1 ===========================
2 drm/i915 Intel GFX Driver
3 ===========================
5 The drm/i915 driver supports all (with the exception of some very early
6 models) integrated GFX chipsets with both Intel display and rendering
7 blocks. This excludes a set of SoC platforms with an SGX rendering unit,
8 those have basic support through the gma500 drm driver.
10 Core Driver Infrastructure
11 ==========================
13 This section covers core driver infrastructure used by both the display
14 and the GEM parts of the driver.
16 Runtime Power Management
17 ------------------------
19 .. kernel-doc:: drivers/gpu/drm/i915/intel_runtime_pm.c
22 .. kernel-doc:: drivers/gpu/drm/i915/intel_runtime_pm.c
25 .. kernel-doc:: drivers/gpu/drm/i915/intel_uncore.c
31 .. kernel-doc:: drivers/gpu/drm/i915/i915_irq.c
32 :doc: interrupt handling
34 .. kernel-doc:: drivers/gpu/drm/i915/i915_irq.c
35 :functions: intel_irq_init intel_irq_init_hw intel_hpd_init
37 .. kernel-doc:: drivers/gpu/drm/i915/i915_irq.c
38 :functions: intel_runtime_pm_disable_interrupts
40 .. kernel-doc:: drivers/gpu/drm/i915/i915_irq.c
41 :functions: intel_runtime_pm_enable_interrupts
43 Intel GVT-g Guest Support(vGPU)
44 -------------------------------
46 .. kernel-doc:: drivers/gpu/drm/i915/i915_vgpu.c
47 :doc: Intel GVT-g guest support
49 .. kernel-doc:: drivers/gpu/drm/i915/i915_vgpu.c
52 Intel GVT-g Host Support(vGPU device model)
53 -------------------------------------------
55 .. kernel-doc:: drivers/gpu/drm/i915/intel_gvt.c
56 :doc: Intel GVT-g host support
58 .. kernel-doc:: drivers/gpu/drm/i915/intel_gvt.c
64 .. kernel-doc:: drivers/gpu/drm/i915/gt/intel_workarounds.c
65 :doc: Hardware workarounds
67 Display Hardware Handling
68 =========================
70 This section covers everything related to the display hardware including
71 the mode setting infrastructure, plane, sprite and cursor handling and
72 display, output probing and related topics.
74 Mode Setting Infrastructure
75 ---------------------------
77 The i915 driver is thus far the only DRM driver which doesn't use the
78 common DRM helper code to implement mode setting sequences. Thus it has
79 its own tailor-made infrastructure for executing a display configuration
85 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_frontbuffer.c
86 :doc: frontbuffer tracking
88 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_frontbuffer.h
91 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_frontbuffer.c
94 Display FIFO Underrun Reporting
95 -------------------------------
97 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_fifo_underrun.c
98 :doc: fifo underrun handling
100 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_fifo_underrun.c
106 This section covers plane configuration and composition with the primary
107 plane, sprites, cursors and overlays. This includes the infrastructure
108 to do atomic vsync'ed updates of all this state and also tightly coupled
109 topics like watermark setup and computation, framebuffer compression and
115 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_atomic_plane.c
116 :doc: atomic plane helpers
118 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_atomic_plane.c
124 This section covers output probing and related infrastructure like the
125 hotplug interrupt storm detection and mitigation code. Note that the
126 i915 driver still uses most of the common DRM helper code for output
127 probing, so those sections fully apply.
132 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_hotplug.c
135 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_hotplug.c
138 High Definition Audio
139 ---------------------
141 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_audio.c
142 :doc: High Definition Audio over HDMI and Display Port
144 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_audio.c
147 .. kernel-doc:: include/drm/i915_component.h
150 Intel HDMI LPE Audio Support
151 ----------------------------
153 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_lpe_audio.c
154 :doc: LPE Audio integration for HDMI or DP playback
156 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_lpe_audio.c
159 Panel Self Refresh PSR (PSR/SRD)
160 --------------------------------
162 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_psr.c
163 :doc: Panel Self Refresh (PSR/SRD)
165 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_psr.c
168 Frame Buffer Compression (FBC)
169 ------------------------------
171 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_fbc.c
172 :doc: Frame Buffer Compression (FBC)
174 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_fbc.c
177 Display Refresh Rate Switching (DRRS)
178 -------------------------------------
180 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dp.c
181 :doc: Display Refresh Rate Switching (DRRS)
183 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dp.c
184 :functions: intel_dp_set_drrs_state
186 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dp.c
187 :functions: intel_edp_drrs_enable
189 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dp.c
190 :functions: intel_edp_drrs_disable
192 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dp.c
193 :functions: intel_edp_drrs_invalidate
195 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dp.c
196 :functions: intel_edp_drrs_flush
198 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dp.c
199 :functions: intel_dp_drrs_init
204 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dpio_phy.c
207 CSR firmware support for DMC
208 ----------------------------
210 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_csr.c
211 :doc: csr support for dmc
213 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_csr.c
216 Video BIOS Table (VBT)
217 ----------------------
219 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_bios.c
220 :doc: Video BIOS Table (VBT)
222 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_bios.c
225 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_vbt_defs.h
231 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_cdclk.c
234 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_cdclk.c
240 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dpll_mgr.c
243 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dpll_mgr.c
246 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dpll_mgr.h
252 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dsb.c
255 .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dsb.c
258 Memory Management and Command Submission
259 ========================================
261 This sections covers all things related to the GEM implementation in the
267 An Intel GPU has multiple engines. There are several engine types.
269 - RCS engine is for rendering 3D and performing compute, this is named
270 `I915_EXEC_RENDER` in user space.
271 - BCS is a blitting (copy) engine, this is named `I915_EXEC_BLT` in user
273 - VCS is a video encode and decode engine, this is named `I915_EXEC_BSD`
275 - VECS is video enhancement engine, this is named `I915_EXEC_VEBOX` in user
277 - The enumeration `I915_EXEC_DEFAULT` does not refer to specific engine;
278 instead it is to be used by user space to specify a default rendering
279 engine (for 3D) that may or may not be the same as RCS.
281 The Intel GPU family is a family of integrated GPU's using Unified
282 Memory Access. For having the GPU "do work", user space will feed the
283 GPU batch buffers via one of the ioctls `DRM_IOCTL_I915_GEM_EXECBUFFER2`
284 or `DRM_IOCTL_I915_GEM_EXECBUFFER2_WR`. Most such batchbuffers will
285 instruct the GPU to perform work (for example rendering) and that work
286 needs memory from which to read and memory to which to write. All memory
287 is encapsulated within GEM buffer objects (usually created with the ioctl
288 `DRM_IOCTL_I915_GEM_CREATE`). An ioctl providing a batchbuffer for the GPU
289 to create will also list all GEM buffer objects that the batchbuffer reads
290 and/or writes. For implementation details of memory management see
291 `GEM BO Management Implementation Details`_.
293 The i915 driver allows user space to create a context via the ioctl
294 `DRM_IOCTL_I915_GEM_CONTEXT_CREATE` which is identified by a 32-bit
295 integer. Such a context should be viewed by user-space as -loosely-
296 analogous to the idea of a CPU process of an operating system. The i915
297 driver guarantees that commands issued to a fixed context are to be
298 executed so that writes of a previously issued command are seen by
299 reads of following commands. Actions issued between different contexts
300 (even if from the same file descriptor) are NOT given that guarantee
301 and the only way to synchronize across contexts (even from the same
302 file descriptor) is through the use of fences. At least as far back as
303 Gen4, also have that a context carries with it a GPU HW context;
304 the HW context is essentially (most of atleast) the state of a GPU.
305 In addition to the ordering guarantees, the kernel will restore GPU
306 state via HW context when commands are issued to a context, this saves
307 user space the need to restore (most of atleast) the GPU state at the
308 start of each batchbuffer. The non-deprecated ioctls to submit batchbuffer
309 work can pass that ID (in the lower bits of drm_i915_gem_execbuffer2::rsvd1)
310 to identify what context to use with the command.
312 The GPU has its own memory management and address space. The kernel
313 driver maintains the memory translation table for the GPU. For older
314 GPUs (i.e. those before Gen8), there is a single global such translation
315 table, a global Graphics Translation Table (GTT). For newer generation
316 GPUs each context has its own translation table, called Per-Process
317 Graphics Translation Table (PPGTT). Of important note, is that although
318 PPGTT is named per-process it is actually per context. When user space
319 submits a batchbuffer, the kernel walks the list of GEM buffer objects
320 used by the batchbuffer and guarantees that not only is the memory of
321 each such GEM buffer object resident but it is also present in the
322 (PP)GTT. If the GEM buffer object is not yet placed in the (PP)GTT,
323 then it is given an address. Two consequences of this are: the kernel
324 needs to edit the batchbuffer submitted to write the correct value of
325 the GPU address when a GEM BO is assigned a GPU address and the kernel
326 might evict a different GEM BO from the (PP)GTT to make address room
327 for another GEM BO. Consequently, the ioctls submitting a batchbuffer
328 for execution also include a list of all locations within buffers that
329 refer to GPU-addresses so that the kernel can edit the buffer correctly.
330 This process is dubbed relocation.
336 This is a description of how the locking should be after
337 refactoring is done. Does not necessarily reflect what the locking
338 looks like while WIP.
340 #. All locking rules and interface contracts with cross-driver interfaces
341 (dma-buf, dma_fence) need to be followed.
343 #. No struct_mutex anywhere in the code
345 #. dma_resv will be the outermost lock (when needed) and ww_acquire_ctx
346 is to be hoisted at highest level and passed down within i915_gem_ctx
349 #. While holding lru/memory manager (buddy, drm_mm, whatever) locks
350 system memory allocations are not allowed
352 * Enforce this by priming lockdep (with fs_reclaim). If we
353 allocate memory while holding these looks we get a rehash
354 of the shrinker vs. struct_mutex saga, and that would be
357 #. Do not nest different lru/memory manager locks within each other.
358 Take them in turn to update memory allocations, relying on the object’s
359 dma_resv ww_mutex to serialize against other operations.
361 #. The suggestion for lru/memory managers locks is that they are small
362 enough to be spinlocks.
364 #. All features need to come with exhaustive kernel selftests and/or
365 IGT tests when appropriate
367 #. All LMEM uAPI paths need to be fully restartable (_interruptible()
368 for all locks/waits/sleeps)
370 * Error handling validation through signal injection.
371 Still the best strategy we have for validating GEM uAPI
373 Must be excessively used in the IGT, and we need to check
374 that we really have full path coverage of all error cases.
376 * -EDEADLK handling with ww_mutex
378 GEM BO Management Implementation Details
379 ----------------------------------------
381 .. kernel-doc:: drivers/gpu/drm/i915/i915_vma_types.h
382 :doc: Virtual Memory Address
384 Buffer Object Eviction
385 ----------------------
387 This section documents the interface functions for evicting buffer
388 objects to make space available in the virtual gpu address spaces. Note
389 that this is mostly orthogonal to shrinking buffer objects caches, which
390 has the goal to make main memory (shared with the gpu through the
391 unified memory architecture) available.
393 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_evict.c
396 Buffer Object Memory Shrinking
397 ------------------------------
399 This section documents the interface function for shrinking memory usage
400 of buffer object caches. Shrinking is used to make main memory
401 available. Note that this is mostly orthogonal to evicting buffer
402 objects, which has the goal to make space in gpu virtual address spaces.
404 .. kernel-doc:: drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
410 .. kernel-doc:: drivers/gpu/drm/i915/i915_cmd_parser.c
411 :doc: batch buffer command parser
413 .. kernel-doc:: drivers/gpu/drm/i915/i915_cmd_parser.c
416 User Batchbuffer Execution
417 --------------------------
419 .. kernel-doc:: drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
420 :doc: User command execution
422 Logical Rings, Logical Ring Contexts and Execlists
423 --------------------------------------------------
425 .. kernel-doc:: drivers/gpu/drm/i915/gt/intel_lrc.c
426 :doc: Logical Rings, Logical Ring Contexts and Execlists
431 .. kernel-doc:: drivers/gpu/drm/i915/i915_vma_types.h
432 :doc: Global GTT views
434 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_gtt.c
437 GTT Fences and Swizzling
438 ------------------------
440 .. kernel-doc:: drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c
443 Global GTT Fence Handling
444 ~~~~~~~~~~~~~~~~~~~~~~~~~
446 .. kernel-doc:: drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c
447 :doc: fence register handling
449 Hardware Tiling and Swizzling Details
450 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
452 .. kernel-doc:: drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c
453 :doc: tiling swizzling details
458 .. kernel-doc:: drivers/gpu/drm/i915/gem/i915_gem_tiling.c
461 .. kernel-doc:: drivers/gpu/drm/i915/gem/i915_gem_tiling.c
462 :doc: buffer object tiling
467 Starting from gen9, three microcontrollers are available on the HW: the
468 graphics microcontroller (GuC), the HEVC/H.265 microcontroller (HuC) and the
469 display microcontroller (DMC). The driver is responsible for loading the
470 firmwares on the microcontrollers; the GuC and HuC firmwares are transferred
471 to WOPCM using the DMA engine, while the DMC firmware is written through MMIO.
479 .. kernel-doc:: drivers/gpu/drm/i915/intel_wopcm.c
485 .. kernel-doc:: drivers/gpu/drm/i915/gt/uc/intel_guc.c
491 .. kernel-doc:: drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h
492 :doc: Firmware Layout
494 GuC Memory Management
495 ~~~~~~~~~~~~~~~~~~~~~
497 .. kernel-doc:: drivers/gpu/drm/i915/gt/uc/intel_guc.c
498 :doc: GuC Memory Management
499 .. kernel-doc:: drivers/gpu/drm/i915/gt/uc/intel_guc.c
500 :functions: intel_guc_allocate_vma
503 GuC-specific firmware loader
504 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
506 .. kernel-doc:: drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
509 GuC-based command submission
510 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
512 .. kernel-doc:: drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
513 :doc: GuC-based command submission
517 .. kernel-doc:: drivers/gpu/drm/i915/gt/uc/intel_huc.c
519 .. kernel-doc:: drivers/gpu/drm/i915/gt/uc/intel_huc.c
520 :functions: intel_huc_auth
522 HuC Memory Management
523 ~~~~~~~~~~~~~~~~~~~~~
525 .. kernel-doc:: drivers/gpu/drm/i915/gt/uc/intel_huc.c
526 :doc: HuC Memory Management
530 The HuC FW layout is the same as the GuC one, see `GuC Firmware Layout`_
534 See `CSR firmware support for DMC`_
539 This sections covers all things related to the tracepoints implemented
542 i915_ppgtt_create and i915_ppgtt_release
543 ----------------------------------------
545 .. kernel-doc:: drivers/gpu/drm/i915/i915_trace.h
546 :doc: i915_ppgtt_create and i915_ppgtt_release tracepoints
548 i915_context_create and i915_context_free
549 -----------------------------------------
551 .. kernel-doc:: drivers/gpu/drm/i915/i915_trace.h
552 :doc: i915_context_create and i915_context_free tracepoints
559 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
560 :doc: i915 Perf Overview
562 Comparison with Core Perf
563 -------------------------
564 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
565 :doc: i915 Perf History and Comparison with Core Perf
567 i915 Driver Entry Points
568 ------------------------
570 This section covers the entrypoints exported outside of i915_perf.c to
571 integrate with drm/i915 and to handle the `DRM_I915_PERF_OPEN` ioctl.
573 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
574 :functions: i915_perf_init
575 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
576 :functions: i915_perf_fini
577 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
578 :functions: i915_perf_register
579 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
580 :functions: i915_perf_unregister
581 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
582 :functions: i915_perf_open_ioctl
583 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
584 :functions: i915_perf_release
585 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
586 :functions: i915_perf_add_config_ioctl
587 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
588 :functions: i915_perf_remove_config_ioctl
593 This section covers the stream-semantics-agnostic structures and functions
594 for representing an i915 perf stream FD and associated file operations.
596 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf_types.h
597 :functions: i915_perf_stream
598 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf_types.h
599 :functions: i915_perf_stream_ops
601 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
602 :functions: read_properties_unlocked
603 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
604 :functions: i915_perf_open_ioctl_locked
605 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
606 :functions: i915_perf_destroy_locked
607 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
608 :functions: i915_perf_read
609 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
610 :functions: i915_perf_ioctl
611 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
612 :functions: i915_perf_enable_locked
613 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
614 :functions: i915_perf_disable_locked
615 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
616 :functions: i915_perf_poll
617 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
618 :functions: i915_perf_poll_locked
620 i915 Perf Observation Architecture Stream
621 -----------------------------------------
623 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf_types.h
624 :functions: i915_oa_ops
626 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
627 :functions: i915_oa_stream_init
628 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
629 :functions: i915_oa_read
630 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
631 :functions: i915_oa_stream_enable
632 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
633 :functions: i915_oa_stream_disable
634 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
635 :functions: i915_oa_wait_unlocked
636 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
637 :functions: i915_oa_poll_wait
639 Other i915 Perf Internals
640 -------------------------
642 This section simply includes all other currently documented i915 perf internals,
643 in no particular order, but may include some more minor utilities or platform
644 specific details than found in the more high-level sections.
646 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
655 i915_perf_add_config_ioctl
656 i915_perf_remove_config_ioctl
657 read_properties_unlocked
658 i915_perf_open_ioctl_locked
659 i915_perf_destroy_locked
660 i915_perf_read i915_perf_ioctl
661 i915_perf_enable_locked
662 i915_perf_disable_locked
663 i915_perf_poll i915_perf_poll_locked
664 i915_oa_stream_init i915_oa_read
665 i915_oa_stream_enable
666 i915_oa_stream_disable
667 i915_oa_wait_unlocked
673 The drm/i915 driver codebase has some style rules in addition to (and, in some
674 cases, deviating from) the kernel coding style.
676 Register macro definition style
677 -------------------------------
679 The style guide for ``i915_reg.h``.
681 .. kernel-doc:: drivers/gpu/drm/i915/i915_reg.h
682 :doc: The i915 register macro definition style guide