linux-dev/drivers/gpu/drm/amd/display/dc/inc, branch master

drm/amd/display: add a license to cursor_reg_cache.h

2022-10-11T15:33:16Z

It's MIT. Fixes: b73353f7f3d434 ("drm/amd/display: Use the same cursor info across features") Reviewed-by: Harry Wentland Signed-off-by: Alex Deucher

drm/amd/display: Use the same cursor info across features

2022-10-10T21:32:55Z

Since different features would need to update cursor registers, However, they would use different approaches. To unify varied methods, this refactor is implemented the same update cursor info method for current varied features. Tested-by: Daniel Wheeler Reviewed-by: Anthony Koo Reviewed-by: Jun Lei Acked-by: Qingqing Zhuo Signed-off-by: Max Tseng Signed-off-by: Alex Deucher

drm/amd/display: Acquire FCLK DPM levels on DCN32

2022-10-10T21:32:55Z

[Why & How] Acquire FCLK DPM levels to properly construct DML clock limits. Further add new logic to keep number of indices for each clock in clk_mgr. Tested-by: Daniel Wheeler Reviewed-by: Jun Lei Acked-by: Qingqing Zhuo Signed-off-by: Dillon Varone Signed-off-by: Alex Deucher

drm/amd/display: Add a helper to map ODM/MPC/Multi-Plane resources

2022-10-10T21:32:38Z

[Why & How] Add a helper to map ODM/MPC/Multi-Plane resources from DC Tested-by: Daniel Wheeler Reviewed-by: Nevenko Stupar Reviewed-by: Chaitanya Dhere Acked-by: Qingqing Zhuo Signed-off-by: Jun Lei Signed-off-by: Alex Deucher

drm/amd/display: Remove OPTC lock check

2022-10-06T16:02:04Z

At some point, we decided to blank HUBP during pixel data blank, and to handle that, we added some OPTC lock checks. Later, we realized that this change caused multiple regression, and we removed it. Nevertheless, we still have some leftovers that might affect some ASIC behavior, and this commit drops those changes to keep the code consistent. Reviewed-by: Aurabindo Pillai Signed-off-by: Rodrigo Siqueira Signed-off-by: Alex Deucher

drm/amd/display: Refactor edp ILR caps codes

2022-10-06T16:00:02Z

We split out ILR config from "global" to "per-panel" config settings. Reviewed-by: Anthony Koo Acked-by: Hamza Mahfooz Signed-off-by: Ian Chen Signed-off-by: Alex Deucher

drm/amd/display: Fix mutex lock in dcn10

2022-10-06T15:54:35Z

Removal of DC_FP_* wrappers from dml (9696679bf7ac) provokes a mutex lock [2] on the amdgpu driver. Re-arrange the dcn10 code to avoid locking the mutex by placing the DC_FP_* wrappers around the proper functions. This fixes the following WARN/stacktrace: BUG: sleeping function called from invalid context at kernel/locking/mutex.c:283 in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 227, name: systemd-udevd preempt_count: 1, expected: 0 CPU: 4 PID: 227 Comm: systemd-udevd Not tainted 6.0.0-rc6-qtec-standard #2 Hardware name: Qtechnology A/S QT5222/QT5221, BIOS v1.0.1 06/07/2021 Call Trace: dump_stack_lvl+0x33/0x42 __might_resched.cold.172+0xa5/0xb3 mutex_lock+0x1a/0x40 amdgpu_dpm_get_clock_by_type_with_voltage+0x38/0x70 [amdgpu] dm_pp_get_clock_levels_by_type_with_voltage+0x64/0xa0 [amdgpu] dcn_bw_update_from_pplib+0x70/0x340 [amdgpu] dcn10_create_resource_pool+0x8c8/0xd20 [amdgpu] ? __kmalloc+0x1c7/0x4a0 dc_create_resource_pool+0xe7/0x190 [amdgpu] dc_create+0x212/0x5d0 [amdgpu] amdgpu_dm_init+0x246/0x370 [amdgpu] ? schedule_hrtimeout_range_clock+0x93/0x120 ? phm_wait_for_register_unequal.part.1+0x4a/0x80 [amdgpu] dm_hw_init+0xe/0x20 [amdgpu] amdgpu_device_init.cold.56+0x1324/0x1653 [amdgpu] ? pci_bus_read_config_word+0x43/0x80 amdgpu_driver_load_kms+0x15/0x120 [amdgpu] amdgpu_pci_probe+0x116/0x320 [amdgpu] pci_device_probe+0x97/0x110 really_probe+0xdd/0x340 __driver_probe_device+0x80/0x170 driver_probe_device+0x1f/0x90 __driver_attach+0xdc/0x180 ? __device_attach_driver+0x100/0x100 ? __device_attach_driver+0x100/0x100 bus_for_each_dev+0x74/0xc0 bus_add_driver+0x19e/0x210 ? kset_find_obj+0x30/0xa0 ? 0xffffffffa0a5b000 driver_register+0x6b/0xc0 ? 0xffffffffa0a5b000 do_one_initcall+0x4a/0x1f0 ? __vunmap+0x28e/0x2f0 ? __cond_resched+0x15/0x30 ? kmem_cache_alloc_trace+0x3d/0x440 do_init_module+0x4a/0x1e0 load_module+0x1cba/0x1e10 ? __do_sys_finit_module+0xb7/0x120 __do_sys_finit_module+0xb7/0x120 do_syscall_64+0x3c/0x80 entry_SYSCALL_64_after_hwframe+0x63/0xcd RIP: 0033:0x7ff2b5f5422d Code: 5d c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d c3 ab 0e 00 f7 d8 64 89 01 48 RSP: 002b:00007ffc44ab28e8 EFLAGS: 00000246 ORIG_RAX: 0000000000000139 RAX: ffffffffffffffda RBX: 0000555c566a9240 RCX: 00007ff2b5f5422d RDX: 0000000000000000 RSI: 00007ff2b60bb353 RDI: 0000000000000019 RBP: 00007ff2b60bb353 R08: 0000000000000000 R09: 0000555c566a9240 R10: 0000000000000019 R11: 0000000000000246 R12: 0000000000000000 R13: 0000000000020000 R14: 0000000000000000 R15: 0000000000000000 Fixes: 9696679bf7ac ("drm/amd/display: remove DC_FP_* wrapper from dml folder") Reviewed-by: Hamza Mahfooz Tested-by: Daniel Wheeler Signed-off-by: Daniel Gomez Signed-off-by: Alex Deucher

drm/amd/display: Refactor LTTPR mode selection

2022-09-29T13:41:46Z

[WHY] Previously, LTTPR mode was decided during detection which makes link training inflexible as mode can't be dynamically changed. [HOW] -Remove lttpr_mode from link struct, and move to link training settings -Defer choosing LTTPR mode until link training Other DP changes included: -Only use fixed vs/pe link training sequence for 8b/10b encoding -Restrict fixed vs aux timeout workaround to Yellow Carp family Reviewed-by: Wenjing Liu Signed-off-by: Michael Strauss Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher

drm/amd/display: Fix various dynamic ODM transitions on DCN32

2022-09-29T13:41:45Z

[Why&How] Several transitions were fixed that will allow Dynamic ODM and MPO transitions to be supported on DCN32. 1) Due to resource limitations, in certain scenarios that require an MPO plane to be split, the features cannot be combined with the current policy. This is due to unsafe transitions being required (OPP instance per MPCC being switched on active pipe is not supported by DCN), to support the split plane with ODM active as it moves across the viewport. Dynamic ODM will now be disabled when MPO is required. 2) When exiting MPO and re-entering ODM, DC assigns an inactive pipe for the next ODM pipe, which under previous power gating policy would result in programming a gated DSC HW block. New policy dynamically gates/un-gates DSC blocks when Dynamic ODM is active to support transitions on DCN32 only. 3) Entry and exit from 3 plane MPO and Dynamic ODM requires a minimal transition so that all pipes which require their MPCC OPP instance to be changed have a full frame to be disabled before reprogramming. To solve this, the Dynamic ODM policy now utilizes minimal state transitions when entering or exiting 3 plane scenarios. 4) Various fixes to DCN32 pipe merge/split algorithm to support Dynamic ODM and MPO transitions. In summary, this commit fixes various transitions to support ODM->MPO and MPO->ODM. Reviewed-by: Martin Leung Reviewed-by: Jun Lei Acked-by: Jasdeep Dhillon Signed-off-by: Dillon Varone Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher

drm/amd/display: fill in clock values when DPM is not enabled

2022-09-29T13:41:45Z

[Why] For individual feature testing, PMFW may not report all clock values back. Driver will default them to 0 but this will cause the BB table to be skipped and default to one state with max clocks. [How] Add helper function to scan through initial clock values and populate them with default clock limits so that BB table can be built. Add dpm_enabled flag to check when DPM is not enabled and to trigger helper function. Reviewed-by: Jun Lei Acked-by: Jasdeep Dhillon Signed-off-by: Samson Tam Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher