All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] drm/amdgpu : Use XGMI mapping when devices on the same hive v3
@ 2018-11-15 19:23 Russell, Kent
       [not found] ` <1542309797-20009-1-git-send-email-kent.russell-5C7GfCeVMHo@public.gmane.org>
  0 siblings, 1 reply; 3+ messages in thread
From: Russell, Kent @ 2018-11-15 19:23 UTC (permalink / raw)
  To: amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW; +Cc: Russell, Kent, Liu, Shaoyun

From: Shaoyun Liu <Shaoyun.Liu@amd.com>

VM mapping will only fall back to P2P if XGMI mapping is not available

V2: Rebase onto 4.20
V3: Remove PCIe-related things

Change-Id: I7a854ab3d5c9958bd45d4fe439ea7e370a092e7a
Signed-off-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Kent Russell <kent.russell@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index dad0e23..5aa60bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2011,6 +2011,8 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 	struct drm_mm_node *nodes;
 	struct dma_fence *exclusive, **last_update;
 	uint64_t flags;
+	uint64_t vram_base_offset = adev->vm_manager.vram_base_offset;
+	struct amdgpu_device *bo_adev;
 	int r;
 
 	if (clear || !bo) {
@@ -2029,9 +2031,15 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 		exclusive = reservation_object_get_excl(bo->tbo.resv);
 	}
 
-	if (bo)
+	if (bo) {
 		flags = amdgpu_ttm_tt_pte_flags(adev, bo->tbo.ttm, mem);
-	else
+		bo_adev = amdgpu_ttm_adev(bo->tbo.bdev);
+		if (adev->gmc.xgmi.hive_id &&
+		    adev->gmc.xgmi.hive_id == bo_adev->gmc.xgmi.hive_id)
+			vram_base_offset = bo_adev->vm_manager.vram_base_offset;
+		else
+			return -EINVAL;
+	} else
 		flags = 0x0;
 
 	if (clear || (bo && bo->tbo.resv == vm->root.base.bo->tbo.resv))
-- 
2.7.4

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

^ permalink raw reply related	[flat|nested] 3+ messages in thread
* RE: [PATCH] drm/amdgpu : Use XGMI mapping when devices on the same hive v3
       [not found] ` <1542309797-20009-1-git-send-email-kent.russell-5C7GfCeVMHo@public.gmane.org>
@ 2018-11-15 19:51   ` Kuehling, Felix
       [not found]     ` <DM5PR12MB170766293DA1DCFC17E06B0792DC0-2J9CzHegvk9TCtO+SvGBKwdYzm3356FpvxpqHgZTriW3zl9H0oFU5g@public.gmane.org>
  0 siblings, 1 reply; 3+ messages in thread
From: Kuehling, Felix @ 2018-11-15 19:51 UTC (permalink / raw)
  To: amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW; +Cc: Russell, Kent, Liu, Shaoyun

Sorry, something is still missing here. The new variable vram_base_offset isn't used anywhere. We have some other changes in amd-kfd-staging to use that vram_base_offset that are probably missing on amd-staging-drm-next. This change won't have any effect as is.

Regards,
  Felix

-----Original Message-----
From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> On Behalf Of Russell, Kent
Sent: Thursday, November 15, 2018 2:23 PM
To: amd-gfx@lists.freedesktop.org
Cc: Russell, Kent <Kent.Russell@amd.com>; Liu, Shaoyun <Shaoyun.Liu@amd.com>
Subject: [PATCH] drm/amdgpu : Use XGMI mapping when devices on the same hive v3

From: Shaoyun Liu <Shaoyun.Liu@amd.com>

VM mapping will only fall back to P2P if XGMI mapping is not available

V2: Rebase onto 4.20
V3: Remove PCIe-related things

Change-Id: I7a854ab3d5c9958bd45d4fe439ea7e370a092e7a
Signed-off-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Kent Russell <kent.russell@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index dad0e23..5aa60bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2011,6 +2011,8 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 	struct drm_mm_node *nodes;
 	struct dma_fence *exclusive, **last_update;
 	uint64_t flags;
+	uint64_t vram_base_offset = adev->vm_manager.vram_base_offset;
+	struct amdgpu_device *bo_adev;
 	int r;
 
 	if (clear || !bo) {
@@ -2029,9 +2031,15 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 		exclusive = reservation_object_get_excl(bo->tbo.resv);
 	}
 
-	if (bo)
+	if (bo) {
 		flags = amdgpu_ttm_tt_pte_flags(adev, bo->tbo.ttm, mem);
-	else
+		bo_adev = amdgpu_ttm_adev(bo->tbo.bdev);
+		if (adev->gmc.xgmi.hive_id &&
+		    adev->gmc.xgmi.hive_id == bo_adev->gmc.xgmi.hive_id)
+			vram_base_offset = bo_adev->vm_manager.vram_base_offset;
+		else
+			return -EINVAL;
+	} else
 		flags = 0x0;
 
 	if (clear || (bo && bo->tbo.resv == vm->root.base.bo->tbo.resv))
-- 
2.7.4

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

^ permalink raw reply related	[flat|nested] 3+ messages in thread
* RE: [PATCH] drm/amdgpu : Use XGMI mapping when devices on the same hive v3
       [not found]     ` <DM5PR12MB170766293DA1DCFC17E06B0792DC0-2J9CzHegvk9TCtO+SvGBKwdYzm3356FpvxpqHgZTriW3zl9H0oFU5g@public.gmane.org>
@ 2018-11-15 20:21       ` Russell, Kent
  0 siblings, 0 replies; 3+ messages in thread
From: Russell, Kent @ 2018-11-15 20:21 UTC (permalink / raw)
  To: Kuehling, Felix, amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW; +Cc: Liu, Shaoyun

Not a problem, I just saw the one missing patch when trying to rebase our KFD branch. I'll just put it on there instead. 

 Kent

-----Original Message-----
From: Kuehling, Felix 
Sent: Thursday, November 15, 2018 2:52 PM
To: Russell, Kent <Kent.Russell@amd.com>; amd-gfx@lists.freedesktop.org
Cc: Russell, Kent <Kent.Russell@amd.com>; Liu, Shaoyun <Shaoyun.Liu@amd.com>
Subject: RE: [PATCH] drm/amdgpu : Use XGMI mapping when devices on the same hive v3

Sorry, something is still missing here. The new variable vram_base_offset isn't used anywhere. We have some other changes in amd-kfd-staging to use that vram_base_offset that are probably missing on amd-staging-drm-next. This change won't have any effect as is.

Regards,
  Felix

-----Original Message-----
From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> On Behalf Of Russell, Kent
Sent: Thursday, November 15, 2018 2:23 PM
To: amd-gfx@lists.freedesktop.org
Cc: Russell, Kent <Kent.Russell@amd.com>; Liu, Shaoyun <Shaoyun.Liu@amd.com>
Subject: [PATCH] drm/amdgpu : Use XGMI mapping when devices on the same hive v3

From: Shaoyun Liu <Shaoyun.Liu@amd.com>

VM mapping will only fall back to P2P if XGMI mapping is not available

V2: Rebase onto 4.20
V3: Remove PCIe-related things

Change-Id: I7a854ab3d5c9958bd45d4fe439ea7e370a092e7a
Signed-off-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Kent Russell <kent.russell@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index dad0e23..5aa60bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2011,6 +2011,8 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 	struct drm_mm_node *nodes;
 	struct dma_fence *exclusive, **last_update;
 	uint64_t flags;
+	uint64_t vram_base_offset = adev->vm_manager.vram_base_offset;
+	struct amdgpu_device *bo_adev;
 	int r;
 
 	if (clear || !bo) {
@@ -2029,9 +2031,15 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 		exclusive = reservation_object_get_excl(bo->tbo.resv);
 	}
 
-	if (bo)
+	if (bo) {
 		flags = amdgpu_ttm_tt_pte_flags(adev, bo->tbo.ttm, mem);
-	else
+		bo_adev = amdgpu_ttm_adev(bo->tbo.bdev);
+		if (adev->gmc.xgmi.hive_id &&
+		    adev->gmc.xgmi.hive_id == bo_adev->gmc.xgmi.hive_id)
+			vram_base_offset = bo_adev->vm_manager.vram_base_offset;
+		else
+			return -EINVAL;
+	} else
 		flags = 0x0;
 
 	if (clear || (bo && bo->tbo.resv == vm->root.base.bo->tbo.resv))
-- 
2.7.4

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

^ permalink raw reply related	[flat|nested] 3+ messages in thread
end of thread, other threads:[~2018-11-15 20:21 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-11-15 19:23 [PATCH] drm/amdgpu : Use XGMI mapping when devices on the same hive v3 Russell, Kent
     [not found] ` <1542309797-20009-1-git-send-email-kent.russell-5C7GfCeVMHo@public.gmane.org>
2018-11-15 19:51   ` Kuehling, Felix
     [not found]     ` <DM5PR12MB170766293DA1DCFC17E06B0792DC0-2J9CzHegvk9TCtO+SvGBKwdYzm3356FpvxpqHgZTriW3zl9H0oFU5g@public.gmane.org>
2018-11-15 20:21       ` Russell, Kent

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.