From: "Christian König" <ckoenig.leichtzumerken-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: "Zeng, Oak" <Oak.Zeng-5C7GfCeVMHo@public.gmane.org>,
"amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org"
<amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org>
Cc: "Kuehling, Felix" <Felix.Kuehling-5C7GfCeVMHo@public.gmane.org>,
"Keely, Sean" <Sean.Keely-5C7GfCeVMHo@public.gmane.org>
Subject: Re: [PATCH 2/8] drm/amdgpu: Remove GWS barriers pre-reservation for gfx
Date: Fri, 10 May 2019 19:41:53 +0200 [thread overview]
Message-ID: <03cdbff1-c757-4831-01f0-354e390a55a1@gmail.com> (raw)
In-Reply-To: <1557504063-1559-2-git-send-email-Oak.Zeng-5C7GfCeVMHo@public.gmane.org>
Am 10.05.19 um 18:01 schrieb Zeng, Oak:
> User space allocated GWS barriers are used. The kernel
> pre-allocated GWS barriers are unused.
I have a patch which goes a step further and also removes the GDS and OA
stuff as well and also cleans up the internal variables.
Give me a moment to dig that up,
Christian.
>
> Change-Id: I7aac259d1f6b7064d02aff231279ff605c29ea34
> Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 2 +-
> drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h | 3 +--
> drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 6 ------
> drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c | 1 -
> drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 1 -
> drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 1 -
> 6 files changed, 2 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
> index 5c79da8..273c16b 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
> @@ -82,7 +82,7 @@ int amdgpu_bo_list_create(struct amdgpu_device *adev, struct drm_file *filp,
>
> kref_init(&list->refcount);
> list->gds_obj = adev->gds.gds_gfx_bo;
> - list->gws_obj = adev->gds.gws_gfx_bo;
> + list->gws_obj = NULL;
> list->oa_obj = adev->gds.oa_gfx_bo;
>
> array = amdgpu_bo_list_array_entry(list, 0);
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h
> index f89f573..7e5b4ba 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h
> @@ -39,13 +39,12 @@ struct amdgpu_gds {
> struct amdgpu_gds_asic_info oa;
> uint32_t gds_compute_max_wave_id;
>
> - /* At present, GDS, GWS and OA resources for gfx (graphics)
> + /* At present, GDS and OA resources for gfx (graphics)
> * is always pre-allocated and available for graphics operation.
> * Such resource is shared between all gfx clients.
> * TODO: move this operation to user space
> * */
> struct amdgpu_bo* gds_gfx_bo;
> - struct amdgpu_bo* gws_gfx_bo;
> struct amdgpu_bo* oa_gfx_bo;
> };
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> index 38ce11e..b0e51e5 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> @@ -1791,12 +1791,6 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
> return r;
> }
>
> - r = amdgpu_bo_create_kernel(adev, adev->gds.gws.gfx_partition_size,
> - 1, AMDGPU_GEM_DOMAIN_GWS,
> - &adev->gds.gws_gfx_bo, NULL, NULL);
> - if (r)
> - return r;
> -
> r = ttm_bo_init_mm(&adev->mman.bdev, AMDGPU_PL_OA,
> adev->gds.oa.total_size);
> if (r) {
> diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
> index a59e0fd..4d73456 100644
> --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
> @@ -4497,7 +4497,6 @@ static int gfx_v7_0_sw_fini(void *handle)
> struct amdgpu_device *adev = (struct amdgpu_device *)handle;
>
> amdgpu_bo_free_kernel(&adev->gds.oa_gfx_bo, NULL, NULL);
> - amdgpu_bo_free_kernel(&adev->gds.gws_gfx_bo, NULL, NULL);
> amdgpu_bo_free_kernel(&adev->gds.gds_gfx_bo, NULL, NULL);
>
> for (i = 0; i < adev->gfx.num_gfx_rings; i++)
> diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
> index 02955e6..e255754 100644
> --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
> @@ -2061,7 +2061,6 @@ static int gfx_v8_0_sw_fini(void *handle)
> struct amdgpu_device *adev = (struct amdgpu_device *)handle;
>
> amdgpu_bo_free_kernel(&adev->gds.oa_gfx_bo, NULL, NULL);
> - amdgpu_bo_free_kernel(&adev->gds.gws_gfx_bo, NULL, NULL);
> amdgpu_bo_free_kernel(&adev->gds.gds_gfx_bo, NULL, NULL);
>
> for (i = 0; i < adev->gfx.num_gfx_rings; i++)
> diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> index 7d7d287..41d1fc3 100644
> --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
> @@ -1785,7 +1785,6 @@ static int gfx_v9_0_sw_fini(void *handle)
> }
>
> amdgpu_bo_free_kernel(&adev->gds.oa_gfx_bo, NULL, NULL);
> - amdgpu_bo_free_kernel(&adev->gds.gws_gfx_bo, NULL, NULL);
> amdgpu_bo_free_kernel(&adev->gds.gds_gfx_bo, NULL, NULL);
>
> for (i = 0; i < adev->gfx.num_gfx_rings; i++)
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
next prev parent reply other threads:[~2019-05-10 17:41 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-10 16:01 [PATCH 1/8] drm/amdkfd: Add gws number to kfd topology node properties Zeng, Oak
[not found] ` <1557504063-1559-1-git-send-email-Oak.Zeng-5C7GfCeVMHo@public.gmane.org>
2019-05-10 16:01 ` [PATCH 2/8] drm/amdgpu: Remove GWS barriers pre-reservation for gfx Zeng, Oak
[not found] ` <1557504063-1559-2-git-send-email-Oak.Zeng-5C7GfCeVMHo@public.gmane.org>
2019-05-10 17:41 ` Christian König [this message]
2019-05-10 16:01 ` [PATCH 3/8] drm/amdkfd: Add interface to alloc gws from amdgpu Zeng, Oak
2019-05-10 16:01 ` [PATCH 4/8] drm/amdkfd: Allocate gws on device initialization Zeng, Oak
2019-05-10 16:01 ` [PATCH 5/8] drm/amdkfd: Add function to set queue gws Zeng, Oak
2019-05-10 16:01 ` [PATCH 6/8] drm/amdkfd: Add function to add/remove gws to kfd process Zeng, Oak
[not found] ` <1557504063-1559-6-git-send-email-Oak.Zeng-5C7GfCeVMHo@public.gmane.org>
2019-05-10 20:17 ` Kuehling, Felix
2019-05-10 16:01 ` [PATCH 7/8] drm/amdkfd: New IOCTL to allocate queue GWS Zeng, Oak
[not found] ` <1557504063-1559-7-git-send-email-Oak.Zeng-5C7GfCeVMHo@public.gmane.org>
2019-05-10 20:28 ` Kuehling, Felix
[not found] ` <c76522c2-115e-a6b6-f136-44fa2a45be2b-5C7GfCeVMHo@public.gmane.org>
2019-05-13 16:03 ` Zeng, Oak
[not found] ` <BL0PR12MB258003620CDCE779957C845B800F0-b4cIHhjg/p/XzH18dTCKOgdYzm3356FpvxpqHgZTriW3zl9H0oFU5g@public.gmane.org>
2019-05-13 18:12 ` Kuehling, Felix
[not found] ` <83257525-4d46-acbd-5853-eeb6913fa398-5C7GfCeVMHo@public.gmane.org>
2019-05-13 19:14 ` Christian König
2019-05-10 16:01 ` [PATCH 8/8] drm/amdkfd: PM4 packets change to support GWS Zeng, Oak
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=03cdbff1-c757-4831-01f0-354e390a55a1@gmail.com \
--to=ckoenig.leichtzumerken-re5jqeeqqe8avxtiumwx3w@public.gmane.org \
--cc=Felix.Kuehling-5C7GfCeVMHo@public.gmane.org \
--cc=Oak.Zeng-5C7GfCeVMHo@public.gmane.org \
--cc=Sean.Keely-5C7GfCeVMHo@public.gmane.org \
--cc=amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org \
--cc=christian.koenig-5C7GfCeVMHo@public.gmane.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