- 07 Jul, 2016 40 commits
-
-
Alexandre Demers authored
Signed-off-by:
Alexandre Demers <alexandre.f.demers@gmail.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Alexandre Demers authored
Signed-off-by:
-
Frank Binns authored
This flag was being set unconditionally at runtime so just set it at compile time instead. Reviewed-by:
Frank Binns <frank.binns@imgtec.com> Signed-off-by:
-
Michel Dänzer authored
If reservation_object_get_fences_rcu failed, we'd previously go directly to the cleanup label, so we'd leave the BO pinned. While we're at it, remove two amdgpu_bo_unreserve calls in favour of two new labels. Reviewed-by:
Christian König <christian.koenig@amd.com> Signed-off-by:
Michel Dänzer <michel.daenzer@amd.com> Signed-off-by:
-
Alex Deucher authored
Reviewed-by:
-
Alex Deucher authored
Reviewed-by:
-
Junwei Zhang authored
Fixed mc stop and resume hardware programming sequence. Signed-off-by:
Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by:
-
Alex Deucher authored
Skip IP blocks that are not enabled. Reviewed-by:
Chunming zhou <david1.zhou@amd.com> Signed-off-by:
-
Alex Deucher authored
Saves a few CPU cycles. Reviewed-by:
-
Alex Deucher authored
Helpers to to call the IP functions for the selected IP. Reviewed-by:
-
Alex Deucher authored
Not enabled yet, but missing the call. Signed-off-by: -
Alex Deucher authored
Use UPDATE_IMMEDIATE (update on next data request boundary) rather than UPDATE_H_RETRACE (update on next line boundary). The data request boundary is less than a scanline, so it update will happen sooner. Cc: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: -
Christian König authored
Now that we can pipeline evictions we need to wait for them to finish when we cleanup a memory domain. Signed-off-by:
-
Christian König authored
It's just overhead to do so and allocating a VMID when we don't need one is actually a bit dangerous. Signed-off-by:
-
Christian König authored
Flush and invalidate the HDP caches. v2: fix typo in comment Signed-off-by:
-
Huang Rui authored
This patch changes pcie_gen_cap magic code to macro to make it more readable. Reviewed-by:
Rex Zhu <Rex.Zhu@amd.com> Signed-off-by:
Huang Rui <ray.huang@amd.com> Cc: Eric Huang <JinHuiEric.Huang@amd.com> Cc: Ken Wang <Qingqing.Wang@amd.com> Signed-off-by:
-
Alex Deucher authored
Not used outside of gfx7. Reviewed-by:
-
Alex Deucher authored
gfx8 already uses them. Remove the direct exports and use the callbacks fpr gfx7. Reviewed-by:
-
Alex Deucher authored
It's gfx IP specific, not asic specific, so move to a gfx callback. Reviewed-by:
-
Alex Deucher authored
It's gfx IP specific, not asic specific, so move to a gfx callback. Reviewed-by:
-
Alex Deucher authored
Only used in the gmc IP modules so just call the local function directly. Reviewed-by:
-
Alex Deucher authored
These are not used outside of the respective gmc ip modules. Reviewed-by:
-
Christian König authored
We don't need to validate them again if the eviction counter didn't changed. Reviewed-by:
-
Christian König authored
Keep track of the number of evictions since boot. Reviewed-by:
-
Chunming Zhou authored
irq need to update when gpu reset happens. Signed-off-by:
-
Arindam Nath authored
This patch is a port of similar patch for amdgpu when PP is disabled. Since the code flow is little different when PP is enabled, we need to make sure the patch is applied for PP enabled path as well. With the current code, when we boot with the amdgpu driver enabled and loaded, the VCE also automatically remains enabled since bootup. This can be verified from the output of amdgpu_pm_info. It does not matter whether we boot into command line directly or into X, the VCE stays enabled the entire time. This patch addresses the issue and makes sure that VCE is turned on only during playback, and remains disabled otherwise. Reviewed-by:
Arindam Nath <arindam.nath@amd.com> Signed-off-by:
-
Arindam Nath authored
With the current code, when we boot with the amdgpu driver enabled and loaded, the VCE also automatically remains enabled since bootup. This can be verified from the output of amdgpu_pm_info. It does not matter whether we boot into command line directly or into X, the VCE stays enabled the entire time. This patch addresses the issue and makes sure that VCE is turned on only during playback, and remains disaled otherwise. Reviewed-by:
-
Alex Deucher authored
No need to stall the pipe when we are using firmware with the fix. Reviewed-by:
-
Nicolai Hähnle authored
v2: do not overwrite register when bitmap is zero Reviewed-by:
Nicolai Hähnle <nicolai.haehnle@amd.com> Signed-off-by:
-
Nicolai Hähnle authored
v2: do not overwrite register when bitmap is zero Reviewed-by:
-
Nicolai Hähnle authored
This parameter will allow disabling individual CUs on module load, e.g. amdgpu.disable_cu=2.0.3,2.0.4 to disable CUs 3 and 4 of SE2. Reviewed-by:
-
Chunming Zhou authored
On workstation cards with ECC vram, the entirety of vram is cleared to 0 on asic init to set the ECC status correctly. On non ECC boards, I don't think they do any explicit clearing, but the vram controller is reset which may cause issues with the data there. Signed-off-by:
-
Chunming Zhou authored
Signed-off-by:
-
Arnd Bergmann authored
In the AMD powerplay driver, a pointer is checked for validity by comparing against an integer '0', which causes a harmless warning when building with "make W=1": drivers/gpu/drm/amd/amdgpu/../powerplay/hwmgr/processpptables.c:1502:16: error: ordered comparison of pointer with integer zero [-Werror=extra] This changes the code to the more conventional "if (pointer)" check. Reviewed-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
Christian König authored
This boosts Xonotic from 38fps to 47fps when artificially limiting VRAM to 256MB for testing. It should improve all CPU bound rendering situations where we have a lot of swapping to/from VRAM. Reviewed-by:
-
Christian König authored
When we pipeline evictions the page directory could already be moving somewhere else when grab_id is called. Reviewed-by:
-
Christian König authored
Free up the memory immediately, remember the last eviction for each domain and make new allocations depend on the last eviction to be completed. Reviewed-by:
-
Christian König authored
As far as I can see no need for a custom implementation any more. Reviewed-by:
-
Christian König authored
Instead of using the flag just remember the fence of the last move operation. This avoids waiting for command submissions pipelined after the move, but before accessing the BO with the CPU again. Reviewed-by:
-
Christian König authored
It isn't used and not waiting for the GPU after scheduling a move is actually quite dangerous. Reviewed-by:
-
