[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAA+WOBvVasy2wRP_wmP-R6Q8y5B4sN08jNYfHuDVjiWXV+m23Q@mail.gmail.com>
Date: Wed, 22 Oct 2025 13:16:47 +0300
From: Mohamed Ahmed <mohamedahmedegypt2001@...il.com>
To: linux-kernel@...r.kernel.org
Cc: dri-devel@...ts.freedesktop.org, Mary Guillemard <mary@...y.zone>,
Faith Ekstrand <faith.ekstrand@...labora.com>, Lyude Paul <lyude@...hat.com>,
Danilo Krummrich <dakr@...nel.org>, Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>,
Maxime Ripard <mripard@...nel.org>, Thomas Zimmermann <tzimmermann@...e.de>,
David Airlie <airlied@...il.com>, Simona Vetter <simona@...ll.ch>, nouveau@...ts.freedesktop.org
Subject: Re: [PATCH 2/5] drm/nouveau/uvmm: Allow larger pages
Hello,
Pinging again re: review and also was asking if we can revert the
select_page_shift() handling back to v1 behavior with a fall-back
path, as it looks like there are some cases where
nouveau_bo_fixup_align() isn't enough;
https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36450#note_3159199.
Thanks!
On Fri, Oct 10, 2025 at 2:39 AM Mohamed Ahmed
<mohamedahmedegypt2001@...il.com> wrote:
>
> From: Mary Guillemard <mary@...y.zone>
>
> Now that everything in UVMM knows about the variable page shift, we can
> select larger values.
>
> The proposed approach relies on nouveau_bo::page unless if it would cause
> alignment issues (in which case we fall back to searching for an
> appropriate shift)
>
> Signed-off-by: Mary Guillemard <mary@...y.zone>
> Co-developed-by: Mohamed Ahmed <mohamedahmedegypt2001@...il.com>
> Signed-off-by: Mohamed Ahmed <mohamedahmedegypt2001@...il.com>
> ---
> drivers/gpu/drm/nouveau/nouveau_uvmm.c | 29 ++++++++++++++++++++++++--
> 1 file changed, 27 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/nouveau/nouveau_uvmm.c b/drivers/gpu/drm/nouveau/nouveau_uvmm.c
> index 2cd0835b05e8..26edc60a530b 100644
> --- a/drivers/gpu/drm/nouveau/nouveau_uvmm.c
> +++ b/drivers/gpu/drm/nouveau/nouveau_uvmm.c
> @@ -454,6 +454,31 @@ op_unmap_prepare_unwind(struct drm_gpuva *va)
> drm_gpuva_insert(va->vm, va);
> }
>
> +static bool
> +op_map_aligned_to_page_shift(const struct drm_gpuva_op_map *op, u8 page_shift)
> +{
> + u64 page_size = 1ULL << page_shift;
> +
> + return op->va.addr % page_size == 0 && op->va.range % page_size == 0 &&
> + op->gem.offset % page_size == 0;
> +}
> +
> +static u8
> +select_page_shift(struct nouveau_uvmm *uvmm, struct drm_gpuva_op_map *op)
> +{
> + struct nouveau_bo *nvbo = nouveau_gem_object(op->gem.obj);
> +
> + /* nouveau_bo_fixup_align() guarantees for us that the page size will be aligned
> + * but just in case, make sure that it is aligned.
> + */
> + if (op_map_aligned_to_page_shift(op, nvbo->page))
> + return nvbo->page;
> +
> + /* This should never happen, but raise a warning and return 4K if we get here. */
> + WARN_ON(1);
> + return PAGE_SHIFT;
> +}
> +
> static void
> nouveau_uvmm_sm_prepare_unwind(struct nouveau_uvmm *uvmm,
> struct nouveau_uvma_prealloc *new,
> @@ -506,7 +531,7 @@ nouveau_uvmm_sm_prepare_unwind(struct nouveau_uvmm *uvmm,
> if (vmm_get_range)
> nouveau_uvmm_vmm_put(uvmm, vmm_get_start,
> vmm_get_range,
> - PAGE_SHIFT);
> + select_page_shift(uvmm, &op->map));
> break;
> }
> case DRM_GPUVA_OP_REMAP: {
> @@ -599,7 +624,7 @@ op_map_prepare(struct nouveau_uvmm *uvmm,
>
> uvma->region = args->region;
> uvma->kind = args->kind;
> - uvma->page_shift = PAGE_SHIFT;
> + uvma->page_shift = select_page_shift(uvmm, op);
>
> drm_gpuva_map(&uvmm->base, &uvma->va, op);
>
> --
> 2.51.0
>
Powered by blists - more mailing lists