El viernes, 11 de julio de 2025 21:11:15 (hora de verano de Europa central),
Adrian Larumbe escribió:
> Hi Caterina,
>
> On 03.07.2025 15:28, Caterina Shablia wrote:
> > From: Boris Brezillon
> >
> > Move the lock/flush_mem operations around the gpuvm_sm_map() calls so
> > we can implement true
Hi Caterina,
On 03.07.2025 15:28, Caterina Shablia wrote:
> From: Boris Brezillon
>
> Move the lock/flush_mem operations around the gpuvm_sm_map() calls so
> we can implement true atomic page updates, where any access in the
> locked range done by the GPU has to wait for the page table updates
>
patch link:
https://lore.kernel.org/r/20250703152908.16702-3-caterina.shablia%40collabora.com
patch subject: [PATCH v2 1/7] drm/panthor: Add support for atomic page table
updates
config: x86_64-buildonly-randconfig-004-20250704
(https://download.01.org/0day-ci/archive/20250704/20250704163
From: Boris Brezillon
Move the lock/flush_mem operations around the gpuvm_sm_map() calls so
we can implement true atomic page updates, where any access in the
locked range done by the GPU has to wait for the page table updates
to land before proceeding.
This is needed for vkQueueBindSparse(), so
From: Boris Brezillon
Move the lock/flush_mem operations around the gpuvm_sm_map() calls so
we can implement true atomic page updates, where any access in the
locked range done by the GPU has to wait for the page table updates
to land before proceeding.
This is needed for vkQueueBindSparse(), so
