Am 13.11.2017 um 03:53 schrieb Chunming Zhou:
Otherwise, they could be optimized by scheduled fence.
Change-Id: I6857eee20aebeaad793d9fe4e1b5222f1be7470e
Signed-off-by: Chunming Zhou
First of all patch is Reviewed-by: Christian König
.
Second do you remember why we did this? I have some br
keep consistency with threshold of swapout
Change-Id: I5b14d18ed7a9f79810cc50c023ac9e240bddf101
Signed-off-by: Roger He
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 11 ---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
b/drivers/g
+ Horace
-Original Message-
From: Colin King [mailto:colin.k...@canonical.com]
Sent: 2017年11月11日 19:51
To: Deucher, Alexander ; Koenig, Christian
; David Airlie ; Liu, Monk
; Yu, Xiangliang ;
amd-gfx@lists.freedesktop.org; dri-de...@lists.freedesktop.org
Cc: kernel-janit...@vger.kernel
Otherwise, they could be optimized by scheduled fence.
Change-Id: I6857eee20aebeaad793d9fe4e1b5222f1be7470e
Signed-off-by: Chunming Zhou
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
b/dr
NAK, you will find the performance drop with this change.
dep_sync should be used by CS dependency/syncobj, I thought which
already be handled.
I will make a patch for using dep_sync in a minute.
Thanks for reminder.
David Zhou
On 2017年11月04日 01:51, Andrey Grodzovsky wrote:
This reverts c
Deleted BOs with the same reservation object can be reaped even if they
can't be reserved.
v2: rebase and we still need to remove/add the BO from/to the LRU.
v3: fix remove/add one more time, cleanup the logic a bit
Signed-off-by: Christian König
---
drivers/gpu/drm/ttm/ttm_bo.c | 55 ++
Needed for the next patch.
v2: actually predicate all unlocks
Signed-off-by: Christian König
---
drivers/gpu/drm/ttm/ttm_bo.c | 59 +---
1 file changed, 33 insertions(+), 26 deletions(-)
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm
A few comments on the code:
+/* Validate bo size is bit bigger then the request domain */
+static inline bool amdgpu_bo_validate_bo_size(struct amdgpu_device *adev,
+ unsigned long size, u32 domain)
Drop the inline keyword and the second _bo_ in the name here.
+{
+ str
Am 10.11.2017 um 08:22 schrieb Chunming Zhou:
On 2017年11月09日 16:59, Christian König wrote:
Deleted BOs with the same reservation object can be reaped even if they
can't be reserved.
v2: rebase and we still need to remove/add the BO from/to the LRU.
Signed-off-by: Christian König
---
drive
Good catch and yeah that is actually the real problem why my original
patch didn't worked as expected.
Going to fix this in v2,
Christian.
Am 10.11.2017 um 10:55 schrieb He, Roger:
static int ttm_bo_cleanup_refs(struct ttm_buffer_object *bo,
bool interruptible, b
Am 10.11.2017 um 19:28 schrieb Piotr Redlewski:
When UVD bo is created, its size is based on the information from firmware
header (ucode_size_bytes). The same value should be be used when programming
UVD mc controller offsets, otherwise it can happen that
(mmUVD_VCPU_CACHE_OFFSET2 + mmUVD_VCPU_CA
Am 10.11.2017 um 19:29 schrieb Tom St Denis:
On 10/11/17 01:20 PM, Christian König wrote:
Am 10.11.2017 um 19:03 schrieb Tom St Denis:
The bottom two bits of the simd value were being put into
the upper bits of the wave value which was likely working due
to the bits being ignored (or aliased).
12 matches
Mail list logo
