[amdgpu][CIK] drm-next-4.9-wip - piglit test - max-texture-size - GPU deadlock

2016-09-22 Thread Shawn Starr
When I try to run piglit, I get a GPU deadlock running max-texture-size, Test: spec/!opengl 1.1/max-texture-size I will begin bisecting kernel driver, but if anyone has some tips on likely commits I can test further. Thanks, Shawn ___ amd-gfx mailin

Re: [PATCH 2/2] drm/amdgpu: bind GTT on demand

2016-09-22 Thread Andy Furniss
Marek Olšák wrote: This breaks Tonga such that it hangs. Reproducible quickly with: R600_DEBUG=testdma glxgears It's a randomized test that runs forever. It should hang within 2 seconds. So what is the status of this now? R600_DEBUG=testdma glxgears isn't a test I've run on my r9285 tonga un

[pull] radeon and amdgpu drm-next-4.9

2016-09-22 Thread Alex Deucher
Hi Dave, Last set of radeon and amdgpu changes for 4.9. This is mostly just the powerplay cleanup for dGPUs. Beyond that, just misc code cleanups and bug fixes. The following changes since commit af1f85ddecfa341e684db950c34a1813d36750db: drm/ttm: remove cpu_address member from ttm_tt (2016-0

[PATCH] drm/amdgpu: Constify tables

2016-09-22 Thread Nils Wallménius
Mark some powerplay tables as 'const' and adjust pointers acessing them to avoid introducing warnings. Signed-off-by: Nils Wallménius --- drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c | 2 +- drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c | 12 ++-- drivers/gpu/drm/amd/po

[PATCH 2/2] drm/radeon: always apply pci shutdown callbacks

2016-09-22 Thread Alex Deucher
We can't properly detect all hypervisors and we need this to properly tear down the hardware. Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/radeon_drv.c | 7 --- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/rade

[PATCH 1/2] drm/amdgpu: always apply pci shutdown callbacks

2016-09-22 Thread Alex Deucher
We can't properly detect all hypervisors and we need this to properly tear down the hardware. Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 7 --- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/

Re: [PATCH 3/3] drm/amd/amdgpu: Various cleanups for DCEv6

2016-09-22 Thread Alex Deucher
On Thu, Sep 22, 2016 at 12:43 PM, Tom St Denis wrote: > Signed-off-by: Tom St Denis Series is: Reviewed-by: Alex Deucher > --- > drivers/gpu/drm/amd/amdgpu/dce_v6_0.c | 46 > --- > 1 file changed, 10 insertions(+), 36 deletions(-) > > diff --git a/drivers/gpu/d

[PATCH 3/3] drm/amd/amdgpu: Various cleanups for DCEv6

2016-09-22 Thread Tom St Denis
Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/dce_v6_0.c | 46 --- 1 file changed, 10 insertions(+), 36 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c index dce5414f4ae3..a4fa30b9924a 100644 --

[PATCH 2/3] drm/amd/amdgpu: Clean up afmt allocation in DCEv6.

2016-09-22 Thread Tom St Denis
Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/dce_v6_0.c | 15 --- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c index e41d583740b5..dce5414f4ae3 100644 --- a/drivers/gpu/drm/a

[PATCH 1/3] drm/amd/amdgpu: Remove division from vblank_wait

2016-09-22 Thread Tom St Denis
Removes division from vblank_wait for DCE v6, v8, and v10. Was already implemented for v11. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/dce_v10_0.c | 8 +--- drivers/gpu/drm/amd/amdgpu/dce_v6_0.c | 8 +--- drivers/gpu/drm/amd/amdgpu/dce_v8_0.c | 8 +--- 3 files chan

Remainder of small cleanups for DCEv6

2016-09-22 Thread Tom St Denis
Small cleanups to sync up DCEv6 with the style of the others ___ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-22 Thread Daniel Vetter
On Wed, Sep 21, 2016 at 06:23:35PM +0200, Christian König wrote: > For a quick workaround I suggest to just serialize all accesses to BO shared > with different drivers, but essentially I think it is a perfectly valid > requirement to have multiple writers to one BO. It is, but it's not possible w

Re: [PATCH] drm/amdgpu: free userptrs even if GTT isn't bound

2016-09-22 Thread Alex Deucher
On Thu, Sep 22, 2016 at 8:53 AM, Christian König wrote: > From: Christian König > > This fixes a memory leak since binding GTT only on demand. > > Signed-off-by: Christian König Reviewed-by: Alex Deucher > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 6 +++--- > 1 file changed, 3 insertio

Re: [PATCH] drm/amdgpu/vce: allow the clock table packet

2016-09-22 Thread Christian König
Am 21.09.2016 um 22:36 schrieb Alex Deucher: This packet allows the user mode driver to specify the required performance for specific use cases. Signed-off-by: Alex Deucher Reviewed-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c | 1 + 1 file changed, 1 insertion(+) di

Re: [PATCH] drm/amdgpu: bump version for new vce packet support

2016-09-22 Thread Christian König
Am 22.09.2016 um 00:07 schrieb Alex Deucher: Signed-off-by: Alex Deucher Reviewed-by: Christian König . --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/a

[PATCH] drm/amdgpu: bump version for new vce packet support

2016-09-22 Thread Alex Deucher
Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 0c5f36d..e7ae6723 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_d

RE: [PATCH] Revert "drm/amdgpu:correct smc fw version error"

2016-09-22 Thread Deucher, Alexander
> -Original Message- > From: Huang Rui [mailto:ray.hu...@amd.com] > Sent: Thursday, September 22, 2016 4:01 AM > To: amd-gfx@lists.freedesktop.org; Deucher, Alexander > Cc: Huang, Ray; Liu, Monk > Subject: [PATCH] Revert "drm/amdgpu:correct smc fw version error" > > From: Huang Rui > > T

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-22 Thread Daniel Vetter
On Thu, Sep 22, 2016 at 12:55 PM, Christian König wrote: > Am 22.09.2016 um 08:36 schrieb Daniel Vetter: >> >> On Wed, Sep 21, 2016 at 06:23:35PM +0200, Christian König wrote: >>> >>> For a quick workaround I suggest to just serialize all accesses to BO >>> shared >>> with different drivers, but e

[PATCH] Revert "drm/amdgpu:correct smc fw version error"

2016-09-22 Thread Huang Rui
From: Huang Rui This reverts commit bb2a2de6e2ea0e99d0d1a20d06536c4dcd66a196. SMC firmware version error should be fixed while ucode is generated. I would do the fix in firmware side. Signed-off-by: Huang Rui Cc: Monk Liu --- drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c | 10 +++--- 1 file ch

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-22 Thread Daniel Vetter
On Thu, Sep 22, 2016 at 2:44 PM, Christian König wrote: >> - explicit fencing: Userspace passes around distinct fence objects for >> any work going on on the gpu. The kernel doesn't insert any stall of >> it's own (except for moving buffer objects around ofc). This is what >> Android. This also se

[PATCH] drm/amdgpu: free userptrs even if GTT isn't bound

2016-09-22 Thread Christian König
From: Christian König This fixes a memory leak since binding GTT only on demand. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/a

Re: [PATCH] drm/amdgpu: fix amdgpu_vm_bo_update param error

2016-09-22 Thread Christian König
Am 22.09.2016 um 05:41 schrieb Flora Cui: Change-Id: I86dc7d3fc4d6197bd13563a420718a8412ce9dbc Signed-off-by: Flora Cui Good catch. That clearly results in some additional overhead if not worse otherwise. Patch is Reviewed-by: Christian König . Regards, Christian. --- drivers/gpu/drm/a

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-22 Thread Christian König
Am 22.09.2016 um 08:36 schrieb Daniel Vetter: On Wed, Sep 21, 2016 at 06:23:35PM +0200, Christian König wrote: For a quick workaround I suggest to just serialize all accesses to BO shared with different drivers, but essentially I think it is a perfectly valid requirement to have multiple writers

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-22 Thread Christian König
Am 22.09.2016 um 14:26 schrieb Daniel Vetter: On Thu, Sep 22, 2016 at 12:55 PM, Christian König wrote: Am 22.09.2016 um 08:36 schrieb Daniel Vetter: On Wed, Sep 21, 2016 at 06:23:35PM +0200, Christian König wrote: For a quick workaround I suggest to just serialize all accesses to BO shared wi

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-22 Thread Christian König
Am 22.09.2016 um 15:05 schrieb Daniel Vetter: On Thu, Sep 22, 2016 at 2:44 PM, Christian König wrote: - explicit fencing: Userspace passes around distinct fence objects for any work going on on the gpu. The kernel doesn't insert any stall of it's own (except for moving buffer objects around ofc

Re: [PATCH] drm/i915: Before pageflip, also wait for shared dmabuf fences.

2016-09-22 Thread Daniel Vetter
On Thu, Sep 22, 2016 at 12:07:24AM +0900, Michel Dänzer wrote: > On 21/09/16 09:56 PM, Daniel Vetter wrote: > > On Wed, Sep 21, 2016 at 1:19 PM, Christian König > > wrote: > >> Am 21.09.2016 um 13:04 schrieb Daniel Vetter: > >>> On Wed, Sep 21, 2016 at 12:30 PM, Christian König > >>> wrote: > >>>

[PATCH] drm/amdgpu: fix amdgpu_vm_bo_update param error

2016-09-22 Thread Flora Cui
Change-Id: I86dc7d3fc4d6197bd13563a420718a8412ce9dbc Signed-off-by: Flora Cui --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 941416a..a7ea9a3 1