Is there an idea how it is going to work? For example, the Intel IGD driver
> and vfio-pci-igd - how should the system pick one? If there is no
> MODULE_DEVICE_TABLE in vfio-pci-xxx, is the user supposed to try binding all
> vfio-pci-xxx drivers until some binds?
We must expose some MODULE_DEVICE_TABLE like thing to userspace.
Compiling everything into one driver and using if statements was only
managable with these tiny drivers - the stuff that is coming are big
things that are infeasible to link directly to vfio_pci.ko
I'm feeling some general consensus around this approach (vs trying to
make a subdriver) so we will start looking at exactly what form that
could take soon.
The general idea would be to have a selection of extended VFIO drivers
for PCI devices that can be loaded as an alternative to vfio-pci and
they provide additional uapi and behaviors that only work on specific
hardware. nvlink is a good example because it does provide new API and
additional HW specific behavior.
A way for userspace to learn about the drivers automatically and sort
out how to load and bind them.
I was thinking about your earlier question about FDT - do you think we
could switch this to a platform_device and provide an of_match_table
that would select correctly? Did IBM enforce a useful compatible
string in the DT for these things?
Jason
next prev parent reply other threads:[~2021-03-11 17:02 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-09 8:33 [PATCH v3 0/9] Introduce vfio-pci-core subsystem Max Gurtovoy
2021-03-09 8:33 ` [PATCH 1/9] vfio-pci: rename vfio_pci.c to vfio_pci_core.c Max Gurtovoy
2021-03-09 8:33 ` [PATCH 2/9] vfio-pci: rename vfio_pci_private.h to vfio_pci_core.h Max Gurtovoy
2021-03-09 8:33 ` [PATCH 3/9] vfio-pci: rename vfio_pci_device to vfio_pci_core_device Max Gurtovoy
2021-03-09 8:33 ` [PATCH 4/9] vfio-pci: introduce vfio_pci_core subsystem driver Max Gurtovoy
2021-03-09 8:33 ` [PATCH 5/9] vfio/pci: introduce vfio_pci_device structure Max Gurtovoy
2021-03-09 8:33 ` [PATCH 6/9] vfio-pci-core: export vfio_pci_register_dev_region function Max Gurtovoy
2021-03-09 8:33 ` [PATCH 7/9] vfio/pci_core: split nvlink2 to nvlink2gpu and npu2 Max Gurtovoy
2021-03-10 8:08 ` Christoph Hellwig
2021-03-09 8:33 ` [PATCH 8/9] vfio/pci: export nvlink2 support into vendor vfio_pci drivers Max Gurtovoy
2021-03-10 6:39 ` Alexey Kardashevskiy
2021-03-10 12:57 ` Max Gurtovoy
2021-03-10 13:02 ` Jason Gunthorpe
2021-03-10 14:24 ` Alexey Kardashevskiy
2021-03-10 19:40 ` Jason Gunthorpe
2021-03-11 1:20 ` Alexey Kardashevskiy
2021-03-11 1:34 ` Jason Gunthorpe
2021-03-11 1:42 ` Alexey Kardashevskiy
2021-03-11 2:00 ` Jason Gunthorpe
2021-03-11 7:54 ` Alexey Kardashevskiy
2021-03-11 9:44 ` Max Gurtovoy
2021-03-11 16:51 ` Jason Gunthorpe
2021-03-11 17:01 ` Jason Gunthorpe [this message]
2021-03-10 14:19 ` Alexey Kardashevskiy
2021-03-11 1:10 ` Max Gurtovoy
2021-03-19 15:23 ` Alex Williamson
2021-03-19 16:17 ` Jason Gunthorpe
2021-03-19 16:20 ` Christoph Hellwig
2021-03-19 16:28 ` Jason Gunthorpe
2021-03-19 16:34 ` Christoph Hellwig
2021-03-19 17:36 ` Alex Williamson
2021-03-19 20:07 ` Jason Gunthorpe
2021-03-19 21:08 ` Alex Williamson
2021-03-19 22:59 ` Jason Gunthorpe
2021-03-20 4:40 ` Alex Williamson
2021-03-21 12:58 ` Jason Gunthorpe
2021-03-22 16:40 ` Alex Williamson
2021-03-23 19:32 ` Jason Gunthorpe
2021-03-24 2:39 ` Alexey Kardashevskiy
2021-03-29 23:10 ` Alex Williamson
2021-04-01 13:04 ` Cornelia Huck
2021-04-01 13:12 ` Jason Gunthorpe
2021-04-01 21:49 ` Alex Williamson
2021-03-22 15:11 ` Christoph Hellwig
2021-03-22 16:44 ` Jason Gunthorpe
2021-03-23 13:17 ` Christoph Hellwig
2021-03-23 13:42 ` Jason Gunthorpe
2021-03-09 8:33 ` [PATCH 9/9] vfio/pci: export igd support into vendor vfio_pci driver Max Gurtovoy
2021-03-10 8:15 ` Christoph Hellwig
2021-03-10 12:31 ` Jason Gunthorpe
2021-03-11 11:37 ` Christoph Hellwig
2021-03-11 12:09 ` Max Gurtovoy
2021-03-11 15:43 ` Jason Gunthorpe
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=20210311170144.GM2356281@nvidia.com \
--to=jgg@nvidia.com \
--cc=ACurrid@nvidia.com \
--cc=aik@ozlabs.ru \
--cc=alex.williamson@redhat.com \
--cc=artemp@nvidia.com \
--cc=aviadye@nvidia.com \
--cc=cjia@nvidia.com \
--cc=cohuck@redhat.com \
--cc=hch@lst.de \
--cc=kvm@vger.kernel.org \
--cc=kwankhede@nvidia.com \
--cc=leonro@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=liranl@nvidia.com \
--cc=mgurtovoy@nvidia.com \
--cc=mjrosato@linux.ibm.com \
--cc=oren@nvidia.com \
--cc=shahafs@nvidia.com \
--cc=tzahio@nvidia.com \
--cc=yarong@nvidia.com \
--cc=yishaih@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;
as well as URLs for NNTP newsgroup(s).