From: Danilo Krummrich <dakr@kernel.org>
To: Jason Gunthorpe <jgg@nvidia.com>
Cc: Zhi Wang <zhiw@nvidia.com>,
kvm@vger.kernel.org, nouveau@lists.freedesktop.org,
alex.williamson@redhat.com, kevin.tian@intel.com,
airlied@gmail.com, daniel@ffwll.ch, acurrid@nvidia.com,
cjia@nvidia.com, smitra@nvidia.com, ankita@nvidia.com,
aniketa@nvidia.com, kwankhede@nvidia.com, targupta@nvidia.com,
zhiwang@kernel.org
Subject: Re: [RFC 00/29] Introduce NVIDIA GPU Virtualization (vGPU) Support
Date: Tue, 24 Sep 2024 00:50:55 +0200 [thread overview]
Message-ID: <ZvHwzzp2F71W8TAs@pollux.localdomain> (raw)
In-Reply-To: <20240923150140.GB9417@nvidia.com>
On Mon, Sep 23, 2024 at 12:01:40PM -0300, Jason Gunthorpe wrote:
> On Mon, Sep 23, 2024 at 10:49:07AM +0200, Danilo Krummrich wrote:
> > > 2. Proposal for upstream
> > > ========================
> >
> > What is the strategy in the mid / long term with this?
> >
> > As you know, we're trying to move to Nova and the blockers with the device /
> > driver infrastructure have been resolved and we're able to move forward. Besides
> > that, Dave made great progress on the firmware abstraction side of things.
> >
> > Is this more of a proof of concept? Do you plan to work on Nova in general and
> > vGPU support for Nova?
>
> This is intended to be a real product that customers would use, it is
> not a proof of concept. There is alot of demand for this kind of
> simplified virtualization infrastructure in the host side.
I see...
> The series
> here is the first attempt at making thin host infrastructure and
> Zhi/etc are doing it with an upstream-first approach.
This is great!
>
> From the VFIO side I would like to see something like this merged in
> nearish future as it would bring a previously out of tree approach to
> be fully intree using our modern infrastructure. This is a big win for
> the VFIO world.
>
> As a commercial product this will be backported extensively to many
> old kernels and that is harder/impossible if it isn't exclusively in
> C. So, I think nova needs to co-exist in some way.
We'll surely not support two drivers for the same thing in the long term,
neither does it make sense, nor is it sustainable.
We have a lot of good reasons why we decided to move forward with Nova as a
successor of Nouveau for GSP-based GPUs in the long term -- I also just held a
talk about this at LPC.
For the short/mid term I think it may be reasonable to start with Nouveau, but
this must be based on some agreements, for instance:
- take responsibility, e.g. commitment to help with maintainance with some of
NVKM / NVIDIA GPU core (or whatever we want to call it) within Nouveau
- commitment to help with Nova in general and, once applicable, move the vGPU
parts over to Nova
But I think the very last one naturally happens if we stop further support for
new HW in Nouveau at some point.
>
> Jason
>
next prev parent reply other threads:[~2024-09-23 22:51 UTC|newest]
Thread overview: 86+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-22 12:49 [RFC 00/29] Introduce NVIDIA GPU Virtualization (vGPU) Support Zhi Wang
2024-09-22 12:49 ` [RFC 01/29] nvkm/vgpu: introduce NVIDIA vGPU support prelude Zhi Wang
2024-09-26 9:20 ` Greg KH
2024-10-14 9:59 ` Zhi Wang
2024-10-14 11:36 ` Greg KH
2024-09-22 12:49 ` [RFC 02/29] nvkm/vgpu: attach to nvkm as a nvkm client Zhi Wang
2024-09-26 9:21 ` Greg KH
2024-10-14 10:16 ` Zhi Wang
2024-10-14 11:33 ` Greg KH
2024-09-22 12:49 ` [RFC 03/29] nvkm/vgpu: reserve a larger GSP heap when NVIDIA vGPU is enabled Zhi Wang
2024-09-22 12:49 ` [RFC 04/29] nvkm/vgpu: set the VF partition count " Zhi Wang
2024-09-26 22:51 ` Jason Gunthorpe
2024-10-13 18:54 ` Zhi Wang
2024-10-15 12:20 ` Jason Gunthorpe
2024-10-15 15:19 ` Zhi Wang
2024-10-15 16:35 ` Jason Gunthorpe
2024-09-22 12:49 ` [RFC 05/29] nvkm/vgpu: populate GSP_VF_INFO " Zhi Wang
2024-09-26 22:52 ` Jason Gunthorpe
2024-09-22 12:49 ` [RFC 06/29] nvkm/vgpu: set RMSetSriovMode " Zhi Wang
2024-09-26 22:53 ` Jason Gunthorpe
2024-10-14 7:38 ` Zhi Wang
2024-10-15 3:49 ` Christoph Hellwig
2024-10-15 12:23 ` Jason Gunthorpe
2024-09-22 12:49 ` [RFC 07/29] nvkm/gsp: add a notify handler for GSP event GPUACCT_PERFMON_UTIL_SAMPLES Zhi Wang
2024-09-22 12:49 ` [RFC 08/29] nvkm/vgpu: get the size VMMU segment from GSP firmware Zhi Wang
2024-09-22 12:49 ` [RFC 09/29] nvkm/vgpu: introduce the reserved channel allocator Zhi Wang
2024-09-22 12:49 ` [RFC 10/29] nvkm/vgpu: introduce interfaces for NVIDIA vGPU VFIO module Zhi Wang
2024-09-22 12:49 ` [RFC 11/29] nvkm/vgpu: introduce GSP RM client alloc and free for vGPU Zhi Wang
2024-09-22 12:49 ` [RFC 12/29] nvkm/vgpu: introduce GSP RM control interface " Zhi Wang
2024-09-22 12:49 ` [RFC 13/29] nvkm: move chid.h to nvkm/engine Zhi Wang
2024-09-22 12:49 ` [RFC 14/29] nvkm/vgpu: introduce channel allocation for vGPU Zhi Wang
2024-09-22 12:49 ` [RFC 15/29] nvkm/vgpu: introduce FB memory " Zhi Wang
2024-09-22 12:49 ` [RFC 16/29] nvkm/vgpu: introduce BAR1 map routines for vGPUs Zhi Wang
2024-09-22 12:49 ` [RFC 17/29] nvkm/vgpu: introduce engine bitmap for vGPU Zhi Wang
2024-09-22 12:49 ` [RFC 18/29] nvkm/vgpu: introduce pci_driver.sriov_configure() in nvkm Zhi Wang
2024-09-26 22:56 ` Jason Gunthorpe
2024-10-14 8:32 ` Zhi Wang
2024-10-15 12:27 ` Jason Gunthorpe
2024-10-15 15:14 ` Zhi Wang
2024-10-14 8:36 ` Zhi Wang
2024-09-22 12:49 ` [RFC 19/29] vfio/vgpu_mgr: introdcue vGPU lifecycle management prelude Zhi Wang
2024-09-22 12:49 ` [RFC 20/29] vfio/vgpu_mgr: allocate GSP RM client for NVIDIA vGPU manager Zhi Wang
2024-09-22 12:49 ` [RFC 21/29] vfio/vgpu_mgr: introduce vGPU type uploading Zhi Wang
2024-09-22 12:49 ` [RFC 22/29] vfio/vgpu_mgr: allocate vGPU FB memory when creating vGPUs Zhi Wang
2024-09-22 12:49 ` [RFC 23/29] vfio/vgpu_mgr: allocate vGPU channels " Zhi Wang
2024-09-22 12:49 ` [RFC 24/29] vfio/vgpu_mgr: allocate mgmt heap " Zhi Wang
2024-09-22 12:49 ` [RFC 25/29] vfio/vgpu_mgr: map mgmt heap when creating a vGPU Zhi Wang
2024-09-22 12:49 ` [RFC 26/29] vfio/vgpu_mgr: allocate GSP RM client when creating vGPUs Zhi Wang
2024-09-22 12:49 ` [RFC 27/29] vfio/vgpu_mgr: bootload the new vGPU Zhi Wang
2024-09-25 0:31 ` Dave Airlie
2024-09-22 12:49 ` [RFC 28/29] vfio/vgpu_mgr: introduce vGPU host RPC channel Zhi Wang
2024-09-22 12:49 ` [RFC 29/29] vfio/vgpu_mgr: introduce NVIDIA vGPU VFIO variant driver Zhi Wang
2024-09-22 13:11 ` [RFC 00/29] Introduce NVIDIA GPU Virtualization (vGPU) Support Zhi Wang
2024-09-23 8:38 ` Danilo Krummrich
2024-09-24 19:49 ` Zhi Wang
2024-09-23 6:22 ` Tian, Kevin
2024-09-23 15:02 ` Jason Gunthorpe
2024-09-26 6:43 ` Tian, Kevin
2024-09-26 12:55 ` Jason Gunthorpe
2024-09-26 22:57 ` Jason Gunthorpe
2024-09-27 0:13 ` Tian, Kevin
2024-09-23 8:49 ` Danilo Krummrich
2024-09-23 15:01 ` Jason Gunthorpe
2024-09-23 22:50 ` Danilo Krummrich [this message]
2024-09-24 16:41 ` Jason Gunthorpe
2024-09-24 19:56 ` Danilo Krummrich
2024-09-24 22:52 ` Dave Airlie
2024-09-24 23:47 ` Jason Gunthorpe
2024-09-25 0:18 ` Dave Airlie
2024-09-25 1:29 ` Jason Gunthorpe
2024-09-25 0:53 ` Jason Gunthorpe
2024-09-25 1:08 ` Dave Airlie
2024-09-25 15:28 ` Jason Gunthorpe
2024-09-25 10:55 ` Danilo Krummrich
2024-09-26 9:14 ` Greg KH
2024-09-26 12:42 ` Jason Gunthorpe
2024-09-26 12:54 ` Greg KH
2024-09-26 13:07 ` Danilo Krummrich
2024-09-26 14:40 ` Jason Gunthorpe
2024-09-26 18:07 ` Andy Ritger
2024-09-26 22:23 ` Danilo Krummrich
2024-09-26 22:42 ` Danilo Krummrich
2024-09-27 12:51 ` Jason Gunthorpe
2024-09-27 14:22 ` Danilo Krummrich
2024-09-27 15:27 ` Jason Gunthorpe
2024-09-30 15:59 ` Danilo Krummrich
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=ZvHwzzp2F71W8TAs@pollux.localdomain \
--to=dakr@kernel.org \
--cc=acurrid@nvidia.com \
--cc=airlied@gmail.com \
--cc=alex.williamson@redhat.com \
--cc=aniketa@nvidia.com \
--cc=ankita@nvidia.com \
--cc=cjia@nvidia.com \
--cc=daniel@ffwll.ch \
--cc=jgg@nvidia.com \
--cc=kevin.tian@intel.com \
--cc=kvm@vger.kernel.org \
--cc=kwankhede@nvidia.com \
--cc=nouveau@lists.freedesktop.org \
--cc=smitra@nvidia.com \
--cc=targupta@nvidia.com \
--cc=zhiw@nvidia.com \
--cc=zhiwang@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox