Michel Dänzer [Mon, 8 Aug 2016 07:23:39 +0000 (16:23 +0900)]
drm: Add DRM_MODE_PAGE_FLIP_TARGET_ABSOLUTE/RELATIVE flags v2
These flags allow userspace to explicitly specify the target vertical
blank period when a flip should take effect.
v2:
* Add new struct drm_mode_crtc_page_flip_target instead of modifying
struct drm_mode_crtc_page_flip, to make sure all existing userspace
code keeps compiling (Daniel Vetter)
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Michel Dänzer [Thu, 4 Aug 2016 03:39:40 +0000 (12:39 +0900)]
drm/radeon: Set MASTER_UPDATE_MODE to 0 again
With the previous change, it's safe to let page flips take effect
anytime during a vertical blank period.
This can avoid delaying a flip by a frame in some cases where we get to
radeon_flip_work_func -> adev->mode_info.funcs->page_flip during a
vertical blank period.
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Michel Dänzer [Thu, 4 Aug 2016 03:39:39 +0000 (12:39 +0900)]
drm/radeon: Provide page_flip_target hook
Now we can program a flip during a vertical blank period, if it's the
one targeted by the flip (or a later one). This allows simplifying
radeon_flip_work_func considerably.
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Michel Dänzer [Thu, 4 Aug 2016 03:39:38 +0000 (12:39 +0900)]
drm/amdgpu: Set MASTER_UPDATE_MODE to 0 again
With the previous change, it's safe to let page flips take effect
anytime during a vertical blank period.
This can avoid delaying a flip by a frame in some cases where we get to
amdgpu_flip_work_func -> adev->mode_info.funcs->page_flip during a
vertical blank period.
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Michel Dänzer [Thu, 4 Aug 2016 03:39:37 +0000 (12:39 +0900)]
drm/amdgpu: Provide page_flip_target hook
Now we can program a flip during a vertical blank period, if it's the
one targeted by the flip (or a later one). This allows simplifying
amdgpu_flip_work_func considerably.
agd: update dce_virtual.c as well.
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Michel Dänzer [Mon, 8 Aug 2016 07:23:03 +0000 (16:23 +0900)]
drm: Add page_flip_target CRTC hook v2
Mostly the same as the existing page_flip hook, but takes an additional
parameter specifying the target vertical blank period when the flip
should take effect.
v2:
* Add curly braces around else statement corresponding to an if block
with curly braces (Alex Deucher)
* Call drm_crtc_vblank_put in the error case (Daniel Vetter)
* Clarify entry point documentation comment (Daniel Vetter)
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Wolfram Sang [Tue, 9 Aug 2016 11:30:30 +0000 (13:30 +0200)]
gpu: drm: radeon: radeon_i2c: don't print error when adding adapter fails
The core will do this for us now.
Signed-off-by: Wolfram Sang <wsa-dev@sang-engineering.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Wolfram Sang [Tue, 9 Aug 2016 11:30:27 +0000 (13:30 +0200)]
gpu: drm: amd: amdgpu: amdgpu_i2c: don't print error when adding adapter fails
The core will do this for us now.
Signed-off-by: Wolfram Sang <wsa-dev@sang-engineering.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Mon, 8 Aug 2016 12:40:18 +0000 (14:40 +0200)]
drm/amdgpu: use more than 64KB fragment size if possible
We align to 64KB, but when userspace aligns even more we can easily use more.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Fri, 5 Aug 2016 11:56:35 +0000 (13:56 +0200)]
drm/amdgpu: flip frag_ptes and update_pts
We can add the fragment params before we split the update for the page tables.
That should save a few CPU cycles for larger updates.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Thu, 4 Aug 2016 13:02:49 +0000 (15:02 +0200)]
drm/amdgpu: add adev to the pte_update_params
No need to carry that forward as a separate parameter.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Thu, 4 Aug 2016 12:52:50 +0000 (14:52 +0200)]
drm/amdgpu: rename amdgpu_vm_update_params
Well those are actually page table entry parameters.
This also makes the variable names used a bit shorter.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Wed, 3 Aug 2016 15:46:42 +0000 (17:46 +0200)]
drm/amdgpu: cleanup VM fragment defines
We can actually do way more than just the 64KB we currently used as default.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Wed, 3 Aug 2016 15:39:26 +0000 (17:39 +0200)]
drm/amdgpu: remove unused VM defines
Not used for a long time.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Tue, 9 Aug 2016 14:13:21 +0000 (10:13 -0400)]
drm/amd/amdgpu: Simplify bitfield operations in gfx v8
This patch introduces a new macro WREG32_FIELD which is used
to write to a register with a new value in a field. It's designed
to replace the pattern:
tmp = RREG32(mmFoo);
tmp &= ~REG__FIELD_MASK;
tmp |= new_value << REG__FIELD__SHIFT;
WREG32(mmFoo, tmp)
with:
WREG32_FIELD(Foo, FIELD, new_value);
Unlike WREG32_P() it understands offsets/masks and doesn't
require the caller to shift the value (or mask properly).
It's applied where suitable in the gfx_v8_0.c driver to start
with.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Tue, 9 Aug 2016 14:10:30 +0000 (10:10 -0400)]
drm/amd/amdgpu: Simplify various gfx v8 functions
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Tue, 9 Aug 2016 14:07:36 +0000 (10:07 -0400)]
drm/amd/amdgpu: Correct whitespace in GFX v8
Fix various whitespace issues in gfx v8 driver.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Wed, 10 Aug 2016 08:01:25 +0000 (16:01 +0800)]
drm/amdgpu: Change the virtual_display type from int to char*.
For virtual display feature, as there may be multiple GPUs,
for user could choose whiche GPU need to enable this feature, change
the type of virtual_display from int to char*. The variable will be set
like this virtual_display="xxxx:xx:xx.x;xxxx:xx:xx.x;".
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Mon, 8 Aug 2016 18:45:29 +0000 (14:45 -0400)]
drm/amdgpu: add virtual dce support for iceland
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Mon, 8 Aug 2016 18:40:04 +0000 (14:40 -0400)]
drm/amdgpu/virtual_dce: add case for topaz for disable_dce
This asic has no DCE block. Also clarify the error message
for unmatched chips.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Mon, 8 Aug 2016 18:35:55 +0000 (14:35 -0400)]
drm/amdgpu: move vsync_timer_enabled setup to dce virtual early_init
Put it in one place.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:37:50 +0000 (11:37 +0800)]
drm/amdgpu: Set ip_blocks according variable amdgpu_virtual_display.
For virtual display feature, if user set the option "amdgpu.virtual_display=1"
when load amdgpu.ko. Then need to set the ip_blocks with virtual display ip
blocks. And when enable virtual display, the amdgpu_dal need to be set to zero.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:37:29 +0000 (11:37 +0800)]
drm/amdgpu: Define one variable for virtual display.
For virtual display feature, define on variable in amdgpu.ko. When want to
enable virtual display feature, need set the option "amdgpu.virtual_display=1".
And then disable vga render and crtc if have DCE engine.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:36:45 +0000 (11:36 +0800)]
drm/amdgpu: Define virtual display ip blocks.
For virtual display feature, define virtual display ip blocks, and set
dce_virtual_ip_funcs to DCE block.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:36:31 +0000 (11:36 +0800)]
drm/amdgpu: Add DRM_MODE_CONNECTOR_VIRTUAL connector in amdgpu_connector_add.
For virtual display feature, add one connector type in amdgpu_connector_add.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:36:07 +0000 (11:36 +0800)]
drm/amdgpu: Call pageflip irq funtion when receiced vsync interrupt.
For virtual display feature, as there is no dce engine, so no pageflip irq
generated. So directly call pageflip irq funtion when received vysn interrupt.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:35:39 +0000 (11:35 +0800)]
drm/amdgpu: Use software timer to generate vsync interrupt.
For virtual display feature, use the software timer to
simulate the vsync interrupt.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:33:11 +0000 (11:33 +0800)]
drm/amdgpu: Disable VGA render and crtc when init GMC.
For virtual display feature, when the GPU has DCE engine, need to disable
the VGA render and CRTC, or it will hang when initialize GMC.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:32:35 +0000 (11:32 +0800)]
drm/amdgpu: To define whether the GPU has DCE engine.
For virtual display feature, when the GPU has DCE engine, need to disable
the VGA render and CRTC, or it will hang when initialize GMC. So first detect
whether the GPU has DCE engine.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:32:22 +0000 (11:32 +0800)]
drm/amdgpu: Initialize dce_virtual_crtc_funcs. (v2)
For virtual display feature, initialize dce_virtual_crtc_funcs.
v2: agd: rebase on upstream
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:32:00 +0000 (11:32 +0800)]
drm/amdgpu: Initialize dce_virtual_crtc_helper_funcs
For virtual display feature, initialize dce_virtual_crtc_helper_funcs.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:31:37 +0000 (11:31 +0800)]
drm/amdgpu: Initialize crtc, pageflip irq funcs (v2)
For virtual display feature, initialize dce_virtual_crtc_irq_funcs,
dce_virtual_pageflip_irq_funcs. As it has no dce engine, so the
pageflip interrupt won't be generated, and the vsync interrupt will
be generated by smu's periodic timer or software timer which will
be implemented later.
v2: agd: rebase on upstream
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:31:13 +0000 (11:31 +0800)]
drm/amdgpu: Initialize dce_virtual_display_funcs.
For virtual display feature, initialize dce_virtual_display_funcs,
which will be used in function dce_virtual_set_display_funcs.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:30:50 +0000 (11:30 +0800)]
drm/amdgpu: Initialize dce_virtual_ip_funcs
For virtual display feature, first need to initialize dce_virtual_ip_funcs,
which will be used when set ip blocks.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Emily Deng [Mon, 8 Aug 2016 03:30:34 +0000 (11:30 +0800)]
drm/amdgpu: Add virtual connector and encoder macros.
For virtual display feature, add virtual connector and encoder macros.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lyude [Sat, 6 Aug 2016 00:30:37 +0000 (20:30 -0400)]
drm/amdgpu: Don't retry 7 times in amdgpu_atombios_dp_get_dpcd()
When this code was written, we didn't retry DP aux transactions on any
error, which required retrying important transactions like this in
individual drivers. Since that's no longer the case, retrying here is
not necessary. As well, we retry any aux transaction on any error 32
times. 7 * 32 = 224, which means this loop causes us to retry grabbing
the dpcd 224 times. This is definitely far more then we actually need to
do.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Lyude <cpaul@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lyude [Sat, 6 Aug 2016 00:30:36 +0000 (20:30 -0400)]
drm/amdgpu: Don't print error on aux transaction timeouts
Since it's normal for DRM to retry our aux transaction helpers multiple
times in a row, up to 32 times for each attempted transaction, we're
making a lot of noise that is no longer necessary now that DRM will just
print the return code we give it.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Lyude <cpaul@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lyude [Sat, 6 Aug 2016 00:30:35 +0000 (20:30 -0400)]
drm/radeon: Don't retry 7 times in radeon_dp_dpcd()
When this code was written, we didn't retry DP aux transactions on any
error, which required retrying important transactions like this in
individual drivers. Since that's no longer the case, retrying here is
not necessary. As well, we retry any aux transaction on any error 32
times. 7 * 32 = 224, which means this loop causes us to retry grabbing
the dpcd 224 times. This is definitely far more then we actually need to
do.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Lyude <cpaul@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lyude [Sat, 6 Aug 2016 00:30:34 +0000 (20:30 -0400)]
drm/radeon: Don't print error on aux transaction timeouts
Since it's normal for DRM to retry our aux transaction helpers multiple
times in a row, up to 32 times for each attempted transaction, we're
making a lot of noise that is no longer necessary now that DRM will just
print the return code we give it.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Lyude <cpaul@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Michel Dänzer [Mon, 8 Aug 2016 03:28:26 +0000 (12:28 +0900)]
drm/ttm: Remove unused parameter evict from ttm_bo_move_memcpy
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Michel Dänzer [Mon, 8 Aug 2016 03:28:25 +0000 (12:28 +0900)]
drm/ttm: Remove unused parameter evict from ttm_bo_move_ttm
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Michel Dänzer [Fri, 5 Aug 2016 09:36:10 +0000 (18:36 +0900)]
drm/ttm: Wait for a BO to become idle before unbinding it from GTT
Fixes hangs under memory pressure, e.g. running the piglit test
tex3d-maxsize concurrently with other tests.
Fixes: 17d33bc9d6ef ("drm/ttm: drop waiting for idle in ttm_bo_evict.")
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Thu, 31 Mar 2016 22:18:28 +0000 (18:18 -0400)]
drm/amdgpu: move all Kconfig options to amdgpu/Kconfig
For consistency.
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Mon, 18 Apr 2016 15:29:32 +0000 (23:29 +0800)]
drm/amd/powerplay: enable powerplay as default on CZ/ST
Enable powerplay as default on Carrizo and Stoney. And it can be
disabled with amdgpu.powerplay=0.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Huang Rui <ray.huang@amd.com>
Cc: Rex Zhu <Rex.Zhu@amd.com>
Cc: Flora Cui <Flora.Cui@amd.com>
Cc: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rex Zhu [Mon, 1 Aug 2016 02:49:00 +0000 (10:49 +0800)]
drm/amd/powerplay: change structure variable name.
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rex Zhu [Wed, 3 Aug 2016 06:37:19 +0000 (14:37 +0800)]
drm/amd/powerplay: delete code set/unset deep_sleep/power_containment.
they were controled by module parameter.
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Chunming Zhou [Wed, 3 Aug 2016 07:15:37 +0000 (15:15 +0800)]
drm/amdgpu: enable VCE soft reset
Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rex Zhu [Wed, 3 Aug 2016 07:10:28 +0000 (15:10 +0800)]
drm/amdgpu: remove RB1_BUSY bit checking
This is a workaround to let VCE soft reset work.
RB1_BUSY bit is always set, so remove its checking now, and we
will depend on RB0_BUSY currently.
After we find the root cause of RB1_BUSY, we can add it back.
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Chunming Zhou [Wed, 3 Aug 2016 06:55:46 +0000 (14:55 +0800)]
drm/amdgpu: fix harvest config checking in vce3 check_soft_reset
Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Chunming Zhou [Wed, 27 Jul 2016 05:15:20 +0000 (13:15 +0800)]
drm/amdgpu: irq resume should be immediately after gpu resume
Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Wed, 3 Aug 2016 16:37:23 +0000 (12:37 -0400)]
drm/amd/amdgpu: Partially revert change to UVD v3 CG
Partially undo changes made by commit:
drm/amd/amdgpu: don't track state in UVD clockgating
To keep bypass even if CG flags are not set.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Wed, 3 Aug 2016 15:34:35 +0000 (11:34 -0400)]
drm/amd/amdgpu: Enable VCE MGCG for Stoney
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Wed, 3 Aug 2016 14:16:17 +0000 (10:16 -0400)]
drm/amd/amdgpu: Enable VCE MGCG for Carrizo
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Wed, 3 Aug 2016 14:15:48 +0000 (10:15 -0400)]
drm/amd/amdgpu: Don't set VCE CG in startup
It's handled by DPM/PP properly.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Wed, 3 Aug 2016 14:15:24 +0000 (10:15 -0400)]
drm/amd/amdgpu: Fix VCE CG order and resume defaults
CG was being enabled in reverse sense from dpm/powerplay.
Also fix the default CLK_EN signal to enable all of the blocks.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Wed, 3 Aug 2016 14:14:44 +0000 (10:14 -0400)]
drm/amd/amdgpu: fix indentation in vce3 CG
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Leo Liu [Wed, 3 Aug 2016 13:25:59 +0000 (09:25 -0400)]
drm/amdgpu: remove the check for sessions being closed
This will make clock and power gated when no block decoded, for example
when paused during the playback.
Signed-off-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Marek Olšák [Fri, 29 Jul 2016 22:48:39 +0000 (00:48 +0200)]
drm/amdgpu: print more accurate error messages on IB submission failure
It's useful for debugging.
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Ken Wang [Wed, 27 Jul 2016 11:18:01 +0000 (19:18 +0800)]
drm/amdgpu: use the kernel zone memory size as the max remote memory in amdgpu
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Ken Wang [Wed, 27 Jul 2016 11:12:15 +0000 (19:12 +0800)]
drm/ttm: Add interface to export kernel_zone max memory size in ttm
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rex Zhu [Sun, 31 Jul 2016 05:46:01 +0000 (13:46 +0800)]
drm/amd/powrplay: delete code no longer in use on Polaris.
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rex Zhu [Fri, 29 Jul 2016 11:26:36 +0000 (19:26 +0800)]
drm/amd/powrplay: workaround Memory EDC Error for certain partner boards.
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rex Zhu [Fri, 29 Jul 2016 11:19:26 +0000 (19:19 +0800)]
drm/amdgpu: AMD SW workaround for certain partner boards
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rex Zhu [Fri, 29 Jul 2016 11:15:52 +0000 (19:15 +0800)]
drm/amd/powerplay: pass sub_device_id and sub_vendor_id to powerplay.
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rex Zhu [Fri, 29 Jul 2016 08:30:04 +0000 (16:30 +0800)]
drm/amd/powerplay: delete useless code in hwmgr.
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Flora Cui [Wed, 20 Jul 2016 06:44:38 +0000 (14:44 +0800)]
drm/amdgpu: expose AMDGPU_GEM_CREATE_VRAM_CLEARED to user space
V2: fix the return value for fill failure and validate bo before
filling data
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Flora Cui [Tue, 19 Jul 2016 08:48:22 +0000 (16:48 +0800)]
drm/amdgpu: implement amdgpu_fill_buffer()
so that bo could be set to some pattern
Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Mon, 1 Aug 2016 18:29:15 +0000 (14:29 -0400)]
drm/amdgpu/gfx8: remove stale function declaration
This got leftover somehow when I cleaned this up.
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Eric Huang [Thu, 28 Jul 2016 21:25:01 +0000 (17:25 -0400)]
drm/amd/amdgpu: change pptable output format from ASCII to binary
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Eric Huang <JinHuiEric.Huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Fri, 29 Jul 2016 22:14:39 +0000 (18:14 -0400)]
drm/amdgpu/ci: add mullins to default case for smc ucode
It's already covered by the default case, but add it for
consistency.
Reviewed-by: Alexandre Demers <alexandre.f.demers@gmail.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Fri, 29 Jul 2016 22:03:42 +0000 (18:03 -0400)]
drm/amdgpu/gmc7: add missing mullins case
Looks like this got missed when we ported the code from radeon.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Rex Zhu [Thu, 28 Jul 2016 09:36:35 +0000 (17:36 +0800)]
drm/amdgpu: use modules parameter to ctrl deep sleep feature in dpm
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rex Zhu [Thu, 28 Jul 2016 08:51:47 +0000 (16:51 +0800)]
drm/amdgpu: add module parameters to ctrl powerplay feature
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rex Zhu [Thu, 7 Jul 2016 11:39:15 +0000 (19:39 +0800)]
drm/amd/powerplay: fix typos of volt/millivolt symbols in comment.
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alexandre Demers <alexandre.f.demers@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:38:45 +0000 (09:38 -0400)]
drm/amd/amdgpu: Enable stoney VCE PG
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:38:29 +0000 (09:38 -0400)]
drm/amd/amdgpu: Enable stoney UVD PG
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:38:13 +0000 (09:38 -0400)]
drm/amd/amdgpu: Enable stoney GFX PG
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:36:26 +0000 (09:36 -0400)]
drm/amd/amdgpu: Enable carrizo VCE PG
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:35:57 +0000 (09:35 -0400)]
drm/amd/amdgpu: Enable carrizo UVD PG
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:33:56 +0000 (09:33 -0400)]
drm/amd/amdgpu: Enable carrizo GFX PG
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:46:00 +0000 (09:46 -0400)]
drm/amd/amdgpu: Add error messages to UVD PG in DPM
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:45:07 +0000 (09:45 -0400)]
drm/amd/amdgpu: enable PG_EN bit in powergating UVD
Enable the PG_EN bit just before the SMU would be tasked
with the PG transition.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:44:49 +0000 (09:44 -0400)]
drm/amd/amdgpu: don't track state in UVD clockgating
There's no need to track CG state anymore.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:44:14 +0000 (09:44 -0400)]
drm/amd/amdgpu: don't set clockgating in uvd_v6_0_start()
This is handled properly by both DPM and PP externally.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:39:22 +0000 (09:39 -0400)]
drm/amd/amdgpu: add pm lock to debugfs mmio entry
Adds support for PM locks around access to registers that might
have race conditions on PG transistions.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom St Denis [Thu, 28 Jul 2016 13:40:07 +0000 (09:40 -0400)]
drm/amd/amdgpu: add mutex locking for both DPM and PP based powergating for UVD/VCE
This adds a mutex lock for both DPM/PP around the changes in
power gating state so that userspace can poll registers without
a race condition on power state.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Thu, 28 Jul 2016 17:35:42 +0000 (13:35 -0400)]
drm/amdgpu/powerplay: enable powerplay by default on TOPAZ
Now that the implementation is complete.
Acked-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Tue, 26 Jul 2016 08:38:36 +0000 (16:38 +0800)]
drm/amd/powerplay: add DPM running checking back
This patch adds DPM running checking back, because the DPM issue is
fixed.
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Tue, 26 Jul 2016 08:33:41 +0000 (16:33 +0800)]
drm/amd/powerplay: add enabling voltage controller back
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Tue, 26 Jul 2016 08:25:57 +0000 (16:25 +0800)]
drm/amd/powerplay: set the platform capability flags for iceland
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Tue, 26 Jul 2016 07:57:07 +0000 (15:57 +0800)]
drm/amd/powerplay: add deep sleep initialization
This patch adds the deep sleep initialization at DPM, it needs send a
message to SMC to enable this feature before enable voltage controller.
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Tue, 26 Jul 2016 08:42:11 +0000 (16:42 +0800)]
drm/amd/powerplay: fix the incorrect checking condition
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Mon, 20 Jun 2016 05:35:28 +0000 (13:35 +0800)]
drm/amdgpu: enable iceland powerplay manually
It's able to enable iceland powerplay manually via the module
parameter. The default state is disabled.
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Tue, 12 Jul 2016 07:45:12 +0000 (15:45 +0800)]
drm/amd/powerplay: add iceland HW manager
This patch introduces the iceland HW manager of powerplay which
includes HW manager, clockpowergating, thermal, and powertune.
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Mon, 11 Jul 2016 16:52:24 +0000 (00:52 +0800)]
drm/amd/powerplay: add iceland_dyn_defaults header
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Sun, 19 Jun 2016 16:42:01 +0000 (00:42 +0800)]
drm/amd/powerplay: add iceland SMU mananger
The system management unit (SMU) is a subcomponent of the northbridge
that is responsible for a variety of system and power management tasks
during boot and runtime for GPU. In powerplay, it will be used on
firmware loading and power task management. This patch adds SMU
mananger for iceland.
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Huang Rui [Tue, 21 Jun 2016 02:26:26 +0000 (10:26 +0800)]
drm/amd/powerplay: add SMU71 header files for iceland (v2)
v2: cleanup headers, add copyright
Signed-off-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Tue, 26 Jul 2016 07:58:45 +0000 (09:58 +0200)]
drm/amdgpu: use amdgpu_bo_create_kernel in amdgpu_ring.c
Saves us quite a bunch of code.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Mon, 25 Jul 2016 16:06:30 +0000 (18:06 +0200)]
drm/amdgpu: use amdgpu_bo_create_kernel in amdgpu_ih.c
Saves us quite a bunch of code.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Mon, 25 Jul 2016 15:37:38 +0000 (17:37 +0200)]
drm/amdgpu: user amdgpu_bo_create_kernel for the UVD BO
Saves us some code.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>