Re: [PATCH v2 0/5] gpu: nova-core: Hopper/Blackwell prerequisites

[John Hubbard <jhubbard@nvidia.com>]

On 11/25/25 5:39 PM, John Hubbard wrote:
> Hi,
> 
> I'm posting this now, instead of with the upcoming full Hopper/Blackwell
> series, because this includes some HAL improvements here that will
> undoubtedly interact "a little bit" with Timur Tabi's Turing support
> patchset [1].
> 

Thanks to everyone (Lyude, Alex C.) who looked at this so far. I'm 
planning to merge it into the upcoming Hopper/Blackwell firmware 
series, due to the timing of things: it no longer helps to have
these prerequisites all by themselves.

thanks,
-- 
John Hubbard

> Changes in v2:
> 
> 0) Rebased on top of today's drm-rust-next.
> 
> 1) Use a new FbRange newtype, in order to clean up the implementation of
> printing ranges with sizes. (Thanks to Alex Courbot's suggestion.)
> 
> 2) Combined the Hopper and Blackwell HAL additions, into a single patch,
> as recommended by Timur Tabi.
> 
> 3) Used a separate patch to change the "use" lines to vertical format.
> That patch is already a commit in drm-rust-next, so this series builds
> on top of that.
> 
> 4) Dropped one patch entirely, because the boot0/boot42 commits fixed
> up everything already.
> 
> As implied above, this is based on top of today's drm-rust-next. There
> is also a branch, for convenience in reviewing:
> 
> https://github.com/johnhubbard/linux/tree/nova-core-blackwell-prereqs-v2
> 
> [1] https://lore.kernel.org/20251114233045.2512853-1-ttabi@nvidia.com
> 
> ************************************************************************
> Here's the original cover letter, edited slightly (removed the link
> to the v0 series, to avoid anyone going there by mistake):
> 
> I've based these Hopper/Blackwell prerequisites on top of Joel's and
> Alex's changes, and also on top of my recent boot0/boot42 changes.
> 
> This makes it easier for both Timur Tabi to post his Turing support
> (which he's about ready to do), and for me to post the actual
> Hopper/Blackwell support, without generating conflicts.
> 
> Testing: This works as expected on Ampere and Blackwell (bare metal),
> on my local test machine.
> 
> John Hubbard (5):
>   gpu: nova-core: print FB sizes, along with ranges
>   gpu: nova-core: add FbRange.len() and use it in boot.rs
>   gpu: nova-core: Hopper/Blackwell: basic GPU identification
>   nova-core: factor .fwsignature* selection into a new
>     get_gsp_sigs_section()
>   gpu: nova-core: use GPU Architecture to simplify HAL selections
> 
>  drivers/gpu/nova-core/falcon/hal.rs   | 19 ++++---
>  drivers/gpu/nova-core/fb.rs           | 72 ++++++++++++++++++++-------
>  drivers/gpu/nova-core/fb/hal.rs       | 18 +++----
>  drivers/gpu/nova-core/firmware/gsp.rs | 30 +++++++++--
>  drivers/gpu/nova-core/gpu.rs          | 22 ++++++++
>  drivers/gpu/nova-core/gsp/boot.rs     |  2 +-
>  6 files changed, 125 insertions(+), 38 deletions(-)
> 
> 
> base-commit: 57dc2ea0b7bdb828c5d966d9135c28fe854933a4