kernel/git/stable/linux.git/Documentation/gpu/drm-usage-stats.rst, branch linux-7.2.y

Documentation/gpu: add some tables of contents to large documents

2026-05-13T08:07:12+00:00

Some of the GPU documentation pages are quite long, with various levels of details. Add document internal tables of contents to the larger documents to make them easier to navigate. The index.rst in the sub-directories have toctrees, which provide similar overviews. Fix one missing newline at the end of drm-uapi.rst while at it, primarily because rst should have it, and secondarily because my editor rst mode refuses to save the file without it. Acked-by: Randy Dunlap Tested-by: Randy Dunlap Link: https://patch.msgid.link/e3f9357c0e8198cc48e69e2a3b8ca072c7ab92ca.1778238671.git.jani.nikula@intel.com Signed-off-by: Jani Nikula

accel/amdxdna: Expose per-client BO memory usage via fdinfo

2026-04-09T20:41:47+00:00

Implement amdxdna_show_fdinfo() to report per-client memory usage, including below driver-specific memory stat: - heap allocation - internal BO allocation - external BO allocation Hook the implementation into the DRM fdinfo infrastructure via drm_driver.show_fdinfo, while continuing to expose standard DRM memory stat through drm_show_memory_stats(). This improves observability of per-process memory usage and aligns with existing fdinfo reporting mechanisms used by other drivers. Suggested-by: Ian Rogers Signed-off-by: Max Zhen Reviewed-by: Mario Limonciello (AMD) Signed-off-by: Lizhi Hou Link: https://patch.msgid.link/20260409152259.176883-1-lizhi.hou@amd.com

Documentation/gpu: Clarify format of driver-specific fidnfo keys

2025-02-07T14:22:46+00:00

This change reflects de facto usage by amdgpu, as exemplified by commit d6530c33a978 ("drm/amdgpu: expose more memory stats in fdinfo"). Cc: Tvrtko Ursulin Acked-by: Tvrtko Ursulin Signed-off-by: Adrián Larumbe Signed-off-by: Boris Brezillon Link: https://patchwork.freedesktop.org/patch/msgid/20250130172851.941597-2-adrian.larumbe@collabora.com

Documentation/gpu: Clarify drm memory stats definition

2024-12-19T15:56:21+00:00

Define how to handle buffers with multiple possible placement so we don't get incompatible implementations. Callout the resident requirement for drm-purgeable- explicitly. Remove the requirement for there to be only drm-memory- or only drm-resident-, it's not what's implemented and having both is better for back-compat. Also re-order the paragraphs to flow better. Signed-off-by: Yunxiang Li Reviewed-by: Tvrtko Ursulin Link: https://patchwork.freedesktop.org/patch/msgid/20241219151411.1150-4-Yunxiang.Li@amd.com Signed-off-by: Christian König

Merge tag 'drm-misc-next-2024-10-31' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-next

2024-11-01T03:46:03+00:00

drm-misc-next for v6.13: All of the previous pull request, with MORE! Core Changes: - Update documentation for scheduler start/stop and job init. - Add dedede and sm8350-hdk hardware to ci runs. Driver Changes: - Small fixes and cleanups to panfrost, omap, nouveau, ivpu, zynqmp, v3d, panthor docs, and leadtek-ltk050h3146w. - Crashdump support for qaic. - Support DP compliance in zynqmp. - Add Samsung S6E88A0-AMS427AP24 panel. Signed-off-by: Dave Airlie From: Maarten Lankhorst Link: https://patchwork.freedesktop.org/patch/msgid/deeef745-f3fb-4e85-a9d0-e8d38d43c1cf@linux.intel.com

Documentation/gpu: Fix Panthor documentation build warnings

2024-10-30T15:25:06+00:00

Fix Panthor documentation build errors uncovered by the makedocs target when building with extra warnings enabled. Signed-off-by: Adrián Larumbe Fixes: 6a797bdfde77 ("drm/panthor: add sysfs knob for enabling job profiling") Reported-by: Stephen Rothwell Reviewed-by: Steven Price Signed-off-by: Boris Brezillon Link: https://patchwork.freedesktop.org/patch/msgid/20241009214346.2308917-1-adrian.larumbe@collabora.com

Merge tag 'amd-drm-next-6.13-2024-10-25' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

2024-10-29T08:25:24+00:00

amd-drm-next-6.13-2024-10-25: amdgpu: - SDMA queue reset support - SMU 13.0.6 updates - Add debugfs interface to help limit jpeg queue scheduling for testing - JPEG 4.0.3 updates - Initial runtime repartitioning support - GFX9 fixes - Misc code cleanups - Rework IP structures to better handle multiple instances of an IP - DML updates - DSC fixes - HDR fixes - Brightness control updates - Runtime pm cleanup - DMCUB fixes - DCN 3.5 updates - Struct drm_edid cleanup - Fetch EDID from _DDC if available - Ring noop optimizations - MES logging fixes - 3DLUT fixes - DCN 4.x fixes - SMU 13.x fixes - Fixes for set_soft_freq_range() - ACPI fixes - SMU 14.x updates - PSR-SU fixes - fdinfo cleanup - DCN documentation updates amdkfd: - Misc code cleanups - Increase event FIFO size - Copy wave state fixes for SDMA radeon: - Fix possible overflow in packet3 check - Late init connector fix - Always set GEM function pointer Documentation: - Update drm-memory documentation From: Alex Deucher Link: https://patchwork.freedesktop.org/patch/msgid/20241025132336.2416913-1-alexander.deucher@amd.com Signed-off-by: Dave Airlie

drm: use drm_file client_name in fdinfo

2024-10-08T08:02:24+00:00

Add an optional drm-client-name field to drm fdinfo's output. Reviewed-by: Christian König Reviewed-by: Tvrtko Ursulin Signed-off-by: Pierre-Eric Pelloux-Prayer Link: https://patchwork.freedesktop.org/patch/msgid/20241003124506.470931-3-pierre-eric.pelloux-prayer@amd.com Reviewed-by: Christian König Signed-off-by: Christian König

Documentation/gpu: Document the situation with unqualified drm-memory-

2024-10-07T18:32:23+00:00

Currently it is not well defined what is drm-memory- compared to other categories. In practice the only driver which emits these keys is amdgpu and in them exposes the current resident buffer object memory (including shared). To prevent any confusion, document that drm-memory- is deprecated and an alias for drm-resident-memory-. While at it also clarify that the reserved sub-string 'memory' refers to the memory region component, and also clarify the intended semantics of other memory categories. v2: * Also mark drm-memory- as deprecated. * Add some more text describing memory categories. (Alex) v3: * Semantics of the amdgpu drm-memory is actually as drm-resident. Reviewed-by: Rob Clark Acked-by: Daniel Vetter Reviewed-by: Alex Deucher Signed-off-by: Tvrtko Ursulin Cc: Alex Deucher Cc: Christian König Cc: Rob Clark Signed-off-by: Alex Deucher

drm/xe/client: Print runtime to fdinfo

2024-05-21T13:33:40+00:00

Print the accumulated runtime for client when printing fdinfo. Each time a query is done it first does 2 things: 1) loop through all the exec queues for the current client and accumulate the runtime, per engine class. CTX_TIMESTAMP is used for that, being read from the context image. 2) Read a "GPU timestamp" that can be used for considering "how much GPU time has passed" and that has the same unit/refclock as the one recording the runtime. RING_TIMESTAMP is used for that via MMIO. Since for all current platforms RING_TIMESTAMP follows the same refclock, just read it once, using any first engine available. This is exported to userspace as 2 numbers in fdinfo: drm-cycles-: drm-total-cycles-: Userspace is expected to collect at least 2 samples, which allows to know the client engine busyness as per: RUNTIME1 - RUNTIME0 busyness = --------------------- T1 - T0 Since drm-cycles- always starts at 0, it's also possible to know if and engine was ever used by a client. It's expected that userspace will read any 2 samples every few seconds. Given the update frequency of the counters involved and that CTX_TIMESTAMP is 32-bits, the counter for each exec_queue can wrap around (assuming 100% utilization) after ~200s. The wraparound is not perceived by userspace since it's just accumulated for all the exec_queues in a 64-bit counter) but the measurement will not be accurate if the samples are too far apart. This could be mitigated by adding a workqueue to accumulate the counters every so often, but it's additional complexity for something that is done already by userspace every few seconds in tools like gputop (from igt), htop, nvtop, etc, with none of them really defaulting to 1 sample per minute or more. Reviewed-by: Umesh Nerlige Ramappa Acked-by: Tvrtko Ursulin Link: https://patchwork.freedesktop.org/patch/msgid/20240517204310.88854-9-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi