- 14 Mar, 2018 40 commits
-
-
Michel Dänzer authored
Turning off the sink in this case causes various issues, because userspace expects it to stay on until it turns it off explicitly. Instead, turn the sink off and back on when a display is connected again. This dance seems necessary for link training to work correctly. Bugzilla: https://bugs.freedesktop.org/105308 Cc: stable@vger.kernel.org Reviewed-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
Michel Dänzer <michel.daenzer@amd.com> Signed-off-by:
-
Colin Ian King authored
The pointer internal_buf is assigned a value but the pointer is never read, hence it is redundant and can be removed. Cleans up clang warning: drivers/gpu/drm/amd/amdgpu/../powerplay/smumgr/smu7_smumgr.c:630:2: warning: Value stored to 'internal_buf' is never read Reviewed-by:
Rex Zhu <Rex.Zhu@amd.com> Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Alex Deucher authored
No need to replicate it in several places. Reviewed-by:
Rex Zhu <rezhu@amd.com> Acked-by:
Christian König <christian.koenig@amd.com> Signed-off-by:
-
Alex Deucher authored
Regardless of whether the user has selected psp fw loading or not. It's still needed for GPU reset among other things. There are already guards in place to avoid setting up the full psp if PSP fw loading is not enabled. Reviewed-by:
-
Alex Deucher authored
The per device firmware load method is limited to what makes sense for that asic rather than whatever arbitrary value may have been set by the user. Reviewed-by:
-
Alex Deucher authored
No need to replicate it in several places. Reviewed-by:
-
Alex Deucher authored
We already checked and returned early in most of the IP functions, fill in the rest as well. Reviewed-by:
-
Shirish S authored
This patch adds an essential check related to the size of the payload to be transferred via aux channel. Without this check dal_ddc_service_read_dpcd_data() is fed with inappropriate payload size leading to deadlocks. Signed-off-by:
Shirish S <shirish.s@amd.com> Reviewed-by:
Harry Wentland <harry.wentland@amd.com> Signed-off-by:
-
Dmytro Laktyushkin authored
We have unused variables being populated when notifying pplib. This change amends that. Signed-off-by:
Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com> Reviewed-by:
Tony Cheng <Tony.Cheng@amd.com> Acked-by:
-
Yongqiang Sun authored
root cause: DMCU try to perform a smoothness brightness change.Incorrect initial brightness level causes the 1 sec dim. Change: Cache brightness level in stream, and clear it when edp backlight on. If brightness level in stream is 0, set brightness with ramp value is 0. DMCU will set the brightness without smoothness transition. v2: squash in null pointer fix (Harry) Signed-off-by:
Yongqiang Sun <yongqiang.sun@amd.com> Reviewed-by:
-
Jerry (Fangzhi) Zuo authored
Signed-off-by:
Jerry (Fangzhi) Zuo <Jerry.Zuo@amd.com> Reviewed-by:
-
Harry Wentland authored
Signed-off-by:
-
Oak Zeng authored
This is preparation for sharing client ID definitions between amdgpu and amdkfd Signed-off-by:
Oak Zeng <Oak.Zeng@amd.com> Reviewed-by:
Chunming Zhou <david1.zhou@amd.com> Acked-by:
-
Dmytro Laktyushkin authored
Bw spreadsheet was updated while dce_calcs was not Signed-off-by:
-
Yongqiang Sun authored
Increase clock, if current dpp div is 0 and request dpp div is 1, request clk is higher than maximum dpp clk as per dpm table. set dispclk to the value of maximum supported dpp clk set div to 1 set dispclk to request value. Decrease clock, currrent dpp div is 1 and request dpp div is 0, current clk is higher than maximum dpp clk as per dpm table. set dispclk to the value of maximum supported dpp clk set div to 0 set dispclk to request value. v2: squash in !DCN build fix Signed-off-by:
-
Samson Tam authored
Signed-off-by:
Samson Tam <Samson.Tam@amd.com> Reviewed-by:
Anthony Koo <Anthony.Koo@amd.com> Acked-by:
-
Shirish S authored
The order of planes is given by the order they are enumerated by kms. Planes with a higher ID appears above planes with a lower ID. Currently the planes are enumerated in the wrong order, putting the nv12 only plane after the two RGBA planes. This patch corrects the plane enumeration order such that all the overlay planes are initialized first then the primary planes. Due to this change in order the dc_add_plane_to_context() shall receive the planes in reverse order hence this patch reverses the parsing of planes in DM side itself. v2: drop local reverse macro for upstream Signed-off-by:
Pratik Vishwakarma <Pratik.Vishwakarma@amd.com> Reviewed-by:
-
Vitaly Prosyak authored
Signed-off-by:
Vitaly Prosyak <vitaly.prosyak@amd.com> Reviewed-by:
-
Tony Cheng authored
Signed-off-by:
-
Anthony Koo authored
Time stamping will be part of surface, and will be updated when address is flipped. FreeSync parameters will be attached to stream, as it adjusts the timing dynamically. Signed-off-by:
-
Anthony Koo authored
Stats will be used for debug purposes Signed-off-by:
-
Eric Yang authored
The work around for hw bug causes S3 resume failure. Don't execute disable vga logic if not in vga mode. Signed-off-by:
Eric Yang <Eric.Yang2@amd.com> Reviewed-by:
-
Mikita Lipski authored
Initializing ABM and DMCU modules for dce 80/81/83/100 as in DCE110 Adding constructors and destructors for each module. Adding register list for DMCU in dce80 as some registers are missing in dce80 from the basic list. DMCU is never used, so it would not have any functional impact. Signed-off-by:
Mikita Lipski <mikita.lipski@amd.com> Reviewed-by:
-
Krunoslav Kovac authored
In MPO scenario when playing SDR clip in HDR desktop mode, Win is boosting desktop and requests driver to boost MPO. But driver boosting is currently done in regamma which is stream property and thus shared between grph and video. Redesigning the boosting in RV: use CM_HDR_MULT register which was added for this scenario. It also has the benefit that it can be done in HIRQL. Signed-off-by:
Krunoslav Kovac <Krunoslav.Kovac@amd.com> Reviewed-by:
-
Bhawanpreet Lakha authored
Use DC_LOGGER macro for logs. Signed-off-by:
Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com> Reviewed-by:
-
Bhawanpreet Lakha authored
These MACROS are only being used by a few files but gets pulled in by dc.h Signed-off-by:
-
Yongqiang Sun authored
This change make sure bandwidth is set properly. For increase bandwidth, set bandwidth before backend and front end programming. For decrease bandwidth, set bandwidth after. To avoid smu hang when reboot and dpms due to 0 disp clk, keep min disp clock as 100Mhz. Signed-off-by:
-
Eric Yang authored
Per discussion with VBIOS team, the orginal check is not correct in all cases on latest VBIOS. Additional check is needed. This change should maintain old behaviour on older VBIOS. Signed-off-by:
Charlene Liu <Charlene.Liu@amd.com> Acked-by:
-
Bhawanpreet Lakha authored
HW Engineer's Notes: During switch from vga->extended, if we set the VGA_TEST_ENABLE and then hit the VGA_TEST_RENDER_START, then the DCHUBP timing gets updated correctly. Then vBIOS will have it poll for the VGA_TEST_RENDER_DONE and unset VGA_TEST_ENABLE, to leave it in the same state as before. Signed-off-by:
-
Jerry (Fangzhi) Zuo authored
Signed-off-by:
Hersen Wu <hersenxs.wu@amd.com> Acked-by:
-
Yongqiang Sun authored
300Mhz disp clk limit was a workaround that was fixed in SMU and is no longer needed. Signed-off-by:
-
Leo (Sunpeng) Li authored
DRM's documentation for the color transform matrix does not specify whether the values are in signed-magnitude, or 2's complement. Therefore, it was assumed to use 2's complement. However, existing usermode implementations use signed-magnitude. Therefore, conform to existing standards, and convert to 2's complement internally. Signed-off-by:
Leo (Sunpeng) Li <sunpeng.li@amd.com> Reviewed-by:
-
SivapiriyanKumarasamy authored
Add null check for stream update Signed-off-by:
SivapiriyanKumarasamy <sivapiriyan.kumarasamy@amd.com> Reviewed-by:
-
SivapiriyanKumarasamy authored
Fix bug and make changes from review 132656 Signed-off-by:
-
SivapiriyanKumarasamy authored
Avoid hanging DMCU by setting abm level only when OTG unblanked Signed-off-by:
-
Leo (Sunpeng) Li authored
Should return -ENOMEM when allocation fails. Also, just return the error code instead of using a variable. Signed-off-by:
-
Leo (Sunpeng) Li authored
While checking plane states for updates during atomic check, we create dc_plane_states in preparation. These dc states should be freed if something errors. Although the input transfer function is also freed by dc_plane_state_release(), we should free it (on error) under the same scope as where it is created. Signed-off-by:
-
Harry Wentland authored
They were used by amdgpu_dm at some point but since it has its own amdgpu_dm_connector now these aren't needed anymore. Signed-off-by:
-
Alex Deucher authored
Save/restore the backlight level scratch register in S3/S4 so the backlight level comes back at the previously requested level. Bug: https://bugzilla.kernel.org/show_bug.cgi?id=199047 Fixes: 4ec6ecf4 (drm/amdgpu: drop scratch regs save and restore from S3/S4 handling) Acked-by:
-
Christian König authored
We unmapped imported DMA-bufs when the GEM handle was dropped, not when the hardware was done with the buffere. Signed-off-by:
-
