- 05 Dec, 2019 40 commits
-
-
Anthony Koo authored
[Why] We want to know DP protocol version [How] In DC create we initialize a cap to indicate the max DP protocol version supported Signed-off-by:
Anthony Koo <Anthony.Koo@amd.com> Reviewed-by:
Aric Cyr <Aric.Cyr@amd.com> Acked-by:
Leo Li <sunpeng.li@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Joseph Gravenor authored
[Why] DF team has produced more optimized latency numbers, for lpddr4 [How] change the p-state laency in the lpddr4 wm table to the new latency number Signed-off-by:
Joseph Gravenor <joseph.gravenor@amd.com> Reviewed-by:
Tony Cheng <Tony.Cheng@amd.com> Acked-by:
-
Reza Amini authored
[Why] Need support for more color management in 10bit surface. [How] Provide support for DePQ for 10bit surface Signed-off-by:
Reza Amini <Reza.Amini@amd.com> Reviewed-by:
Krunoslav Kovac <Krunoslav.Kovac@amd.com> Acked-by:
-
Nikola Cornij authored
[why] During mode transition steer fifo could overflow. Quite often it recovers by itself, but sometimes it doesn't. [how] Add steer fifo reset before unblanking the stream. Also add a short delay when resetting dig resync fifo to make sure register writes don't end up back-to-back, in which case the HW might miss the reset request. Signed-off-by:
Nikola Cornij <nikola.cornij@amd.com> Reviewed-by:
-
Dmytro Laktyushkin authored
Bring this calculation in line with HW programming guide. Signed-off-by:
Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com> Reviewed-by:
-
Michael Strauss authored
[WHY] Previous Renoir chroma viewport workaround fixed an MPO flicker by increasing the chroma viewport size. However, when the MPO plane is rotated 180 degrees, the viewport is read in reverse. Since the workaround increases viewport size, when reading in reverse it causes a vertical chroma offset. [HOW] Pass rotation value to viewport set functions Temporarily disable the chroma viewport w/a when hubp is rotated 180 degrees Signed-off-by:
Michael Strauss <michael.strauss@amd.com> Reviewed-by:
-
abdoulaye berthe authored
Signed-off-by:
abdoulaye berthe <abdoulaye.berthe@amd.com> Reviewed-by:
Wenjing Liu <Wenjing.Liu@amd.com> Acked-by:
-
Brandon Syu authored
[Why] The variable mismatch assignment error. [How] To use uint32_t replace it. Signed-off-by:
Brandon Syu <Brandon.Syu@amd.com> Reviewed-by:
Charlene Liu <Charlene.Liu@amd.com> Acked-by:
-
Leo (Hanghong) Ma authored
[why] DP spec requires 1000 symbols delay between the end of link training and enabling FEC in the stream. Currently we are using 1 miliseconds delay which is not accurate. [how] One lane RBR should have the maximum time for transmitting 1000 LL codes which is 6.173 us. So using 7 microseconds delay instead of 1 miliseconds. Signed-off-by:
Leo (Hanghong) Ma <hanghong.ma@amd.com> Reviewed-by:
Harry Wentland <Harry.Wentland@amd.com> Reviewed-by:
-
Aric Cyr authored
Signed-off-by:
-
Eric Yang authored
[Why] Before was using HW counter value to determine the dprefclk. Which take into account ss, but has large variation, not good enough for generating audio dto. Also, the bios parser code to get the ss percentage was not working. [How] After this change, dprefclk is hard coded, same as on RV. We don't expect this to change on Renoir. Modified bios parser code to get the right ss percentage. Signed-off-by:
Eric Yang <Eric.Yang2@amd.com> Reviewed-by:
-
Joseph Gravenor authored
[Why] new sr and pstate latencies are optimized for the case when we are not using lpddr4 memory [How] have two different wm tables, one for the lpddr case and one for non lpddr case Signed-off-by:
-
David Galiffi authored
[Why] In dc_link_is_dp_sink_present, if dal_ddc_open fails, then dal_gpio_destroy_ddc is called, destroying pin_data and pin_clock. They are created only on dc_construct, and next aux access will cause a panic. [How] Instead of calling dal_gpio_destroy_ddc, call dal_ddc_close. Signed-off-by:
David Galiffi <David.Galiffi@amd.com> Reviewed-by:
-
Joseph Gravenor authored
[Why] When video_memory_type bw_params->vram_type is assigned, wedistinguish between Ddr4MemType and LpDdr4MemType. Because of this we will never report that we are using LpDdr4MemType and never re-purpose WM set D [How] populate bios integrated info for renoir by adding the revision number for renoir and use that integrated info table instead of of asic_id to get the vram type Signed-off-by:
-
Nicholas Kazlauskas authored
[Why] On dcn21 this is programmed for tracebuffer support but isn't being programmed on dcn20. DMCUB execution hits an undefined address 65000000 on tracebuffer access. [How] Program CW5. Signed-off-by:
Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Reviewed-by:
-
Nicholas Kazlauskas authored
[Why] Not having support for autoload isn't an error. If the DMUB firmware doesn't support it then don't return DMUB_STATUS_INVALID. [How] Return DMUB_STATUS_OK when ->is_auto_load_done is NULL. Signed-off-by:
-
Nicholas Kazlauskas authored
[Why] The wait for PHY init won't finish if the firmware doesn't support it. [How] Only hook this functionality up on DCN21 and move it out of DCN20. For ASIC without support then this should return OK so we don't hang while waiting in DC. Signed-off-by:
-
Yongqiang Sun authored
Signed-off-by:
Yongqiang Sun <yongqiang.sun@amd.com> Reviewed-by:
-
Lucy Li authored
[Why] Black screen seen after display is disabled then re-enabled. Caused by difference in link settings when switching between different resolutions. [How] In PnP case, or whenever the display is still enabled but the driver is unloaded, disable link before re-enabling with new link settings. Signed-off-by:
Lucy Li <lucy.li@amd.com> Reviewed-by:
-
Paul Hsieh authored
[Why] Link training failed randomly when plugging USB-C display in/out. [How] If link training failed, reset PHY in link re-training. Signed-off-by:
Paul Hsieh <paul.hsieh@amd.com> Reviewed-by:
-
Jaehyun Chung authored
[Why] Wrong guards were causing the debug option not to run. [How] Changed the guard to the correct one, matching the rq, ttu, dlg regs struct members that need to be guarded. Also log a message when validation starts. Signed-off-by:
Jaehyun Chung <jaehyun.chung@amd.com> Reviewed-by:
Alvin Lee <Alvin.Lee2@amd.com> Acked-by:
-
Joseph Gravenor authored
[why] pstate_latency_us never gets updated from the hard coded value in rn_clk_mgr.c [how] update the wm table's values before we do calculations with them Signed-off-by:
-
Nikola Cornij authored
[why] On ASICs where number of DSCs is the same as OPPs there's no need for DSC resource management. Mappping 1-to-1 fixes mode-set- or S3- -related issues for such platforms. [how] Map DSC resources 1-to-1 to pipes only if number of OPPs is the same as number of DSCs. This will still keep other ASICs working. A follow-up patch to fix mode-set issues on those ASICs will be required if testing shows issues with mode set. Signed-off-by:
-
Michael Strauss authored
[WHY] Dali is currently being misinterpreted as Renoir, as a result uses wrong clk mgr constructor [HOW] Add check to init Dali as Raven2 before it can be misidentified Clean up & fix Raven2 & Dali ASIC checks Signed-off-by:
-
Anthony Koo authored
[Why] Some function pointers in the hwss function pointer table are meant to be hw sequencer entry points to be called from dc. However some of those function pointers are not meant to be entry points, but instead used as a code reuse/inheritance tool called directly by other hwss functions, not by dc. Therefore, we want a more clear separation of which functions we determine to be interface functions vs the functions we use within hwss. [How] DC interface functions will be stored in: struct hw_sequencer_funcs Functions used within HWSS will be stored in: struct hwseq_private_funcs Signed-off-by: -
Anthony Koo authored
[Why] First, to make code more consistent Second, to get rid of those scenario where we create a second local pointer to dc when it's already passed in. [How] Rename core_dc to dc Remove duplicate local pointers to dc Signed-off-by:
-
Eric Yang authored
[Why] DF team has produced more optimized latency numbers. [How] Add sr latencies to the wm table, use different latencies for different wm sets. Also fix bb override from registery key for these latencies. Signed-off-by:
-
Le Ma authored
Move the print out of uvd instance loop in amdgpu_uvd_suspend v2: drop unnecessary brackets v3: grab ras_intr state once for multiple times use Signed-off-by:
Le Ma <le.ma@amd.com> Reviewed-by:
Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by:
-
Le Ma authored
Otherwise next err_event_athub error cannot call gpu reset. And following resume sequence will not be affected by this flag. v2: create function to clear amdgpu_ras_in_intr for modularity of ras driver Signed-off-by:
-
Le Ma authored
This athub fatal error can be recovered by baco without system-level reboot, so add a mode to use baco for the recovery. Not affect the default psp reset situations for now. Signed-off-by:
-
Le Ma authored
Currently each XGMI node reset wq does not run in parrallel if bound to same cpu. Make change to bound the xgmi_reset_work item to different cpus. XGMI requires all nodes enter into baco within very close proximity before any node exit baco. So schedule the xgmi_reset_work wq twice for enter/exit baco respectively. To use baco for XGMI, PMFW supported for baco on XGMI needs to be involved. The case that PSP reset and baco reset coexist within an XGMI hive never exist and is not in the consideration. v2: define use_baco flag to simplify the code for xgmi baco sequence Signed-off-by:
-
Le Ma authored
This operation is needed when baco entry/exit for ras recovery Signed-off-by:
-
Le Ma authored
This should be cleared during every nbif uncorrectable error cleanup work. Signed-off-by:
-
Le Ma authored
To fix issue that ras controller interrupt cannot be triggered anymore after one time nbif uncorrectable error. And error count is stored in nbif ras object for query. Signed-off-by:
-
Le Ma authored
Change it to external interface. Signed-off-by:
-
Le Ma authored
v2: add notification when ras controller interrupt generates Signed-off-by:
-
Pierre-Eric Pelloux-Prayer authored
The same workaround is used for gfx7. Both PAL and Mesa use it for gfx8 too, so port this commit to gfx_v8_0_ring_emit_fence_gfx. Signed-off-by:
Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Reviewed-by:
-
Alex Deucher authored
So the output is consistent with other asics. Reviewed-by:
Evan Quan <evan.quan@amd.com> Signed-off-by:
-
Yong Zhao authored
Up to this point, those functions are all the same for all ASICs, so no need to call them by functions pointers. Removing the function pointers will greatly increase the code readablity. If there is ever need for those function pointers, we can add it back then. Signed-off-by:
Yong Zhao <Yong.Zhao@amd.com> Reviewed-by:
Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by:
-
James Zhu authored
SPI limits total CS waves in flight per SE to no more than 32 * num_cu and we need to stuff 40 waves on a CU to completely clean the SGPR. This is accomplished in the WR by cleaning the SE in two steps, half of the CU per step. Signed-off-by:
James Zhu <James.Zhu@amd.com> Reviewed-by:
-
