dri-devel Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Christian König" <christian.koenig@amd.com>
To: Dave Airlie <airlied@gmail.com>, dri-devel@lists.freedesktop.org
Subject: Re: [PATCH 1/7] drm/ttm: move some move binds into the drivers
Date: Tue, 20 Oct 2020 12:20:27 +0200	[thread overview]
Message-ID: <54966de9-5cb0-43a3-9cb0-37cce05ec977@amd.com> (raw)
In-Reply-To: <20201020010319.1692445-2-airlied@gmail.com>

Am 20.10.20 um 03:03 schrieb Dave Airlie:
> From: Dave Airlie <airlied@redhat.com>
>
> This just gives the driver control over some of the bind paths.
>
> Signed-off-by: Dave Airlie <airlied@redhat.com>

Reviewed-by: Christian König <christian.koenig@amd.com>

> ---
>   drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c |  7 ++++++-
>   drivers/gpu/drm/nouveau/nouveau_bo.c    | 10 +++++++---
>   drivers/gpu/drm/radeon/radeon_ttm.c     | 11 ++++++++---
>   drivers/gpu/drm/ttm/ttm_bo_util.c       |  1 -
>   4 files changed, 21 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> index 91b20aa2294d..4af4891264e1 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> @@ -605,10 +605,15 @@ static int amdgpu_move_ram_vram(struct ttm_buffer_object *bo, bool evict,
>   	}
>   
>   	/* move/bind old memory to GTT space */
> -	r = ttm_bo_move_to_new_tt_mem(bo, ctx, &tmp_mem);
> +	r = ttm_tt_populate(bo->bdev, bo->ttm, ctx);
> +	if (unlikely(r))
> +		return r;
> +
> +	r = amdgpu_ttm_backend_bind(bo->bdev, bo->ttm, &tmp_mem);
>   	if (unlikely(r)) {
>   		goto out_cleanup;
>   	}
> +
>   	ttm_bo_assign_mem(bo, &tmp_mem);
>   	/* copy to VRAM */
>   	r = amdgpu_move_blit(bo, evict, new_mem, old_mem);
> diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
> index fec7a901865e..7e604340b780 100644
> --- a/drivers/gpu/drm/nouveau/nouveau_bo.c
> +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
> @@ -931,9 +931,13 @@ nouveau_bo_move_flips(struct ttm_buffer_object *bo, bool evict,
>   	if (ret)
>   		return ret;
>   
> -	ret = ttm_bo_move_to_new_tt_mem(bo, ctx, &tmp_reg);
> -	if (ret)
> -		goto out;
> +	ret = ttm_tt_populate(bo->bdev, bo->ttm, ctx);
> +	if (unlikely(ret != 0))
> +		return ret;
> +
> +	ret = nouveau_ttm_tt_bind(bo->bdev, bo->ttm, &tmp_reg);
> +	if (unlikely(ret != 0))
> +		return ret;
>   
>   	ttm_bo_assign_mem(bo, &tmp_reg);
>   	ret = nouveau_bo_move_m2mf(bo, true, ctx, new_reg);
> diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
> index ec713f0e55e5..24ef25665249 100644
> --- a/drivers/gpu/drm/radeon/radeon_ttm.c
> +++ b/drivers/gpu/drm/radeon/radeon_ttm.c
> @@ -279,10 +279,15 @@ static int radeon_move_ram_vram(struct ttm_buffer_object *bo,
>   	if (unlikely(r)) {
>   		return r;
>   	}
> -	r = ttm_bo_move_to_new_tt_mem(bo, ctx, &tmp_mem);
> -	if (unlikely(r)) {
> +
> +	r = ttm_tt_populate(bo->bdev, bo->ttm, ctx);
> +	if (unlikely(r))
>   		goto out_cleanup;
> -	}
> +
> +	r = radeon_ttm_tt_bind(bo->bdev, bo->ttm, &tmp_mem);
> +	if (unlikely(r))
> +		goto out_cleanup;
> +
>   	ttm_bo_assign_mem(bo, &tmp_mem);
>   	r = radeon_move_blit(bo, true, new_mem, old_mem);
>   	if (unlikely(r)) {
> diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
> index e4bd381a8962..c8ca6694cc1c 100644
> --- a/drivers/gpu/drm/ttm/ttm_bo_util.c
> +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
> @@ -64,7 +64,6 @@ int ttm_bo_move_to_new_tt_mem(struct ttm_buffer_object *bo,
>   
>   	return 0;
>   }
> -EXPORT_SYMBOL(ttm_bo_move_to_new_tt_mem);
>   
>   int ttm_bo_move_to_system(struct ttm_buffer_object *bo,
>   			  struct ttm_operation_ctx *ctx)

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

  reply	other threads:[~2020-10-20 10:20 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-20  1:03 [PATCH 0/7] drm/ttm: get rid of bind/unbind Dave Airlie
2020-10-20  1:03 ` [PATCH 1/7] drm/ttm: move some move binds into the drivers Dave Airlie
2020-10-20 10:20   ` Christian König [this message]
2020-10-20  1:03 ` [PATCH 2/7] drm/ttm: minor cleanup to move to system Dave Airlie
2020-10-20 10:22   ` Christian König
2020-10-20  1:03 ` [PATCH 3/7] drm/ttm: add move to system into drivers Dave Airlie
2020-10-20 12:03   ` Christian König
2020-10-20  1:03 ` [PATCH 4/7] drm/ttm: drop unbind callback Dave Airlie
2020-10-20 12:04   ` Christian König
2020-10-20  1:03 ` [PATCH 5/7] drm/ttm: remove move to new and inline into remainging place Dave Airlie
2020-10-20 12:06   ` Christian König
2020-10-20  1:03 ` [PATCH 6/7] drm/ttm: drop move notify around move Dave Airlie
2020-10-20 12:15   ` Christian König
2020-10-20  1:03 ` [PATCH 7/7] drm/ttm: move last binding into the drivers Dave Airlie
2020-10-20 12:16   ` Christian König

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=54966de9-5cb0-43a3-9cb0-37cce05ec977@amd.com \
    --to=christian.koenig@amd.com \
    --cc=airlied@gmail.com \
    --cc=dri-devel@lists.freedesktop.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