- 14 Sep, 2022 3 commits
-
-
Alex Deucher authored
This mirrors what we do for other asics and this way we are sure the sdma doorbell range is properly initialized. There is a comment about the way doorbells on gfx9 work that requires that they are initialized for other IPs before GFX is initialized. However, the statement says that it applies to multimedia as well, but the VCN code currently initializes doorbells after GFX and there are no known issues there. In my testing at least I don't see any problems on SDMA. This is a prerequisite for fixing the Unsupported Request error reported through AER during driver load. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216373 The error was unnoticed before and got visible because of the commit referenced below. This doesn't fix anything in the commit below, rather fixes the issue in amdgpu exposed by the commit. The reference is only to associate this commit with below one so that both go together. Fixes: 8795e182 ("PCI/portdrv: Don't disable AER reporting in get_port_device_capability()") Acked-by:
Christian König <christian.koenig@amd.com> Reviewed-by:
Lijo Lazar <lijo.lazar@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Alex Deucher authored
This mirrors what we do for other asics and this way we are sure the ih doorbell range is properly initialized. There is a comment about the way doorbells on gfx9 work that requires that they are initialized for other IPs before GFX is initialized. In this case IH is initialized before GFX, so there should be no issue. This is a prerequisite for fixing the Unsupported Request error reported through AER during driver load. Link: https://bugzilla.kernel.org/show_bug.cgi?id=216373 The error was unnoticed before and got visible because of the commit referenced below. This doesn't fix anything in the commit below, rather fixes the issue in amdgpu exposed by the commit. The reference is only to associate this commit with below one so that both go together. Fixes: 8795e182 ("PCI/portdrv: Don't disable AER reporting in get_port_device_capability()") Acked-by:
-
Li Ma authored
Add new gpu_metrics_v2_3 to acquire average temperature info from SMU metrics. To acquire average temp info from gpu_metrics interface, but gpu_metrics_v2_2 only has members to show current temp info. --- v1: Only add average_temperature_gfx in gpu_metrics_v2_3. v2: Add average temp members for soc, core and l3 in gpu_metrics_v2_3 and put these new members at the end of gpu_metrics_v2_3. Add operation to read average temp info from metrics table. v3: Merge v1 and v2 and rename the patch. v4: Merge v3. Add firmware version judgment in vangogh_common_get_gpu_metrics to maintain backward compatibility and rename the patch. "return ret" on error scenario in smu_cmn_get_smc_version. Signed-off-by:
Li Ma <li.ma@amd.com> Reviewed-by:
Evan Quan <evan.quan@amd.com> Signed-off-by:
-
- 13 Sep, 2022 37 commits
-
-
Xu Panda authored
soc15_common.h is included more than once. Reported-by:
Zeal Robot <zealci@zte.com.cn> Signed-off-by:
Xu Panda <xu.panda@zte.com.cn> Signed-off-by:
-
Jiapeng Chong authored
The function copy_stream_update_to_stream() is defined in the notif.c file, but not called elsewhere, so delete this unused function. drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c:2852:6: warning: no previous prototype for ‘dc_reset_state’. Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=2113Reported-by:
Abaci Robot <abaci@linux.alibaba.com> Signed-off-by:
Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Signed-off-by:
-
Jiapeng Chong authored
These two functions are not used outside the function dcn32_mpc.c, so the modification is defined as static. drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_mpc.c:704:6: warning: no previous prototype for ‘mpc32_program_shaper’. drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_mpc.c:900:6: warning: no previous prototype for ‘mpc32_program_3dlut’. Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=2115Reported-by:
-
Jiapeng Chong authored
drm/amd/display: make dcn32_link_encoder_is_in_alt_mode and dcn32_link_encoder_get_max_link_cap static These two functions are not used outside the function dcn32_dio_link_encoder.c, so the modification is defined as static. drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_dio_link_encoder.c:121:6: warning: no previous prototype for ‘dcn32_link_encoder_is_in_alt_mode’. drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_dio_link_encoder.c:136:6: warning: no previous prototype for ‘dcn32_link_encoder_get_max_link_cap’. Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=2117Reported-by:
-
Jiapeng Chong authored
This symbol is not used outside of dcn32_dpp.c, so marks it static. drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_dpp.c:34:6: warning: no previous prototype for ‘dscl32_calc_lb_num_partitions’. Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=2118Reported-by:
-
Jiapeng Chong authored
This symbol is not used outside of dcn32_dio_stream_encoder.c, so marks it static. drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_dio_stream_encoder.c:63:6: warning: no previous prototype for ‘enc32_stream_encoder_dvi_set_stream_attribute’. Link:https://bugzilla.openanolis.cn/show_bug.cgi?id=2119Reported-by:
-
Rafael Mendonca authored
If construction of the array of work queues to handle hpd_rx_irq offload work fails, we need to unwind. Destroy all the created workqueues and the allocated memory for the hpd_rx_irq_offload_work_queue struct array. Fixes: 8e794421 ("drm/amd/display: Fork thread to offload work of hpd_rx_irq") Signed-off-by:
Rafael Mendonca <rafaelmendsr@gmail.com> Signed-off-by:
-
Alex Deucher authored
for imu_v11_0_3_program_rlc_ram(). Include imu_v11_0_3.h in imu_v11_0_3.c. Reviewed-by:
Hawking Zhang <Hawking.Zhang@amd.com> Reported-by:
kernel test robot <lkp@intel.com> Signed-off-by:
-
Christian König authored
Sort the functions in the order they are called Signed-off-by:
-
Christian König authored
Cleanup the coding style and function names to represent the data they process. Only initialize and cleanup the CS structure in init/fini. Check the size of the IB chunk in pass1. v2: fix job initialisation order and use correct scheduler instance v3: try to move all functional changes into a separate patch. v4: move reordering and pass2 out of this patch as well Signed-off-by:
-
Christian König authored
Use DMA_RESV_USAGE_BOOKKEEP for VM page table updates and KFD preemption fence. v2: actually update all usages for KFD Signed-off-by:
-
Christian König authored
This reverts commit 94f4c496. We found that the bo_list is missing a protection for its list entries. Since that is fixed now this workaround can be removed again. Signed-off-by:
-
Christian König authored
Move setting the job resources into amdgpu_job.c Signed-off-by:
Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by:
Luben Tuikov <luben.tuikov@amd.com> Reviewed-by:
-
Christian König authored
We should not have any different CS constrains based on the execution environment. Signed-off-by:
-
Nathan Chancellor authored
This function consumes a lot of stack space and it blows up the size of dml30_ModeSupportAndSystemConfigurationFull() with clang: drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn30/display_mode_vba_30.c:3542:6: error: stack frame size (2200) exceeds limit (2048) in 'dml30_ModeSupportAndSystemConfigurationFull' [-Werror,-Wframe-larger-than] void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib) ^ 1 error generated. Commit a0f7e7f7 ("drm/amd/display: fix i386 frame size warning") aimed to address this for i386 but it did not help x86_64. To reduce the amount of stack space that dml30_ModeSupportAndSystemConfigurationFull() uses, mark UseMinimumDCFCLK() as noinline, using the _for_stack variant for documentation. While this will increase the total amount of stack usage between the two functions (1632 and 1304 bytes respectively), it will make sure both stay below the limit of 2048 bytes for these files. The aforementioned change does help reduce UseMinimumDCFCLK()'s stack usage so it should not be reverted in favor of this change. Link: https://github.com/ClangBuiltLinux/linux/issues/1681Reported-by:"Sudip Mukherjee (Codethink)" <sudipm.mukherjee@gmail.com> Tested-by:
Maíra Canal <mairacanal@riseup.net> Reviewed-by:
Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Signed-off-by:
Nathan Chancellor <nathan@kernel.org> Signed-off-by:
-
Nathan Chancellor authored
Most of the arguments are identical between the two call sites and they can be accessed through the 'struct vba_vars_st' pointer. This reduces the total amount of stack space that dml31_ModeSupportAndSystemConfigurationFull() uses by 112 bytes with LLVM 16 (1976 -> 1864), helping clear up the following clang warning: drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn31/display_mode_vba_31.c:3908:6: error: stack frame size (2216) exceeds limit (2048) in 'dml31_ModeSupportAndSystemConfigurationFull' [-Werror,-Wframe-larger-than] void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib) ^ 1 error generated. Link: https://github.com/ClangBuiltLinux/linux/issues/1681Reported-by: -
Nathan Chancellor authored
drm/amd/display: Reduce number of arguments of dml31's CalculateWatermarksAndDRAMSpeedChangeSupport() Most of the arguments are identical between the two call sites and they can be accessed through the 'struct vba_vars_st' pointer. This reduces the total amount of stack space that dml31_ModeSupportAndSystemConfigurationFull() uses by 240 bytes with LLVM 16 (2216 -> 1976), helping clear up the following clang warning: drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn31/display_mode_vba_31.c:3908:6: error: stack frame size (2216) exceeds limit (2048) in 'dml31_ModeSupportAndSystemConfigurationFull' [-Werror,-Wframe-larger-than] void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib) ^ 1 error generated. Link: https://github.com/ClangBuiltLinux/linux/issues/1681Reported-by: -
Nathan Chancellor authored
Several of the arguments are identical between the two call sites and they can be accessed through the 'struct vba_vars_st' pointer. This reduces the total amount of stack space that dml32_ModeSupportAndSystemConfigurationFull() uses by 208 bytes with LLVM 16 (1936 -> 1728), helping clear up the following clang warning: drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/display_mode_vba_32.c:1721:6: error: stack frame size (2152) exceeds limit (2048) in 'dml32_ModeSupportAndSystemConfigurationFull' [-Werror,-Wframe-larger-than] void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib) ^ 1 error generated. Additionally, while modifying the arguments to dml32_CalculatePrefetchSchedule(), use 'v' consistently, instead of 'v' mixed with 'mode_lib->vba'. Link: https://github.com/ClangBuiltLinux/linux/issues/1681Reported-by: -
Nathan Chancellor authored
drm/amd/display: Reduce number of arguments of dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport() Most of the arguments are identical between the two call sites and they can be accessed through the 'struct vba_vars_st' pointer created at the top of dml32_ModeSupportAndSystemConfigurationFull(). This reduces the total amount of stack space that dml32_ModeSupportAndSystemConfigurationFull() uses by 216 bytes with LLVM 16 (2152 -> 1936), helping clear up the following clang warning: drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/display_mode_vba_32.c:1721:6: error: stack frame size (2152) exceeds limit (2048) in 'dml32_ModeSupportAndSystemConfigurationFull' [-Werror,-Wframe-larger-than] void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib) ^ 1 error generated. Additionally, while modifying the arguments to dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(), use 'v' consistently, instead of 'v' mixed with 'mode_lib->vba'. Link: https://github.com/ClangBuiltLinux/linux/issues/1681Reported-by: -
Aric Cyr authored
This version brings along following fixes: *Fixed register definitions for DCN32/321 *Adding log for spread_spectrum_info *Reverted DSC programming sequence change *Correct I2C register offset *Updated DCN30 header files, viewport position for phantom pipes *Enable dlg and vba compilation for dcn314 Tested-by:
Daniel Wheeler <daniel.wheeler@amd.com> Acked-by:
Pavle Kotarac <Pavle.Kotarac@amd.com> Signed-off-by:
Aric Cyr <aric.cyr@amd.com> Signed-off-by:
-
Aurabindo Pillai authored
[Why & How] Fix the instatiation sequence for MPC registers and add a few other missing register definitions that were ommited erroneously when copying them over to enable runtime initialization of reigster offsets for DCN32/321 Tested-by:
Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by:
-
Alvin Lee authored
Refactor calculation to remove floating point operations from dmub_srv. To ensure that 32-bit compilation works well, we use the div64 family of macros to do integer division for SubVP-related timing parameters. Cc: Maíra Canal <mairacanal@riseup.net> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Isabella Basso <isabbasso@riseup.net> Cc: Magali Lemes <magalilemes00@gmail.com> Tested-by:
Samson Tam <Samson.Tam@amd.com> Acked-by:
Tom Chung <chiahsuan.chung@amd.com> Signed-off-by:
Alvin Lee <alvin.lee2@amd.com> Co-developed-by:
-
JeromeHong authored
[why] Call commit_minimal_transition_state wrongly in case of surface_count equal to 0. [how] Add a condition to filter case of surface_count equal to 0. Tested-by:
JeromeHong <jerome.hong@amd.com> Signed-off-by:
-
Aric Cyr authored
[why] Do not need to spend time reprogramming DRR if there were no updates to the parameters. [how] Compare the current stream state to the requested one to determine if an update is required. In amdgpu_dm the timing_changed flag is set but never used so can remove it. Similarly, the stream update for VRR is done after dc_commit and should not update its adjust field until after the update is completed. The adjust field is managed by dc_stream_adjust_vmin_vmax and should not be manually updated in amdgpu_dm. Tested-by:
Anthony Koo <Anthony.Koo@amd.com> Acked-by:
-
Leo Chen authored
[Why & How] Enable logging for spread_spectrum_percentage in spread_spectrum_info to facilitate debugging for audio compliance issues Tested-by:
Leo Chen <SanChuan.Chen@amd.com> Reviewed-by:
Charlene Liu <Charlene.Liu@amd.com> Acked-by:
-
Yao Wang1 authored
[Why] For HDR mode, we get total 512 tf_point and after switching to SDR mode we actually get 400 tf_point and the rest of points(401~512) still use dirty value from HDR mode. We should limit the rest of the points to max value. [How] Limit the value when coordinates_x.x > 1, just like what we do in translate_from_linear_space for other re-gamma build paths. Tested-by:
Krunoslav Kovac <Krunoslav.Kovac@amd.com> Reviewed-by:
Yao Wang1 <Yao.Wang1@amd.com> Signed-off-by:
-
Aurabindo Pillai authored
[Why&How] Kernel does not have a means to tell the userspace to use software cursor. Due to lack of this functionality, reducing the max cursor size is the only way to ensure that power savings of Subview port feature is utilized for asics that support it. The workaround could be removed after cursor caching is fixed while a subviewport config is active. Tested-by:
-
Lee, Alvin authored
[Description] Regkey option for forcing num ways for subvp for debug purposes Tested-by:
Jun Lei <Jun.Lei@amd.com> Acked-by:
-
Alvin Lee authored
[Description] In some cases the viewport position of the main pipes can change without triggering a full update. In this case the subvp phantom viewports must be updated accordingly. Tested-by:
-
Taimur Hassan authored
[Why & How] When calculating cursor size for MALL allocation, the cursor width should be the actual width rounded up to 64 alignment. Additionally, the bit depth should vary depending on color format. Tested-by:
Taimur Hassan <Syed.Hassan@amd.com> Signed-off-by:
-
Duncan Ma authored
[Why] The interpretation of the number of memory channels differ by memory type, and this affects channel width for the DML input. [How] Set dram channel width according to memory type for dcn314. Tested-by:
Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com> Reviewed-by:
Duncan Ma <duncan.ma@amd.com> Signed-off-by:
-
Nicholas Kazlauskas authored
[Why] HW can support the display swizzle modes for video, and those are preferable over standard or linear for decode use. [How] Remove the check for DCN314. Tested-by:
-
Nicholas Kazlauskas authored
[Why & How] Add support for the DML314 functions and hook up DCN314 to use them. This has some necessary additions for calculating Max VSTARTUP for future features, but there's also some changes that we have to make for pixel format/swizzle support. That will come in a following patch to make this transition easier to bisect. Tested-by:
-
Rodrigo Siqueira authored
We were not using the VBA and DLG files for DCN314, but the next sequence of changes for DCN314 will require those files. This commit adds the necessary files to the Makefile. Cc: Roman Li <roman.li@amd.com> Tested-by:
-
Rodrigo Siqueira authored
We have some compilation errors in some DML files from DCN314 that we never noticed because we were not compiling some of the DML files. This commit fixes those syntax errors before we enable the compilation. Cc: Roman Li <roman.li@amd.com> Tested-by:
-
Aric Cyr authored
[why] Incorrectly using MicroTileWidth instead of MacroTileWidth for calculations. [how] Remove all unused references to MicroTile and change them to MacroTile. Tested-by:
-
Leo Chen authored
[Why & How] DIG_FIFO_READ_START_LEVEL should only be set to default value (7) by software. Removed all instances of resetting the register to 0 Tested-by:
Leo Chen <sancchen@amd.com> Signed-off-by:
-
