- 12 Jul, 2022 2 commits
-
-
Alex Deucher authored
For GMC 10 parts which support scatter/gather display (display from system memory), we should allocate a larger gart size to better handler larger displays. This mirrors what we already do for GMC 9 parts. v2: fix typo (Alex) Acked-by:
Christian König <christian.koenig@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Jack Xiao authored
Need reserve buffers before unmap mes ctx bo va. v2: fix removal of dma_resv_excl_fence() (Alex) v3: fix dma_resv_usage (Alex) Signed-off-by:
Jack Xiao <Jack.Xiao@amd.com> Reviewed-by:
Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by:
-
- 08 Jul, 2022 6 commits
-
-
Jack Xiao authored
For some cases (accessing registers, unmap legacy queue), it needs access mes in atomic context. Use spinlock to protect agaist mes ring buffer race condition. Signed-off-by:
-
Kenneth Feng authored
drop the thermal_controller_type check since it's not relevant. Signed-off-by:
Kenneth Feng <kenneth.feng@amd.com> Reviewed-by:
Evan Quan <evan.quan@amd.com> Signed-off-by:
-
Kenneth Feng authored
fixed the issue: gpu runs in dc mode but it is expected to be in ac mode. this causes the lower performance on smu_v13_0 Signed-off-by:
-
Aurabindo Pillai authored
[Why&How] Output Data Mapping is a power saving feature that allows us to run at reduced DPP and DISP clocks compared to what could be achieved with a single pipe. Set the default policy for single display use case to use 2 to 1 ODM combine. The options are queried by DC and appropriate register programming sequence is initiated to enable this feature. Fixes: 235c6763 ("drm/amd/display: add DCN32/321 specific files for Display Core") Signed-off-by:
Aurabindo Pillai <aurabindo.pillai@amd.com> Acked-by:
-
Aurabindo Pillai authored
[Why&How] Add a missing callback to set DIG FIFO output pixel mode. This is used when ODM combine is activated. Fixes: 235c6763 ("drm/amd/display: add DCN32/321 specific files for Display Core") Signed-off-by:
-
Alex Deucher authored
There are several things wrong here. First, none of these numbers are FP, so there is no need to cast to double. Next make sure to use proper 64 bit division helpers. Fixes: 85f4bc0c ("drm/amd/display: Add SubVP required code") Reviewed-by:
Harry Wentland <harry.wentland@amd.com> Signed-off-by:
-
- 07 Jul, 2022 14 commits
-
-
Yang Li authored
Eliminate the follow smatch warnings: drivers/gpu/drm/amd/amdgpu/../display/dc/bios/bios_parser2.c:405 get_bios_object_from_path_v3() warn: inconsistent indenting drivers/gpu/drm/amd/amdgpu/../display/dc/bios/bios_parser2.c:611 bios_parser_get_hpd_info() warn: inconsistent indenting drivers/gpu/drm/amd/amdgpu/../display/dc/bios/bios_parser2.c:818 bios_parser_get_device_tag() warn: inconsistent indenting drivers/gpu/drm/amd/amdgpu/../display/dc/bios/bios_parser2.c:1599 bios_parser_is_device_id_supported() warn: inconsistent indenting Signed-off-by:
Yang Li <yang.lee@linux.alibaba.com> Signed-off-by:
-
Alex Deucher authored
Need to protect FP DMCUB code with CONFIG_DRM_AMD_DC_DCN. Fixes build failures like the following on arm64: ERROR: modpost: "__floatunsidf" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined! ERROR: modpost: "__divdf3" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined! ERROR: modpost: "fma" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined! ERROR: modpost: "__adddf3" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined! ERROR: modpost: "__fixdfsi" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined! ERROR: modpost: "__muldf3" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined! ERROR: modpost: "__floatsidf" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined! ERROR: modpost: "__fixunsdfsi" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined! Fixes: 85f4bc0c ("drm/amd/display: Add SubVP required code") Reported-by:
Nathan Chancellor <nathan@kernel.org> Tested-by:
-
Stylon Wang authored
[Why] Changes from "Fix for dmub outbox notification enable" need to land in DM or DMUB outbox notification would be disabled. [How] Enable outbox notification only after interrupt are enabled and IRQ handlers registered. Any pending notification will be sent by DMUB once outbox notification is enabled. Fixes: ed720870 ("drm/amd/display: Fix for dmub outbox notification enable") Reviewed-by:
Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com> Acked-by:
Solomon Chiu <solomon.chiu@amd.com> Signed-off-by:
Stylon Wang <stylon.wang@amd.com> Acked-by:
-
xinhui pan authored
Queue would be freed when create_queue_cpsch fails So lets do queue cleanup otherwise various list and memory issues happen. Signed-off-by:
xinhui pan <xinhui.pan@amd.com> Reviewed-by:
Philip Yang <Philip.Yang@amd.com> Signed-off-by:
-
Mateusz Jończyk authored
I was getting the following message on boot on Linux 5.19-rc5: radeon 0000:01:05.0: vram limit (0) must be a power of 2 (I didn't use any radeon.vramlimit commandline parameter). This is caused by commit 8c2d34eb ("drm/radeon: use kernel is_power_of_2 rather than local version") which removed radeon_check_pot_argument() and converted its users to is_power_of_2(). The two functions differ in its handling of 0, which is the default value of radeon_vram_limit: radeon_check_pot_argument() "incorrectly" considered it a power of 2, while is_power_of_2() does not. An appropriate conditional silences the warning message. It is not necessary to add a similar test to other callers of is_power_of_2() in radeon_device.c. The matching commit in amdgpu: commit 76117507 ("drm/amdgpu: use kernel is_power_of_2 rather than local version") is unaffected by this bug. Tested on Radeon HD 3200. Not ccing stable, this is not serious enough. Fixes: 8c2d34eb ("drm/radeon: use kernel is_power_of_2 rather than local version") Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: "Pan, Xinhui" <Xinhui.Pan@amd.com> Cc: David Airlie <airlied@linux.ie> Cc: Daniel Vetter <daniel@ffwll.ch> Cc: Jonathan Gray <jsg@jsg.id.au> Signed-off-by:Mateusz Jończyk <mat.jonczyk@o2.pl> Signed-off-by:
-
Guo Zhengkui authored
Remove a repeated "#include <drm/drm_print.h>" Reviewed-by:
Guo Zhengkui <guozhengkui@vivo.com> Signed-off-by:
-
Zhongjun Tan authored
Remove duplicated argument to || Signed-off-by:
Zhongjun Tan <tanzhongjun@coolpad.com> Signed-off-by:
-
Alex Deucher authored
It's not used outside of dcn30_resource.c. Reported-by:
kernel test robot <lkp@intel.com> Reviewed-by:
-
Alex Deucher authored
Fixes the following: drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_hwseq.c:428:33: warning: variable 'old_pipe' set but not used Reported-by:
-
Eric Huang authored
It is to avoid unnecessary queue eviction when range is not mapped to gpu. Signed-off-by:
Eric Huang <jinhuieric.huang@amd.com> Reviewed-by:
Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by:
-
Evan Quan authored
[Why] The return value indicates whether the operation(disable/enable) succeeded or not. The existing logic reports wrong result even if the disablement was performed successfully. That will make succeeding reenablement abandoned as dc->idle_optimizations_allowed is always true. [How] Correct the return value to reflect the real result of disablement. Fixes: 235c6763 ("drm/amd/display: add DCN32/321 specific files for Display Core") Signed-off-by:
-
Aurabindo Pillai authored
[Why&How] The debug option to disable idle power optimization can be dropped Fixes: 235c6763 ("drm/amd/display: add DCN32/321 specific files for Display Core") Signed-off-by:
-
Eric Huang authored
Adding always evict queues when flag is set to KFD_IOCTL_SVM_FLAG_GPU_ALWAYS_MAPPED as if XNACK off. Signed-off-by:
-
Eric Huang authored
It is to add new option for always keeping gpu mapping. Signed-off-by:
-
- 05 Jul, 2022 18 commits
-
-
Alex Deucher authored
Seems to break hibernation. Disable for now until we can root cause it. Fixes: 087451f3 ("drm/amdgpu: use generic fb helpers instead of setting up AMD own's.") Bug: https://bugzilla.kernel.org/show_bug.cgi?id=216119Acked-by:
-
Alex Deucher authored
Was dropped when we converted to the generic helpers. Fixes: 087451f3 ("drm/amdgpu: use generic fb helpers instead of setting up AMD own's.") Acked-by:
-
Maíra Canal authored
Some variables from the struct vba_vars_st are not referenced in any other place on the codebase. As they are not used, this commit removes those variables. Signed-off-by:
Maíra Canal <mairacanal@riseup.net> Signed-off-by:
-
Maíra Canal authored
The function CalculateBytePerPixelAnd256BBlockSizes was defined four times: on display_mode_vba_30.c, display_rq_dlg_calc_30.c, display_mode_vba_31.c and display_rq_dlg_calc_31.c. In order to avoid code duplication, the CalculateBytePerPixelAnd256BBlockSizes is defined on display_mode_vba_30.h and used across dcn30 and dcn31. Signed-off-by:
-
Yang Li authored
Eliminate the follow smatch warning: drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c:3151 commit_planes_for_stream() warn: inconsistent indenting Signed-off-by:
-
Maíra Canal authored
The function Calculate256BBlockSizes always returns true, regardless of the parameters. As any file checks the return of the function, this commit changes the return value to void. Signed-off-by:
-
André Almeida authored
To avoid code repetition, unify the function exit path when possible. No functional changes. Acked-by:
André Almeida <andrealmeid@igalia.com> Signed-off-by:
-
Jianglei Nie authored
if amdgpu_mes_ctx_alloc_meta_data() fails, we should call amdgpu_vm_fini() to handle amdgpu_vm_init(). Add a new lable before amdgpu_vm_init() and goto this lable when amdgpu_mes_ctx_alloc_meta_data() fails. Signed-off-by:
Jianglei Nie <niejianglei2021@163.com> Signed-off-by:
-
Alvin Lee authored
[Description] Program audio DTO before wall dto for audio Reviewed-by:
Martin Leung <Martin.Leung@amd.com> Acked-by:
Qingqing Zhuo <qingqing.zhuo@amd.com> Signed-off-by:
Alvin Lee <Alvin.Lee2@amd.com> Tested-by:
Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by:
-
Aric Cyr authored
This DC patchset brings improvements in multiple areas. In summary, we highlight: - Program ACP-related registers - Fixes for DMUB, DPIA, PSR, and others - Improvements in the pipe split - Add SubVP code - Add basic setup for FAMS support - Improve BB capabilities Acked-by:
Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Signed-off-by:
Aric Cyr <aric.cyr@amd.com> Tested-by:
-
Hamza Mahfooz authored
hdmi_frl_pcon_support has been the source of confusion. So, rename it to dp_hdmi21_pcon_support. Signed-off-by:
Hamza Mahfooz <hamza.mahfooz@amd.com> Tested-by:
-
Hamza Mahfooz authored
Generic PCON SST support already exists and works for newer ASICs. So, enable it by default. Acked-by:
-
Chris Park authored
[Why] Integrate OVT timing from DM to DC logic to update info frame and mode management to report the resolution to the OS. [How] Reflect RID and Frame Rate to AVI InfoFrame Version 5. Define new Timing Standard for OVT timing. Reviewed-by:
Charlene Liu <Charlene.Liu@amd.com> Acked-by:
Alan Liu <HaoPing.Liu@amd.com> Signed-off-by:
Chris Park <Chris.Park@amd.com> Tested-by:
-
Jack Xiao authored
MES fw updated to support unmapping legacy gfx/compute queue. Signed-off-by:
-
Michael Strauss authored
[WHY] lt_settings' pointers remain uninitialized but nonzero if display fails to light up with no DPCD/EDID info populated, leading to a hang on access Reviewed-by:
Michael Strauss <michael.strauss@amd.com> Tested-by:
-
Jimmy Kizito authored
[Why] Uninitialized variable causes diag compilation build failure. [How] - Ensure that variable in question is always initialized before being used. - The variable in question is the USB4 DP training pattern. In case an unsupported training pattern has been requested, update status accordingly and abort current link training attempt. Reviewed-by:
Meenakshikumar Somasundaram <Meenakshikumar.Somasundaram@amd.com> Acked-by:
Jimmy Kizito <Jimmy.Kizito@amd.com> Tested-by:
-
Harry Wentland authored
Move all linux includes into OS types. Acked-by:
-
Eric Bernstein authored
[Why] For some customer blending transition cases, the available pipe for second stream is a pipe index that is greater than the number of timing generators, which can cause a problem in acquire_first_free_pipe since it assumes same index for pipe and timing generator [How] Added logic to use last timing generator index if the pipe index is greater than number of timing generators. Acked-by:
Eric Bernstein <eric.bernstein@amd.com> Tested-by:
-
