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/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/intel_frontbuffer.c
86 :doc: frontbuffer tracking
88 .. kernel-doc:: drivers/gpu/drm/i915/intel_frontbuffer.h
91 .. kernel-doc:: drivers/gpu/drm/i915/intel_frontbuffer.c
94 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem.c
95 :functions: i915_gem_track_fb
97 Display FIFO Underrun Reporting
98 -------------------------------
100 .. kernel-doc:: drivers/gpu/drm/i915/intel_fifo_underrun.c
101 :doc: fifo underrun handling
103 .. kernel-doc:: drivers/gpu/drm/i915/intel_fifo_underrun.c
109 This section covers plane configuration and composition with the primary
110 plane, sprites, cursors and overlays. This includes the infrastructure
111 to do atomic vsync'ed updates of all this state and also tightly coupled
112 topics like watermark setup and computation, framebuffer compression and
118 .. kernel-doc:: drivers/gpu/drm/i915/intel_atomic_plane.c
119 :doc: atomic plane helpers
121 .. kernel-doc:: drivers/gpu/drm/i915/intel_atomic_plane.c
127 This section covers output probing and related infrastructure like the
128 hotplug interrupt storm detection and mitigation code. Note that the
129 i915 driver still uses most of the common DRM helper code for output
130 probing, so those sections fully apply.
135 .. kernel-doc:: drivers/gpu/drm/i915/intel_hotplug.c
138 .. kernel-doc:: drivers/gpu/drm/i915/intel_hotplug.c
141 High Definition Audio
142 ---------------------
144 .. kernel-doc:: drivers/gpu/drm/i915/intel_audio.c
145 :doc: High Definition Audio over HDMI and Display Port
147 .. kernel-doc:: drivers/gpu/drm/i915/intel_audio.c
150 .. kernel-doc:: include/drm/i915_component.h
153 Intel HDMI LPE Audio Support
154 ----------------------------
156 .. kernel-doc:: drivers/gpu/drm/i915/intel_lpe_audio.c
157 :doc: LPE Audio integration for HDMI or DP playback
159 .. kernel-doc:: drivers/gpu/drm/i915/intel_lpe_audio.c
162 Panel Self Refresh PSR (PSR/SRD)
163 --------------------------------
165 .. kernel-doc:: drivers/gpu/drm/i915/intel_psr.c
166 :doc: Panel Self Refresh (PSR/SRD)
168 .. kernel-doc:: drivers/gpu/drm/i915/intel_psr.c
171 Frame Buffer Compression (FBC)
172 ------------------------------
174 .. kernel-doc:: drivers/gpu/drm/i915/intel_fbc.c
175 :doc: Frame Buffer Compression (FBC)
177 .. kernel-doc:: drivers/gpu/drm/i915/intel_fbc.c
180 Display Refresh Rate Switching (DRRS)
181 -------------------------------------
183 .. kernel-doc:: drivers/gpu/drm/i915/intel_dp.c
184 :doc: Display Refresh Rate Switching (DRRS)
186 .. kernel-doc:: drivers/gpu/drm/i915/intel_dp.c
187 :functions: intel_dp_set_drrs_state
189 .. kernel-doc:: drivers/gpu/drm/i915/intel_dp.c
190 :functions: intel_edp_drrs_enable
192 .. kernel-doc:: drivers/gpu/drm/i915/intel_dp.c
193 :functions: intel_edp_drrs_disable
195 .. kernel-doc:: drivers/gpu/drm/i915/intel_dp.c
196 :functions: intel_edp_drrs_invalidate
198 .. kernel-doc:: drivers/gpu/drm/i915/intel_dp.c
199 :functions: intel_edp_drrs_flush
201 .. kernel-doc:: drivers/gpu/drm/i915/intel_dp.c
202 :functions: intel_dp_drrs_init
207 .. kernel-doc:: drivers/gpu/drm/i915/intel_dpio_phy.c
210 CSR firmware support for DMC
211 ----------------------------
213 .. kernel-doc:: drivers/gpu/drm/i915/intel_csr.c
214 :doc: csr support for dmc
216 .. kernel-doc:: drivers/gpu/drm/i915/intel_csr.c
219 Video BIOS Table (VBT)
220 ----------------------
222 .. kernel-doc:: drivers/gpu/drm/i915/intel_bios.c
223 :doc: Video BIOS Table (VBT)
225 .. kernel-doc:: drivers/gpu/drm/i915/intel_bios.c
228 .. kernel-doc:: drivers/gpu/drm/i915/intel_vbt_defs.h
234 .. kernel-doc:: drivers/gpu/drm/i915/intel_cdclk.c
237 .. kernel-doc:: drivers/gpu/drm/i915/intel_cdclk.c
243 .. kernel-doc:: drivers/gpu/drm/i915/intel_dpll_mgr.c
246 .. kernel-doc:: drivers/gpu/drm/i915/intel_dpll_mgr.c
249 .. kernel-doc:: drivers/gpu/drm/i915/intel_dpll_mgr.h
252 Memory Management and Command Submission
253 ========================================
255 This sections covers all things related to the GEM implementation in the
261 An Intel GPU has multiple engines. There are several engine types.
263 - RCS engine is for rendering 3D and performing compute, this is named
264 `I915_EXEC_RENDER` in user space.
265 - BCS is a blitting (copy) engine, this is named `I915_EXEC_BLT` in user
267 - VCS is a video encode and decode engine, this is named `I915_EXEC_BSD`
269 - VECS is video enhancement engine, this is named `I915_EXEC_VEBOX` in user
271 - The enumeration `I915_EXEC_DEFAULT` does not refer to specific engine;
272 instead it is to be used by user space to specify a default rendering
273 engine (for 3D) that may or may not be the same as RCS.
275 The Intel GPU family is a family of integrated GPU's using Unified
276 Memory Access. For having the GPU "do work", user space will feed the
277 GPU batch buffers via one of the ioctls `DRM_IOCTL_I915_GEM_EXECBUFFER2`
278 or `DRM_IOCTL_I915_GEM_EXECBUFFER2_WR`. Most such batchbuffers will
279 instruct the GPU to perform work (for example rendering) and that work
280 needs memory from which to read and memory to which to write. All memory
281 is encapsulated within GEM buffer objects (usually created with the ioctl
282 `DRM_IOCTL_I915_GEM_CREATE`). An ioctl providing a batchbuffer for the GPU
283 to create will also list all GEM buffer objects that the batchbuffer reads
284 and/or writes. For implementation details of memory management see
285 `GEM BO Management Implementation Details`_.
287 The i915 driver allows user space to create a context via the ioctl
288 `DRM_IOCTL_I915_GEM_CONTEXT_CREATE` which is identified by a 32-bit
289 integer. Such a context should be viewed by user-space as -loosely-
290 analogous to the idea of a CPU process of an operating system. The i915
291 driver guarantees that commands issued to a fixed context are to be
292 executed so that writes of a previously issued command are seen by
293 reads of following commands. Actions issued between different contexts
294 (even if from the same file descriptor) are NOT given that guarantee
295 and the only way to synchronize across contexts (even from the same
296 file descriptor) is through the use of fences. At least as far back as
297 Gen4, also have that a context carries with it a GPU HW context;
298 the HW context is essentially (most of atleast) the state of a GPU.
299 In addition to the ordering guarantees, the kernel will restore GPU
300 state via HW context when commands are issued to a context, this saves
301 user space the need to restore (most of atleast) the GPU state at the
302 start of each batchbuffer. The non-deprecated ioctls to submit batchbuffer
303 work can pass that ID (in the lower bits of drm_i915_gem_execbuffer2::rsvd1)
304 to identify what context to use with the command.
306 The GPU has its own memory management and address space. The kernel
307 driver maintains the memory translation table for the GPU. For older
308 GPUs (i.e. those before Gen8), there is a single global such translation
309 table, a global Graphics Translation Table (GTT). For newer generation
310 GPUs each context has its own translation table, called Per-Process
311 Graphics Translation Table (PPGTT). Of important note, is that although
312 PPGTT is named per-process it is actually per context. When user space
313 submits a batchbuffer, the kernel walks the list of GEM buffer objects
314 used by the batchbuffer and guarantees that not only is the memory of
315 each such GEM buffer object resident but it is also present in the
316 (PP)GTT. If the GEM buffer object is not yet placed in the (PP)GTT,
317 then it is given an address. Two consequences of this are: the kernel
318 needs to edit the batchbuffer submitted to write the correct value of
319 the GPU address when a GEM BO is assigned a GPU address and the kernel
320 might evict a different GEM BO from the (PP)GTT to make address room
321 for another GEM BO. Consequently, the ioctls submitting a batchbuffer
322 for execution also include a list of all locations within buffers that
323 refer to GPU-addresses so that the kernel can edit the buffer correctly.
324 This process is dubbed relocation.
326 GEM BO Management Implementation Details
327 ----------------------------------------
329 .. kernel-doc:: drivers/gpu/drm/i915/i915_vma.h
330 :doc: Virtual Memory Address
332 Buffer Object Eviction
333 ----------------------
335 This section documents the interface functions for evicting buffer
336 objects to make space available in the virtual gpu address spaces. Note
337 that this is mostly orthogonal to shrinking buffer objects caches, which
338 has the goal to make main memory (shared with the gpu through the
339 unified memory architecture) available.
341 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_evict.c
344 Buffer Object Memory Shrinking
345 ------------------------------
347 This section documents the interface function for shrinking memory usage
348 of buffer object caches. Shrinking is used to make main memory
349 available. Note that this is mostly orthogonal to evicting buffer
350 objects, which has the goal to make space in gpu virtual address spaces.
352 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_shrinker.c
358 .. kernel-doc:: drivers/gpu/drm/i915/i915_cmd_parser.c
359 :doc: batch buffer command parser
361 .. kernel-doc:: drivers/gpu/drm/i915/i915_cmd_parser.c
367 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_batch_pool.c
370 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_batch_pool.c
373 User Batchbuffer Execution
374 --------------------------
376 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_execbuffer.c
377 :doc: User command execution
379 Logical Rings, Logical Ring Contexts and Execlists
380 --------------------------------------------------
382 .. kernel-doc:: drivers/gpu/drm/i915/intel_lrc.c
383 :doc: Logical Rings, Logical Ring Contexts and Execlists
385 .. kernel-doc:: drivers/gpu/drm/i915/intel_lrc.c
391 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_gtt.c
392 :doc: Global GTT views
394 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_gtt.c
397 GTT Fences and Swizzling
398 ------------------------
400 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_fence_reg.c
403 Global GTT Fence Handling
404 ~~~~~~~~~~~~~~~~~~~~~~~~~
406 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_fence_reg.c
407 :doc: fence register handling
409 Hardware Tiling and Swizzling Details
410 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
412 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_fence_reg.c
413 :doc: tiling swizzling details
418 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_tiling.c
421 .. kernel-doc:: drivers/gpu/drm/i915/i915_gem_tiling.c
422 :doc: buffer object tiling
430 .. kernel-doc:: drivers/gpu/drm/i915/intel_wopcm.c
436 GuC-specific firmware loader
437 ----------------------------
439 .. kernel-doc:: drivers/gpu/drm/i915/intel_guc_fw.c
442 GuC-based command submission
443 ----------------------------
445 .. kernel-doc:: drivers/gpu/drm/i915/intel_guc_submission.c
446 :doc: GuC-based command submission
448 .. kernel-doc:: drivers/gpu/drm/i915/intel_guc_submission.c
454 .. kernel-doc:: drivers/gpu/drm/i915/intel_guc_fwif.h
455 :doc: GuC Firmware Layout
460 .. kernel-doc:: drivers/gpu/drm/i915/intel_guc.c
461 :doc: GuC Address Space
466 This sections covers all things related to the tracepoints implemented
469 i915_ppgtt_create and i915_ppgtt_release
470 ----------------------------------------
472 .. kernel-doc:: drivers/gpu/drm/i915/i915_trace.h
473 :doc: i915_ppgtt_create and i915_ppgtt_release tracepoints
475 i915_context_create and i915_context_free
476 -----------------------------------------
478 .. kernel-doc:: drivers/gpu/drm/i915/i915_trace.h
479 :doc: i915_context_create and i915_context_free tracepoints
484 .. kernel-doc:: drivers/gpu/drm/i915/i915_trace.h
485 :doc: switch_mm tracepoint
492 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
493 :doc: i915 Perf Overview
495 Comparison with Core Perf
496 -------------------------
497 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
498 :doc: i915 Perf History and Comparison with Core Perf
500 i915 Driver Entry Points
501 ------------------------
503 This section covers the entrypoints exported outside of i915_perf.c to
504 integrate with drm/i915 and to handle the `DRM_I915_PERF_OPEN` ioctl.
506 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
507 :functions: i915_perf_init
508 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
509 :functions: i915_perf_fini
510 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
511 :functions: i915_perf_register
512 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
513 :functions: i915_perf_unregister
514 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
515 :functions: i915_perf_open_ioctl
516 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
517 :functions: i915_perf_release
518 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
519 :functions: i915_perf_add_config_ioctl
520 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
521 :functions: i915_perf_remove_config_ioctl
526 This section covers the stream-semantics-agnostic structures and functions
527 for representing an i915 perf stream FD and associated file operations.
529 .. kernel-doc:: drivers/gpu/drm/i915/i915_drv.h
530 :functions: i915_perf_stream
531 .. kernel-doc:: drivers/gpu/drm/i915/i915_drv.h
532 :functions: i915_perf_stream_ops
534 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
535 :functions: read_properties_unlocked
536 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
537 :functions: i915_perf_open_ioctl_locked
538 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
539 :functions: i915_perf_destroy_locked
540 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
541 :functions: i915_perf_read
542 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
543 :functions: i915_perf_ioctl
544 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
545 :functions: i915_perf_enable_locked
546 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
547 :functions: i915_perf_disable_locked
548 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
549 :functions: i915_perf_poll
550 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
551 :functions: i915_perf_poll_locked
553 i915 Perf Observation Architecture Stream
554 -----------------------------------------
556 .. kernel-doc:: drivers/gpu/drm/i915/i915_drv.h
557 :functions: i915_oa_ops
559 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
560 :functions: i915_oa_stream_init
561 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
562 :functions: i915_oa_read
563 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
564 :functions: i915_oa_stream_enable
565 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
566 :functions: i915_oa_stream_disable
567 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
568 :functions: i915_oa_wait_unlocked
569 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
570 :functions: i915_oa_poll_wait
572 All i915 Perf Internals
573 -----------------------
575 This section simply includes all currently documented i915 perf internals, in
576 no particular order, but may include some more minor utilities or platform
577 specific details than found in the more high-level sections.
579 .. kernel-doc:: drivers/gpu/drm/i915/i915_perf.c
585 The drm/i915 driver codebase has some style rules in addition to (and, in some
586 cases, deviating from) the kernel coding style.
588 Register macro definition style
589 -------------------------------
591 The style guide for ``i915_reg.h``.
593 .. kernel-doc:: drivers/gpu/drm/i915/i915_reg.h
594 :doc: The i915 register macro definition style guide