On Tue, Jul 26, 2016 at 11:38 AM, Tom St Denis wrote:
> Because of the ip_blocks init order powerplay would power down
> the UVD block before UVD start is called. This results in a VCPU
> hang.
>
> This patch prevents power down before UVD is initialized.
>
> Also correct the power up order so cl
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Christian König
> Sent: Tuesday, July 26, 2016 4:00 AM
> To: amd-gfx@lists.freedesktop.org
> Subject: [PATCH 6/6] drm/amdgpu: use amdgpu_bo_create_kernel in
> amdgpu_ring.c
>
> From: Christia
Reviewed-by: Alexandre Demers
Alexandre Demers
On 2016-07-27 16:44, Alex Deucher wrote:
The message is harmless and confusing. On PX systems,
there is one ATIF method, but potentially multiple GPUs
leading to an error on the GPU with no backlight control.
Bug: https://bugzilla.kernel.org/sho
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Christian König
> Sent: Monday, July 25, 2016 10:44 AM
> To: amd-gfx@lists.freedesktop.org
> Subject: [PATCH] drm/amdgpu: enable UVD VM only on polaris
>
> From: Christian König
>
> Stoney
We're also looking into a hang with a KFD unit test that allocates lots
of memory and fragments it deliberately, without mapping it all at once.
It's a new problem for us as we're rebasing on amd-staging-4.6.
Something weird seems to be happening with evictions, but I haven't been
able to figure it
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Christian König
> Sent: Friday, July 22, 2016 9:10 AM
> To: amd-gfx@lists.freedesktop.org; dri-de...@lists.freedesktop.org
> Cc: Lin, Amber; Kuehling, Felix; funfunc...@folklore1984.net
> Subj
On Wed, Jul 27, 2016 at 2:36 PM, Slava Grigorev wrote:
> Signed-off-by: Slava Grigorev
Reviewed-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/ci_dpm.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
> b/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
Ok I'll split them up and re-send when I have the VCE content more stable.
Tom
From: Alex Deucher
Sent: Wednesday, July 27, 2016 16:16
To: Tom St Denis
Cc: amd-gfx list; StDenis, Tom
Subject: Re: [PATCH 3/4] drm/amd/amdgpu: Add GFX and UVD PG for Stoney
On Wed
On Mon, Jul 18, 2016 at 11:41 AM, Lyude wrote:
> Just about all of amdgpu's connector probing functions try to acquire
> runtime PM refs. If we try to do this in the context of
> amdgpu_resume_kms by calling drm_helper_hpd_irq_event(), we end up
> deadlocking the system.
>
> Since we're guaranteed
The message is harmless and confusing. On PX systems,
there is one ATIF method, but potentially multiple GPUs
leading to an error on the GPU with no backlight control.
Bug: https://bugzilla.kernel.org/show_bug.cgi?id=115011
Signed-off-by: Alex Deucher
Cc: Hans de Goede
---
drivers/gpu/drm/rade
Just early engineering samples are rev0.
Alex
On Tue, Jul 26, 2016 at 2:49 PM, StDenis, Tom wrote:
> I don't have rev0 hardware. I can edit the comment but I can't test the
> switch to the logic.
>
>
> Is there even rev0 hardware in retail?
>
>
> Tom
>
>
> --
> *Fro
On Wed, Jul 27, 2016 at 12:21 PM, Tom St Denis wrote:
> Enables GFX PG and UVD static PG on Stoney class
> asics.
>
> Signed-off-by: Tom St Denis
> ---
> drivers/gpu/drm/amd/amdgpu/vi.c | 6 ++
> 1 file changed, 6 insertions(+)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/d
If we do it at enable time, it's too late for the feature
checks.
Signed-off-by: Alex Deucher
Cc: Peter Wu
---
drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
b/drivers/gpu/d
If we do it at enable time, it's too late for the feature
checks.
bug: https://bugzilla.kernel.org/show_bug.cgi?id=115321
Signed-off-by: Alex Deucher
Cc: Peter Wu
---
drivers/gpu/drm/radeon/radeon_atpx_handler.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/d
Some of the checks didn't handle frev 2 tables properly.
Signed-off-by: Alex Deucher
Cc: sta...@vger.kernel.org
---
drivers/gpu/drm/radeon/radeon_atombios.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c
b/drivers/gpu/drm/radeon/
Some of the checks didn't handle frev 2 tables properly.
amdgpu doesn't support any tables pre-frev 2, so drop
the checks.
Signed-off-by: Alex Deucher
Cc: sta...@vger.kernel.org
---
drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c | 25 -
1 file changed, 8 insertions(+), 17 d
On Fri, Jul 15, 2016 at 2:31 PM, Alex Deucher wrote:
> Hi Dave,
>
> A few more patches for 4.8. Mostly bug fixes and some prep work
> for iceland powerplay support. I have a couple polaris patches and
> Edward's misc cleanups that require a merge with Linus'. I don't know
> if you are planning
Signed-off-by: Slava Grigorev
---
drivers/gpu/drm/amd/amdgpu/ci_dpm.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
b/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
index 573bc97..5d4ec41 100644
--- a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
+++ b/drivers/gpu/drm/amd/
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Christian König
> Sent: Thursday, July 21, 2016 6:01 AM
> To: amd-gfx@lists.freedesktop.org
> Subject: [PATCH 7/7] drm/amdgpu: add a fence timeout for the IB tests v2
>
> From: Christian Köni
Enables GFX PG and UVD static PG on Stoney class
asics.
Signed-off-by: Tom St Denis
---
drivers/gpu/drm/amd/amdgpu/vi.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 0786c68bc25e..3af4f204810f 100644
--- a/driver
This patch removes the enable_clock_power_gatings_tasks
task from the initialize sequence which would have powered
off the UVD (and other) blocks before init.
The patch also fixes the power up order to set the clockgating
after ungating power.
Finally, the patch enables UVD PG with powerplay enab
Enables GFX PG and UVD static PG on Carrizo class
asics.
Signed-off-by: Tom St Denis
---
drivers/gpu/drm/amd/amdgpu/vi.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 9ba64989f092..0786c68bc25e 100644
--- a/dr
This stream changes a comment in patch #2 and re-works the init
sequence for powerplay in patch #4 by removing the
enable_clock_power_gatings_tasks task from the initialize task list.
In theory that should be better as now other blocks won't be powered
off before init.
___
Move the CG state into the adev.uvd structure to allow
for parallel UVD instances. Also prevent uvd start if
in the PG state (will be enhanced later to support
powerplay). Also avoid setting CG state in start
since DPM will call set_clockgating_state anyways.
Adds static PG to the DPM implementa
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Rex Zhu
> Sent: Monday, July 18, 2016 1:00 PM
> To: amd-gfx@lists.freedesktop.org
> Cc: Zhu, Rex
> Subject: [PATCH 1/3] drm/amdgpu: add bypass mode for vce3.0
>
> Change-Id: I68aa5431146b21
Change-Id: I9f30b54365492b234a7f0887dd0c67a9817c3705
Signed-off-by: Ken Wang
---
drivers/gpu/drm/ttm/ttm_memory.c | 6 ++
include/drm/ttm/ttm_memory.h | 1 +
2 files changed, 7 insertions(+)
diff --git a/drivers/gpu/drm/ttm/ttm_memory.c b/drivers/gpu/drm/ttm/ttm_memory.c
index a1803fb..2
Change-Id: Ibf193cc2d9e20c3aefa1ce8ff24241dfbb6768ff
Signed-off-by: Ken Wang
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 12 +++-
drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c | 7 ++-
drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 7 ++-
27 matches
Mail list logo
