- 31 Jul, 2019 19 commits
-
-
Andrey Grodzovsky authored
HW requires for caching to be unset for scanout BO mappings when the BO placement is in GTT memory. Usually the flag to unset is passed from user mode but for FB mode this was missing. v2: Keep all BO placement logic in amdgpu_display_supported_domains Suggested-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by:
Shirish S <shirish.s@amd.com> Signed-off-by:
-
Jonathan Kim authored
monitor xgmi errors via mc pie status through fica registers. Signed-off-by:
Jonathan Kim <Jonathan.Kim@amd.com> Reviewed-by:
Kent Russell <Kent.Russell@amd.com> Signed-off-by:
-
Jonathan Kim authored
adding perfmon and fica atomic operations to adhere to data fabrics finite state machine requirements for indirect register access. Signed-off-by:
-
Jay Cornwall authored
Signed-off-by:
Jay Cornwall <jay.cornwall@amd.com> Acked-by:
-
Jay Cornwall authored
Previously submitted code was taken from an incorrect branch and was non-functional. Cc: Oak Zeng <oak.zeng@amd.com> Signed-off-by:
Oak Zeng <oak.zeng@amd.com> Signed-off-by:
-
Jay Cornwall authored
If the trap is entered due to MODE.DEBUG_EN=1 and SAVECTX is raised concurrently the handler cannot identify the source of the exception. This causes the debugger to lose single step exception notification when a context save request arrives at the same time. When MODE.DEBUG_EN=1 and STATUS.HALT=0 (exception not already handled) jump to the second-level trap handler upon entering the trap. The second-level trap will set STATUS.HALT=1 and return to the shader. If SAVECTX was raised then control flow will return to the trap, which will then handle the context save request. Cc: Tony Tye <tony.tye@amd.com> Cc: Laurent Morichetti <laurent.morichetti@amd.com> Cc: Qingchuan Shi <qingchuan.shi@amd.com> Signed-off-by:
Laurent Morichetti <laurent.morichetti@amd.com> Signed-off-by:
-
Jay Cornwall authored
When a wavefront raises TRAPSTS.XNACK_ERROR with STATUS.ALLOW_REPLAY=0 subsequent memory instructions have undefined behavior. In practice SQC stores continue to work but TCP stores do not. Context save is permitted to fail after XNACK error because the wavefront will be halted and subsequently terminated. However the debugger has an interest in retrieving the wavefront VGPR/LDS state. Detect the out-of-spec case and use SQC stores during context save in place of TCP stores. Signed-off-by:
-
tiancyin authored
navi14 share same PTE format with navi10. Reviewed-by:
Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by:
tiancyin <tianci.yin@amd.com> Signed-off-by:
-
James Zhu authored
Since we are using the signed FW now, and also using PSP firmware loading, but it's still potential to break driver when loading FW directly instead of PSP, so we should add offset. Signed-off-by:
James Zhu <James.Zhu@amd.com> Reviewed-by:
Leo Liu <leo.liu@amd.com> Signed-off-by:
-
Kevin Wang authored
in this patch, drm/amd/powerplay: add callback function of get_thermal_temperature_range the driver missed temperature granularity change on other temperature. Signed-off-by:
Kevin Wang <kevin1.wang@amd.com> Reviewed-by:
Evan Quan <evan.quan@amd.com> Reviewed-by:
Kenneth Feng <kenneth.feng@amd.com> Signed-off-by:
-
Evan Quan authored
Drop redundant check, duplicate check, duplicate setting and fix the return value. Signed-off-by:
-
Evan Quan authored
Honor the 'dpm' module parameter setting on SW SMU routine as what we did on previous ASICs. SMU FW loading is still proceeded even with "dpm=0". Signed-off-by:
-
Colin Ian King authored
Currently the allocation of config may fail and a null pointer dereference on config can occur. Fix this by added a null check on a failed allocation of config. Addresses-Coverity: ("Dereference null return") Fixes: c2cd9d04 ("drm/amd/display: Hook up calls to do stereo mux and dig programming to stereo control interface") Signed-off-by:Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Chuhong Yuan authored
Instead of using to_pci_dev + pci_get_drvdata, use dev_get_drvdata to make code simpler. Signed-off-by:
Chuhong Yuan <hslester96@gmail.com> Signed-off-by:
-
Chuhong Yuan authored
Instead of using to_pci_dev + pci_get_drvdata, use dev_get_drvdata to make code simpler. Signed-off-by:
-
Chuhong Yuan authored
Instead of using to_pci_dev + pci_get_drvdata, use dev_get_drvdata to make code simpler. Signed-off-by:
-
Nick Desaulniers authored
arch/x86/Makefile disables SSE and SSE2 for the whole kernel. The AMDGPU drivers modified in this patch re-enable SSE but not SSE2. Turn on SSE2 to support emitting double precision floating point instructions rather than calls to non-existent (usually available from gcc_s or compiler_rt) floating point helper routines for Clang. This was originally landed in: commit 10117450 ("drm/amd/display: add -msse2 to prevent Clang from emitting libcalls to undefined SW FP routines") but reverted in: commit 193392ed ("Revert "drm/amd/display: add -msse2 to prevent Clang from emitting libcalls to undefined SW FP routines"") due to bugreports from GCC builds. Add guards to only do so for Clang. Link: https://bugs.freedesktop.org/show_bug.cgi?id=109487 Link: https://github.com/ClangBuiltLinux/linux/issues/327Suggested-by:
Sedat Dilek <sedat.dilek@gmail.com> Suggested-by:
Sami Tolvanen <samitolvanen@google.com> Signed-off-by:
Nick Desaulniers <ndesaulniers@google.com> Signed-off-by:
-
Kent Russell authored
This was missed during the addition of VegaM support Reviewed-by:
-
Kevin Wang authored
1. the thermal temperature is asic related data, move the code logic to xxx_ppt.c. 2. replace data structure PP_TemperatureRange with smu_temperature_range. 3. change temperature uint from temp*1000 to temp (temperature uint). Signed-off-by:
Huang Rui <ray.huang@amd.com> Signed-off-by:
-
- 22 Jul, 2019 4 commits
-
-
Alex Deucher authored
On vega20, there is an SMU message to query it. On navi, it's fetched from the metrics table. Reviewed-by:
-
Hawking Zhang authored
Otherwise, it will cause driver access non-existing sdma registers in gpu reset code path Signed-off-by:
Le Ma <Le.Ma@amd.com> Signed-off-by:
-
Kevin Wang authored
remove redundacy debug log in smu. eg: [ 6897.969447] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled [ 6897.969448] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled [ 6897.969448] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled [ 6899.024114] amdgpu: [powerplay] Unsupported SMU message: 38 [ 6899.024151] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled [ 6899.024151] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled [ 6899.024152] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled [ 6900.078296] amdgpu: [powerplay] Unsupported SMU message: 38 [ 6900.078332] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled [ 6900.078332] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled [ 6900.078333] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled [ 6901.133230] amdgpu: [powerplay] Unsupported SMU message: 38 Signed-off-by:
-
Kevin Wang authored
v2: add function smu_default_set_performance_level as default dpm level handler. change function name smu_set_performance_level to smu_asic_set_performance_level v1: 1.NAVI10_PEAK_SCLK_XTX 1830 Mhz 2.NAVI10_PEAK_SCLK_XT 1755 Mhz 3.NAVI10_PEAK_SCLK_XL 1625 Mhz Signed-off-by:
Jack Gui <Jack.Gui@amd.com> Signed-off-by:
-
- 18 Jul, 2019 17 commits
-
-
Dale Zhao authored
[Why]: Some active dongles have DP++ port and DP port at the same time. Current code doesn't cover DP++ case and processes as default DVI case, in which audio is disabled. Because of dual mode, DP case is also treat as DVI case for the other port. [How]: According DP 1.4 spec, add DP++ procedure similar with HDMI case. Also add None dongle type for DP case. Signed-off-by:
Dale Zhao <dale.zhao@amd.com> Reviewed-by:
Wenjing Liu <wenjing.liu@amd.com> Acked-by:
Leo Li <sunpeng.li@amd.com> Signed-off-by:
-
Nicholas Kazlauskas authored
[Why] DC configures the GSL group for the pipe when pipe_split is enabled and we're switching flip types (buffered <-> immediate flip) on DCN2. In order to record what GSL group the pipe is using DC stores it in the pipe's stream_res. DM is not aware of this internal grouping, nor is DC resource. So when DM creates a dc_state context and passes it to DC the current GSL group is lost - DM never knew about it in the first place. After 3 immediate flips we run out of GSL groups and we're no longer able to correctly perform *any* flip for multi-pipe scenarios. [How] The gsl_group needs to be copied to the new context. DM has no insight into GSL grouping and could even potentially create a brand new context without referencing current hardware state. So this makes the most sense to have happen in DC. There are two places where DC can apply a new context: - dc_commit_state - dc_commit_updates_for_stream But what's shared between both of these is apply_ctx_for_surface. This logic only matters for DCN2, so it can be placed in dcn20_apply_ctx_for_surface. Before doing any locking (where the GSL group is setup) we can copy over the GSL groups before committing the new context. Signed-off-by:
Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Reviewed-by:
Hersen Wu <hersen.wu@amd.com> Acked-by:
-
Lewis Huang authored
Add a flag to dc_debug_options to determine if a disconnected edp link should be destroyed. Signed-off-by:
Lewis Huang <Lewis.Huang@amd.com> Reviewed-by:
Yongqiang Sun <yongqiang.sun@amd.com> Acked-by:
-
Tony Cheng authored
Signed-off-by:
Tony Cheng <tony.cheng@amd.com> Reviewed-by:
-
Fatemeh Darbehani authored
For some platforms, we need to know SMU version for driver/SMU compatibility. This change adds that field. Signed-off-by:
Fatemeh Darbehani <fatemeh.darbehani@amd.com> Reviewed-by:
-
Jun Lei authored
[why] Existing support in DC for pstate only accounts for a single latency. This is sufficient when the variance of latency is small, or that pstate support isn't necessary for correct ASIC functionality. Newer ASICs violate both existing assumptions. PState support is mandatory of correct ASIC functionality, but not all latencies have to be supported. Existing code supports a "full p state" which allows memory clock to change, but is hard for DCN to support (as it requires very large buffers). New code will now fall back to a "dummy p state" support when "full p state" cannot be support. This easy p state support should always be allowed. [how] Define a new latency in socBB. Add fallback logic to support it. Note DML is also updated to ensure that fallback will always work. Signed-off-by:
Jun Lei <Jun.Lei@amd.com> Reviewed-by:
Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com> Acked-by:
-
Eric Yang authored
[Why] In DCN2x asic, the MPCC status register definition changed, and our logic for assert is incorrect. disabled is valid state, where we should see idle and not busy, where as in not disabled state, we should see not idle. [How] Change assert condition to be more sensible. Signed-off-by:
Eric Yang <Eric.Yang2@amd.com> Reviewed-by:
-
Eric Yang authored
[Why] Previously assume eDP sink present if connector present. Do not need to enforce this restriction. Fix issue where driver attempt to read link setting even though no edp connected. {How] Only read link setting after reading connection status. Signed-off-by: -
Tai Man authored
[Why] The audios array defined in "struct resource_pool" is only 6 (MAX_PIPES) but the max number of audio devices (num_audio) is 7. In some projects, it will run out of audios array. [How] Incraese the audios array size to 7. Signed-off-by:
Tai Man <taiman.wong@amd.com> Reviewed-by:
Joshua Aberback <Joshua.Aberback@amd.com> Acked-by:
-
Aric Cyr authored
Signed-off-by:
Aric Cyr <aric.cyr@amd.com> Acked-by:
-
Zhan Liu authored
[Why] For boards that support eDP but do not have a physical eDP display connected an ASSERT will be thrown. This is not a critical failure and shouldn't be treated as such. [How] Drop the assertion. Signed-off-by:
Zhan Liu <zhan.liu@amd.com> Reviewed-by:
-
Nicholas Kazlauskas authored
[Why] In an effort to stop redundant calls to dce110_disable_audio_stream the audio->enabled flag was added to the audio resource struct. While this state probably shouldn't have been tracked on the audio struct itself it still works fine for some sequences. However, it does not work for cases where we're freeing the audio resource (such as hotplugs) or when dynamic audio is enabled. In these cases the pipe_ctx->stream_res.audio = NULL before we can set audio->enabled = false. The next time we acquire the audio resource such as on hotplug the audio will not be enabled for the stream since DC thinks it's still enabled. Audio state tracking should cover this sequence. [How] Set audio->enabled = false at the start as long as we have pipe_ctx->stream_res.audio. Signed-off-by:
-
Julian Parkin authored
[Why] Code to enable DCN20 dynamic metadata feature is duplicated in two places and was added to DCE110 enable stream. [How] Create DCN20 specific enable stream function for clarity, and add a hardware sequencer function to program dynamic metadata to avoid the duplicate code. Signed-off-by:
Julian Parkin <julian.parkin@amd.com> Reviewed-by:
Charlene Liu <Charlene.Liu@amd.com> Acked-by:
-
Alvin Lee authored
[Why] In dm_helpers_parse_edid_caps, there is a corner case where no speakers can be allocated even though the audio mode count is greater than 0. Enabling audio when no speaker allocations exists can cause issues in the video stream. [How] Add a check to not enable audio unless one or more speaker allocations exist (since doing this can cause issues in the video stream). Signed-off-by:
Alvin Lee <alvin.lee2@amd.com> Reviewed-by:
-
Julian Parkin authored
[Why] It is possible (but very unlikely) that constructing dc fails before current_state is created. We support 666 color depth in some scenarios, but this isn't handled in get_norm_pix_clk. It uses exactly the same pixel clock as the 888 case. [How] Check for non null current_state before destructing. Add case for 666 color depth to get_norm_pix_clk to avoid assertion. Signed-off-by:
-
Jun Lei authored
[why] update_bounding_box calculates intermediate bb states based on clock relationship however, the last state doesn't need to maintain a minimum relationship, but should actually contain maximum of every clock. otherwise maximum clocks are not usable [how] once the calculated bb is built, override the last state with max values Signed-off-by:
-
Dmytro Laktyushkin authored
Add dual plane dcc programming support for surfaces. Removes unions from plane size and dcc params as they serve no practical purpose only making our code more convoluted. This results in easy dual plane dcc and surface size programming. Temporary diags_dm code is used to handle the interface change without breaking functionality as a diags change needs to be applied after this one. Signed-off-by:
-
