From: "Danilo Krummrich" <dakr@kernel.org>
To: "Zhi Wang" <zhiw@nvidia.com>
Cc: <kvm@vger.kernel.org>, <alex.williamson@redhat.com>,
<kevin.tian@intel.com>, <jgg@nvidia.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>, <acourbot@nvidia.com>,
<joelagnelf@nvidia.com>, <apopple@nvidia.com>,
<jhubbard@nvidia.com>, <nouveau@lists.freedesktop.org>
Subject: Re: [RFC v2 03/14] vfio/nvidia-vgpu: introduce vGPU type uploading
Date: Sat, 06 Sep 2025 12:34:40 +0200 [thread overview]
Message-ID: <DCLNF4ADS21M.1WJ4YMSZ4RNF1@kernel.org> (raw)
In-Reply-To: <20250904174213.00003c38@nvidia.com>
On Thu Sep 4, 2025 at 5:43 PM CEST, Zhi Wang wrote:
> On Thu, 04 Sep 2025 11:41:03 +0200
> "Danilo Krummrich" <dakr@kernel.org> wrote:
>
>> (Cc: Alex, John, Joel, Alistair, nouveau)
>>
>> On Thu Sep 4, 2025 at 11:37 AM CEST, Danilo Krummrich wrote:
>> > nova-core won't provide any firmware specific APIs, it is meant to serve as a
>> > hardware and firmware abstraction layer for higher level drivers, such as vGPU
>> > or nova-drm.
>> >
>> > As a general rule the interface between nova-core and higher level drivers must
>> > not leak any hardware or firmware specific details, but work on a higher level
>> > abstraction layer.
>> >
>
> It is more a matter of where we are going to place vGPU specific
> functionality in the whole picture. In this case, if we are thinking about
> the requirement of vGPU type loading, which requires the GSP version
> number and checking. Are we leaning towards putting some vGPU specific
> functionality also in nova-core?
As much as needed to abstract firmware (and hardware) API details.
> Regarding not leaking any of the hardware details, is that doable?
> Looking at {nv04 * _fence}.c {chan*}.c in the current NVIF interfaces, I
> think we will expose the HW concept somehow.
I don't really mean that vGPU must be entirely unaware of the hardware, it's
still a driver of course. But for the API between nova-core and client drivers
we want to abstract how the firmware and hardware is programmed, i.e. not leak
any (version specific) RM structures or provide APIs that consume raw register
values to write, etc.
next prev parent reply other threads:[~2025-09-06 10:34 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-09-03 22:10 [RFC v2 00/14] Introduce NVIDIA GPU Virtualization (vGPU) Support Zhi Wang
2025-09-03 22:10 ` [RFC v2 01/14] vfio/nvidia-vgpu: introduce vGPU lifecycle management prelude Zhi Wang
2025-09-03 22:10 ` [RFC v2 02/14] vfio/nvidia-vgpu: allocate GSP RM client for NVIDIA vGPU manager Zhi Wang
2025-09-03 22:11 ` [RFC v2 03/14] vfio/nvidia-vgpu: introduce vGPU type uploading Zhi Wang
2025-09-04 9:37 ` Danilo Krummrich
2025-09-04 9:41 ` Danilo Krummrich
2025-09-04 12:15 ` Jason Gunthorpe
2025-09-04 12:45 ` Danilo Krummrich
2025-09-04 13:58 ` Jason Gunthorpe
2025-09-04 15:43 ` Zhi Wang
2025-09-06 10:34 ` Danilo Krummrich [this message]
2025-09-03 22:11 ` [RFC v2 04/14] vfio/nvidia-vgpu: allocate vGPU channels when creating vGPUs Zhi Wang
2025-09-03 22:11 ` [RFC v2 05/14] vfio/nvidia-vgpu: allocate vGPU FB memory " Zhi Wang
2025-09-03 22:11 ` [RFC v2 06/14] vfio/nvidia-vgpu: allocate mgmt heap " Zhi Wang
2025-09-03 22:11 ` [RFC v2 07/14] vfio/nvidia-vgpu: map mgmt heap when creating a vGPU Zhi Wang
2025-09-03 22:11 ` [RFC v2 08/14] vfio/nvidia-vgpu: allocate GSP RM client when creating vGPUs Zhi Wang
2025-09-03 22:11 ` [RFC v2 09/14] vfio/nvidia-vgpu: bootload the new vGPU Zhi Wang
2025-09-03 22:11 ` [RFC v2 10/14] vfio/nvidia-vgpu: introduce vGPU host RPC channel Zhi Wang
2025-09-03 22:36 ` Timur Tabi
2025-09-03 22:11 ` [RFC v2 11/14] vfio/nvidia-vgpu: introduce NVIDIA vGPU VFIO variant driver Zhi Wang
2025-09-03 22:11 ` [RFC v2 12/14] vfio/nvidia-vgpu: scrub the guest FB memory of a vGPU Zhi Wang
2025-09-03 22:11 ` [RFC v2 13/14] vfio/nvidia-vgpu: introduce vGPU logging Zhi Wang
2025-09-03 22:11 ` [RFC v2 14/14] vfio/nvidia-vgpu: add a kernel doc to introduce NVIDIA vGPU Zhi Wang
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=DCLNF4ADS21M.1WJ4YMSZ4RNF1@kernel.org \
--to=dakr@kernel.org \
--cc=acourbot@nvidia.com \
--cc=acurrid@nvidia.com \
--cc=airlied@gmail.com \
--cc=alex.williamson@redhat.com \
--cc=aniketa@nvidia.com \
--cc=ankita@nvidia.com \
--cc=apopple@nvidia.com \
--cc=cjia@nvidia.com \
--cc=daniel@ffwll.ch \
--cc=jgg@nvidia.com \
--cc=jhubbard@nvidia.com \
--cc=joelagnelf@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