aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/gpu
AgeCommit message (Collapse)AuthorFilesLines
2021-10-28drm/amd/display: 3.2.159Aric Cyr1-1/+1
This new DC version brings improvements in the following areas: - Improvements for USB4; - Isolate FPU code for DCN20, DCN301, and DSC; - Fixes on Linking training; - Refactoring some parts of the code, such as PSR; Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Aric Cyr <aric.cyr@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: [FW Promotion] Release 0.0.90Anthony Koo1-2/+2
Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Anthony Koo <Anthony.Koo@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: Remove unused macrosAnson Jacob3-8/+0
fixed16_to_double & fixed16_to_double_to_cpu are not used. Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Anson Jacob <Anson.Jacob@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: allow windowed mpo + odmDmytro Laktyushkin3-17/+28
This change adds a config flag to allow non fullscreen MPO during ODM. Scaling calculation will still fail configurations where video is only one one side of the screen. Reviewed-by: Aric Cyr <Aric.Cyr@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: set Layout properly for 8ch audio at timing validationGuo, Bing1-2/+4
Why: For audio packet type 0x02, there are 2 Layouts: Layout = 0 for 2 channels and Layout = 1 for > 2 channels. Layout will affect bandwidth check. Currently, for HDMI FRL, Layout field isn't set and has a default value of 0, so theoretically only 2-channel audio for audio packet type 0x02 is supported now. How: 1. Set Layout properly according to maximum audio channel numbers for audios with audio packet type 0x02. 2. 8ch LPCM audio is not supported for timing modes with v_active <= 576. Reviewed-by: Chris Park <chris.park@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Bing Guo <Bing.Guo@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: Fix 3DLUT skipped programmingAric Cyr2-0/+5
We found a compilation error that we thought was caused by the 3DLUT patch; later on, we figured out the root cause of the problem, but we already applied the revert in the wrong patch. This commit brings it back the 3DLUT fix. Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Aric Cyr <aric.cyr@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: 3.2.158Aric Cyr1-1/+1
Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Aric Cyr <aric.cyr@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: [FW Promotion] Release 0.0.89Anthony Koo1-3/+6
- Add flag to control root clock gating in init_hw - Add flag to indicate a diags environment is being used Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Anthony Koo <Anthony.Koo@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: Handle I2C-over-AUX write channel status updateAric Cyr4-10/+44
[Why] When writing long AUX commands some sinks will respond will write status update requiring source to read status. [How] When a write request is replied with data (AUX_ACK_M), retry a read of write status to determine when the write is completed. Reviewed-by: Martin Leung <Martin.Leung@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Aric Cyr <aric.cyr@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: Add comment for preferred_training_settingsGeorge Shen1-0/+4
[Why] Currently the naming of preferred_training_settings is ambiguous and has caused confusion regarding its purpose and usage. [How] Add comment to clarify the intention. Reviewed-by: Wenjing Liu <Wenjing.Liu@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: George Shen <george.shen@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: Implement fixed DP drive settingsGeorge Shen6-1/+27
[Why] Currently there are use cases that require DP link to maintain fixed VS and PE in HW regardless of what the sink requests. BIOS integrated info table will specify whether we need to use the fixed drive settings, and the drive settings to use. [How] Implement changes to parse the integrated info table and set the fixed drive settings accordingly. Reviewed-by: Wenjing Liu <Wenjing.Liu@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: George Shen <george.shen@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: restyle dcn31 resource header inline with other asicsDmytro Laktyushkin2-2/+12
Style change for better consistency across codebase Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: clean up dcn31 revision checkDmytro Laktyushkin1-8/+2
This is unnecessary in clk_mgr Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: Defer GAMCOR and DSCL power down sequence to vupdateMichael Strauss6-6/+38
[WHY] Every other CM LUT power down sequence is deferred to next vupdate as memory powerdown updates immediately while selecting LUTs is double buffered. Previous update to defer LUT power down missed GAMCOR and DSCL, causing some visible flicker when entering/exiting fullscreen video playback. [HOW] Update dpp deferred update loop to check for valid DPPs in res_pool instead of referencing dcn_ip which turns out to not be populated during runtime. Move GAMCOR and DSCL powerdown to dpp deferred updates. Reviewed-by: Haonan Wang <Haonan.Wang2@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Michael Strauss <michael.strauss@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: Set i2c memory to light sleep during hw initMichael Strauss3-2/+9
[WHY] i2c memory doesn't get set to light sleep on hw init as intended [HOW] Set i2c to light sleep after reg gets zeroed, ensuring memory power control doesn't get disabled for any other DIO memory Reviewed-by: Haonan Wang <Haonan.Wang2@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Michael Strauss <michael.strauss@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: fix a crash on USB4 over C20 PHYAhmad Othman1-5/+20
[Why] Crash when USB4 is connected. [How] Added an ASIC specific code guard. Reviewed-by: Nikola Cornij <Nikola.Cornij@amd.com> Reviewed-by: Wenjing Liu <Wenjing.Liu@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Ahmad Othman <Ahmad.Othman@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: move FPU associated DSC code to DML folderQingqing Zhuo8-338/+389
[Why & How] As part of the FPU isolation work documented in https://patchwork.freedesktop.org/series/93042/, isolate code that uses FPU in DSC to DML, where all FPU code should locate. This change does not refactor any functions but move code around. Cc: Christian König <christian.koenig@amd.com> Cc: Hersen Wu <hersenxs.wu@amd.com> Cc: Anson Jacob <Anson.Jacob@amd.com> Cc: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Tested-by: Anson Jacob <Anson.Jacob@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: Add support for USB4 on C20 PHY for DCN3.1Ahmad Othman4-6/+64
[Why] Created new fields that matches new B0 structs On DCN31 the mapping of DIO output to PHY differs from A0 to B0 boards with new PHY C20 & this new mapping needed to be handled. [How] Mapped new structure based on new structs Added logic for mapping over A0 and B0 boards Hooked all new structs together. Reviewed-by: Wenjing Liu <Wenjing.Liu@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Ahmad Othman <Ahmad.Othman@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: dc_link_set_psr_allow_active refactoringRobin Chen10-23/+125
[Why] To expose new power optimization flags to PSR interface. It allows the PSR related power features can be enabled separately base on different use scenarios. Reviewed-by: Anthony Koo <Anthony.Koo@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Robin Chen <po-tchen@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: Get ceiling for v_total calcGuo, Bing1-3/+12
Updating certain variable blanking calculations to use ceiling function. Reviewed-by: Chris Park <chris.park@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Bing Guo <Bing.Guo@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: dcn20_resource_construct reduce scope of FPU enabledAnson Jacob1-7/+9
Limit when FPU is enabled to only functions that does FPU operations for dcn20_resource_construct, which gets called during driver initialization. Enabling FPU operation disables preemption. Sleeping functions(mutex (un)lock, memory allocation using GFP_KERNEL, etc.) should not be called when preemption is disabled. Fixes the following case caught by enabling CONFIG_DEBUG_ATOMIC_SLEEP in kernel config [ 1.338434] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:281 [ 1.347395] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 197, name: systemd-udevd [ 1.356356] CPU: 7 PID: 197 Comm: systemd-udevd Not tainted 5.13.0+ #3 [ 1.356358] Hardware name: System manufacturer System Product Name/PRIME X570-PRO, BIOS 3405 02/01/2021 [ 1.356360] Call Trace: [ 1.356361] dump_stack+0x6b/0x86 [ 1.356366] ___might_sleep.cold+0x87/0x98 [ 1.356370] __might_sleep+0x4b/0x80 [ 1.356372] mutex_lock+0x21/0x50 [ 1.356376] smu_get_uclk_dpm_states+0x3f/0x80 [amdgpu] [ 1.356538] pp_nv_get_uclk_dpm_states+0x35/0x50 [amdgpu] [ 1.356711] init_soc_bounding_box+0xf9/0x210 [amdgpu] [ 1.356892] ? create_object+0x20d/0x340 [ 1.356897] ? dcn20_resource_construct+0x46f/0xd30 [amdgpu] [ 1.357077] dcn20_resource_construct+0x4b1/0xd30 [amdgpu] ... Tested on: 5700XT (NAVI10 0x1002:0x731F 0x1DA2:0xE410 0xC1) Cc: Christian König <christian.koenig@amd.com> Cc: Hersen Wu <hersenxs.wu@amd.com> Cc: Anson Jacob <Anson.Jacob@amd.com> Cc: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Anson Jacob <Anson.Jacob@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amd/display: Align bw context with hw config when system resumeLewis Huang1-0/+6
[Why] When the vbios config and driver config are different, if we update clock to lower before call program_timing and program_pixel_clk, garbage appear. [How] Align bw context with hw config when system resume Reviewed-by: Anthony Koo <Anthony.Koo@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Acked-by: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by: Lewis Huang <Lewis.Huang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdgpu/pm: look up current_level for asics without pm callbackAlex Deucher1-1/+3
For asics without a callback, use the current level rather than 0xff. This can avoid an unnecesary forced level set on older asics when set by the user. Reviewed-by: Kevin Wang <kevinyang.wang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdgpu/display: add quirk handling for stutter modeAlex Deucher1-0/+33
Stutter mode is a power saving feature on GPUs, however at least one early raven system exhibits stability issues with it. Add a quirk to disable it for that system. Bug: https://bugzilla.kernel.org/show_bug.cgi?id=214417 Fixes: 005440066f929b ("drm/amdgpu: enable gfxoff again on raven series (v2)") Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdgpu: remove GPRs init for ALDEBARAN in gpu reset (v3)Tao Zhou1-3/+3
Remove GPRs init for ALDEBARAN in gpu reset temporarily, will add the init once the algorithm is stable. v2: Only remove GPRs init in gpu reset. v3: Suspend needs it, only skip it in gpu reset. Signed-off-by: Tao Zhou <tao.zhou1@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdkfd: Remove cu mask from struct queue_properties(v2)Lang Yu8-54/+57
Actually, cu_mask has been copied to mqd memory and does't have to persist in queue_properties. Remove it from queue_properties. And use struct mqd_update_info to store such properties, then pass it to update queue operation. v2: * Rename pqm_update_queue to pqm_update_queue_properties. * Rename struct queue_update_info to struct mqd_update_info. * Rename pqm_set_cu_mask to pqm_update_mqd. Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Lang Yu <lang.yu@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdkfd: Add an optional argument into update queue operation(v2)Lang Yu9-35/+52
Currently, queue is updated with data in queue_properties. And all allocated resource in queue_properties will not be freed until the queue is destroyed. But some properties(e.g., cu mask) bring some memory management headaches(e.g., memory leak) and make code complex. Actually they have been copied to mqd and don't have to persist in queue_properties. Add an argument into update queue to pass such properties, then we can remove them from queue_properties. v2: Don't use void *. Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Lang Yu <lang.yu@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdgpu: skip GPRs init for some CU settings on ALDEBARANTao Zhou1-0/+5
Skip GPRs init in specific condition since current GPRs init algorithm only works for some CU settings. Signed-off-by: Tao Zhou <tao.zhou1@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdgpu: Update TA version output in driverCandice Li7-26/+26
TA version should only be displayed in firmware version column. Signed-off-by: Candice Li <candice.li@amd.com> Reviewed-by: John Clements <john.clements@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdgpu: fix a potential memory leak in amdgpu_device_fini_sw()Lang Yu1-1/+1
amdgpu_fence_driver_sw_fini() should be executed before amdgpu_device_ip_fini(), otherwise fence driver resource won't be properly freed as adev->rings have been tore down. Fixes: 72c8c97b1522 ("drm/amdgpu: Split amdgpu_device_fini into early and late") Signed-off-by: Lang Yu <lang.yu@amd.com> Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdkfd: Separate pinned BOs destruction from general routineLang Yu5-37/+118
Currently, all kfd BOs use same destruction routine. But pinned BOs are not unpinned properly. Separate them from general routine. v2 (Felix): Add safeguard to prevent user space from freeing signal BO. Kunmap signal BO in the event of setting event page error. Just kunmap signal BO to avoid duplicating the code. Signed-off-by: Lang Yu <lang.yu@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdkfd: restore userptr ignore bad address errorPhilip Yang2-10/+20
The userptr can be unmapped by application and still registered to driver, restore userptr work return user pages will get -EFAULT bad address error. Pretend this error as succeed. GPU access this userptr will have VM fault later, it is better than application soft hangs with stalled user mode queues. v2: squash in warning fix (Alex) Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdgpu: Add kernel parameter support for ignoring bad page thresholdKent Russell3-5/+13
When a GPU hits the bad_page_threshold, it will not be initialized by the amdgpu driver. This means that the table cannot be cleared, nor can information gathering be performed (getting serial number, BDF, etc). If the bad_page_threshold kernel parameter is set to -2, continue to initialize the GPU, while printing a warning to dmesg that this action has been done v2: squash in Luben's fix to restore RAS info reporting Cc: Luben Tuikov <luben.tuikov@amd.com> Cc: Mukul Joshi <Mukul.Joshi@amd.com> Signed-off-by: Kent Russell <kent.russell@amd.com> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Luben Tuikov <luben.tuikov@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/amdgpu: Warn when bad pages approaches 90% thresholdKent Russell1-0/+7
dmesg doesn't warn when the number of bad pages approaches the threshold for page retirement. WARN when the number of bad pages is at 90% or greater for easier checks and planning, instead of waiting until the GPU is full of bad pages. Cc: Luben Tuikov <luben.tuikov@amd.com> Cc: Mukul Joshi <Mukul.Joshi@amd.com> Signed-off-by: Kent Russell <kent.russell@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Luben Tuikov <luben.tuikov@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-10-28drm/i915: Fix icl+ combo phy static lane power down setupVille Syrjälä1-1/+1
Our lane power down defines already include the necessary shift, don't shift them a second time. Fortunately we masked off the correct bits, so we accidentally left all lanes powered up all the time. Bits 8-11 where we end up writing our misdirected lane mask are documented as MBZ, but looks like you can actually write there so they're not read only bits. No idea what side effect the bogus register write might have. Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/4151 Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20211006204937.30774-17-ville.syrjala@linux.intel.com Reviewed-by: Imre Deak <imre.deak@intel.com>
2021-10-28drm/i915: Fix type1 DVI DP dual mode adapter heuristic for modern platformsVille Syrjälä1-22/+63
Looks like we never updated intel_bios_is_port_dp_dual_mode() when the VBT port mapping became erratic on modern platforms. This is causing us to look up the wrong child device and thus throwing the heuristic off (ie. we might end looking at a child device for a genuine DP++ port when we were supposed to look at one for a native HDMI port). Fix it up by not using the outdated port_mapping[] in intel_bios_is_port_dp_dual_mode() and rely on intel_bios_encoder_data_lookup() instead. Cc: stable@vger.kernel.org Tested-by: Randy Dunlap <rdunlap@infradead.org> Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/4138 Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20211025142147.23897-1-ville.syrjala@linux.intel.com Reviewed-by: Jani Nikula <jani.nikula@intel.com>
2021-10-28drm/i915/pmu: Connect engine busyness stats from GuC to pmuUmesh Nerlige Ramappa13-28/+453
With GuC handling scheduling, i915 is not aware of the time that a context is scheduled in and out of the engine. Since i915 pmu relies on this info to provide engine busyness to the user, GuC shares this info with i915 for all engines using shared memory. For each engine, this info contains: - total busyness: total time that the context was running (total) - id: id of the running context (id) - start timestamp: timestamp when the context started running (start) At the time (now) of sampling the engine busyness, if the id is valid (!= ~0), and start is non-zero, then the context is considered to be active and the engine busyness is calculated using the below equation engine busyness = total + (now - start) All times are obtained from the gt clock base. For inactive contexts, engine busyness is just equal to the total. The start and total values provided by GuC are 32 bits and wrap around in a few minutes. Since perf pmu provides busyness as 64 bit monotonically increasing values, there is a need for this implementation to account for overflows and extend the time to 64 bits before returning busyness to the user. In order to do that, a worker runs periodically at frequency = 1/8th the time it takes for the timestamp to wrap. As an example, that would be once in 27 seconds for a gt clock frequency of 19.2 MHz. Note: There might be an over-accounting of busyness due to the fact that GuC may be updating the total and start values while kmd is reading them. (i.e kmd may read the updated total and the stale start). In such a case, user may see higher busyness value followed by smaller ones which would eventually catch up to the higher value. v2: (Tvrtko) - Include details in commit message - Move intel engine busyness function into execlist code - Use union inside engine->stats - Use natural type for ping delay jiffies - Drop active_work condition checks - Use for_each_engine if iterating all engines - Drop seq locking, use spinlock at GuC level to update engine stats - Document worker specific details v3: (Tvrtko/Umesh) - Demarcate GuC and execlist stat objects with comments - Document known over-accounting issue in commit - Provide a consistent view of GuC state - Add hooks to gt park/unpark for GuC busyness - Stop/start worker in gt park/unpark path - Drop inline - Move spinlock and worker inits to GuC initialization - Drop helpers that are called only once v4: (Tvrtko/Matt/Umesh) - Drop addressed opens from commit message - Get runtime pm in ping, remove from the park path - Use cancel_delayed_work_sync in disable_submission path - Update stats during reset prepare - Skip ping if reset in progress - Explicitly name execlists and GuC stats objects - Since disable_submission is called from many places, move resetting stats to intel_guc_submission_reset_prepare v5: (Tvrtko) - Add a trylock helper that does not sleep and synchronize PMU event callbacks and worker with gt reset v6: (CI BAT failures) - DUTs using execlist submission failed to boot since __gt_unpark is called during i915 load. This ends up calling the GuC busyness unpark hook and results in kick-starting an uninitialized worker. Let park/unpark hooks check if GuC submission has been initialized. - drop cant_sleep() from trylock helper since rcu_read_lock takes care of that. v7: (CI) Fix igt@i915_selftest@live@gt_engines - For GuC mode of submission the engine busyness is derived from gt time domain. Use gt time elapsed as reference in the selftest. - Increase busyness calculation to 10ms duration to ensure batch runs longer and falls within the busyness tolerances in selftest. v8: - Use ktime_get in selftest as before - intel_reset_trylock_no_wait results in a lockdep splat that is not trivial to fix since the PMU callback runs in irq context and the reset paths are tightly knit into the driver. The test that uncovers this is igt@perf_pmu@faulting-read. Drop intel_reset_trylock_no_wait, instead use the reset_count to synchronize with gt reset during pmu callback. For the ping, continue to use intel_reset_trylock since ping is not run in irq context. - GuC PM timestamp does not tick when GuC is idle. This can potentially result in wrong busyness values when a context is active on the engine, but GuC is idle. Use the RING TIMESTAMP as GPU timestamp to process the GuC busyness stats. This works since both GuC timestamp and RING timestamp are synced with the same clock. - The busyness stats may get updated after the batch starts running. This delay causes the busyness reported for 100us duration to fall below 95% in the selftest. The only option at this time is to wait for GuC busyness to change from idle to active before we sample busyness over a 100us period. Signed-off-by: John Harrison <John.C.Harrison@Intel.com> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com> Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20211027004821.66097-2-umesh.nerlige.ramappa@intel.com
2021-10-28drm/i915/pmu: Add a name to the execlists statsUmesh Nerlige Ramappa3-46/+53
In preparation for GuC pmu stats, add a name to the execlists stats structure so that it can be differentiated from the GuC stats. Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Signed-off-by: John Harrison <John.C.Harrison@Intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20211027004821.66097-1-umesh.nerlige.ramappa@intel.com
2021-10-28nouveau/svm: Use kvcalloc() instead of kvzalloc()Gustavo A. R. Silva1-1/+1
Use 2-factor argument form kvcalloc() instead of kvzalloc(). Link: https://github.com/KSPP/linux/issues/162 Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org> Reviewed-by: Kees Cook <keescook@chromium.org>
2021-10-28drm/i915/adlp: Extend PSR2 support in transcoder BJosé Roberto de Souza1-1/+3
PSR2 is supported in transcoder A and B on Alderlake-P. v2: - explicity checking for transcoder A and B to avoid invalid transcoder BSpec: 49185 Reviewed-by: Jouni Högander <jouni.hogander@intel.com> # v1 Cc: Jani Nikula <jani.nikula@intel.com> Cc: Mika Kahola <mika.kahola@intel.com> Cc: Jouni Hogander <jouni.hogander@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com> # v2 Link: https://patchwork.freedesktop.org/patch/msgid/20211027180545.55660-1-jose.souza@intel.com
2021-10-28drm/bridge: sn65dsi86: ti_sn65dsi86_read_u16() __maybe_unusedBjorn Andersson1-2/+2
When built without CONFIG_PWM there are no references to ti_sn65dsi86_read_u16(), avoid the W=1 build warning by marking the function as __maybe_unused. __maybe_unused is used insted of a #ifdef guard as it looks slighly cleaner and it avoids issues if in the future other permutations of the config options would use the function. Reported-by: kernel test robot <lkp@intel.com> Fixes: cea86c5bb442 ("drm/bridge: ti-sn65dsi86: Implement the pwm_chip") Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org> Reviewed-by: Robert Foss <robert.foss@linaro.org> Link: https://patchwork.freedesktop.org/patch/msgid/20211028163548.273736-1-bjorn.andersson@linaro.org Signed-off-by: Robert Foss <robert.foss@linaro.org>
2021-10-28drm/i915/dsc: demote noisy drm_info() to drm_kms_dbg()Jani Nikula1-16/+16
The PPS, RC_RANGE_PARAM, and RC_BUF_THRESH logging are clearly for debugging, and should not be info level messages. Reviewed-by: Manasi Navare <manasi.d.navare@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20211027135900.6329-1-jani.nikula@intel.com
2021-10-28drm/i915: Revert 'guc_id' from i915_request tracepointJoonas Lahtinen1-5/+2
Avoid adding backend specific data to the tracepoints outside of the LOW_LEVEL_TRACEPOINTS kernel config protection. These bits of information are bound to change depending on the selected submission method per platform and are not necessarily possible to maintain in the future. Fixes: dbf9da8d55ef ("drm/i915/guc: Add trace point for GuC submit") Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: John Harrison <john.c.harrison@intel.com> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Matt Roper <matthew.d.roper@intel.com> Reviewed-by: Matthew Auld <matthew.auld@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20211027093255.66489-1-joonas.lahtinen@linux.intel.com (cherry picked from commit 64512a66b67e6546e2db15192b3603cd6d58b75c) Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
2021-10-28Merge tag 'drm-misc-fixes-2021-10-28' of git://anongit.freedesktop.org/drm/drm-misc into drm-fixesDave Airlie1-0/+6
One patch to fix the default screen orientation on the GPD Win3 Signed-off-by: Dave Airlie <airlied@redhat.com> From: Maxime Ripard <maxime@cerno.tech> Link: https://patchwork.freedesktop.org/patch/msgid/20211028072300.b4gqexq6zfhby24g@gilmour
2021-10-28drm: Add R10 and R12 FourCCLaurent Pinchart1-0/+2
Add FourCCs for 10- and 12-bit red formats with padding to 16 bits. They correspond to the V4L2 10- and 12-bit greyscale (V4L2_PIX_FMT_Y10 and V4L2_PIX_FMT_Y12) formats, as well as the Bayer formats with the same bit depth (V4L2_PIX_FMT_SBGGR{10,12} and all other Bayer pattern permutations). These formats are not used by any kernel driver at this point, but need to be exposed to applications by libcamera, which uses DRM FourCCs for pixel formats. Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/20211027233140.12268-1-laurent.pinchart@ideasonboard.com
2021-10-28Merge tag 'drm-misc-fixes-2021-10-26' of git://anongit.freedesktop.org/drm/drm-misc into drm-fixesDave Airlie3-0/+8
drm-misc-fixes for v5.15-rc8: - Fix fence leak in ttm_transfered_destroy. - Add quirk for Aya Neo 2021 - Reset property count for each drm damage selftest so full run will work correctly. Signed-off-by: Dave Airlie <airlied@redhat.com> From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/4a133970-ff4b-aa62-d346-b269b1b9236e@linux.intel.com
2021-10-28Merge tag 'drm-msm-next-2021-10-26' of https://gitlab.freedesktop.org/drm/msm into drm-nextDave Airlie63-1161/+1213
* eDP support in DP sub-driver (for newer SoCs with native eDP output) * dpu irq handling cleanup * CRC support for making igt happy * Support for NO_CONNECTOR bridges * dsi: 14nm phy support for msm8953 * mdp5: support for msm8x53, sdm450, sdm632 * various smaller fixes and cleanups Signed-off-by: Dave Airlie <airlied@redhat.com> From: Rob Clark <robdclark@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/CAF6AEGsH9EwcpqGNNRJeL99NvFFjHX3SUg+nTYu0dHG5U9+QuA@mail.gmail.com
2021-10-28BackMerge tag 'v5.15-rc7' into drm-nextDave Airlie81-436/+1032
The msm next tree is based on rc3, so let's just backmerge rc7 before pulling it in. Signed-off-by: Dave Airlie <airlied@redhat.com>
2021-10-27drm/amd/display: Fix deadlock when falling back to v2 from v3Nicholas Kazlauskas1-4/+2
[Why] A deadlock in the kernel occurs when we fallback from the V3 to V2 add_topology_to_display or remove_topology_to_display because they both try to acquire the dtm_mutex but recursive locking isn't supported on mutex_lock(). [How] Make the mutex_lock/unlock more fine grained and move them up such that they're only required for the psp invocation itself. Fixes: bf62221e9d0e ("drm/amd/display: Add DCN3.1 HDCP support") Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Reviewed-by: Aric Cyr <aric.cyr@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2021-10-27drm/amd/display: Fallback to clocks which meet requested voltage on DCN31Michael Strauss1-3/+10
[WHY] On certain configs, SMU clock table voltages don't match which cause parser to behave incorrectly by leaving dcfclk and socclk table entries unpopulated. [HOW] Currently the function that finds the corresponding clock for a given voltage only checks for exact voltage level matches. In the case that no match gets found, parser now falls back to searching for the max clock which meets the requested voltage (i.e. its corresponding voltage is below requested). Signed-off-by: Michael Strauss <michael.strauss@amd.com> Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
Copyright © 1996 – 2026 Jason A. Donenfeld. All Rights Reverse Engineered.