Lines Matching +full:shared +full:- +full:glapi

1 Mesa 22.0.0 Release Notes / 2022-03-09
20 ---------------
24    e6c41928b5b9917485bd67cec22d15e62cad7a358bf4c711a647979987601250  mesa-22.0.0.tar.xz
28 ------------
30 - lavapipe,radv,anv KHR_dynamic_rendering
31 - radv EXT_image_view_min_lod
32 - VK_KHR_synchronization2 on RADV.
33 - OpenSWR has been moved to the Amber branch
34 - radeonsi, zink ARB_sparse_texture
35 - d3d12 GLES3.1 (shader storage buffers, images, compute, indirect draw, draw params, ARB_framebuff…
36 - radeonsi, zink ARB_sparse_texture2
37 - zink EXT_memory_object, EXT_memory_object_fd, EXT_semaphore, EXT_semaphore_fd
38 - anv VK_VALVE_mutable_descriptor_type
39 - Vulkan 1.3 on RADV,Anv.
40 - radeonsi, zink ARB_sparse_texture_clamp
44 ---------
46 - anv: Unable to import I915_FORMAT_MOD_Y_TILED surfaces produced by iHD vaapi driver
47 - anv doesn't always resolve aux buffers with private bindings on transition to external queue
48 - Redraw freeze after upgrade to Xwayland 21.1.3
49 - [ANV] Rendering corruption in DOOM Eternal
50 - lavapipe: dEQP-VK.spirv_assembly.instruction.compute.float16.arithmetic_3.step fails
51 - radv: When using VS+PS only, primitive ID is mixed up after NGG culling
52 - [22.0.0_rc{1,2}] radeonsi: broken support for HD7850 (radeon 0000:0b:00.0: ring 0 stalled for mor…
53 - RADV: Texture seams in The Evil Within 2 (regression)
54 - ANV: Bad output from TransformFeedback . Regression from Mesa 21. Something to do with VB+XFB -> …
55 - radv: CullDistance fail
56 - Changing the buffer modifer via Wayland dmabuf-feedback apparently does not reliably reset the bu…
57 - radv: causes hang untl gpu reset on navy_flounder
58 - Crocus related crashes when running QT apps
59 - Please add missing ray tracing Vulkan extensions to features.txt
60 - [anv][bisected][regression][tgl] dEQP-VK.dynamic_rendering.suballocation.multisample_resolve.* fa…
61 - radv/aco: DXVKs strict float emulation causes black screen/graphical corruption on RDNA1
62 - intel: env vars doc out of date
63 - [REGRESSION][BISECTED] iris: Qutebrowser/QtWebEngine sporadically flashes the window in white
64 - Flickering Intel Uhd 620 Graphics
65 - Vulkan Wayland WSI returns empty surface formats
66 - [radeon, rs690] piglit spec@glsl-1.10@execution@built-in-functions@fs-atan-float + 10 other 'atan…
67 - [bisected] artifacts in vaapi h.264 playback
68 - [bisected] artifacts in vaapi h.264 playback
69 - radv: msaa regression
70 - [DG2] dEQP-GLES3.functional.shaders.derivate.fwidth* failures
71 - FIFO present mode doesn't wait
72 - Dota2 lockup wsi/x11: Avoid a class of deadlocks in the WSI queue thread
73 - [vulkan][intel] Implement VK_VALVE_mutable_descriptor_type
74 - Broken Terraria & Glitches in Forza Horizon 4
75 - radv: Q2RTX with cswave32 crashes the GPU
76 - CopelliaSim crashes on adding vision sensor to a scene on Ubuntu 20+, runs fine on Ubuntu 18.04
77 - radv: renderpass regression
78 - Dirt Rally: Flickering glitches on certain foliage going from Mesa 21.2.5 to 21.3.0
79 - FrontFacing input is broken on Intel/Vulkan
80 - glGetActiveUniform failing with GL_INVALID_VALUE for no reason
81 - llvmpipe: Unimplemented get_driver_uuid/get_device_uuid causes segfaults in e.g. Wine
82 - All vulkan apps terminate with VK_ERROR_DEVICE_LOST
83 - Texture rendering broken on Vega 56, probably due to MR 14171 (tested on Battlefield 1 / vkd3d-pr…
84 - Major graphic distortions with RADV in DX11/DX12 with Mesa-git as of January 12th 2022
85 - Setting mesa_glthread to true crashes Minecraft when Sodium optimisation mod is enabled.
86 - ANV: deviceUUID assumes a single device per system
87 - MESA Overlay Vulkan layer version
88 - [radeonsi, regression, bisected]: Rendering issues with Factorio
89 - Distorted graphics when running Battle.net app under Wine with Radeon hardware
90 - [r300g] Battle.net app window content rotated and stretched
91 - Freedreno reporting 10 MB of Video memory that's causing issues with WineD3D on Poco F1
92 - Panfrost - RK3399 - FACTORIO - glitches everywhere
93 - Baldurs Gate 3 (Patch 6) - ribbon-like artifacts on textures
94 - mesa >= 21.x (with llvmpipe) will break the game JVGS
95 - Game Starsector crashes under certain circumstances with mesa 21.3.+
96 - FreeSpace models incorrectly rendered on Polaris cards, causing system freeze
97 - ci: Test with Skqp
98 - ci: Test with Skqp
99 - Metro Exodus (native) fails to launch
100 - Poor performance on RX 580 (Polaris20)
101 - VAAPI H.264 encoding is slow on Polaris vs. HEVC and Windows AMF
102 - v3dv: Missing gem handle reference tracking logic
103 - Incomplete evaluation of nested DEFINE macros
104 - [r300g, bisected] piglit glsl-fs-discard-04 fails
105 - Panfrost G52 Firefox terrible glitches on youtube playback
106 - prim restart broken on crocus by prim restart fix for d3d12
107 - RADV: Invalid fast clear triggered
108 - Assassin's Creed Syndicate crashes with Mesa 21.3.0+ ACO
109 - anv: dEQP-VK.graphicsfuzz.spv-stable-pillars-volatile-nontemporal-store fails
110 - seg-fault with soft fp64 on iris after register allocation change
111 - Taking RGP traces on VANGOGH results in hangs
112 - Flickering and blackscreen on Mpv and Clapper (and also low performance on glxgears)
113 - intel/vec4: Rework texture handling to not use \`ir_texture_opcode`
114 - lavapipe leaks introduced in eb7eccc76f0a02b3a3291adaa39788b157979dc5
115 - DXVK SIGBUS with Turnip on Poco F1 at loading to open world.
116 - RADV: IsHelperInvocationEXT query is not considered volatile in ACO
117 - [GraphicsFuzz] dEQP-VK.graphicsfuzz.stable-binarysearch-tree-nested-if-and-conditional
118 - [bisected] Mesa 21.3.x breaks GBM with NVIDIA closed source driver 495.44
119 - [DG2] dEQP-GL[45|ES31].functional.shaders.builtin_functions.pack_unpack.packhalf2x16_compute fail
120 - Steam reports fatal error with missing OpenGL GLX extension not supported by display since Decemb…
121 - SNK HEROINES Tag Team Frenzy stuck on white screen
122 - Run-time dependency dri found: NO
123 - Rise of the Tomb Raider segfault
124 - glsl/compiler: suspcious snippet of code (Unused variable)
125 - GPU Crash in Yuzu 6600xt 5.15
126 - anv: gfxbench 5.0 vulkan fails on android
127 - [spirv-fuzz]  lower_trivial_continues_block: Assertion \`!first_instr || instr_is_continue(first_…
128 - [RADV] Crash in Metro Exodus in Caspain chapter and Sam's Story
129 - [nir][radv] more undefined shifts caught by gcc's sanitizer
130 - NIR validation failed after nir_copy_prop
131 - lima: Corrupted Android-12 UI on Allwinner A64
132 - RADV/ACO: Rendering glitches in Forza Horizon 5 windshields
133 - NV_pixel_buffer_object extension can be available in a GLES2 context but the associated glGet con…
134 - ThreadSanitizer: data races inside libvulkan_radeon.so
135 - [radeonsi] Commit b1b491cdbbac1ccb785f5eb1802006f76ec253f5 leads to memory leak
136 - dEQP-GLES31.*imulextended* compiling fp64 glsl 4.00 shader
137 - anv: vkMapMemory fails "Memory object already mapped" with VA-API (mpv --gpu-api=vulkan --hwdec=v…
138 - RADV: crash in radv_GetPhysicalDeviceImageFormatProperties2
139 - RADV/ACO: Monster Hunter Rise Demo renders wrong results
140 - radv: Odd lack of implicit host memory invalidation
141 - Regression/Bisected: Crash in Godot games when steam overlay enabled
142 - RADV: IsHelperInvocationEXT query is not considered volatile in ACO
143 - ANV: error submitting with the same semaphore for wait and signal - regression?
144 - Vulkan does not work when 10-bit colors ( DefaultDepth 30) are enabled under Xorg on AMD/RADV
145 - [TGL+] anv: some dEQP-VK.drm_format_modifiers.* fails
146 - radv: SQTT generates trace with no data
147 - Mesa 21.3rc3 - compile failure
148 - iris: subslice assertion failure on some hardware during initialization
149 - Final Fantasy V (Old version): Random text characters are not displayed
150 - Diagonal rendering artifacts in Tomb Raider
151 - vkCmdSetViewport sometimes ignored on Ivy Bridge, Haswell
152 - dota2 black squares corruption
153 - [hsw][bisected][regression] dEQP-VK.reconvergence.*nesting* failures
154 - Piglit.spec.arb_shader_image_load_store.compiler.expression-allowed-1d_vert failed.
155 - anv: dEQP-VK.wsi.wayland.<various> failures
156 - freedreno: Crash in fd_bo_cache_alloc()
157 - radv_android.c: build errors due to commit 49c3a88
158 - dEQP-EGL.functional.sharing.gles2.multithread.* regression with Iris
159 - [Turnip] Forsaken Remastered rendering issue
160 - [radeonsi] Euro Truck Simulator 2: broken mimimap
161 - Old Linux Rocket League version failing to start on Poco F1.
162 - [regression][bisected] Launching Valheim OpenGL game leads to GPU Hang
163 - Android Meson build regression: hardware/system information apps crash on Raspberry Pi 4
164 - radv: format properties are broken with modifiers
165 - anv: dEQP-VK.graphicsfuzz.cov-multiple-one-iteration-loops-global-counter-write-matrices fails
169 -------
173 - clover: implement CL_IMAGE_BUFFER
174 - clover/image: add dimension property
178 - mesa/x11: Remove the swrast-classic-based fake libGL
179 - mesa: Remove unused _mesa_all_buffers_are_unmapped
180 - mesa: Remove unused _mesa_get_render_format
181 - mesa: Remove unused _mesa_DrawTexx{,v}
182 - mesa: Remove unused _check_*TexGen*OES
183 - mesa: Remove unused _mesa_apply_ci_transfer_ops
184 - mesa: Remove unused _mesa_compressed_image_address
185 - mesa: Remove unused _es_color4ub
186 - mesa: Remove unused _es_RenderbufferStorageEXT
187 - mesa: Remove unused _es_{,Get}TexGenfv
188 - mesa: Remove unused _vbo_current_binding
189 - mesa/vbo: Always use buffer objects for storage
190 - mesa/program: Dead code cleanup
191 - mesa: Remove unused execmem code
192 - mesa: Make _mesa_generate_mipmap_level static
193 - glapi: Remove remnants of EXT_paletted_texture and the imaging subset
194 - docs: Remove no-longer-accurate text about the xlib driver
195 - mesa: Remove unused _mesa_initialize_visual
196 - dri: Remove unused driContextSetFlags
197 - dri: Remove unused driUpdateFramebufferSize
198 - dri: Remove unused driGetRendererString
199 - glx: Fix GLX_NV_float_buffer fbconfig handling
200 - mesa: Remove unused _mesa_allow_light_in_model
201 - mesa: Remove unused _mesa_AllocTextureStorage_sw
202 - mesa: Remove unused _mesa_bind_texture
203 - mesa: Remove unused _mesa_format_fallback_rgbx_to_rgba
204 - mesa: Remove unused _mesa_get_linear_format_srgb
205 - mesa: Remove unused _mesa_convert_colors
206 - mesa: Remove unused _mesa_all_varyings_in_vbos
207 - mesa: Remove unused _mesa_delete_nameless_texture
208 - mesa/math: Remove unused m_translate.c
209 - mesa: Remove unused _mesa_is_alpha_to_coverage_enabled
210 - mesa: Remove unused _mesa_is_front_buffer_{draw,read}ing
211 - mesa: Remove unused _mesa_set_sampler_{filters,srgb_decode,wrap}
212 - mesa: Remove unused src/mesa/x86-64
213 - wsi/x11: Avoid a class of deadlocks in the WSI queue thread
214 - dri_interface: Remove the remaining DRI1 API definitions
218 - v3dv/pipeline: don't clone the nir shader at pipeline_state_create_binning
219 - v3d/clif: add support for dumping GS shader state
220 - broadcom/compiler: remove unused macro and function definition
221 - v3d: remove unused include
222 - v3d: remove static v3d_start_binning
223 - gallium/u_blitter: clean up texcoords ZW when filling up just XY
224 - v3d,v3dv: move TFU register definition to a common header
225 - vulkan: move common format helpers to vk_format
226 - broadcom/compiler: make shaderdb debug output compatible with shaderdb's report tool
227 - meson: bump meson requirement to 0.53.0
228 - v3d: support for texture buffer objects
229 - v3d: add support for no buffer object bound
230 - v3d: restrict formats supported for PIPE_BIND_SHADER_IMAGE
231 - v3d: enable ARB_texture_buffer_object and ARB_texture_buffer_range
232 - broadcom/compiler: avoid unneeded sint/unorm clamping when lowering stores
233 - v3dv: simplify v3dv_debug_ignored_stype
234 - vulkan: return default string for undefined enum
235 - v3dv: remove unused v3dv_descriptor_map_get_texture_format
236 - v3dv: check correct format when load/storing on a depth/stencil buffer
237 - vc4/nir_lower_blend: update write mask when we update num components
241 - Use initial-exec TLS for glibc only, enable TLS elsewhere
242 - Auto-enable TLSDESC support
243 - meson: check for lld split TLSDESC bug (fixes #5665)
244 - Fix TSD stubs for non-initial-exec case (fixes #5667).
245 - meson: tlsdesc: minor reformatting, add comments
249 - panfrost: Don't allow rendering/texturing 48-bit
250 - panfrost: Detect implementations support AFBC
251 - panfrost,panvk: Use dev->has_afbc instead of quirks
252 - panfrost: Remove unused MIDGARD_NO_AFBC quirk
253 - panfrost: Fix gl_FragColor lowering
254 - panfrost: Workaround ISSUE_TSIX_2033
255 - panfrost: Add internal afbc_formats
256 - panfrost: Decompress for incompatible AFBC formats
257 - panfrost: Enable AFBC on v7
258 - panfrost: Remove ancient TODO
259 - panfrost: Remove duplicated #if
260 - panfrost: Rename depth bias fields
261 - panfrost: Add alpha_zero_nop/one_store predicate
262 - panfrost: Test alpha_zero_nop/one_store predicates
263 - panfrost: Pass through alpha_zero_nop/one_store
264 - panvk: Pass through alpha_zero_nop/one_store flags
265 - nir/lower_blend: Use correct clamp for SNORM
266 - mesa: Require MRT support for GL3/ES3
267 - nir/lower_pntc_ytransform: Support PointCoordIsSysval
268 - pan/bi: Fix typo in helper invocation analysis
269 - pan/bi: Make bi_index padding explicit
270 - pan/bi: Add secondary staging count
271 - pan/bi: Add second destination to TEXC
272 - pan/bi: Add bifrost_texture_operation_mode enum
273 - pan/bi: Add bifrost_dual_texture_operation struct
274 - pan/bi: Use BIFROST_TEXTURE_OPERATION_SINGLE enum
275 - pan/bi: Support dual texture scheduling
276 - pan/bi: Add bi_dual_tex_as_u32 helper
277 - pan/bi: Fix up dual texturing registers
278 - pan/bi: Add dual texture fusing pass
279 - pan/bi: Test dual texture fusing
280 - pan/bi: Enable dual texture fusing pass
281 - pan/mdg: Remove duplicate compiler option
282 - pan/bi: Annotate Valhall instructions with units
283 - pan/bi: Add full form of Valhall MUX instruction
284 - pan/bi: Add sqrt form of Valhall FREXPM
285 - pan/bi: Add Valhall's special FMA_RSCALE instructions
286 - pan/bi: Add XML for assembling Valhall image stores
287 - pan/bi: Forbid unaligned staging registers on Valhall
288 - pan/bi: Confirm IDP unit on Valhall
289 - pan/bi: Suppress uniform validation for LD_BUFFER
290 - pan/bi: Add XML for LD_BUFFER
291 - asahi: Rename PANDECODE->AGXDECODE
292 - asahi: Deflake addresses
293 - asahi: Allocate special scratch buffers
294 - asahi: Fix agx_map_* structures
295 - asahi: Remove silly magic numbers
296 - asahi: Remove obnoxious workaround
297 - asahi: Fix BIND_PIPELINE sizing and alignment
298 - panfrost: Collapse 0 parameters in drm-shim
299 - panfrost: Handle AFBC_FEATURES in drm-shim
300 - panfrost: Add empty tile flags to GenXML
301 - panfrost: Only build GPU indirect kernels for v7
302 - pan/bi: Link with Valhall disassembler
303 - pan/va: Only hex dump when verbosely disassembling
304 - panfrost: Add "hex" type to GenXML
305 - panfrost: Add XML for Valhall data structures
306 - panfrost: Zero initialize disassembler stats
307 - panfrost: Don't shadow Mesa's fui()
308 - panfrost: Add Valhall support to pandecode
309 - pan/va: Improve assembler unit test output
310 - pan/va: Disambiguate sign of CSEL instructions
311 - pan/va: Add more assembler tests
312 - pan/va: Make LD_VAR index more fine-grained
313 - pan/va: Add sample/update modes to LD_VAR
314 - pan/va: Add table parameter to LD_ATTR_IMM
315 - pan/mdg: Fix definition of UBO unpack
316 - docs/macos: Update for recent Mesa changes
317 - pan/indirect_draw: Don't upload garbage UBO
318 - pan/indirect_draw: Split out update_dcd
319 - pan/indirect_draw: Support IDVS jobs
320 - panfrost: Track preloaded registers
321 - panfrost: Set preload descriptor more accurately
322 - panfrost: Remove unused shader info bits
323 - panfrost: Fix Secondary Shader field
324 - panfrost: Treat IDVS jobs as tiler for scoreboarding
325 - panfrost: Add IDVS fields to shader_info
326 - panfrost: Split out regalloc/preload helpers
327 - panfrost: Remove regalloc from v6.xml
328 - panfrost: Set secondary_* fields for IDVS
329 - panfrost: Extract panfrost_draw_emit_vertex_section
330 - panfrost: Emit IDVS jobs
331 - panfrost: Extract panfrost_batch_skip_rasterization
332 - panfrost: Skip rasterizer discard draws without side effects
333 - panfrost: Align instance size for IDVS
334 - panfrost: Add panfrost_compile_inputs->no_idvs option
335 - pan/bi: Extract bi_finalize_nir
336 - pan/bi: Allow UBO pushing to run multiple times
337 - pan/bi: Add IDVS mode to bi_context
338 - pan/bi: Remove the "wrong" stores in IDVS variants
339 - pan/bi: Split out varying store paths
340 - pan/bi: Use position shader ST_CVT path
341 - pan/bi: Add helper to decide if IDVS should be used
342 - pan/bi: Specialize shaders for IDVS
343 - pan/va: Remove extra LD_VAR_IMM_F32 source
344 - pan/va: Rename LEA_ATTR to LEA_VARY
345 - pan/va: Add .signed bit to right shift instructions
346 - pan/va: Generalize LD_VAR_IMM_* to support flat varyings
347 - pan/bi: Don't call useless NIR passes
348 - panfrost: Make pan_merge macro more robust
349 - pan/bi: Don't read base for combined stores
350 - pan/mdg: Don't read base for combined stores
351 - nir: Eliminate store_combined_output_pan BASE
352 - panfrost: Simplify blend lowering pass
353 - nir: Extend store_combined_output_pan
354 - panfrost: Combine dual source blends
355 - panfrost: Remove pan_nir_reorder_writeout
356 - pan/bi: Use is_staging_src helper
357 - pan/bi: Allow an extra staging source
358 - pan/bi: Use fused dual source blending
359 - pan/bi: Fix load_const of 1-bit booleans
360 - gallium/util: Add pixel->blocks box helper
361 - lima,panfrost: Correct pixel vs block mismatches
362 - v3d: Use u_box_pixels_to_blocks helper
363 - vc4: Use u_box_pixels_to_blocks helper
364 - pan/decode: Remove hierarchy mask check
365 - pan/decode: Include addresses for jobs
366 - pan/decode: Track mmaps with a red-black tree
367 - pan/decode: Add hexdump helper
368 - pan/decode: Add pandecode_dump_mappings
369 - pan/decode: Decode Valhall surface descriptor
370 - panfrost: Remove FBD pointer on Bifrost XML
371 - pan/decode: Don't print Preload twice
372 - pan/decode: Disassemble Bifrost quietly
373 - pan/bi: Schedule around blend shader register clobbering
374 - asahi: Fake more CAPs with dEQP hacks mode
375 - asahi: Warn when hacks mode is enabled
376 - asahi: Add XML for unknown 0x4a packet
377 - asahi: Break out Fragment Parameters word
378 - asahi: Rectify confusing XML comment
379 - asahi: Route sample mask from shader
380 - agx: Add sample_mask instruction
381 - agx: Handle discard intrinsics
382 - agx: Lower UBO loads to use per-element indexing
383 - panfrost: Use u_reduced_prim for primitive checks
384 - panfrost: Only cull polygons
385 - pan/bi: Pull BLEND precolouring out of per-dest loop
386 - pan/decode: Fix missing newlines in error messages
387 - panfrost: Remove HAS_SWIZZLES quirk
388 - panfrost: Remove MIDGARD_SFBD quirk
389 - panfrost: Remove MIDGARD_BROKEN_FP16 quirk
390 - panfrost: Remove NO_TILE_ENABLE_MAP quirk
391 - panfrost: Remove MIDGARD_{NO_TYPED_BLEND_STORES,MISSING_LOADS}
392 - panfrost: Don't set NO_BLEND_PACKS on Bifrost
393 - panfrost: Simplify format class selection
394 - panfrost: Remove NO_BLEND_PACKS quirk
395 - panfrost: Make primary_shader boolean
396 - panfrost: Fix v9 "Stencil from shader" bit
397 - panfrost: Don't pass quirks to pan_lower_framebuffer
398 - panfrost: Centralize our model list
399 - panfrost: Replace panfrost_model_name with model->name
400 - panfrost: Make the GPU allowlist implicit
401 - panfrost: Get performance counters from table
402 - pan/bi: Clean up quirks
403 - pan/bi: Assume future Valhall is 16-wide warps
404 - panfrost: Add Mali-G51 support
405 - docs/panfrost: Add new Midgard/Bifrost chips
406 - pan/va: Add .absolute bit to BRANCHZI
407 - pan/va: Add missing <clamp/> to V2F32_TO_V2F16
408 - pan/va: Add missing fields to LD_TILE
409 - pan/va: Test LD_TILE assembly
410 - pan/va: Fix MKVEC.v2i16 lane select
411 - pan/va: Fix lane select for [US]_TO_[USF]32
412 - pan/va: Vectorize 8->16-bit conversions
413 - pan/va: Add 2-channel 8-bit swizzles for conversions
414 - pan/va: Add lots of swizzle assembler tests
415 - pan/va: Handle shift lanes in assembler
416 - pan/va: Add ARM_shader_framebuffer_fetch asm test
417 - pan/bi: Avoid \*FADD.v2f16 hazard in optimizer
418 - pan/bi: Avoid \*FADD.v2f16 hazard in scheduler
419 - pan/bi: Lower swizzles on CSEL.i32/MUX.i32
420 - panvk: Use more reliable assert for UBO pushing
421 - pan/bi: Specialize IDVS in NIR
422 - panfrost: Fix FD resource_get_handle
423 - panfrost: Fix set_sampler_views for big GL
424 - panfrost: Handle NULL sampler views
425 - panfrost: Handle NULL samplers
426 - panfrost: Flush resources when shadowing
427 - panfrost: Push twice as many uniforms
431 - lima: Fix drawing wide lines
435 - gallium: add a link shader hook
436 - virgl: Link shader program
440 - genxml/gen125: Update 3DSTATE_TE fields
441 - iris: Enable geometry distribution
442 - anv: Enable geometry distribution
443 - iris: Enable tessellation redistribution
444 - anv: Enable tessellation redistribution
445 - anv, iris: Add Wa_16011773973 for DG2
446 - anv, iris: Add Wa_22011440098 for DG2
450 - radv: Fix modifier property query.
451 - radv: Add bufferDeviceAddressMultiDevice support.
452 - amd/rgp: Use VGH clocks for RGP workaround.
453 - radv: Disable coherent L2 optimization on cards with noncoherent L2.
454 - ci: Bump libdrm for the android image.
455 - ci: Add libelf to the Android image.
456 - amd/addrlib: Use alternative ADDR_C_ASSERT definition.
457 - radv: Remove android build warning.
458 - radv: Always use linker script when possible.
459 - aco: Remove useless sub-expr.
460 - meson: Check arguments before adding.
461 - amd/addrlib: Ignore self-assign warnings.
462 - util: Add support for clang::fallthrough.
463 - ci: Add RADV to Android CI.
464 - radv: Fix memory corruption loading RT pipeline cache entries.
465 - radv: Add more checking of cache sizes.
466 - radv: Don't crash if VkExternalImageFormatProperties isn't provided.
467 - radv: Avoid using a new subpass for ds clears.
468 - radv: Stop using a subpass for color clears.
469 - radv: Remove the skipping of framebuffer emission if we don't have a framebuffer.
470 - radv: Set RB+ registers correctly without framebuffer.
471 - radv: Don't emit framebuffer state if there is no renderpass active.
472 - radv: Add named constants for max framebuffer width/height.
473 - radv: Do not use VK_FORMAT_UNDEFINED in meta passes.
474 - radv: Support VK_KHR_dynamic_rendering for pipeline creation.
475 - radv: Support Begin/EndRendering.
476 - radv: Support dynamic rendering inheritance info.
477 - radv: Expose the VK_KHR_dynamic_rendering extension.
478 - radv/amdgpu: Add a syncobj per queue.
479 - radv/winsys: Add queue family param to submit.
480 - radv/amdgpu: Add support for submitting 0 commandbuffers.
481 - radv: Use the winsys 0 cmdbuffer submission support.
482 - radv: Set up ETC2 emulation wiring.
483 - radv: Use the correct base format for reintepretation.
484 - radv: Add extra plane for decoding ETC images with emulation.
485 - radv: Add ETC2 decode shader.
486 - radv: Deal with border colors with emulated ETC2.
487 - radv: Expose the ETC2 emulation.
488 - radv/amdgpu: Only wait on queue_syncobj when needed.
489 - radv: Add safety check for RGP traces on VanGogh.
490 - radv: Use correct buffer size for query pool result copies.
491 - radv: Skip wait timeline ioctl with 0 handles.
492 - radv/amdgpu: Use VkResult for wait_timeline_syncobj.
493 - vulkan/runtime: Refactor queue submit to take an argument struct.
494 - vulkan/runtime: Add sparse bind support.
495 - radv: Use dispatch table for QueueWaitIdle in the SQTT layer.
496 - radv: Use dispatch table for wsi_display.c
497 - radv: Use vulkan runtime for device lost.
498 - meson: Bump libdrm_amdgpu version req to 2.4.109.
499 - radv: Initialize vk device drm fd.
500 - radv: Add function to allow WSI signalling fences/semaphores.
501 - radv: Set horizontal sync types.
502 - radv: Use vk_command_buffer for preambles.
503 - radv: Add new cs_submit2 winsys call.
504 - radv: Add new submission path for use by the common sync framework.
505 - radv: Delete lots of sync code.
506 - radv: Remove syncobj reset mechanism.
507 - radv: Rename submit2->submit.
508 - radv: Remove dependencies on vk_common entrypoints.
509 - radv: Set optimal copy alignment to 1.
510 - radv: Add common entrypoint dependency.
511 - radv: Use MAX_PUSH_CONSTANTS_SIZE for saved push constants.
512 - radv: Use 16-bits to store push constant indices.
513 - radv: 256 byte push constants.
514 - Revert "nir/algebraic: distribute fmul(fadd(a, b), c) when b and c are constants"
515 - util/fossilize_db: Fix double free in error handling.
516 - vulkan/wsi/display: Add common implementation of VK_EXT_display_control.
517 - anv: Remove VK_EXT_display_control support in favor of common impl.
518 - radv: Remove VK_EXT_display_control support in favor of common impl.
519 - radv/amdgpu: Fix handling of IB alignment > 4 words.
520 - radv/amdgpu: Use aligned sizing for IB buffers.
521 - radv: Handle SDMA for padding.
522 - radv: Fix preamble argument order.
526 - kmsro: Add 'rcar-du' driver support
530 - vulkan: Fix weak symbol emulation when compiling with MSVC
531 - vulkan: Set unused entrypoints to vk_entrypoint_stub when compiling with MSVC
532 - vulkan: Fix entrypoint generation when compiling for x86 with MSVC
533 - microsoft/compiler: Fix dxil_nir_create_bare_samplers()
534 - gallium/d3d12: Don't use designated initializers
535 - d3d12: Fix "use of designated initializers requires at least '/std:c++20'" error
536 - microsoft/compiler: Fix sampler/texture array emission
537 - microsoft/compiler: Skip images in redirect_texture_derefs()
538 - microsoft/compiler: textureLoad() doesn't take a LOD on MS textures
539 - microsoft/spirv_to_dxil: Remove dead variables after the struct split pass
540 - microsoft/spirv_to_dxil: Allow dumping NIR
541 - microsoft/spirv_to_dxil: Support [un]conditional YZ-flip
542 - microsoft/spirv_to_dxil: Discard PSIZ accesses
543 - microsoft/spirv_to_dxil: Lower atomics to their dxil variants
544 - microsoft/spirv_to_dxil: Make sure the SampleMask is a uint
545 - microsoft/spirv_to_dxil: Define idep_libspirv_to_dxil
549 - radeon/vcn: update vcn1 enc interface
550 - radeon/vcn: update vcn2 enc interface
551 - radeon/vcn: remove redundancy for vcn2 enc
552 - radeon/vcn: combine encode params func
553 - radeon/vcn: combine session init func
557 - nir/schedule: Handle nir_intrisic_scoped_barrier
558 - nir: Add nir_var_mem_image
559 - intel/fs: Consider nir_var_mem_image for TGM fences
560 - anv: Use nir_foreach_image_variable
561 - spirv: Use nir_var_mem_image
562 - intel/compiler: Rename vec4 test fixtures
563 - intel/compiler: Build all tests in a single binary
564 - spirv: Use a single binary for gtests
565 - nir: Use a single binary for gtests
566 - util: Consolidate existing gtests in a single binary
567 - util: Move tests in single file directories to tests/
568 - util: Convert u_atomic_test to use gtest
569 - util: Convert blob_test to use gtest
570 - util: Convert mesa-sha1_test to use gtest
571 - util: Convert rb_tree_test to gtest
572 - util: Convert roundeven_test to use gtest
573 - util: Convert sparse array multithread test to use gtest
574 - util: Move test sources to tests/ directory
575 - intel/compiler: Add helpers to select SIMD for compute shaders
576 - intel/compiler: Use SIMD selection helpers for CS
577 - intel/compiler: Use SIMD selection helpers for variable workgroup size
578 - intel/compiler: Don't use SIMD larger than needed for workgroup
579 - anv: Get rid of "may be used initialized" warning in anv_QueueSubmit2KHR
580 - anv: Make shaders array in anv_graphics_pipeline fit Task/Mesh
581 - anv: Process FS last when compiling graphics pipeline
582 - intel/compiler: Make brw_nir_populate_wm_prog_data() static
583 - intel/compiler: Use gl_shader_stage_uses_workgroup() helpers
584 - .mailmap: Simplify my name
585 - intel/genxml: Add Mesh Shading structures
586 - intel/genxml: Inline the BODY structs into the instructions
587 - intel/dev: Add an intel_device_info::has_mesh_shading bit
588 - intel/blorp: Add option to emit packets that disable Mesh
589 - gtest: Add mesa-gtest-extras.h with array ASSERT/EXPECT macros
590 - util: Change blob_test to use macro from mesa-gtest-extras.h
591 - pan/bi: Make some headers compilable with C++
592 - pan/bi: Use gtest for test-scheduler-predicates
593 - pan/bi: Use gtest for test-packing
594 - pan/bi: Use gtest for test-pack-formats
595 - pan/bi: Use gtest for test-optimizer
596 - pan/bi: Use gtest for test-constant-fold
597 - pan/bi: Drop unused test helpers
598 - util/ra: Add simple test for register set serialization
599 - util/ra: Fix deserialization of register sets
600 - gtest: Fix output of array ASSERT/EXPECT macros
601 - intel: Add INTEL_DEBUG=task,mesh
602 - intel/compiler: Properly lower WorkgroupId for Task/Mesh
603 - intel/compiler: Handle per-primitive inputs in FS
604 - intel/compiler: Don't stage Task/Mesh outputs in registers
605 - intel/compiler: Don't lower Mesh/Task I/O to temporaries
606 - intel/compiler: Add structs to hold TUE/MUE
607 - intel/compiler: Make MUE available when setting up FS URB access
608 - intel/compiler: Export brw_nir_lower_simd
609 - intel/compiler: Add backend compiler basics for Task/Mesh
610 - intel/compiler: Lower Task/Mesh local_invocation_{id,index}
611 - intel/compiler: Implement Task Output and Mesh Input
612 - intel/compiler: Implement Mesh Output
613 - util: Convert cache test to use gtest
614 - util: Use ralloc for strings in cache test
615 - intel/compiler: Use a struct for brw_compile_tcs parameters
616 - intel/compiler: Use a struct for brw_compile_tes parameters
617 - intel/compiler: Use a struct for brw_compile_gs parameters
618 - intel/compiler: Use a struct for brw_compile_bs parameters
619 - nir: Initialize nir_register::divergent
620 - meson: Bump version required for gtest protocol
621 - Revert "nir: disable a NIR test due to undebuggable & locally unreproducible CI failures"
622 - ci/windows: Remove line numbers of SPIR-V errors in spirv2dxil tests
623 - spirv: Update headers and metadata to SPIR-V 1.6, revision 1
624 - nir: Handle volatile semantics for loading HelperInvocation builtin
625 - spirv: Identify non-temporal image operand added in SPIR-V 1.6
626 - spirv: Use the incorporated names
627 - anv/blorp: Split blorp_exec into a render and compute
628 - anv: Refactor dirty masking in cmd_buffer_flush_state
629 - anv: Simplify assertions related to graphics stages
630 - anv: Add another case to INTEL_DEBUG=pc output
631 - anv: Use pending pipe control mechanism in flush_pipeline_select()
632 - anv/blorp: Apply pending pipe flushes after PIPELINE_SELECT
633 - intel/fs/xehp: Add unit test for handling of RaR deps across multiple pipelines.
634 - anv: SPIR-V 1.6 shaders imply ALLOW_VARYING_SUBGROUP_SIZE
635 - anv: Fix subgroupSupportedStages physical property
636 - intel: Only reserve space for Compute Engine out of URB in Gfx12LP
637 - intel/compiler: Have specific mesh handling in calculate_urb_setup()
638 - intel/compiler: Merge Per-Primitive attribute handling in Mesh case
639 - compiler, intel: Add gl_shader_stage_is_mesh()
640 - intel: Add INTEL_URB_DEREF_BLOCK_SIZE_MESH
641 - intel/common: Add helper for URB allocation in Mesh pipeline
642 - anv: Add boilerplate for VK_NV_mesh_shader
643 - anv: Implement Mesh Shading pipeline
644 - intel/dev: Enable Mesh Shading for DG2
645 - anv: Add experimental support for VK_NV_mesh_shader
649 - vulkan/overlay: support Vulkan 1.2
650 - vulkan/overlay: revise and reformat README
654 - panfrost: Add GPU G76 to the set of known ids
655 - kmsro: Add komeda DPU
656 - panfrost: Don't double-free when handling error for unsupported GPU
660 - zink: Enable VK_KHR_image_format_list for VK_KHR_imageless_framebuffer
661 - zink: Output PackHalf2x16 to uint not float
662 - zink: Avoid redundant cast to uint on PackHalf2x16 result
663 - zink: Set vertex binding stride without dynamic state extensions
664 - Revert "zink: handle vertex buffer offset overflows"
668 - radv: Update description of vk_icdNegotiateLoaderICDInterfaceVersion
669 - v3dv: Update LoaderICDInterfaceVersion to v4
670 - panvk: Export vk_icdGetPhysicalDeviceProcAddr
671 - freedreno, tu: Export vk_icdGetPhysicalDeviceProcAddr
672 - radv: Update LoaderICDInterfaceVersion to v5
673 - panvk: Update LoaderICDInterfaceVersion to v5
674 - anv: Update LoaderICDInterfaceVersion to v5
675 - lavapipe: Update LoaderICDInterfaceVersion to v5
676 - freedreno, tu: Update LoaderICDInterfaceVersion to v5
677 - v3dv: Update LoaderICDInterfaceVersion to v5
681 - mesa: fix misaligned pointer returned by dlist_alloc
683 Cherser-s (1):
685 - radv: handle VK_DESCRIPTOR_TYPE_SAMPLER in VK_VALVE_mutable_descriptor_type extension
687 Chia-I Wu (23):
689 - venus: fix vn_instance_wait_roundtrip when seqno wraps
690 - venus: prefer VIRTGPU_BLOB_MEM_HOST3D for shmems
691 - venus: fix vn_buffer_get_max_buffer_size
692 - venus: add vn_renderer_util.[ch]
693 - venus: add vn_renderer_shmem_pool
694 - venus: use vn_renderer_shmem_pool for reply shmems
695 - venus: add vn_cs_encoder_storage_type
696 - venus: add VN_CS_ENCODER_STORAGE_SHMEM_POOL for VkCommandBuffer
697 - venus: cache shmems
698 - venus: add some trace points
699 - virgl: disable texture uploads with copy transfers
700 - vulkan/wsi: add wsi_common_get_image
701 - anv,lavapipe,v3dv: use wsi_common_get_image
702 - freedreno/drm, turnip: set DRM_RDWR for exported dma-bufs
703 - venus: fix VK_KHR_driver_properties
704 - venus: format with clang-format
705 - venus: remember the memory bound to a swapchain image
706 - venus: handle VkBindImageMemorySwapchainInfoKHR
707 - vulkan/wsi/x11: fix x11_image_init return value on errors
708 - venus: updates to the doc
709 - turnip: respect buf->bo_offset in transform feedback
710 - glthread: call _mesa_glthread_BindBuffer unconditionally
711 - venus: update venus-protocol to 1.3.204
715 - ci/etnaviv: add manual piglit testing
716 - ci/bare-metal: armhf: move BM_ROOTFS to generic place
717 - ci/etnaviv: armhf: switch to .baremetal-test-armhf
718 - ci/etnaviv: no need to force nir anymore
719 - ci/bare-metal: add .baremetal-test-arm64
720 - ci/bare-metal: switch to common .baremetal-test-arm64
721 - panfrost/ci: update piglit fails
722 - etnaviv/ci: update piglit fails
723 - i915g/ci: update piglit fails
724 - iris/ci: update piglit fails
725 - ci: Uprev piglit to af1785f31
726 - mesa: always support occlusion queries
727 - broadcom/ci: use .test-manual-mr
728 - vc4: remove not needed lie about PIPE_CAP_OCCLUSION_QUERY
729 - i915: remove not needed lie about PIPE_CAP_OCCLUSION_QUERY
730 - lima: remove not needed lie about PIPE_CAP_OCCLUSION_QUERY
731 - isaspec: Add support for special {:align=} field
732 - nir: make lower_sample_tex_compare a common pass
733 - nir/nir_lower_tex_shadow: support tex_instr without deref src
734 - etnaviv: make use of nir_lower_tex_shadow
735 - etnaviv: fix FRONT_AND_BACK culling
736 - etnaviv: add multiply_with_8 flag
737 - etnaviv: use bytes for read TX data
738 - etnaviv: add two new HI related perfmon counter
739 - etnaviv: add support for INTEL_blackhole_render
743 - anv: don't advertise vk conformance on GPUs that aren't conformant
747 - vk/format, v3dv: Add a vulkan -> pipe swizzle helper
748 - freedreno/fdl: Constify fdl6_get_ubwc_blockwidth()
749 - freedreno/fdl: Add mip_level to fdl_layout
750 - freedreno/fdl: Add fdl6_view
751 - tu: Use fdl6_view in tu_image_view and cross-check
752 - tu: Switch clear/blit to fdl6_view and cross-check
753 - tu: Remove cross-check scaffolding
754 - tu/clear_blit: Stop creating a franken-image for staging blits
755 - ir3/cse: Support mov instructions
756 - ir3: Use stp/ldp base offset for {load,store}_scratch
757 - tu/clear_blit: Move around copy_format()/tu6_plane_format()
758 - freedreno/a6xx: Rename GRAS_2D_BLIT_INFO
759 - tu: Emit GRAS_LRZ_MRT_BUF_INFO_0
760 - tu: Always write GRAS_LRZ_MRT_BUF_INFO_0
761 - freedreno/a6xx: Emit GRAS_LRZ_MRT_BUF_INFO_0
762 - ir3: Emit barriers for images again
763 - ir3: Don't emit barriers for make_available/make_visible
764 - ir3/spill: Mark root as non-spillable after inserting
765 - ir3/spill: Initial implementation of rematerialization
766 - tu: Enable subgroupBroadcastDynamicId
767 - tu: Add VK_KHR_buffer_device_address stubs
768 - tu: Expose Vulkan 1.2
769 - util/dag: Make edge data a uintptr_t
770 - util/dag: Add dag_add_edge_max_data
771 - ir3/sched: Rewrite delay handling
772 - ir3/postsched: Fix copy-paste mistake
773 - ir3/postsched: Handle sync dependencies better
774 - ir3/delay: Ignore earlier definitions to the same register
775 - ir3/postsched: Rewrite delay handling
776 - ir3/postsched: Only prefer tex/sfu if they are soft-ready
777 - ir3: Stop inserting nops during scheduling
778 - ir3/ra: Consider reg file size when swapping killed sources
779 - ir3/ra: Add missing asserts to ra_push_interval()
780 - ir3/spill: Support larger spill slot offset
781 - ir3/lower_pcopy: Fix shr.b illegal copy lowering
782 - ir3/lower_pcopy: Fix bug with "illegal" copies and swaps
783 - ir3/lower_subgroups: Fix potential infinite loop
784 - ir3/ra: Fix logic bug in compress_regs_left
785 - ir3: Bump type mismatch penalty to 3
786 - ir3: Introduce systall metric and new helper functions
787 - ir3: Use new (sy)/(ss) stall helpers in the compiler
788 - ir3/sched: Rename tex/sfu to sy/ss
789 - ir3/postsched: Rename tex/sfu to sy/ss
790 - ir3: Use (ss) for instructions writing shared regs
791 - ir3, freedreno: Add options struct for ir3_shader_from_nir()
792 - ir3: Pass shader to ir3_nir_post_finalize()
793 - ir3: Add wavesize control
794 - tu, ir3: Support runtime gl_SubgroupSize in FS
795 - tu: Implement VK_EXT_subgroup_size_control
796 - freedreno/fdl: Fix reinterpreting "size-compatible" formats
797 - nir: Reorder ffma and fsub combining
798 - freedreno/a6xx: Name texture descriptor bit
799 - tu/blit: Don't set CLAMPENABLE in sampler for 3d path
800 - tu: Report code size in pipeline statistics
801 - tu: Initial link-time optimizations
802 - nir/lower_subgroups: Rename lower_shuffle to lower_relative_shuffle
803 - nir: Add support for lowering shuffle to a waterfall loop
804 - ir3: Fix copy-paste mistakes in ir3_block_remove_physical_predecessor()
805 - ir3: Rewrite (jp) insertion
806 - ir3/cp: ir3: Prevent propagating shared regs out of loops harder
807 - ir3,tu: Enable subgroup shuffles and relative shuffles
808 - ir3/spill: Fix simplify_phi_nodes with multiple loop nesting
809 - ir3: Use CAN_REORDER instead of NON_WRITEABLE
810 - ir3: Don't always set bindless_tex with readonly images
811 - ir3/nir: Fix 1d array readonly images
815 - virgl: Disable cache for VIRGL_BIND_SAMPLER_VIEW
819 - ci: Uprev deqp-runner to 0.11.0
820 - ci: Support building and installing deqp-runner from source
821 - ci: Do not remove cmake
822 - ci: Create results folder before starting virgl_test_server
823 - virgl/ci: Force crosvm error when exit code file is missing
824 - virgl/ci: Prevent static link of virglrenderer inside crosvm
825 - virgl/ci: Do not hide crosvm output messages
826 - virgl/ci: Fix identification of dEQP binary paths
827 - iris/ci: Fix whl dEQP expectations
828 - iris/ci: Fix piglit tests expectations on amly
829 - panfrost/ci: Fix piglit tests expectations on G52
830 - freedreno/ci: Fix dEQP tests expectations on A530
834 - nir/fold_16bit_sampler_conversions: skip sparse residency tex instructions
835 - aco: add more D16 load/store instructions to RA and validator
836 - aco: workaround GFX9 hardware bug for D16 image instructions
837 - aco: implement D16 texture loads
838 - radv: use nir_fold_16bit_sampler_conversions()
839 - aco/optimizer: fix fneg modifier propagation on VOP3P
840 - aco: change fneg for VOP3P to use fmul with +1.0
841 - aco/optimizer: propagate and fold inline constants on VOP3P instructions
842 - nir/opt_algebraic: lower fneg_hi/lo to fmul
843 - aco/ra: fix get_reg_for_operand() in case of stride mismatches
844 - aco: don't allow SDWA on VOP3P instructions
845 - aco/optimizer: keep instr_mod_labels after applying extract
846 - aco/optimizer: apply extract from p_extract_vector
847 - aco/optimizer: optimize extract(extract())
848 - aco/optimizer: apply extract from subdword p_split_vector
849 - aco: use explicit zero-padding for 64bit image loads in expand_vector()
850 - aco: use p_create_vector(v2b,v2b) in get_alu_src_vop3p()
851 - aco: don't split VOP3P definitions
852 - aco: validate VOP3P opsel correctly
853 - nir: refactor nir_opt_move
854 - nir/opt_if: merge two break statements from both branch legs
855 - nir/opt_if: also merge break statements with ones after the branch
859 - CI: Disable Windows jobs
860 - Revert "CI: Disable Windows jobs"
861 - ci: Use common build script for libwayland
862 - ci: Consistently build Wayland and protocols
863 - ci: Upgrade to libdrm 2.4.109
864 - zink/ci: Add GL4.6 tessellation flake
865 - CI: Don't stream wget directly into bash
866 - Revert "gitlab-ci: disable radv-fossils"
867 - Revert "ci: disable vs2019 windows build"
868 - egl/wayland: Reset buffer age when destroying buffers
872 - ir3/freedreno: account for component in build_tessfactor_base
873 - turnip: add support for dirconf
874 - driconf: add vk_dont_care_as_load workaround option
875 - turnip: implement vk_dont_care_as_load workaround
876 - drirc: Apply vk_dont_care_as_load workaround to Forsaken Remastered
877 - nir/lower_amul: do not lower 64bit amul to imul24
878 - ir3/freedreno: add 64b undef lowering
879 - freedreno/ir3: disallow immediate addr/offset for ldg/ldg.a
880 - freedreno/ir3: set proper dst size for {store,load}_{global,shared}_ir3
881 - freedreno/ir3: use stg.a/ldg.a only if offset is reg or doesn't fit
882 - isaspec: inherite parent's bitset gpu gen requirements
883 - nir/serialize: Make more space for intrinsic_op allowing 1024 ops
884 - ir3/ra: Check register file upper bound when updating preferred_reg
885 - tu: fix rast state allocation size on a6xx gen4
886 - freedreno/computerator: Support A660 gpu
887 - vulkan/util: Handle depth-only formats in vk_att_ref_stencil_layout
888 - ir3: print half-dst/src for ldib.b/stib.b
889 - freedreno/ir3: add a6xx global atomics and separate atomic opcodes
890 - freedreno/ir3: handle global atomics
891 - turnip: implement VK_KHR_buffer_device_address
892 - ir3/cp: Prevent setting an address on subgroup macros
893 - freedreno,tu: Limit the amount of instructions preloaded into icache
894 - ir3: Add gen4 new subgroup instructions
895 - ir3: Use getfiberid for SubgroupInvocationID on gen4
896 - ir3,turnip: Enable subgroup ops support in all stages on gen4
897 - ir3,turnip: Add support for GL_KHR_shader_subgroup_quad
898 - turnip: Fix operator precedence in address calculation macros for queries
899 - ir3: Be able to reduce register limit for RA when CS has barriers
900 - ir3: Assert that we cannot have enough concurrent waves for CS with barrier
901 - tu: fix workaround for depth bounds test without depth test
902 - ir3: New cat3 instructions
903 - nir/algebraic: Separate has_dot_4x8 into has_sdot_4x8 and has_udot_4x8
904 - ir3: Make nir compiler options a part of ir3_compiler
905 - tu,ir3: Implement VK_KHR_shader_integer_dot_product
906 - vulkan/wsi: create a common function to compare drm devices
907 - tu: implement wsi hook to decide if we can present directly on device
908 - tu: support VK_EXT_primitive_topology_list_restart
909 - tu: expose VK_KHR_copy_commands2
910 - tu: add reference counting for descriptor set layouts
911 - turnip/perfetto: Optimize timestamp synchronization
912 - turnip/trace: refactor creation and usage of trace flush data
913 - turnip: rename tu_drm_get_timestamp into tu_device_get_gpu_timestamp
914 - turnip/trace: process u_trace chunks on queue submission
915 - turnip/perfetto: handle gpu timestamps being non-monotonic
916 - turnip/trace: Delete unused start/end_resolve tracepoints
917 - turnip: Drop references to layout of all sets on pool reset/destruction
918 - tu: implement sysmem vs gmem autotuner
919 - freedreno: Update gmem/sysmem debug options to be in line with turnip
920 - tu: add debug option to force gmem
921 - ci/freedreno: properly test sysmem and gmem paths
922 - ir3: opt_deref in opt loop to remove unnecessary tex casts
923 - turnip: Unconditionaly remove descriptor set from pool's list on free
924 - ir3: Limit the maximum imm offset in nir_opt_offset for shared vars
925 - turnip: Use LATE_Z when there might be depth/stencil feedback loop
929 - brw/nir: remove unused function prototypes.
930 - crocus: Delete the MI_COPY_MEM_MEM resource_copy_region implementation.
931 - llvmpipe: fix userptr for texture resources.
932 - clover: use max shader sampler view/images queries for clover.
933 - llvmpipe: swizzle image stores for CL BGRA
934 - radv: fence->user_ptr and ctx->fence_map are now totally unused.
935 - lavapipe: drop EXT_acquire_xlib_display
936 - vulkan/wsi: set correct bits for host allocations/exports for images.
937 - vulkan/include: import the video codec headers.
938 - vulkan: add new image types undef beta define to switch statements.
939 - meson: allow building with vulkan beta extensions enabled.
940 - llvmpipe: disable 64-bit integer textures.
941 - llvmpipe: fix compressed image sizes.
942 - intel/genxml: cleanup video xml collisions.
943 - intel/genxml: fix some missing address from the 75 xml
944 - intel/genxml: align QM field names across gens.
945 - intel/genxml: fix Picure->Picture typo
946 - intel/genxml: fix gen6 LD->VLD typo.
947 - intel/genxml: generate video headers
948 - util/vl: move gallium vl_vlc.h and vl_rbsp.h to shared code.
949 - mtypes: drop some context pointers that are unused now
950 - mesa: drop unused sw extensions init
951 - mesa/dd: burn a bunch of legacy driver interfaces down
952 - mesa/dd: remove some fbo driver hooks.
953 - mesa/dd: remove NewSamplerObject
954 - mesa/light: make _mesa_light static do_light.
955 - iris/ci: comment out iris-cml-traces-performance due to hw unavailable
956 - mesa/query: remove all the mesa queryobj code.
957 - mesa/syncobj: drop unused syncobj code.
958 - mesa: remove unused buffer object code.
959 - mesa/transformfeedback: remove unused transform feedback code
960 - mesa/barrier: remove unused barrier functions
961 - mesa/externalobject: delete unused functions
962 - intel/compiler: drop glsl options from brw_compiler
963 - meson: make mesa/tests/glx depend on gallium
964 - mesa/dd/st: move get strings pointer out of dd.h
965 - mesa/dd/st: direct wire queries/timestamp/condrender.
966 - mesa/st: direct call sync object functions
967 - mesa/st: move barriers to direct call
968 - mesa/st: move transformfeedback to direct calls
969 - mesa/st: move external objects to direct calls
970 - mesa/crocus/iris/blorp: drop minify macro in favour of u_minify
971 - mesa: move _mesa_varying_slot_in_fs to shader_enums
972 - anv: include futex.h explicitly in allocator.
973 - brw/compiler: drop mtypes.h from compiler
974 - intel/compiler: drop shader_info.h from compiler header
975 - intel/crocus: push main/macros.h out to the users
976 - mesa/st: move rendermode to direct call
977 - mesa/st: move Enable to direct call
978 - mesa/st: move query memory info to direct call
979 - mesa/st: move perf query to direct call
980 - mesa/st: move perfomance monitor to direct call
981 - mesa/dd: drop purgeable interface
982 - mesa/st: move fbo code to direct calling
983 - mesa/st: move texture APIs to direct st calls
984 - mesa/st: move Clear to new direct call
985 - mesa/st: move pixel/bitmap functions to direct call
986 - mesa/st: move clear/flush/finish to direct call
987 - mesa/st: move some context functions to direct calls
988 - mesa/st: move viewport to direct call
989 - mesa/st: move copy image sub data to direct call
990 - mesa/st: move program calls to direct call
991 - mesa/st: replace most of buffer funcs with direct calls.
992 - mesa/st: move blit function to direct call
993 - mesa/st: convert the non-optional egl image to direct calls
994 - mesa/st: convert DrawTex to direct call
995 - mesa/st: move msaa functions to direct call
996 - mesa/st: move compute to direct call
997 - mesa/st: move draw indirect and xfb to direct calls.
998 - mesa/st: drop Draw from dd function table.
999 - treewide: drop mtypes/macros includes from main
1000 - crocus: cleanup bo exports for external objects
1001 - mesa/st: merge st buffer object into GL
1002 - mesa/st: start moving bufferobject alloc/free/reference to main.
1003 - mesa: add a pipe_context pointer to gl context
1004 - mesa: add a pointer to st_config_options to gl_context
1005 - mesa: add pointer to cso_context to gl_context
1006 - mesa/st: migrate most of state tracker buffer objects into mesa
1007 - mesa/st: make static the buffer object funcs that can be
1008 - mesa/bufferobj: rename bufferobj functions to be more consistent.
1009 - bufferobj: make sw clear buffer static, move it and rename it
1010 - bufferobj: inline buffer clearing
1011 - bufferobj: inline page commitment
1012 - bufferobj: cleanup subdata copies
1013 - mesa/st: rename access flag to transfer flag function
1014 - mesa/bufferobj: move invalidate buffer to optional feature
1015 - mesa/st: remove st_cb_bufferobjects*
1016 - mesa: inline mesa_initialize_buffer_object.
1017 - mesa/st: refactor compute dispatch to fill grid info earlier.
1018 - mesa/st: migrate compute dispatch to mesa
1019 - mesa/compute: refactor compute launch to look more like draw
1020 - mesa/st: move get sample position code to static in mesa
1021 - mesa/st: move msaa functionality into multisample.c
1022 - mesa/st: migrate barrier code into mesa
1023 - mesa/st: move st strings handling into mesa
1024 - mesa: drop texformat code this isn't used.
1025 - mesa/st: use has_stencil_export instead of querying screen cap.
1026 - mesa: drop unused new renderbuffer code.
1027 - mesa: drop unused _mesa_new_program.
1028 - mesa/dd: drop unused InvalidateBufferSubData entry.
1029 - intel/compiler: remove gfx6 gather wa from backend.
1030 - intel/compiler: don't lower swizzles in backend.
1031 - intel/compiler: drop unused decleration
1032 - mesa: remove StripTextureBorder option.
1033 - mesa/draw: drop the multi draw with indices fallback.
1034 - mesa/st: move default enabled extensions into mesa.
1035 - mesa: drop optional tex/tnl maintains mode.
1036 - mesa/st: merge NewDepthClamp state flag
1037 - mesa/st: drop the new array driver state bit
1038 - mesa/st: drop the rasterizer driver flags
1039 - mesa/st: remove the viewport driver state flags
1040 - mesa/st: drop NewBlend driver state flags
1041 - mesa/st: drop new depth/stencil state bits
1042 - mesa/st: drop poly stipple driver state bit
1043 - mesa/st: drop new tess state driver bit
1044 - mesa/st: drop new uniform driver state bit
1045 - mesa: drop unused transform feedback state driver flags
1046 - mesa/st: drop ssbo, image and sampler driver state flags bits
1047 - mesa/st: drop scissor/window rect driver state bits
1048 - mesa/st: drop clip plane driver state bits
1049 - mesa/st: drop new framebuffer srgb driver state bit
1050 - mesa/st: drop multisample mask/locations state drivers bits
1051 - mesa/st: remove conditionals for driver state bits that are always set.
1052 - intel/genxml/gen4-5: fix more Raster Operation in BLT to be a uint
1053 - crocus: fail resource allocation properly.
1054 - crocus: don't create staging resources > half aperture
1055 - gallium/mesa: enhance PIPE_CAP_CLIP_PLANES to support override number
1056 - crocus: set max clip planes to 6 for gen4.
1057 - mapi: generate correct dispatch for EXT_draw_instanced
1058 - glsl/linker: cleanup passing gl_context unnecessarily
1059 - glsl: don't pass gl_context to lower shared references.
1060 - glsl/linker: don't pass gl_context just for constants in xfb code
1061 - glsl/linker: move more ctx->Consts to consts.
1062 - glsl/linker: remove direct gl_context usage in favour of consts/exts/api
1063 - glsl/linker: get rid of gl_context from link varyings
1064 - glsl/linker: get rid of gl_context from uniform assign paths
1065 - glsl/linker: get rid of gl_context from atomic counters paths
1066 - glsl/linker: remove gl_context from check image resources
1067 - glsl/linker: remove gl_context usage from more places.
1068 - glsl/nir/linker: avoid passing gl_context inside gl_nir linker
1069 - glsl/linker/uniform_blocks: don't pass gl_context around.
1070 - glsl/linker: drop unused gl_context.
1071 - glsl/linker: remove a bunch more gl_context references.
1072 - glsl/nir: don't pass gl_context to the convertor routine.
1073 - mesa/program: don't use CALLOC_STRUCT for instructions.
1074 - mesa: rebalance the CALLOC_STRUCT/FREE force.
1075 - mesa/st/perfmon: rebalance CALLOC_STRUCT/FREE
1076 - mesa/st: get rid of ST_CALLOC_STRUCT use CALLOC_STRUCT
1077 - intel/compiler: add clamp_pointside to vs/tcs/tes keys.
1078 - crocus: only clamp point size on last stage.
1079 - amd: move vcn decoding regs + structs to a common file.
1080 - amd: move uvd decode definitions to common place
1081 - mesa/\*: use an internal enum for tessellation primitive types.
1082 - mesa/\*: add a shader primitive type to get away from GL types.
1083 - includes: add windows lean and mean guard.
1084 - nir: remove gl.h include from nir headers.
1085 - intel/brw: drop gl header from the brw backend.
1086 - glsl/parser: extract consts/exts/api out of context at start.
1087 - mesa: drop unused context parameter to shader program data reference.
1088 - mtypes: split gl extensions and consts out into a separate header
1089 - mesa/mtypes: move matrix enums to shader_enums.h
1090 - glsl: remove some deps on mtypes.h.
1091 - mesa: move some gl shader types to shader_types.h.
1092 - glsl: avoid rebuilding builtin functions on mtypes.h changes.
1093 - glsl: move ast_function.cpp off mtypes.h
1094 - mesa: move ast_to_hir.cpp off mtypes.h
1095 - mtypes: move gl_linked_shader and gl_shader_program to new shader_types.h
1096 - mtypes: move gl_program to shader_types.h
1097 - mtypes: more gl_active_atomic_buffer to shader_types.h
1098 - mtypes: move transform feedback internal structs to shader_types.h
1099 - mtypes: move uniform shader types to shader_types.h
1100 - mtypes: move bindless image/sampler objects to shader_types.h
1101 - mtypes: move gl_shader_variable to shader_types.h
1102 - glsl: move off mtypes.h in lots of places.
1103 - glsl/fp64: move context.h dependent checks into main.
1104 - glsl: drop some more context.h/mtypes.h interactions
1105 - vbo: drop unused mtypes.h
1106 - docs: update docs for new extension header.
1107 - mesa: more mtypes.h cleanups
1108 - mesa: split struct gl_config into it's own header.
1109 - glsl: drop glheader.h include.
1110 - mesa/st: merge memoryobjects code from st into mesa
1111 - mesa/st: merge semaphore objects from st into mesa
1112 - mesa/st: merge the syncobj code from st into mesa
1113 - mesa/st: merge queryobj code from st into mesa.
1114 - mesa/st: merge condrender code from st into mesa.
1115 - mesa/st: merge st transform feedback object into gl one.
1116 - mesa/st: merge transform feedback code from st into mesa
1117 - mesa/st: merge perfmon object from st into mesa
1118 - mesa/st: merge perfmon counters/groups objects from st into mesa
1119 - mesa/st: merge perfmon groups init/cleanup from st into mesa
1120 - mesa/st: move perfmon code from st into mesa
1121 - mesa/st: cleanup last bits of st perfmon code.
1122 - mesa/st: merge texture object/image structs into mesa
1123 - mesa/st: merge texture obj/image alloc/free into mesa
1124 - intel/perf: use a function to do common allocations
1125 - meson: start building intel earlier.
1126 - mesa/st: move intel blackhole noop enable to frontend
1127 - mesa/st: remove st_context from debug callback
1128 - mesa/st: migrate debug callback code into mesa
1129 - mesa/st: drop last user of st_Enable.
1130 - mesa/st: directly call the uuid get funcs.
1131 - mesa/st: drop emit string marker device table entry.
1132 - mesa/st: move pin l3 cache to direct check/call.
1133 - mesa/dd: drop GetProgramBinaryDriverSHA1
1134 - mesa/st: drop useless tex parameter calls.
1135 - mesa/st: move st_TexParameter into mesa
1136 - mesa/st: drop release all sampler views wrapper
1137 - mesa/st/vdpau: direct call the vdpau functions.
1138 - mesa/st: move evaluate depth values into mesa
1139 - mesa/ctx: store screen pointer in ctx as well
1140 - mesa/st: inline st_max_shader_compiler_threads
1141 - mesa/st: move shader completion into mesa
1142 - mesa/st: move memory query into mesa.
1143 - mesa/st: drop some bindless wrappers
1144 - mesa/st: drop lots of perfquery wrappers
1145 - mesa/st: move perf query test to st_context, drop files.
1146 - mesa/st: merge st_renderbuffer into gl_renderbuffer.
1147 - mesa/st: move some renderbuffer code into mesa
1148 - mesa/st: merge framebuffer objects from st to mesa
1149 - mesa/st: move render/finish_render texture in to mesa.
1150 - mesa/st: move validate/discard framebuffer into mesa
1151 - mesa/st: move st_ReadBuffer functionality into mesa
1152 - mesa/st: move DrawBufferAllocate into mesa.
1153 - mesa/st: move st renderbuffer code into mesa renderbuffer
1154 - mesa/st: move map/unmap renderbuffer code into mesa
1155 - mesa/st: move st_new_renderbuffer_fb to manager
1156 - mesa/st: move some fbo helpers around.
1157 - mesa/st: fixup viewport drawable invalidation
1158 - mesa/st: migrate blit code into mesa
1159 - mesa/st: move last of renderbuffer functionality into mesa
1160 - mesa/st: move renderbuffer format choosing wrapper into mesa.
1161 - mesa/st: move manager colorbuffer interface to gl_context.
1162 - mesa/st: move invalidate_on_gl_viewport to ctx
1163 - mesa/st: move st_fb_orientation into a mesa function
1164 - mesa: consolidate setting no error state and checking suid.
1165 - mesa/st: collapse st_program into gl_program object.
1166 - mesa/st: move new ati fragment shader to mesa
1167 - mesa/st: move st_vertex_program to gl_vertex_program in mesa
1168 - mesa/st: move program new/delete into mesa
1169 - mesa/st: refactor program translation into one file.
1170 - crocus: find correct relocation target for the bo.
1171 - crocus: fix leak on gen4/5 stencil fallback blit path.
1172 - draw/so: don't use pre clip pos if we have a tes either.
1173 - lavapipe: fix sampler + sampler view leaks.
1174 - lavapipe: reference gallium fences correctly.
1175 - vulkan/wsi: handle queue families properly for non-concurrent sharing mode.
1176 - crocus: change the line width workaround for gfx4/5
1177 - gallivm/nir: extract a valid texture index according to exec_mask.
1181 - egl/wayland: Properly clear stale buffers on resize
1182 - panfrost: Support planar formats for scanout
1183 - panfrost: support PIPE_RESOURCE_PARAM_NPLANES query
1187 - docs: add release notes for 21.2.4
1188 - docs: add sha256 sum for 21.2.4 release
1189 - docs: update calendar and link releases notes for 21.2.4
1190 - meson: use gtest protocol for gtest based tests when possible
1191 - docs: add release notes for 21.2.5
1192 - docs: add sha256 sums for 21.2.5
1193 - docs: update calendar and link releases notes for 21.2.5
1194 - turnip: don't use mesa/macros.h to get utils/rounding.h
1195 - docs: add release notes for 21.2.6
1196 - docs: add sha256 sums for 21.2.6 relnotes
1197 - docs: update calendar and link releases notes for 21.2.6
1198 - docs/release-calendar: remove additional 21.2 releases
1199 - docs: Add calendar entries for 22.0 release candidates.
1200 - mesa/tests: ensure that util_cpu_detect has been called
1201 - classic/r100: Delete driver
1202 - classic/r200: Delete driver
1203 - classic/nouveau: Remove driver
1204 - classic/i915: Remove driver
1205 - classic/i965: Remove driver
1206 - mesa/dri: remove mega driver stub
1207 - include/pci_ids: Move PCI ids supported by both i965 and iris to iris
1208 - mesa/main/tests: remove dispatch sanity
1209 - mesa: Delete libmesa_classic
1210 - mesa: Merge libmesa_gallium and libmesa_common
1211 - mesa: move common/dri into gallium
1212 - meson: restore dri.pc file
1213 - CODEOWNERS: remove OpenSWR
1214 - new_features: Add OpenSWR removal
1215 - fixup! gallium/swr: Remove driver source
1216 - docs: move the release for 22.0 out
1217 - VERSION: bump for 22.0.0-rc1 release
1218 - .pick_status.json: Update to 0447a2303fb06d6ad1f64e5f079a74bf2cf540da
1219 - .pick_status.json: Update to 8335fdfeafbe1fd14cb65f9088bbba15d9eb00dc
1220 - .pick_status.json: Update to 5e9df85b1a4504c5b4162e77e139056dc80accc6
1221 - VERSION: bump version for 22.0.0-rc2
1222 - .pick_status.json: Update to 7955df28a6660d8dff77c79c345aa28aa7fa859c
1223 - .pick_status.json: Update to 22fc53493092a7507c1e2db47b0c8763158d7b2d
1224 - .pick_status.json: Update to 108e6eaa83eed3eb356f3cce835c5f5e3a836b8e
1225 - .pick_status.json: Mark b07372312d7053f2ef5c858ceb1fbf9ade5e7c52 as denominated
1226 - .pick_status.json: Update to 3759a16d8a883355effa435f46867951ce712dbe
1227 - .pick_status.json: Mark a8418abd748e8e761dda9c3594e29e560833d9ff as denominated
1228 - .pick_status.json: Update to 22efeec399fb55dcb364dabd65afab137d6f3fc3
1229 - .pick_status.json: Update to f030b75b7d2c359b90c18ee4ed83fa05265c12e0
1230 - .pick_status.json: Update to fd3451babd6cded6794561d74c8919576ba1f97d
1231 - llvmpipe: remove test that passes on 22.0 for some reason from the fail list
1232 - .pick_status.json: Update to b21e7e1ef7ba2c2cf97f073a8a01d6bc43835622
1233 - d3d12: mark incorrectly passing tests as expected failures
1234 - VERSION: bump for 22.0.0-rc3
1235 - .pick_status.json: Update to 48b3ef625e1909c1be31fbe10adb53734af38eb4
1236 - .pick_status.json: Update to 48b3ef625e1909c1be31fbe10adb53734af38eb4
1237 - ci/zink: mark tests fixed by previous commit as fixed
1238 - ci/zink: mark more expected passes as such
1239 - Revert "panfrost: Fix set_sampler_views for big GL"
1240 - .pick_status.json: Update to 342e6f83321a91816358dee82178809b2a8aeeaa
1244 - clover/images: Add array_size to implement CL_IMAGE_ARRAY_SIZE
1245 - clover: Implement CL_MEM_OBJECT_IMAGE2D_ARRAY
1246 - clover: Implement CL_MEM_OBJECT_IMAGE1D_BUFFER
1247 - clover: Implement CL_MEM_OBJECT_IMAGE1D_ARRAY
1251 - dri_drawable: missing header
1255 - v3dv: Implement VK_KHR_create_renderpass2
1256 - vulkan: Allow RegisterDisplayEventEXT before first page flip
1258 Ella-0 (2):
1260 - v3dv: implement VK_EXT_host_query_reset
1261 - v3dv: enable VK_KHR_swapchain_mutable_format
1265 - ci/lvp: Skip some slow tests under ASan.
1266 - loader: Avoid enumerating drm devices just to get an fd's PCI ID.
1267 - freedreno/afuc: Disable the disassembler on 32-bit builds.
1268 - ci/deqp-runner: Drop SUMMARY_LIMIT env var.
1269 - ci/deqp-runner: Simplify the --jobs argument setup.
1270 - ci/deqp-runner: Use new deqp-runner's built-in renderer/version checks.
1271 - ci/deqp-runner: Drop silly CSV env vars.
1272 - ci/deqp-runner: Move remaining asan runs to --env LD_PRELOAD=
1273 - ci/deqp-runner: Drop LD_LIBRARY_PATH=/usr/local for libkms workaround.
1274 - ci/deqp-runner: Don't start GPU hang detection for making junit results.
1275 - ci/deqp-runner: Move more non-suite logic under the non-suite 'if'.
1276 - ci/piglit-runner: Fix funny indentation of the piglit-runner command.
1277 - turnip: Switch tu6_format_texture() to a pipe_format.
1278 - turnip: Switch tu6_format_color() to a pipe_format.
1279 - turnip: Switch format_to_ifmt() to take a pipe_format.
1280 - util: Move freedreno's snorm-to-unorm to util/, adding remaining cases.
1281 - turnip: Make copy_format() and tu6_plane_format() return pipe_format
1282 - gallium/u_blitter: Read MSAA z/s from sampler's .x instead of .y or .z.
1283 - turnip: Drop the assertion about the temporary bit in sync fd imports.
1284 - ci/radeonsi: Use a deqp-runner suite suite for stoney.
1285 - ci/deqp-runner: Rename the deqp-drivername-\*.txt files to drivername-\*.txt
1286 - ci/piglit-runner: Merge piglit-driver-\*.txt files into driver-\*.txt.
1287 - ci: Enable testing radeonsi's libva using libva-util unit tests.
1288 - ci/etnaviv: Fix the dependency for the build artifacts.
1289 - ci/etnaviv: Add some more deqp flakes I've seen in recent runs.
1290 - etnaviv: Switch to the NIR compiler by default.
1291 - i915g: Check for negate/swizzle on TGSI_OPCODE_KILL_IF's src.yzw.
1292 - i915g: Make sure we consider negates/swizzles on bias/shadow coords.
1293 - i915g: Improve the explanation for the 1D Y swizzle.
1294 - ci: Uprev VK-GL-CTS to 1.2.7.2, and pull in piglit while I'm here.
1295 - freedreno: Fix gmem invalidating the depth or stencil of packed d/s.
1296 - freedreno/a6xx: Emit a null descriptor for unoccupied IBO slots.
1297 - freedreno/fdl6: Add an interface for setting up buffer descriptors.
1298 - turnip: Use the new shared buffer-view descriptor creation function.
1299 - turnip: Remove buffer-view cross-check code.
1300 - freedreno/fdl6: Add support for texture swizzles of A/L/I/LA/RGBx.
1301 - freedreno/a6xx: Use the fdl buffer view setup for img/ssbo descriptors.
1302 - freedreno: Fix layered rendering to just Z/S and not color.
1303 - freedreno/a6xx: Fix partial z/s clears with sysmem.
1304 - freedreno/a5xx: Use the defined names for 2D_BLIT_CNTL regs.
1305 - freedreno/a618: Mark a flaky test that triggers hangcheck.
1306 - freedreno/a6xx: Don't try to generate mipmaps for SNORM with our blitter.
1307 - freedreno/a5xx+: Fix missing LA formats.
1308 - freedreno/a5xx: Diff reduction in fd5_layout to fd6_layout.
1309 - freedreno/a6xx: Try to fix drawing to z/s miplevel/layer offsets.
1310 - freedreno/a5xx: Remove bogus assertion about BO size.
1311 - freedreno/a5xx: Try to fix drawing to z/s miplevel/layer offsets.
1312 - freedreno/a5xx: Clean up a little bit of blitter array pitch setup.
1313 - freedreno/ir3: Fix off-by-one in prefetch safety assert.
1314 - turnip: Claim 1.2.7.1 CTS conformance.
1315 - freedreno: Fix the uniform/nonuniform handling for cat5 bindful modes.
1316 - freedreno: Fix constant-index assumptions in IBO loads.
1317 - ci/etnaviv: Mark the rest of uniform_api.random as flaky.
1318 - ci/etnaviv: Add more texturing flakes.
1319 - ci/bare-metal: Close serial and join serial threads before exit.
1320 - util: Rename PIPE_FORMAT_G8_B8_R8_420_UNORM.
1321 - util/format: Add G8_B8R8_420_UNORM to match Vulkan.
1322 - freedreno/fdl6: Skip redundant setting of TILE_ALL for NV12.
1323 - freedreno: Set layer_first on (2D) resource imports.
1324 - freedreno/a6xx: Create a fd6_view at sampler view update time.
1325 - freedreno/a6xx: Switch to relying on fd6_view for our texture descriptors.
1326 - freedreno/a6xx: Use fd6_view for non-buffer image descriptors, too.
1327 - freedreno/a6xx: Clean up sysmem fb read patching using fd6_view.
1328 - freedreno/a6xx: Drop an unused tile_mode arg.
1329 - freedreno/a6xx: Inline remaining fd6_tex_const_0() call.
1330 - mesa/st: Disable NV_copy_depth_to_color on non-doubles-capable HW.
1331 - freedreno/a6xx: Add some notes about piglit failures.
1332 - freedreno/ir3: Fix an off-by-one in so->outputs_count safety assert.
1333 - ci/freedreno: Enable the tes-input/tcs-input tests.
1334 - freedreno/a6xx: Do sparse setup of the TFB program.
1335 - ci/freedreno: Add notes explaining the KHR-GL* failures.
1336 - ci/freedreno: Add a link to the issue for color_depth_attachments.
1337 - freedreno/fdl: Dump the generated layout when a layout test fails.
1338 - freedreno: Fix the texturator unit test script.
1339 - freedreno/cffdump: Fix 64-bit reg decode in script mode.
1340 - freedreno/fdl: Add support for unit testing 3D texture array strides.
1341 - freedreno/cffdump: Fix up formatting of texturator unit test script output.
1342 - freedreno/cffdump: Handle the TILE_ALL flag in unit test generation.
1343 - freedreno/a6xx: Fix a bunch of 3D texture layout to match blob behavior.
1344 - freedreno: Stop exposing MSAA image load/store on desktop GL.
1345 - freedreno/a6xx: Disable sample averaging on z/s or integer blits.
1346 - freedreno/a6xx: Disable sample averaging on non-ubwc z24s8 MSAA blits.
1347 - freedreno/a5xx: Define a5xx_2d_surf_info like a6xx has.
1348 - freedreno/a5xx: Document the sRGB bit on RB_2D_SRC/DST info.
1349 - freedreno/a5xx: Emit MSAA state for sysmem rendering, too.
1350 - nir: Un-inline nir_builder_alu_instr_finish_and_insert()
1351 - nir: Un-inline more of nir_builder.h.
1352 - freedreno/ir3: Make a shared helper for the tess factor stride.
1353 - turnip: Move CP_SET_SUBDRAW_SIZE to vkCmdBindPipeline() time.
1354 - nir: Uninline a bunch of nir.h functions.
1355 - nir: Make nir_build_alu() variants per 1-4 arg count.
1356 - freedreno/a6xx: Allocate a fixed-size tess factor BO.
1357 - freedreno/a6xx: Skip emitting tess BO pointers past the shader's constlen.
1358 - freedreno/a6xx: Set the tess BO ptrs in the program stateobj.
1359 - mesa/st: Remove GL_ARB_depth_clamp emulation support.
1360 - r300: Add deqp expectations for RV515.
1361 - r300: Turn a comment about presub into an assert.
1362 - r300: Fix mis-optimization turning -1 - x into 1 - x.
1363 - nouveau/nir: Use the address reg for indirect scratch access.
1364 - nir/algebraic: Move some generated-code algebraic opt args into a struct.
1365 - nir/algebraic: Drop the check for cache == None.
1366 - nir/algebraic: Replace relocations for nir_search values with a table.
1367 - nir/algebraic: Remove array-of-cond code
1368 - nir/algebraic: Move relocations for expression conds to a table.
1369 - nir/algebraic: Move relocations for variable conds to a table.
1370 - nir/algebraic: Pack various bitfields in the nir_search_value_union.
1371 - nir/algebraic: Mark the automaton's filter tables as const.
1372 - nir/algebraic: Move all the individual transforms to a common table.
1373 - r300: Move the instruction filter for r500_transform_IF() to the top.
1374 - r300: Cache the var list in the peephole_mul_omod() loop.
1375 - r300: Ensure that immediates have matching negate flags too.
1376 - r300: Also consider ALU condition modifiers for loop DCE.
1377 - r300: Remove the non_normalized_coords from the shader key.
1378 - r300: Precompile the FS at shader creation time.
1379 - r300: Route shader stats output to ARB_debug_output.
1380 - r300/ci: Update loop expectations
1381 - loader: Restore i915g support.
1382 - r300/ci: Add some piglit expectations.
1383 - nir/nir_to_tgsi: Add support for "if" statements with !native_integers
1384 - nir_to_tgsi: Make !native_integers front face input match glsl_to_tgsi.
1385 - r300: Disable loop unrolling on r500.
1386 - r300: Request NIR shaders from mesa/st and use NIR-to-TGSI.
1387 - nir_to_tgsi: Enable nir_opt_move.
1388 - nir/nir_opt_move,sink: Include load_ubo_vec4 as a load_ubo instr.
1389 - nir_to_tgsi: Set the TGSI Precise flag for exact ALU instructions.
1390 - r300: Remove some dead compiler code.
1391 - r300: Remove support for SCS.
1392 - r300: Remove unused RC_OPCODE_ABS.
1393 - r300: Remove unused RC_OPCODE_XPD.
1394 - r300: Remove unused RC_OPCODE_SWZ.
1395 - r300: Remove unused RC_OPCODE_CLAMP.
1396 - r300: Remove unused RC_OPCODE_SFL
1397 - r300: Remove unused RC_OPCODE_DPH
1398 - ci/freedreno: Add known flakes from the last month.
1399 - ci/crocus: Add support for manual CI runs on my G41.
1400 - crocus: Clamp VS point sizes to the HW limits as required.
1401 - glsl: Delete the vectorization opt pass.
1402 - glsl: Delete the optimize_redundant_jumps pass.
1403 - glsl: Remove dead prototype for old do_discard_simplification().
1404 - glsl: Remove comment about non-existing DFREXP_TO_ARITH
1405 - glsl: Retire unused modes for lower_64bit_integer_instructions.
1406 - ci/r300: Add another xfail on the main branch.
1407 - r300/vs: Allocate temps we see a use as a source, too.
1408 - r300/vs: Reuse rc_match_bgnloop().
1409 - r300/vs: Fix flow control processing just after an endloop.
1410 - ci: Enable a build with MSan.
1411 - glcpp: Disable the valgrind tests.
1412 - softpipe: Drop the quad pstipple stage.
1413 - softpipe: Use the draw module's poly stipple handling, like llvmpipe.
1414 - softpipe: Drop duplicate decl of softpipe_find_fs_variant
1415 - ci/crocus: Add manual CI for the new HSW box I have at home.
1416 - ci: Enable reporting to the flakes IRC channel for i915g and crocus.
1417 - ci/i915g: Add a couple more recent regressions.
1418 - nir_to_tgsi: Use the same address reg mappings as GLSL-to-TGSI did.
1419 - gallium: Delete PIPE_CAP_TGSI_ANY_REG_AS_ADDRESS
1420 - tgsi/exec: Simplify indirects now that they always use the ADDR file.
1421 - i915g: Turn off FP16 in the vertex shaders.
1422 - r300: Use uif() instead of pointer aliasing in program printing.
1423 - ci/r300: Update xfails from a full dEQP run.
1424 - r300: Fix omod failing to increase the number of channels stored.
1425 - nir_to_tgsi: Enable fdot_replicates flag.
1426 - ci: Shrink container/rootfs sizes.
1427 - nir_to_tgsi: Fix a bug in TXP detection after backend lowering.
1428 - ci/llvmpipe: Sort the list of traces.
1429 - ci/llvmpipe: Add a trace for the game JVGS, which got regressed recently.
1430 - ci: Add paraview traces to several drivers.
1431 - freedreno/ir3: Use nir_opt_offset for removing constant adds for shared vars.
1432 - nir: Apply nir_opt_offsets to nir_intrinsic_load_uniform as well.
1433 - r300: Drop unused r300_get_stats() call.
1434 - r300: Add consts (uniforms) count to the shader-db output.
1435 - nir/lower_dynamic_bo_access: Use copy_inst_indices for our cloned instrs.
1436 - nir: Add a .base field to nir_load_ubo_vec4.
1437 - nir/opt_offsets: Disable unsigned wrap checks on non-native-integers HW.
1438 - nir/opt_offsets: Also apply the max offset to top-level constant folding.
1439 - nir_to_tgsi: Use nir_opt_offsets for load_ubo_vec4.
1440 - nir/opt_offsets: Use nir_ssa_scalar to chase offset additions.
1441 - softpipe: respect !independent_blend_enable for color masks.
1442 - softpipe: Request that st fix up DST_ALPHA blending for RGB render targets.
1443 - ci/softpipe: Drop the GS sampling known-flakes.
1444 - nir/lower_locals_to_regs: Do an ad-hoc copy propagate on our generated MOV.
1445 - tgsi_exec: Fix shared var stores for >1 real invocation, and overflow checks.
1446 - softpipe: Improve some local var naming in compute shaders.
1447 - softpipe: Initialize the CS dispatch mask at machine setup time.
1448 - softpipe: Dispatch 4 CS invocations per tgsi_exec thread.
1449 - Revert "ci: freedreno: Update a530 dEQP fail expectation list"
1450 - ci/llvmpipe: Drop the skip of piglit edgeflag test.
1451 - ci/llvmpipe,softpipe: Switch piglit testing to piglit-runner.
1452 - ci/traces: Rename the piglit/run.sh script to piglit-traces.sh.
1453 - ci/traces: Drop the PIGLIT_PROFILES setting for traces replay.
1454 - ci/traces: Clean up the failure report message.
1455 - ci/traces: Drop the baseline file creation for trace results.
1456 - ci/traces: Drop PIGLIT_REPLAY_UPLOAD_TO_MINIO.
1457 - ci/traces: Always generate the junit XML.
1458 - ci/freedreno: Update hashes for closed traces.
1459 - vulkan: Fix leak of error messages
1460 - ci: Use a dlclose-disabling preload library for leak checking in Vulkan.
1461 - ci/freedreno: Switch 2 default a630 VK jobs to being GLES and VK ASan jobs.
1462 - turnip: Store the computed iova in the tu_buffer.
1463 - turnip: Store the computed iova in the tu_image.
1464 - r300: Disable fp16 and int16 in swtcl vertex shaders.
1465 - r300/r600: Add drm-shim support.
1466 - freedreno/isaspec: Add missing dep of encode.py/decode.py calls on isa.py
1467 - intel: Add missing dep of gen_*_header.py on utils.py.
1468 - i915g: Initialize the rest of the "from_nir" temporary VS struct.
1472 - microsoft/compiler: Shadow tex instructions always use shadow samplers
1473 - microsoft/spirv_to_dxil: Add DXIL validation to spirv2dxil
1474 - ci/windows: Add validation tests for spriv_to_dxil
1478 - VERSION: bump to 21.4
1479 - docs: reset new_features.txt
1480 - VERSION: bump to 22.0
1481 - docs: update calendar for 21.3.0-rc1
1482 - docs: update calendar for 21.3.0-rc2
1483 - docs: update calendar for 21.3.0-rc3
1484 - docs: update calendar for 21.3.0-rc4
1485 - release-calendar: fix date for next 21.3 rc
1486 - meson: drop duplicate addition of surfaceless & drm to the list of platforms
1487 - meson: move \`egl_native_platform` definition inside the \`with_egl` block
1488 - meson: drop impossible \`if no platform` branch
1489 - meson: always define \`HAVE_{X11,XCB}_PLATFORM` when it's enabled
1490 - meson: automatically define \`HAVE_{some}_PLATFORM`
1491 - docs: update calendar for 21.3.0-rc5
1492 - docs/submittingpatches: add link to section describing how to make a backport MR
1493 - docs/submittingpatches: add formatting around the release branches names
1494 - docs/submittingpatches: mention use of the \`-x` flag of \`git cherry-pick` when backporting a co…
1495 - docs: update branchpoint instructions
1496 - docs: add release notes for 21.3.0
1497 - docs: update calendar and link releases notes for 21.3.0
1498 - docs: add 21.3.x release schedule
1499 - docs: add 22.0 branchpoint date for perspective
1500 - docs: add release notes for 21.3.1
1501 - docs: update calendar and link releases notes for 21.3.1
1502 - docs: add release notes for 21.3.2
1503 - docs: update calendar and link releases notes for 21.3.2
1504 - docs: add release notes for 21.3.3
1505 - docs: update calendar and link releases notes for 21.3.3
1506 - docs: add release notes for 21.3.4
1507 - docs: update calendar and link releases notes for 21.3.4
1508 - docs: add release notes for 21.3.5
1509 - docs: update calendar and link releases notes for 21.3.5
1510 - docs/release-calendar: add another 21.3.x since 22.0 has been delayed a bit
1514 - ci: temporarily disable lima CI
1515 - mesa: fix GL_MAX_SAMPLES with GLES2
1516 - lima/gpir: avoid invalid write in regalloc
1517 - lima/ppir: initialize slots array for dummy/undef
1518 - lima/ppir: refactor bitcopy to use unsigned char
1520 Erik Faye-Lund (9):
1522 - docs: update trademark disclaimer
1523 - CODEOWNERS: remove ownership of deleted code
1524 - ci: remove testing of deleted code
1525 - docs: remove mentions of deleted code
1526 - docs: remove stale notice about deleted dir
1527 - ensure csv-files are crlf on disk
1528 - bin/gen_calendar_entries: fix newlines on windows
1529 - docs: use http-redirect when possible
1530 - docs: remove incorrect drivers from extension
1534 - anv: increase binding table pool size to 64KB
1535 - pps: increase intel.cfg buffer size
1539 - r300: improve precission of linear interpolation
1540 - r300: stub derivatives on r300 and r400 hardware
1541 - nir: assert that variables in optimize_atomic are initialized
1542 - glsl: fix trivial strict aliasing warning
1543 - radv: dont call calloc when BVH is empty
1544 - iris: fix mapping compressed textures
1545 - r300: fix handling swizzle in transform_source_conflicts
1546 - r300: replace recursive calls with loops
1550 - intel/fs/xehp: Teach SWSB pass about the exec pipeline of FS_OPCODE_PACK_HALF_2x16_SPLIT.
1551 - intel/fs: Add physical fall-through CFG edge for unconditional BREAK instruction.
1552 - intel/dev: Fix size of device info num_subslices array.
1553 - intel/dev: Add support for pixel pipe subslice accounting on multi-slice GPUs.
1554 - intel/dev: Implement DG2 restrictions requiring additional DSSes to be disabled.
1555 - intel/xehp: Implement XeHP workaround Wa_14013910100.
1556 - intel/xehp: Implement XeHP workaround Wa_14014148106.
1557 - intel/xehp: Update 3DSTATE_PS maximum number of threads per PSD.
1558 - intel/fs: Don't assume packed dispatch for fragment shaders on XeHP.
1559 - intel/blorp/gfx12+: Drop unnecessary state cache invalidation from binding table setup.
1560 - intel/genxml: Fix SLICE_HASH_TABLE struct on XeHP.
1561 - iris: Merge gfx11\_ and gfx12_upload_pixel_hashing_tables() into the same function.
1562 - intel: Move pixel hashing table computation into common header file.
1563 - intel: Minimal calculation of pixel hash table for arbitrary number of pixel pipes.
1564 - intel: Rename intel_compute_pixel_hash_table() to intel_compute_pixel_hash_table_3way().
1565 - iris: Program pixel hashing tables on XeHP.
1566 - anv: Program pixel hashing tables on XeHP.
1567 - intel/xehp: Switch to coarser cross-slice pixel hashing with table permutation.
1568 - iris/xehp: Implement workaround for 3D texturing+anisotropic filtering.
1569 - intel/fs/xehp: Merge repeated in-order read dependencies instead of replacement.
1570 - intel/fs: Move legal exec type calculation into helper function in lower_regioning pass.
1571 - intel/fs: Teach the lower_regioning pass how to split instructions of unsuported exec type.
1572 - intel/fs: Take into account region strides during SIMD lowering decision of SHUFFLE.
1573 - intel/fs: Fix destination suboffset calculations for non-trivial strides in SHUFFLE codegen.
1574 - intel/fs: Perform 64-bit SHUFFLE lowering in the lower_regioning pass.
1575 - intel/fs: Perform 64-bit SEL_EXEC lowering in the lower_regioning pass.
1576 - intel/fs: Honor strided source regions specified by the IR for CLUSTER_BROADCAST.
1577 - intel/fs: Perform 64-bit CLUSTER_BROADCAST lowering in the lower_regioning pass.
1581 - meson: Use get_supported_arguments more often.
1582 - meson: Remove some unnecessary loops.
1583 - amd/addrlib: Use get_supported_arguments to get compiler args.
1584 - radv: Increase maxFragmentCombinedOutputResources.
1585 - vulkan/wsi/wayland: Fix add_wl_shm_format alpha/opaqueness.
1586 - vulkan/wsi/wayland: Convert  missing vulkan formats to shm formats.
1587 - vulkan/wsi/wayland: Add modifiers for RGB formats.
1588 - vulkan/wsi/wayland: Fix add_drm_format_modifier aplha/opaqueness.
1592 - virgl: Add driconf tweak to force-enable reading back R8_SRGB textures
1593 - virgl: obtain supported number of shader sampler views from host
1594 - ci: pin virglrenderer version
1595 - virgl: Enable higher compatibility profiles if host supports it
1596 - util/primconvert: map only index buffer part that is needed
1597 - nir_lower_io: propagate the "invariant" flag to outputs
1598 - ntt: Set the output invariant flag according to the semantics
1602 - util: __getProgramName: remove check for ancient FreeBSD versions, simplify ifdefs
1603 - util: make util_get_process_exec_path work on FreeBSD w/o procfs
1607 - etnaviv/drm: Use etna_mesa_debug for debugging messages
1608 - etnaviv/drm: Add some bo debug output
1609 - etnaviv/drm: Print gpu model at debug verbosity
1610 - etnaviv/drm: Use mesa_log* for debugging
1611 - entaviv/drm: Use same log format as gallium bits
1612 - etnaviv: Use mesa_log*
1616 - ci: Update linux kernel to v5.15
1617 - iris/ci: Fix traces for amly and deqp list for whl
1618 - ci/freedreno: Add maxcpus=2 to the kernel cmdline on a530
1619 - panfrost/ci: update piglit tests expectations on G52
1620 - ci: Update ci-fairy to version with --token-file support
1621 - ci: Uprev piglit
1622 - ci: Use ci-fairy minio login via token file
1623 - ci: Build skqp on ARM64 images
1624 - ci: Add a630_skqp jobs
1625 - ci: skqp: Add documentation on how to maintain skqp jobs
1626 - ci: Uprev Kernel to v5.16
1627 - ci: freedreno: Update a530 dEQP fail expectation list
1628 - ci: Add docs for Linux Kernel uprevs
1629 - virgl/ci: make crosvm-runner pass variables in a secure way
1633 - zink: Fix building on macOS
1637 - intel/compiler/test: Fix build with GCC 7
1641 - zink/codegen: support platform tags
1642 - zink/codegen: remove core_since in constructor
1643 - zink/codegen: remove bogus print statement
1647 - turnip: expose VK_KHR_driver_properties
1648 - anv: Fix to honor the spec to get stencil layout.
1649 - radv: Fix to honor the spec to get stencil layout.
1650 - vulkan/util: Move helper functions for depth/stencil images to vk_iamge
1651 - turnip: Enable VK_KHR_separate_depth_stencil_layouts
1652 - turnip: Use the new common device lost tracking
1653 - vulkan: fix typo
1654 - turnip: Porting to common vulkan implementation for synchronization.
1655 - turnip: Porting to common implementation for timeline semaphore
1656 - turnip: fix leaks of submit requests.
1660 - v3dv: fix TLB buffer to image copy path for 3D images
1661 - v3dv: enable Vulkan 1.1
1662 - broadcom/compiler: disallow tsy barrier in thrsw delay slots
1663 - broadcom/compiler: fix assert that current instruction must be in current block
1664 - v3dv: refactor TFU jobs
1665 - broadcom/compiler: rework simultaneous peripheral access checks
1666 - broadcom/compiler: fix condition encoding bug
1667 - broadcom/compiler: padding fixes to QPU assembly dumps
1668 - broadcom/compiler: make opt passes set current block
1669 - broadcom/compiler: check that sig packing is valid when pipelining ldvary
1670 - broadcom/compiler: copy packing when converting add to mul
1671 - v3dv,v3d: don't store swizzle pointer in shader/pipeline keys
1672 - v3d: use V3D_MAX_DRAW_BUFFERS instead of hardcoded constant
1673 - v3dv: account for multisampling when computing subpass granularity
1674 - v3dv: don't use a global constant for default pipeline dynamic state
1675 - v3d,v3dv: move tile size calculation to a common helper
1676 - v3dv: fix internal bpp of D/S formats
1677 - broadcom/compiler: fix early fragment tests setup
1678 - broadcom/compiler: don't allow RF writes from signals after thrend
1679 - broadcom/compiler: fix scoreboard locking checks
1680 - broadcom/compiler: don't move ldvary earlier if current instruction has ldunif
1681 - broadcom/compiler: allow color TLB writes in last instruction
1682 - broadcom/compiler: relax restriction on VPM inst in last thread end slot
1683 - broadcom/compiler: emit passthrough Z write if shader reads Z
1684 - broadcom/compiler: track passthrough Z writes
1685 - v3d,v3dv: don't disable EZ for passthrough Z writes
1686 - broadcom/compiler: improve documentation for Z writes
1687 - broadcom/compiler: improve thrsw merge
1688 - v3dv: add a refcount mechanism to BOs
1689 - v3dv: add swizzle helpers to identify formats wit R/B swap and reverse flags
1690 - v3dv: handle formats with reverse flag
1691 - v3dv: implement VK_EXT_4444_formats
1692 - v3dv: implement double-buffer mode
1693 - v3d: implement double-buffer mode
1694 - docs/features: flag VK_KHR_create_renderpass2 as implemented for v3dv
1695 - broadcom/simulator: handle DRM_V3D_PARAM_SUPPORTS_MULTISYNC_EXT
1696 - v3dv: implement VK_KHR_driver_properties
1697 - broadcom/compiler: add lowering pass to scalarize non 32-bit general load/store
1698 - broadcom/compiler: better document vectorization implications
1699 - broadcom/compiler: implement TMU general 16-bit load/store
1700 - broadcom/compiler: lower packing after vectorization
1701 - broadcom/compiler: support ldunifa with some 16-bit loads
1702 - broadcom/compiler: use ldunifa with unaligned constant offset
1703 - broadcom/compiler: implement 32-bit/16-bit conversion opcodes
1704 - broadcom/compiler: support f32 to f16 RTZ and RTE rounding modes
1705 - v3dv: support VK_KHR_16_bit_storage
1706 - broadcom/compiler: support 16-bit uniforms
1707 - v3dv: expose storagePushConstant16 feature from VK_KHR_16bit_storage
1708 - broadcom/compiler: support 8-bit general store access
1709 - broadcom/compiler: handle to/from 8-bit integer conversions
1710 - broadcom/compiler: support 8-bit loads via ldunifa
1711 - broadcom/compiler: allow vectorization to larger scalar type
1712 - broadcom/compiler: update comment on load_uniform fast-path
1713 - v3dv: support VK_KHR_8bit_storage
1714 - v3dv: drop signature of undefined function
1715 - v3dv: implement VK_KHR_imageless_framebuffer
1716 - v3dv: rework Vulkan 1.2 feature queries
1717 - v3dv: document why we don't expose VK_EXT_scalar_block_layout
1718 - v3dv: support resolving depth/stencil attachments
1719 - v3dv: fallback to blit resolve if render area is not aligned to tile boundaries
1720 - v3dv: expose VK_KHR_depth_stencil_resolve
1721 - v3dv: don't submit noop job if there is nothing to wait on or signal
1722 - broadcom/compiler: fix offset alignment for ldunifa when skipping
1726 - nir/loop_unroll: Always unroll loops that iterate at most once
1727 - glsl/nir: Don't build soft float64 when it cannot be used
1728 - nir/constant_folding: Optimize txb with bias of constant zero to tex
1729 - intel/compiler: Don't predicate a WHILE if there is a CONT
1730 - intel/compiler: Don't store "scalar stage" bits on Gfx8 or Gfx9
1731 - intel/stub: Suppress warnings about DRM_I915_QUERY_PERF_CONFIG
1732 - intel/stub: Implement DRM_I915_QUERY_ENGINE_INFO
1733 - intel/stub: Implement DRM_I915_QUERY_MEMORY_REGIONS
1734 - intel/stub: Implement I915_PARAM_HAS_USERPTR_PROBE
1735 - intel/fs: Use HF as destination type for F32TOF16 in fquantize2f16
1736 - mesa: OpenGL 1.3 feature GL_ARB_texture_border_clamp is not optional
1737 - mesa: OpenGL 1.3 feature GL_ARB_texture_cube_map is not optional
1738 - mesa: OpenGL 1.3 feature GL_ARB_texture_env_combine is not optional
1739 - mesa: OpenGL 1.3 feature GL_ARB_texture_env_dot3 is not optional
1740 - mesa: OpenGL 1.3 and OpenGL ES 1.0 are not optional
1741 - intel/stub: Silence "initialized field overwritten" warning
1742 - intel/stub: Implement shell versions of DRM_I915_GEM_GET_TILING and DRM_I915_SEM_GET_TILING
1743 - intel/fs: Fix gl_FrontFacing optimization on Gfx12+
1744 - ntt: Extend ntt_compile::addr_declared and ntt_compile::addr_reg
1745 - mesa: OpenGL 1.4 feature GL_ARB_depth_texture is not optional
1746 - mesa: OpenGL 1.4 feature GL_ARB_texture_env_crossbar is not optional
1747 - mesa: OpenGL 1.4 feature GL_EXT_blend_color is not optional
1748 - mesa: OpenGL 1.4 feature GL_EXT_blend_func_separate is not optional
1749 - mesa: OpenGL 1.4 feature GL_EXT_blend_minmax is not optional
1750 - mesa: OpenGL 1.4 feature GL_EXT_point_parameters is not optional
1751 - mesa: OpenGL ES 1.1 is not optional
1752 - gallivm/nir: Call nir_lower_bool_to_int32 after nir_opt_algebraic_late
1753 - nir: All set-on-comparison opcodes can take all float types
1754 - intel/fs: Don't optimize out 1.0*x and -1.0*x
1755 - spriv: Produce correct result for GLSLstd450Step with NaN
1756 - spirv: Produce correct result for GLSLstd450Modf with Inf
1757 - spirv: Produce correct result for GLSLstd450Tanh with NaN
1758 - nir: Properly handle various exceptional values in frexp
1759 - nir: Produce correct results for atan with NaN
1760 - nir: Add missing dependency on nir_opcodes.py
1764 - panfrost: Set PIPE_CAP_QUADS_FOLLOW_PROVOKING_VERTEX_CONVENTION
1765 - pan/bi: Add interference between destinations
1766 - pan/bi: Check dependencies of both destinations of instructions
1767 - panfrost: Set dirty state in set_shader_buffers
1768 - panfrost: Re-emit descriptors after resource shadowing
1769 - pan/bi: Make disassembler build reproducibly
1770 - panfrost: Fix ubo_mask calculation
1771 - pan/bi: Skip psuedo sources in ISA.xml
1772 - pan/bi: Don't assign slots for the blend second source
1776 - freedreno: support lua54
1777 - meson: build freedreno tools when other parts of freedreno not enabled
1778 - freedreno: check batch size after the fallback blitter clear
1779 - freedreno/ir3: fix setting the max tf vertex when there are no outputs
1780 - freedreno/a4xx: fix min/max/bias lod sampler settings
1781 - mesa: add missing state to state string computation
1782 - nv50,nvc0: expose R8/RG8_SRGB formats for texturing
1783 - mesa: add just a tiny bit of debug info to some _mesa_problem calls
1784 - freedreno: prefer float immediates when float values are involved
1785 - freedreno/a[345]xx: add R8/RG8 SRGB formats
1786 - a5xx: remove astc srgb workaround logic
1787 - freedreno/ir3: fix image-to-tex flags, remove 3d -> array hack
1788 - nir: remove double-validation of src component counts
1789 - freedreno/a4xx: add proper buffer texture support
1790 - freedreno/a5xx: re-express buffer textures more logically
1791 - freedreno/a6xx: re-express buffer textures more logically
1792 - freedreno/a4xx: fix stencil-textured border colors
1793 - freedreno/a4xx: indicate whether outputs are uint/sint
1794 - freedreno/a4xx: include guesses from a3xx for some of the constid's
1795 - freedreno/a4xx: hook up sample mask/id, used to determine helper invocs
1796 - freedreno/ir3: adjust condition for when to use ldib
1797 - mesa: check target/format for Tex(ture)StorageMem*
1798 - mesa: move around current texture object fetching
1799 - freedreno/a4xx: add missing SNORM formats to help tests pass
1800 - freedreno/a4xx: add some missing legacy formats to help TBOs
1801 - freedreno/a5xx: add missing L8A8_UNORM format to support TBOs
1802 - freedreno/a[345]xx: fix unorm/snorm blend factors when they're "over"
1803 - nir: always keep the clip distance array size updated
1804 - nir: apply interpolated input intrinsics setting when lowering clipdist
1805 - freedreno/ir3: remove unused actual_in counting
1806 - nir/lower_clip: replace bogus comment about gl_ClipDistance reading in GL
1807 - nir/lower_clip: location offset goes into offset, not base
1808 - nir/lower_clip: increment num_inputs/outputs by appropriate amount
1809 - nir/lower_clip: support clipdist array + no vars
1810 - freedreno/ir3: indicate that clipdist arrays are in use
1811 - freedreno/ir3: get the post-lowering clip/cull mask
1812 - ci: move windowoverlap exclusion to all-skips
1813 - freedreno/ci: add piglit runs for a306
1814 - freedreno/ci/a306: add additional skip which hangchecks
1815 - freedreno/a3xx: add some legacy formats
1816 - freedreno/ci/a306: separate msaa fails
1817 - freedreno/ci/a306: split off the f32 blend / texturing failures
1818 - freedreno/ci/a306: split off snorm blending failures
1819 - freedreno/a5xx: enable OES_gpu_shader5
1820 - gitlab-ci: serial close can leave an active read
1821 - gitlab-ci: detect a3xx gpu hang recovery failure
1822 - nv50,nvc0: add new caps to list
1823 - nv50: don't claim support for format-less stores
1824 - freedreno/ci/a306: add more skips
1825 - freedreno/ci/a306: increase concurrency
1829 - virgl/vtest: use correct resource stride in flush_frontbuffer
1830 - virgl/vtest: implement resource_create_front
1831 - virgl: flush cmd buffer when flushing frontbuffer
1832 - drisw: do an MSAA resolve when copying the backbuffer
1836 - intel/nir: also allow unknown format for getting the size of a storage image
1837 - anv: Remove unused struct member
1838 - anv: add functions to set up fake render passes
1839 - anv: allocate fake render pass on pipeline creation
1840 - anv: Split out state attachments allocation
1841 - anv: allocate fake render pass for continuation command buffers
1842 - anv: Split attachment clearing code into their own functions
1843 - anv: split end_subpass into more discrete components
1844 - anv: implement the meat of VK_KHR_dynamic_rendering
1845 - anv: Enable VK_KHR_dynamic_rendering
1846 - anv: Handle resolveImageLayout on dynamic rendering
1847 - anv: Report the right conformance version
1848 - anv: make the pointer valid before we assign stuff into it
1849 - anv: only advertise 64b atomic floats if 64b floats are supported
1850 - intel/compiler: make CLUSTER_BROADCAST always deal with integers
1854 - vulkan-device-select: Don't leak drmDevicePtr
1858 - gbm: Don't pass default usage flags on ABIs < 1
1862 - vulkan, radv: Support backslash in ICD paths
1863 - aco: Work around MSVC restrict in c99_compat.h
1864 - ac: Align ADDR_FASTCALL with addrlib
1865 - radv: Match function definitions to declarations
1866 - meson: Update libelf wrap for Windows
1870 - meson: disable -Werror=thread-safety on FreeBSD
1874 - gallium/swr: Remove driver source
1875 - gallium/swr: Remove common code and build options
1876 - gallium/swr: clean up the documentation after SWR removal from main
1880 - vulkan/log: Tweak our handling of a couple error enums
1881 - vulkan/log: Don't assert on non-client-visible objects
1882 - vulkan/wsi/win32: Delete the wrapper entrypoints
1883 - i965: Emit a NULL surface for buffer textures with no buffer
1884 - nir: Add a nir_foreach_image_variable() iterator
1885 - i965/uniforms: Handle images as a separate pass
1886 - amd/llvm/nir: Add support for nir_var_mem_image
1887 - aco: Add support for nir_var_mem_image
1888 - glsl/nir_linker: nir_var_mem_image is also a GL uniform
1889 - nir/gl_nir_lower_samplers_as_deref: Support nir_var_mem_image
1890 - nir/gl_nir_lower_images: Support nir_var_mem_image
1891 - st/nir: Assign uniform locations to nir_var_mem_image vars
1892 - ntt: Separate image and sampler handling
1893 - llvmpipe: Support image variables living in nir_var_mem_image
1894 - nir/lower_readonly_images_to_tex: Also rewrite variable modes
1895 - clover: Insert dummy uniform variables for images
1896 - clover: Use nir_foreach_image_variable for images
1897 - zink: Images can live in nir_var_mem_image now
1898 - ir3: Check for nir_var_mem_image in shared_barrier handling
1899 - spirv: Assert that OpTypeForwardPointer only points to structs
1900 - glsl: Use nir_var_mem_image for images
1901 - ttn: Use nir_var_mem_image
1902 - st/pbo: Use nir_var_mem_image for images
1903 - nir: Validate image variable modes
1904 - aco: Split var_mem_image barrier handling from global/ssbo
1905 - intel/fs: Stop emitting TGM fences for nir_var_mem_ssbo
1906 - nir/gl_nir_lower_images: Require nir_var_mem_image
1907 - ir3: Images are always nir_var_mem_image
1908 - nir: Drop our attempt at typed-based image mode validation
1909 - nir: s/nir_var_mem_image/nir_var_image/g
1910 - nir: Re-arrange the variable modes
1911 - nir/serialize: Pack deref modes better
1912 - compiler/types: Combine image and sampler type serialization
1913 - compiler/types: Unify the guts of get_sampler/image_count
1914 - compiler/types: Add a texture type
1915 - nir: Allow texture types
1916 - nir/deref: Also optimize samplerND -> textureND casts
1917 - lavapipe: Allow for texture types
1918 - clover/nir: Don't remove texture variables
1919 - spirv: Use texture types for sampled images
1920 - compiler/types: Add a wrap_in_arrays helper
1921 - anv: Implement VK_EXT_global_priority_query
1922 - glsl/nir/linker: Also remove image variables
1923 - vulkan/vk_extensions_gen: Drop support for extra includes
1924 - vulkan/vk_extensions_gen: Stop including vk_object.h
1925 - v3dv: Use vk_instance_get_proc_addr_unchecked for WSI
1926 - lavapipe: Use vk_instance_get_proc_addr_unchecked for WSI
1927 - vulkan: Rework mako error handling in python generators
1928 - vulkan: Drop unnecessary [en]coding comments from python generators
1929 - vulkan/dispatch_table: EntrypointBase doesn't need to derive from object
1930 - vulkan: Break entrypoint parsing into its own file
1931 - vulkan: Move trampoline code-gen to its own file
1932 - vulkan: Move all the common object code to runtime/
1933 - lavapipe: Don't wrap errors returned from vk_device_init in vk_error
1934 - intel: Add has_bit6_swizzle to devinfo
1935 - anv,iris: Advertise a max 3D workgroup size of 1024^3
1936 - anv/allocator: Add a couple of helpers
1937 - anv: Pull aperture size from devinfo
1938 - anv/allocator: Use anv_device_release_bo in anv_block_pool_finish
1939 - anv: Rename anv_bo::index to exec_obj_index
1940 - anv: Add get/set_tiling helpers
1941 - anv: Add a couple more checks in MapMemory
1942 - anv: Fix FlushMappedMemoryRanges for odd mmap offsets
1943 - anv: Add helpers in anv_allocator for mapping BOs
1944 - anv: Always set bindless surface base on SKL+
1945 - anv: Int64 atomics don't need to depend on softpin
1946 - anv: Add a anv_use_relocations helper and use it
1947 - anv: Add a use_relocations physical device bit
1948 - anv: Stop checking for HAS_EXEC_FENCE
1949 - anv: Add an anv_bo_is_pinned helper
1950 - anv: Also disallow CCS_E for multi-LOD images
1951 - anv: Move device memory maps back to anv_device_memory
1952 - vulkan: Pull the device lost framework from ANV
1953 - anv: Use the new common device lost tracking
1954 - vulkan/device: Add a check_status hook
1955 - anv: Wire up the new status check
1956 - vulkan: Add a vk_sync base class
1957 - vulkan/meson: Re-arrange libvulkan_util deps a bit
1958 - vulkan/util: Include stdlib.h
1959 - vulkan/vk_device: Add a drm_fd field
1960 - vulkan: Add a common vk_drm_syncobj struct
1961 - c11/threads: Re-align return values for timed waits
1962 - vulkan: Add an emulated timeline sync type
1963 - vulkan: Add a common implementation of VkFence
1964 - vulkan: Add a common implementation of VkSemaphore
1965 - vulkan: Add common implementations of vkQueueSubmit and vkQueueWaitIdle
1966 - vulkan/device: Log the timeline mode when lost
1967 - vulkan/wsi: Drop wsi_common_get_current_time()
1968 - wsi/display: Rework wsi_fence a bit
1969 - vulkan/wsi/display: Wrap wsi_display_fence in a vk_sync
1970 - anv: Delete ANV_SEMAPHORE_TYPE_DUMMY
1971 - anv: Add a BO sync type
1972 - anv: Remove the last remnants of in/out fences
1973 - anv: Use helpers in util/os_time.h in the query code
1974 - anv: Convert to the common sync and submit framework
1975 - vulkan: Add an emulated binary vk_sync type
1976 - anv: Remove unnecessary syncobj wrappers
1977 - anv: Simplify submit_simple_batch()
1978 - vulkan,anv: Auto-detect syncobj features
1979 - vulkan: Add a dummy sync type
1980 - vulkan/sync: Rework asserts a bit
1981 - crocus: wm_prog_key::key_alpha_test uses GL enums
1982 - intel/fs,vec4: Drop uniform compaction and pull constant support
1983 - intel/fs,vec4: Drop support for shader time
1984 - intel/blorp: Stop depending on prog_data binding tables
1985 - intel/fs,vec4: Drop prog_data binding tables
1986 - intel/compiler: Get rid of wm_prog_key::frag_coord_adds_sample_pos
1987 - intel/fs: Drop high_quality_derivatives
1988 - anv: Stop doing too much per-sample shading
1989 - intel/dev: Add gtt_size to devinfo
1990 - Revert "anv: Stop doing too much per-sample shading"
1991 - vulkan/queue: Handle WSI memory signal information
1992 - anv: Implement vk_device::create_sync_for_memory
1993 - anv: Drop anv_sync_create_for_bo
1994 - radeonsi/nir: Check for VARYING_SLOT_PRIMITIVE_ID not SYSTEM_VALUE
1995 - nir: Add a new sample_pos_or_center system value
1996 - intel/fs: Return fs_reg directly from builtin setup helpers
1997 - intel/fs: Rework emit_samplepos_setup()
1998 - intel/fs: Implement the sample_pos_or_center system value
1999 - anv,nir: Use sample_pos_or_center in lower_wpos_center
2000 - anv/pipeline: Get rid of sample_shading_enable
2001 - intel/fs: Be more conservative in split_virtual_grfs
2002 - intel/fs: Use OPT() for split_virtual_grfs
2003 - intel/eu: Don't double-loop as often in brw_set_uip_jip
2004 - Revert "intel/fs: Do cmod prop again after scheduling"
2005 - intel/fs: Reset instruction order before re-scheduling
2006 - intel/fs: Add a NONE scheduling mode
2007 - vulkan/runtime: Validate instance version on 1.0 implementations
2008 - anv,radv,v3dv: Move AcquireImageANDROID to common code
2009 - radv: Move QueueSignalReleaseImageANDROID to common code
2010 - anv: Use the common QueueSignalReleaseImageANDROID from RADV
2011 - v3dv: Use the common QueueSignalReleaseImageANDROID from RADV
2012 - turnip: Use vk_common_AcquireImageANDROID
2013 - turnip: Use vk_common_QueueSignalReleaseImageANDROID for DRM
2014 - intel/compiler: Stop using GLuint in brw_compiler.h
2015 - intel/fs: Use compare_func for wm_prog_key::alpha_test_func
2016 - spirv,radv: Fix some GL enum comments
2017 - vulkan/runtime: Implement 1.3 features/properties
2018 - anv/pass: Don't set first_subpass_layout for stencil-only attachments
2019 - .mailmap: Switch Jason Ekstrand to @collabora.com
2020 - anv: Implement 1.3 features/properties
2021 - anv: Advertise Vulkan 1.3
2022 - vulkan/wsi: Set MUTABLE_FORMAT_BIT in the prime path
2023 - vulkan/wsi/drm: Break create_native_image in pieces
2024 - vulkan/wsi: Add a helper for the configure/create/bind pattern
2025 - vulkan/wsi/drm: Break create_prime_image in pieces
2026 - vulkan/wsi/x11: Split image creation
2027 - vulkan/wsi/wayland: Split image creation
2028 - vulkan/wsi/display: Split image creation
2029 - vulkan/wsi/win32: Delete unnecessary copy+paste from DRM
2030 - vulkan/wsi/win32: Break create_win32_image in pieces
2031 - vulkan/wsi/drm: Drop wsi_create_native/prime_image
2032 - wsi/common: Set VK_IMAGE_CREATE_ALIAS_BIT
2033 - vulkan/wsi: Add image create and bind helpers
2034 - anv/image: Add some asserts when binding swapchain images
2035 - anv/image: Call into WSI to create swapchain images
2036 - anv: Call vk_command_buffer_finish if create fails
2037 - anv: Don't assume depth/stencil attachments have depth
2041 - Fix typo
2045 - microsoft/clc: Images use nir_var_mem_image
2046 - util/hash_table: Clear special 0/1 entries for u64 hash table too
2047 - microsoft/compiler: Use textures for SRVs
2048 - u_prim_restart: Fix index scanning with start offset
2049 - d3d12: Don't accumulate timestamp queries
2050 - u_threaded_context: Support including from C++
2051 - d3d12: Inherit from threaded_query
2052 - d3d12: Resources inherit from threaded_resource
2053 - d3d12: Inherit from threaded_transfer
2054 - d3d12: Use thread safe slab allocators in transfer_map handling
2055 - d3d12: Pass explicit context to pre/post draw surface blits
2056 - d3d12: Hook up threaded context
2057 - d3d12: Fully init primconvert config
2058 - d3d12: Fix Linux fence wait return value
2059 - d3d12: Handle non-infinite wait timeouts > 49.7 days as infinite
2060 - util/libsync: Fix timeout handling if poll() wakes up early
2061 - d3d12: Don't wait for \*all* batches when synchronizing a resource
2062 - d3d12: Don't wait for GPU reads to do CPU reads
2063 - mesa/main: Fix use of alloca() without #include "c99_alloca.h"
2064 - clc: Use kernel_arg_type_qual string to add const type qualifier to arg metadata
2065 - microsoft/clc: Add a test for arg metadata
2066 - d3d12: Fix incorrect hash table usage
2067 - meson: Don't override built-in cpp_rtti option, error if it's invalid
2068 - meson: Allow mismatching RTTI for MSVC
2069 - android: Add a BOARD CFlags option so build can be customized
2070 - mesa/main, android: Log errors to logcat
2071 - android,d3d12: Support using DirectX-Headers dependency from AOSP
2072 - android: Allow forcing softpipe
2073 - d3d12: Support BGRA 555 and 565 formats
2074 - d3d12: Support PIPE_CAP_MIXED_COLOR_DEPTH_BITS
2075 - d3d12: Support RGBX formats mapped to RGBA
2076 - microsoft/compiler: Handle GLES external textures
2077 - gallium, windows: Use HANDLE instead of FD for external objects
2078 - winsys/d3d12: Populate winsys handle format
2079 - d3d12: Handle import/export of fd shared handles
2080 - d3d12: Make format list all use macros
2081 - d3d12: Generate format table using a macro list
2082 - d3d12: Generate a pipe format -> typeless mapping table too
2083 - d3d12: Validate opened D3D12 resource matches pipe template
2084 - CI/windows: Uprev piglit
2085 - CI/windows: Upload result.txt as an artifact
2086 - CI/windows: Move D3D12 test YML to D3D12 driver folder
2087 - CI/windows: Move SPIRV-to-DXIL test YML to microsoft folder
2088 - CI/windows: Move reference files to relevant ci subdirectories
2089 - CI/d3d12: Add a quick_shader run
2090 - d3d12: Support compat level 330
2091 - windows: Use TLS context/dispatch with shared-glapi
2092 - d3d12: Handle depth readback on drivers that require full-resource copies for depth
2093 - nir: Add an 'external' texture type for parity with samplers
2094 - d3d12: Force emulation of all YUV formats using per-plane formats
2095 - d3d12: Handle opening planar resources
2096 - d3d12: Allow creating planar resources
2097 - d3d12: Use overall resource format + plane format to get format info
2098 - microsoft/compiler: Implement inot
2099 - microsoft/compiler: Remove algebaric pass for inot
2100 - ci/windows: Remove line numbers from assertions in spirv2dxil tests
2101 - glapi: Never use dllimport/dllexport for TLS vars on Windows
2102 - microsoft/compiler: Support lowered io (nir_intrinsic_load_input/store_output)
2103 - microsoft/compiler: Lower io
2104 - microsoft/compiler: Delete non-sysval deref load/store code
2105 - microsoft/compiler: Load synthesized sysvals via lowered io
2106 - d3d12: Fix NV12 resource importing
2107 - softpipe: Add a dummy field to sp_fragment_shader_variant_key
2108 - CI: Trigger Windows build on softpipe changes
2109 - microsoft/compiler: Emit SSBOs from 0 -> count for GL (non-kernel, non-Vulkan) shaders
2110 - microsoft/compiler: Hook up uavs-at-every-stage flag
2111 - microsoft/compiler: Handle write masks in SSBO lowering pass
2112 - d3d12: Support SSBOs in root signatures
2113 - d3d12: Always create buffers as UAV-capable
2114 - d3d12: Support setting SSBOs on the context and turning them into descriptors
2115 - d3d12: Use DXIL load/store lowering pass
2116 - d3d12: Set SSBO support caps
2117 - d3d12: Avoid a debug warning trying to unmap a not-mapped resource
2118 - d3d12: Replace pipe cap literals with D3D12 defines when available
2119 - d3d12: Enable cubemap arrays
2120 - microsoft/compiler: Position should always be no-perspective
2121 - d3d12: Handle cubemap gather on int cubemaps
2122 - d3d12: Enable texture gather
2123 - microsoft/compiler: Fix LOD instruction to return 2 values
2124 - gallium/aux: Move index offsetting from prim restart to primconvert
2125 - microsoft/compiler: Change vulkan_environment bool to an enum
2126 - microsoft/compiler: Put SSBO and image handles in separate arrays
2127 - microsoft/compiler: Emit GL images in descriptor space 1 with driver_location instead of binding
2128 - microsoft/compiler: Unify handle retrieval between images and UBO/SSBO
2129 - microsoft/compiler: Emit SRVs/UAVs as arrays
2130 - microsoft/compiler: Fix array-of-array handling for derefs of textures/images
2131 - microsoft/compiler: Handle images as derefs for GL
2132 - microsoft/compiler: Implement atomic image ops
2133 - microsoft/compiler: Handle forced early depth
2134 - microsoft/compiler: Hook up memory/control barriers
2135 - microsoft/compiler: Fix handling of fp16-in-32bit-val ops to handle high bits
2136 - d3d12: Shrink 2D array size so that max-layer cube arrays can be created
2137 - d3d12: Fix format table typeless-ness for A8 and RGBA1010102
2138 - d3d12: Rename UAV -> SSBO to disambiguate with image UAVs
2139 - d3d12: Add missed SSBO binding enum value
2140 - d3d12: Figure out if we can support GL shader images
2141 - d3d12: Handle format support queries for shader images
2142 - d3d12: Init null UAVs
2143 - d3d12: Retrieve shader image dimensions during shader compiles
2144 - d3d12: Handle images in the root signature
2145 - d3d12: Handle set_shader_images
2146 - d3d12: Create textures as UAV-capable when appropriate
2147 - d3d12: Fill out shader image descriptor tables
2148 - d3d12: Lower cube images to 2D arrays via existing int cubemap lowering pass
2149 - d3d12: Handle memory barriers
2150 - d3d12: Handle bitcasting of shader images
2151 - d3d12: Set appropriate caps for shader images
2152 - nir_opt_dead_cf: Remove dead ifs
2153 - shader_info: tess.spacing needs to be unsigned
2154 - microsoft/compiler: Move workgroup_size lowering from clc
2155 - microsoft/compiler: Handle more GL memory barriers
2156 - d3d12: Limit sampler view count to 32
2157 - d3d12: Keep state vars last in the per-stage root parameters
2158 - d3d12: Remove draw_info from selection_context
2159 - d3d12: Stop trying to set D3D12_DIRTY_SHADER during bindings
2160 - d3d12: Compile, bind, and cache compute PSOs
2161 - d3d12: Support compute root signatures
2162 - d3d12: Hook up compute shader variations
2163 - d3d12: Implement launch_grid
2164 - d3d12: Implement num workgroups as a state var
2165 - d3d12: Handle indirect dispatch
2166 - d3d12: Run DXIL shared atomic lowering pass
2167 - d3d12: Enable compute
2168 - docs: Update d3d12 extension list and new_features.txt
2169 - d3d12/ci: Skip flaky tex-miplevel-selection and timestamp tests
2170 - mesa/st: Assert that NIR drivers that support tess use tess levels as inputs
2171 - d3d12: Export d3d12_get_state_var from d3d12_nir_passes.c
2172 - d3d12: Fix re-enabling predication after temporary disablement
2173 - d3d12: Predication fix: re-enable after restarting a batch if needed
2174 - d3d12: Predication fix: For boolean queries used for predication, D3D12 uses uint64, so clear at …
2175 - d3d12: Declare support for inverted conditional render
2176 - d3d12: Upgrade first vertex state var into all vertex draw params
2177 - d3d12: Enable base instance and draw params extensions
2178 - d3d12: Add a command signature cache for indirect draws
2179 - d3d12: Handle draw indirect and multi-draw indirect
2180 - d3d12: Handle indirect twoface draws
2181 - d3d12: Add a compute transformation to handle indirect draws that need draw params
2182 - d3d12: Enable draw and multi-draw indirect
2183 - docs: Update d3d12 features
2184 - d3d12: When no framebuffer attachments are present, use ForcedSampleCount instead of SampleDesc.C…
2185 - d3d12: When no framebuffer attachments are present, the viewport must be clamped to framebuffer s…
2186 - d3d12: Support ARB_framebuffer_no_attachments
2187 - docs: Update d3d12 feature list
2188 - ci/windows: Use 2 container stages
2189 - microsoft/compiler: Handle variables declared per-sample
2190 - microsoft/compiler: Handle load_sample_pos_at_id
2191 - microsoft/compiler: Always have at least one GS active stream
2192 - microsoft/compiler: Handle 'pull model' explicit interpolation intrinsics
2193 - microsoft/compiler: Handle textureGatherCmp
2194 - microsoft/compiler: Handle input coverage
2195 - microsoft/compiler: Handle tex texture/sampler offset srcs
2196 - microsoft/compiler: Handle load_invocation_id for GS and HS
2197 - microsoft/compiler: Emit samplers as array types
2198 - microsoft/compiler: Handle bitfield_insert
2199 - microsoft/compiler: Use ibfe/ubfe for bitfield extract instead of lowering to shifts
2200 - microsoft/compiler: Handle msb/lsb/bfrev
2201 - microsoft/compiler: Lower helper invocations
2202 - d3d12: Sample mask output needs to be uint-typed
2203 - d3d12: Lower load_sample_pos to load_sample_pos_at_id
2204 - d3d12: Report sample positions
2205 - d3d12: Modify shaders when MSAA is disabled
2206 - d3d12: Relax multisampling direct copy requirements
2207 - d3d12: Temp resources for same-resource copies can be MSAA too
2208 - d3d12: Report number of GS streams
2209 - d3d12: Apply GS point sprite lowering to fixed-function point size too
2210 - d3d12: Run point sprite lowering pass on multi-stream GS when safe
2211 - d3d12: Support dynamic UBO/SSBO indexing
2212 - d3d12: When mapping a non-directly-mappable resource for write, readback first
2213 - d3d12: Set sample-rate shading and GLSL 400 caps
2214 - docs: Update d3d12 features
2215 - mesa/st: Lower user clip planes for tess eval too
2216 - microsoft/compiler: Force integer I/O vars to use flat/constant interpolation
2217 - microsoft/compiler: Use driver_location instead of location for inter-stage varying index in GL
2218 - microsoft/compiler: Semantic table should be de-duped for multi-row semantics too
2219 - microsoft/compiler: Multi-row output semantics need to write multiple never_writes_masks
2220 - microsoft/compiler: Getting a builtin function with an undeclared signature should be unreachable
2221 - microsoft/compiler: Add mapping from MESA_SHADER_* to DXIL_*_SHADER for tessellation
2222 - microsoft/compiler: Fix typo in enum entry
2223 - microsoft/compiler: Emit statically-indexed resource handles and scratch later
2224 - microsoft/compiler: Support emitting multiple functions into a DXIL module
2225 - microsoft/compiler: Emit functions with actual function names
2226 - microsoft/compiler: Emit all NIR functions into the DXIL module
2227 - microsoft/compiler: Handle store_per_vertex_output for HS outputs
2228 - microsoft/compiler: Split hull (tess ctrl) shaders into main and patch constant funcs
2229 - microsoft/compiler: Delete misleading TODO comments about semantic table
2230 - microsoft/compiler: Emit HS PSV validation and entrypoint metadata
2231 - microsoft/compiler: Fix I/O signatures for tess shaders
2232 - microsoft/compiler: Overlap patch and non-patch varyings so both are separately 0-indexed
2233 - microsoft/compiler: When sorting patch varyings, adjust location to be in normal varying range
2234 - microsoft/compiler: Gather patch const signature and handle tess factor in it
2235 - microsoft/compiler: Add patch constant signature into PSV and as container blob
2236 - microsoft/compiler: Add a pass for hull and domain shaders to shrink tess level vars
2237 - microsoft/compiler: For store_output from HS, use storePatchConstant
2238 - microsoft/compiler: For load_input from DS, use loadPatchConstant
2239 - microsoft/compiler: Handle load_per_vertex_output as LoadOutputControlPoint
2240 - microsoft/compiler: Handle load_output in the HS stage as reading a previously written patch cons…
2241 - microsoft/compiler: Handle domain location intrinsic
2242 - microsoft/compiler: Emit DS PSV validation and entrypoint metadata
2243 - microsoft/compiler: Primitive ID should only be added as a sysval in geometry shaders
2244 - microsoft/compiler: Location_frac needs to be included in sort order
2245 - microsoft/compiler: Handle clip/cull distance as an input to tess shaders
2246 - d3d12: Enable PIPE_CAP_TGSI_TEXCOORD
2247 - d3d12: Initial plumbing for tesselation
2248 - d3d12: Link tesselation control and eval shaders
2249 - d3d12: Handle patch_vertices and patch topology
2250 - d3d12: Handle passthrough TCS in the case where eval is bound
2251 - d3d12: Add a state variable for patch_vertices_in
2252 - d3d12: Update varying creation logic to handle location_frac
2253 - d3d12: Handle input clip array size in the shader key
2254 - d3d12: Set caps for tesselation
2255 - microsoft/compiler: Fix UAV resource ID counting for static indexed handles
2256 - d3d12: Fix compute transform for multi-draw indirect with dynamic count + state vars
2257 - d3d12: Add UAV barriers for UAVs that are being used by compute transforms
2258 - d3d12: Include SO buffer count as a PSO dirty bit
2259 - d3d12: Support transform feedback pause/resume
2260 - d3d12: Move indirect compute to real indirect dispatches
2261 - d3d12: SO buffer filled size is only 32-bit
2262 - d3d12: Add a comment for what the existing compute transform does
2263 - d3d12: Add a couple compute transforms for "fake" SO buffers
2264 - d3d12: Add a compute transform for draw auto
2265 - d3d12: Move compute transform state save/restore to compute_transforms.cpp
2266 - d3d12: Move "fake" SO buffer handling to compute transforms instead of CPU readback
2267 - d3d12: Implement DrawAuto aka DrawTransformFeedback
2268 - d3d12: Compute transform UBO0 is actually binding 1
2269 - d3d12: Rewrite subquery logic
2270 - d3d12: Switch primitives-generated query to use XFB, GS, and IA data
2271 - d3d12: ARB_transform_feedback2
2272 - microsoft/compiler: Correctly support I/O on variables with location_frac
2273 - microsoft/compiler: Support multiple GS output streams
2274 - d3d12: Unpack multi-stream varyings
2275 - d3d12: Fix xfb varying matching for vars with location_frac
2276 - d3d12: Handle indexed queries
2277 - d3d12: ARB_transform_feedback3
2278 - microsoft/compiler: Only prep phis for the current function
2279 - microsoft/compiler: Only treat tess level location as special if it's a patch constant
2280 - tc: CPU storage needs to be freed with align_free
2284 - intel: provide pci bus and dev info in base device struct
2285 - intel: use PCI info to compute device uuid
2286 - anv: remove private pci fields
2287 - intel: dump PCI info in intel_dev_info
2288 - intel: remove chipset_id
2289 - intel: add swizzle flag into driver uuid
2290 - anv: refactor queue chain
2294 - util: unbreak non-linux mips64 build
2295 - util: fix util_cpu_detect_once() build on OpenBSD
2296 - radv: use MAJOR_IN_SYSMACROS for sysmacros.h include
2297 - util/u_atomic: fix build on clang archs without 64-bit atomics
2298 - util: fix build with clang 10 on mips64
2299 - util: use correct type in sysctl argument
2303 - freedreno/layout: Fix the UBWC block size for the Y plane
2304 - turnip: enable UBWC for NV12
2305 - turnip: use SUBDRAW_SIZE and constant sized tess bos
2309 - turnip: Update the msm_kgsl.h header with the sanitized 4.19 version
2313 - intel/genxml: Update genxml to support tessellation/geometry distribution
2314 - intel/dev/test: Assert (verx10 / 10) == ver
2315 - Revert "iris: Disable I915_FORMAT_MOD_Y_TILED_GEN12* on adl-p/display 13"
2316 - iris: Use mi_builder in iris_load_indirect_location()
2317 - intel/genxml/125: Update COMPUTE_WALKER POSTSYNC_DATA struct
2318 - anv,blorp,iris: Set MOCS for COMPUTE_WALKER post-sync operation
2319 - intel/dev: Add platform enum with DG2 G10 & G11
2320 - intel: Add intel_gem_count_engines
2321 - intel: Add intel_gem_create_context_engines
2322 - iris: Add iris_init_batches
2323 - iris/batch: Move kernel context init to iris_init_non_engine_contexts
2324 - iris/batch: Add exec_flags field
2325 - iris: Move away from "hw" for some context terminology
2326 - iris: Destroy all batches with a new iris_destroy_batches() function
2327 - iris: Make iris_kernel_context_get_priority() public
2328 - iris/batch: Add support for engines contexts
2329 - intel/l3: Make DG1 urb-size exception more generic
2330 - iris: Not all gfx12+ have aux_map_ctx
2331 - anv: Align buffer VMA to 2MiB for XeHP
2332 - iris: Align buffer VMA to 2MiB for XeHP
2333 - nir/lower_tex: Add filter for tex offset lowering
2334 - intel/compiler: Use nir_lower_tex_options::lower_offset_filter for tg4 on XeHP
2335 - intel/genxml/12.5: Remove bt-pool enable from 3DSTATE_BINDING_TABLE_POOL_ALLOC
2336 - anv: Add BINDING_TABLE_POOL_BLOCK_SIZE
2337 - intel/compiler: Adjust TCS instance-id for dg2+
2338 - isl: Don't enable HDC:L1 caches on DG2
2339 - intel: Add device info for DG2
2340 - intel: Add \*disabled* device ids for DG2
2341 - intel/devinfo: Adjust L3 banks for DG2
2342 - iris: Use mi_builder to set 3DPRIM registers for draws
2343 - iris: Use mi_builder for load/store reg/mem/imm functions
2344 - intel/dev: Add max_threads_per_psd field to devinfo for gfx8+
2345 - anv,blorp,crocus,i965,iris: Use devinfo->max_threads_per_psd for gfx8+
2346 - intel/dev: Add intel_hwconfig_types.h from random post on the internet
2347 - intel/dev: Add intel_print_hwconfig_table()
2348 - intel/dev: Print urb size with intel_dev_info
2349 - intel/dev: Add intel_device_info::apply_hwconfig
2350 - intel/dev: Set intel_device_info::apply_hwconfig for DG2
2351 - intel/dev: Apply settings from hwconfig if devinfo::apply_hwconfig is set
2352 - intel/dev: Recalculate max_cs_threads after applying hwconfig changes
2353 - intel/gem: Return length from intel_i915_query_alloc
2354 - intel/dev: Add DG1 PCI id 0x4909
2355 - intel/dev: Add device ids for ADL-N
2356 - intel/dev: Add device info for RPL
2357 - intel/genxml: Extend length of 3DSTATE_WM_HZ_OP for gfx12.5
2358 - intel/genxml: Extend length of 3DSTATE_DEPTH_BUFFER for gfx12.5
2359 - isl: Enable compression with Tile4
2360 - intel/fs: Assert that old pull-const code is not used if devinfo->has_lsc
2361 - anv: Align GENERAL_STATE_POOL_MIN_ADDRESS to 2MiB
2362 - anv: Align state pools to 2MiB on XeHP
2366 - radv: Always inline descriptor writes
2367 - radv: Split off cmd_buffer variant of descriptor set updates
2368 - radv: Split off cmd_buffer variant of descriptor set template updates
2369 - nvc0: Fix uninitialized width/height/depth warning.
2370 - radv: Refactor S_FIXED to radv_float_to_{s,u}fixed
2371 - radv: Expose min_lod in \*_make_texture_descriptor
2372 - vulkan: Update the XML and headers to 1.2.199
2373 - radv: Implement VK_EXT_image_view_min_lod
2377 - d3d10umd: Rename Dxgi.h to DxgiFns.h.
2378 - d3d10umd: Update for transfer interface changes.
2379 - d3d10umd: Fix MSVC build.
2380 - d3d10umd: Update for set_sampler_views take_ownership parameter.
2384 - vc4/ci: update expected results
2385 - v3dv/ci: update expected results
2386 - broadcom/compiler: handle array of structs in GS/FS inputs
2387 - broadcom/compiler: apply constant folding on early GS lowering
2388 - nir: add NIR_DEBUG envvar
2389 - mesa: allow TEXTURE_BUFFER target for ARB_texture_buffer_range
2390 - st/pbo: do not use GS for NIR preferred shaders
2391 - gallium/util: add helper to clamp colors to valid range
2392 - v3d: clamp clear color
2393 - gallium: add new PIPE_CAP_IMAGE_STORE_FORMATTED
2394 - st/pbo: add the image format in the download FS
2395 - st/pbo: set layer coord for array textures
2396 - v3d: enable ARB_texture_view
2397 - tgsi-to-nir: initialize NIR_DEBUG envvar
2398 - nir: use call_once() to init debug variable
2399 - broadcom/ci: restructure expected results
2400 - softpipe: enable PIPE_CAP_IMAGE_STORE_FORMATTED
2401 - d3d12: enable PIPE_CAP_IMAGE_STORE_FORMATTED
2402 - mesa/st: do not expose ARB_shader_image_load_store if not fully implemented
2403 - mesa: fix MAX_GEOMETRY_IMAGE_UNIFORMS check support
2404 - v3d/doc: do not expose ARB_shader_image_load_store
2405 - v3d: keep clear color untouched
2409 - spirv: Don't add 0.5 to array indicies for OpImageSampleExplicitLod
2410 - clover/image: add templated basic_image class to simplify image subclassing
2411 - clover/format: Full rework on how we declare supported images.
2412 - clover/formats: pass in cl_mem_flags for better format checking
2413 - clover/api: fix clGetMemObjectInfo for images
2417 - intel: Drop Tigerlake revision 0 workarounds
2418 - crocus: Replace devinfo->ver[x10] checks with GFX_VER[x10]
2419 - intel/genxml: Fix Indirect Object Access Upper Bound on Gfx4
2420 - intel/genxml: Add an "mbz" data type
2421 - intel/genxml: Drop "Hierarchical Depth Buffer MOCS" field
2422 - intel/genxml: Change 3DSTATE_CONSTANT_XS::MOCS to be MBZ on Gfx8.
2423 - isl: Fill in MOCS even for SURFTYPE_NULL surfaces.
2424 - isl: Fill in MOCS for NULL depth, stencil, and HiZ buffers.
2425 - blorp: Fill in MOCS even for SURFTYPE_NULL surfaces.
2426 - blorp: Fill in MOCS for null depth/stencil/HiZ buffers.
2427 - blorp: Use a non-zero MOCS for disabled constant buffers
2428 - iris: Drop unnecessary parenthesis
2429 - iris: Set Bindless Sampler State MOCS
2430 - iris: Set default MOCS for NULL depth/stencil/HiZ buffers
2431 - iris: Set MOCS on 3DSTATE_CONSTANT_XS on Gfx9+
2432 - iris: Set MOCS on 3DSTATE_CONSTANT_ALL packets that disable all buffers
2433 - iris: Set MOCS on NULL vertex buffers
2434 - iris: Set MOCS on NULL stream output buffers
2435 - iris: Fix MOCS for buffer copies
2436 - anv: Set default MOCS for NULL depth/stencil/HiZ buffers
2437 - anv: Set MOCS for 3DSTATE_CONSTANT_XS on Gfx7.x as well
2438 - anv: Set MOCS in 3DSTATE_CONSTANT_XS even if there isn't a buffer.
2439 - anv: Set MOCS on NULL vertex buffers
2440 - anv: Set MOCS on NULL stream output buffers
2441 - crocus: Set MOCS for most state base addresses on pre-Gen8
2442 - crocus: Tidy the ifdefs for emitting STATE_BASE_ADDRESS
2443 - crocus: Set MOCS for index buffers on Gen6+
2444 - crocus: Set MOCS on NULL stream output buffers
2445 - crocus: Set default MOCS for NULL depth/stencil/HiZ buffers
2446 - crocus: Set MOCS for push constant buffers where possible
2447 - crocus: Set MOCS for 3DSTATE_SO_BUFFERS on Gfx7.x too
2448 - crocus: Fix MOCS for buffer copies.
2449 - i965: Use ISL for MOCS rather than open coding it everywhere
2450 - i965: Set default MOCS for NULL depth/stencil/HiZ buffers
2451 - i965: Set MOCS for push constant buffers on Haswell and Gfx9+
2452 - i965: Set MOCS on NULL stream output buffers
2453 - i965: Set MOCS for 3DSTATE_SO_BUFFERS on Gfx7.x too
2454 - i965: Set MOCS for 3DSTATE_INDEX_BUFFER on Gfx6/7 as well.
2455 - i965: Fix MOCS for BLORP buffer copies
2456 - i965: Port STATE_BASE_ADDRESS to genxml and fix bugs
2457 - i965: Set MOCS for Bindless Surface/Sampler State base addresses
2458 - intel/genxml: Add an field option for nonzero="true"
2459 - intel/genxml: Assert that all MOCS fields are non-zero on Gfx7+
2460 - intel/genxml: Include blitter commands in gen*_pack.h
2461 - intel/genxml: Allow MI_FLUSH_DW on the blitter
2462 - intel/genxml: Add XY_BLOCK_COPY_BLT on Tigerlake and later.
2463 - iris: Fix parameters to iris_copy_region in reallocate_resource_inplace
2464 - intel/genxml: Simplify prefix handling for field value lists
2465 - intel/genxml: Collapse leading underscores on prefixed value defines
2466 - intel/genxml: Fix MI_FLUSH_DW to actually specify the length properly
2467 - intel/genxml: Fix XY_BLOCK_COPY_BLT destination tiling field type
2468 - intel/genxml: Decode VALIGN/HALIGN values in XY_BLOCK_COPY_BLT
2469 - iris: Make a helper function for cross-batch dependency flushing
2470 - iris: Check for cross-batch flushing whenever a buffer is newly written.
2471 - iris: Tidy code in iris_use_pinned_bo a bit
2472 - blorp: Fix compute-blits for rectangles not aligned to the workgroup
2473 - blorp: Don't try to use the 3D stencil write hardware for compute
2474 - blorp: Assert that BLORP_BATCH_PREDICATE_ENABLE isn't set for compute
2475 - blorp: Disallow multisampling for BLORP compute blits and copies.
2476 - iris: Rename is_render_target to is_dest in a few blit functions
2477 - isl: Move some genxml surface state helpers into an include file
2478 - intel/vec4: Use ir_texture_opcode less in emit_texture()
2479 - intel/vec4: Use nir_texop in emit_texture instead of translating
2480 - intel/vec4: Inline emit_texture and move helpers to brw_vec4_nir.cpp
2481 - intel/compiler: Use uppercase enum values in brw_ir_performance.cpp
2482 - intel/fs: Reuse the same FS input slot for VUE header fields.
2483 - iris: Use prog_data->inputs rather than shader info in SBE code.
2484 - iris: Do primitive ID overrides in 3DSTATE_SBE not SBE_SWIZ
2485 - iris: Directly access BOs rather than using iris_resource_bo(...)
2486 - intel: Allow copy engine class in intel_gem_create_context_engines()
2487 - intel/genxml: Add XY_BLOCK_COPY_BLT Color Depth enum values
2488 - intel/dev: Add a has_flat_ccs flag
2489 - blorp: Add a blorp_address::local_hint flag
2490 - isl: Add isl_dev->mocs.blitter_{src,dst} fields
2491 - blorp: Add support for blorp_copy via XY_BLOCK_COPY_BLT
2492 - iris: Create an IRIS_BATCH_BLITTER for using the BLT command streamer
2493 - iris: Only have one blorp_batch_init/finish in iris_copy_region()
2494 - iris: Set BLORP_BATCH_USE_{COMPUTE,BLITTER} flags for the target batch
2495 - iris: Implement iris_blorp_exec() for the blitter engine
2496 - iris: Fix and refactor check for clear color being fully zero
2497 - iris: Make an iris_foreach_batch macro that skips unsupported batches
2498 - blorp: Add blorp_measure hooks to the blitter codepaths
2502 - v3dv: account for 64bit time_t on 32bit arches
2506 - lavapipe: Fixed maxFragmentCombinedOutputResources
2507 - anv: Fixed maxFragmentCombinedOutputResources
2508 - turnip: Fixed maxFragmentCombinedOutputResources
2509 - panvk: Fixed maxFragmentCombinedOutputResources
2513 - util/u_trace: Replace Flag with IntEnum to support python3.5
2514 - util/ra: use adjacency matrix for undirected graph
2515 - util/ra: Fix numeric overflow during bitset allocation
2519 - frontends/va: use un-padded width/height in ExportSurfaceHandle
2520 - frontend/va: Setting the size of VADRMPRIMESurfaceDescriptor
2524 - egl: remove unnecessary spaces after types
2525 - egl/wayland: replace EGL_DRI2_MAX_FORMATS by EGL_DRI2_NUM_FORMATS
2526 - egl/wayland: deprecate drm_handle_format() and drm_handle_capabilities()
2527 - egl/wayland: do not try to access memory if allocation failed
2528 - egl/wayland: move formats and modifiers to a separate struct
2529 - egl/wayland: remove unused constant EGL_DRI2_NUM_FORMATS
2530 - loader: add function to get render node from dev_t
2531 - egl/wayland: add initial dma-buf feedback support
2532 - egl/wayland: move loader_dri_create_image() calls to separate functions
2533 - egl/wayland: use surface dma-buf feedback to allocate surface buffers
2534 - egl/wayland: do not try to bind to wl_drm if not advertised
2535 - egl/wayland: fix surface dma-buf feedback error exits
2539 - driconf: Fix unhandled tags in static conf
2543 - vulkan/wsi/wayland: don't expose surface formats not fully supported
2544 - anv: fix push constant lowering with bindless shaders
2545 - pps: remove counter_ids fields
2546 - pps: add an intel config file
2547 - docs: put a list of commands to setup perfetto
2548 - intel/dev: printout timestamp period
2549 - intel/pps: provide accurate min sampling period
2550 - intel/pps: reuse timestamp_frequency from intel_device_info
2551 - intel/dev: fix HSW GT3 number of subslices in slice1
2552 - intel/dev: don't forget to set max_eu_per_subslice in generated topology
2553 - intel/dev: reuse internal functions to set mask
2554 - intel/dev: fix subslice/eu total computations with some fused configurations
2555 - intel/perf: fix perf equation subslice mask generation for gfx12+
2556 - intel/devinfo: use compatible type for ARRAY_SIZE
2557 - intel/devinfo: fix wrong offset computation
2558 - intel: remove 2 preproduction pci-id for ADLS
2559 - intel: move away from booleans to identify platforms
2560 - intel/dev: also test crocus & i915 pci-ids
2561 - anv: don't forget to add scratch buffer to BO list
2562 - anv: fix multiple wait/signal on same binary semaphore
2563 - anv: don't try to close fd = -1
2564 - anv: initialize anv_bo_sync base fields
2565 - intel/fs: fix shader call lowering pass
2566 - pps: allow drivers to report timestamps in their own time domain
2567 - intel/perf: add a helper to read timestamp from reports
2568 - pps: fixup sporadic missing counters
2569 - intel/ds: drop timestamp correlation code
2570 - intel/perf: track end timestamp of queries
2571 - intel/ds: drop unused constructors
2572 - intel/ds: isolate intel/perf from the pps-producer
2573 - intel/pps: tweak intel config some more
2574 - intel/ds: remove verbose messages
2575 - intel: move timestamp scaling helper to intel/perf
2576 - anv: fix execbuf syncobjs/syncobj_values array leak
2577 - util/u_trace: refcount payloads
2578 - nir/opt_deref: don't try to cast empty structures
2579 - util/u_trace: add end_of_pipe property to tracepoints
2580 - util/u_trace/perfetto: add new env variable to enable perfetto
2581 - intel/nir: preserve access value when duping intrinsic
2582 - nir/lower_io: include the variable access in the lowered intrinsic
2583 - nir/print: printout ACCESS_STREAM_CACHE_POLICY
2584 - nir: add a new access flag to allow access in helper invocations
2585 - nir: add intrinsics for ray queries
2586 - nir: track variables representing ray queries
2587 - nir: add a ray query optimization pass
2588 - spirv: handle ray query intrinsics
2589 - intel/debug: reclaim 7 unused bits from classic driver
2590 - genxml: fix compilation with P/I defines
2591 - genxml: protect _length defines in genX_bits.h
2592 - docs/envvars: update after INTEL_DEBUG cleanup
2593 - util/u_vector: prevent C++ warning on cast from void* to something else
2594 - vulkan: fix missing handling of WSI memory signal
2595 - anv,wsi: simplify WSI synchronization
2596 - gitlab-ci: disable radv-fossils
2597 - intel/devinfo: adjust subslice array size
2598 - util/u_trace: protect against reentrant calls
2599 - anv: don't leave anv_batch fields undefined
2600 - anv: limit compiler valid color outputs using NIR variables
2601 - intel/dev: fixup chv workaround
2602 - ci: disable vs2019 windows build
2603 - intel/devinfo: printout pixel pipes in info printout
2604 - intel/devinfo: printout devinfo struct size
2605 - intel/devinfo: add a helper to check for slice availability
2606 - intel/devinfo: drop num_eus_per_subslice field
2607 - anv: fix perf queries
2608 - intel/dev: extract slice/subslice total computation
2609 - intel/devinfo: split out l3/pixelpipes counting
2610 - intel/devinfo: deal with i915 topology query change
2611 - intel/fs: disable VRS when omask is written
2612 - blorp: add description & helpers to printout ops
2613 - isl: add helpers to printout ops
2614 - anv: expose a couple of emit helper to build utrace buffer copies
2615 - intel/dev,perf: Use a single timescale function
2616 - intel/blorp: add measure_end entry point
2617 - anv: implement u_trace support
2618 - intel/ds: reuse intel_ioctl()
2619 - intel/ds: allow user to select metric set at start time
2620 - intel/ds: don't forget to reset upper dword timestamp read
2621 - intel/ds: use the right i915_drm.h include location
2622 - intel/ds: use a per GPU clock ID
2623 - util/u_process: protect entrypoints for c++
2624 - anv: add perfetto source
2625 - pps: enable anv source in example config file
2626 - tools/pps: limit intel cfg to 250ms of sampling
2627 - iris: utrace/perfetto support
2628 - intel/dev: fix ppipe_mask computation
2629 - anv: fix missing descriptor copy of bufferview/surfacestate content
2630 - genxml: reduce amount of generated code
2631 - anv: verify that the format supports multisampling
2632 - anv: switch a bunch of struct/enum to 1.3 versions
2633 - relnotes/features: updates for Vulkan 1.3
2634 - intel/ci: expected failure for 1.3 with older CTS
2635 - docs: start some documentation on Anv
2636 - docs/anv: list environment variables
2637 - docs/anv: add descriptor memory layout
2638 - anv: tidy long lines in descriptor code
2639 - docs: update INTEL_DEBUG environment variable documentation
2640 - intel/tracepoint: simplify tracepoint descriptions
2641 - util/u_trace: make mako conditional code easier to read
2642 - util/utrace: make generated code a tiny bit nicer to look at
2643 - compiler: add VARYING bit for primitive shading rate
2644 - genxml: gen12.5 changes for CPS
2645 - genxml: add new 3DSTATE_PS_EXTRA bit
2646 - intel/dev: details CPS feature support
2647 - isl: add support for coarse pixel control surfaces
2648 - isl: disable CPB surface compression
2649 - nir/builder: add ishl_imm helper
2650 - intel/compiler: add a new pass to lower shading rate into HW format
2651 - intel/compiler: add primitive rate output support
2652 - anv: force primitive shading rate write in last geometry stage
2653 - anv/pass: rely on precomputed dynamic rendering pass/subpass more
2654 - anv: Update VK_KHR_fragment_shading_rate for newer HW
2655 - intel/fs: don't set allow_sample_mask for CS intrinsics
2656 - intel/nir: fix shader call lowering
2657 - anv: fix conditional render for vkCmdDrawIndirectByteCountEXT
2658 - nir: fix lower_memcpy
2659 - anv/genxml/intel/fs: fix binding shader record entry
2660 - anv: fix fast clear type value with external images
2661 - intel/fs: fix total_scratch computation
2665 - frontends/va: Return error in vaRenderPicture if decoder is NULL
2667 Louis-Francis Ratté-Boulianne (7):
2669 - microsoft/compiler: add support for load_layer_id
2670 - microsoft/compiler: Add subpass input types
2671 - microsoft/compiler: Use SRVs for read-only images
2672 - microsoft/spirv_to_dxil: check for variables r/w access
2673 - microsoft/spirv_to_dxil: lower input attachments
2674 - microsoft/spirv_to_dxil: Lower push constant loads to UBO loads
2675 - microsoft/spirv_to_dxil: Allow passing a vulkan -> d3d12 binding mapping table
2679 - etnaviv: fix alpha blend with dither on older GPUs
2680 - etnaviv: initialize vertex attributes on context reset
2681 - egl/wayland: break double/tripple buffering feedback loops
2682 - etnaviv: drm: fix size limit in etna_cmd_stream_realloc
2683 - etnaviv: drm: properly handle reviving BOs via a lookup
2684 - etnaviv: drm: export BO idle check function
2685 - etnaviv: drm: rename _etna_bo_del
2686 - etnaviv: drm: defer destruction of softpin BOs
2687 - egl/dri2: remove superfluous flush when changing the context
2688 - egl/dri2: short-circuit dri2_make_current when possible
2689 - etnaviv: draw: only mark resources as read/written when the state changed
2693 - nouveau/nir: Use natural alignment for scalars
2694 - nouveau/nir: Lower 64-bit phis
2698 - panvk: Add check for null fence
2699 - panvk: Fix pointer corruption in panvk_add_wait_event_syncobjs
2703 - egl: set TSD as NULL after deinit
2707 - gbm: add missing R16 case in gbm_bo_get_bpp
2708 - gbm: add GBM_FORMAT_GR1616 and RG1616
2712 - iris: fix scratch address patching for TESS_EVAL stage
2713 - intel: fix INTEL_DEBUG environment variable on 32-bit systems
2714 - intel/decoder: Dump Task/Mesh shaders
2715 - spirv: handle SpvOpMemberName
2716 - intel/compiler: extract brw_nir_load_global_const out of rt code
2717 - intel/compiler: Get mesh_global_addr from the Inline Parameter for Task/Mesh
2718 - intel/compiler: Load draw_id from XP0 in Task/Mesh shaders
2719 - nir: limit lower_clip_cull_distance_arrays input to traditional stages
2720 - nir/print: simplify printing of IO semantics
2721 - nir/print: expand printing of io semantics.gs_streams
2722 - nir/print: compact printing of intrinsic indices
2723 - nir/print: move print_load_const_instr up
2724 - nir/print: group hex and float vectors together
2725 - nir/print: print const value near each use of const ssa variable
2726 - intel/compiler: disable workaround not applicable to gfx >= 11
2727 - spirv: handle ViewportMaskNV builtin/cap from SPV_NV_mesh_shader
2728 - compiler: add new MESH_VIEW_COUNT/MESH_VIEW_INDICES system values
2729 - spirv: add MeshViewCountNV/MeshViewIndidcesNV builtins from SPV_NV_mesh_shader
2730 - nir: add load_mesh_view_count and load_mesh_view_indices intrinsics
2731 - spirv: handle multiview bits of SPV_NV_mesh_shader
2732 - nir: remove invalid assert affecting per-view variables
2733 - spirv: mark [Clip|Cull]DistancePerViewNV variables as compact
2734 - nir: handle per-view clip/cull distances
2735 - freedreno/rnn: normalize line endings in rules-ng.xsd
2736 - microsoft/compiler: normalize line endings
2737 - ci/windows: normalize line endings
2738 - radv/ci: add line endings exception for files generated with wine
2739 - Add new rules to .gitattributes
2740 - intel/compiler: handle gl_[Clip|Cull]Distance in mesh shaders
2741 - intel/compiler: handle gl_[Clip|Cull]Distance from mesh in fragment shaders
2742 - intel/compiler: Use Task/Mesh InlineData for the first few push constants
2743 - anv: Enable conditional rendering in vkCmdDrawMeshTasksNV
2744 - anv: Add support for non-zero firstTask in vkCmdDrawMeshTasksNV
2745 - anv: Add support for UBOs, SSBOs and push constants in Mesh pipeline
2746 - anv: Implement indirect dispatch for Mesh pipeline
2747 - anv: tell the hardware about gl_[Clip|Cull]Distance in mesh shaders
2748 - anv: include ClipDistance array in mesh shader per-vertex output
2749 - anv: Put first few push constants directly into Task/Mesh InlineData
2750 - intel/compiler: fix array & struct IO lowering in mesh shaders
2751 - anv: don't set color state when input state was requested
2752 - intel/compiler: ignore per-primitive attrs when calculating flat input mask
2756 - gallium/util: add some extern "C" guards
2757 - radeonsi: si_state_shaders.c -> cpp
2758 - radeonsi: split si_shader_key into ps and ge parts to minimize memcmp overhead
2759 - radeonsi: don't memcmp inlined uniform values if uniform inlining is disabled
2760 - radeonsi: don't pass NULL into si_get_nir_shader
2761 - radeonsi: replace the GS prolog with a monolithic shader variant
2762 - radeonsi: enable shader culling on Navi1x consumer SKUs as well
2763 - ac,radeonsi: print a lowercase codename in the renderer string
2764 - radeonsi: reorder and don't print patch level DRM version in the renderer string
2765 - ac/llvm: use fmac instead of mul+sub in face culling
2766 - ac/llvm: add helper ac_build_is_inf_or_nan
2767 - ac/llvm: accept primitives whose face culling determinant is Inf or NaN
2768 - gallium,vbo: add PIPE_BIND_VERTEX_STATE for display lists
2769 - gallium/u_threaded: implement pipelined partial buffer uploads using CPU storage
2770 - mesa: fix crashes in the no_error path of glUniform
2771 - radeonsi: add SI_MAX_VRAM_MAP_SIZE definition
2772 - radeonsi: add an option to use CPU storage uploads for threaded context
2773 - radeonsi: change bind_history to track usage in each shader stage
2774 - radeonsi: rebind a buffer only in shader stages where it's been bound
2775 - radeonsi: don't sync PS or CS before (clear|copy)_buffer based on bind history
2776 - radeonsi: don't update bind_history for internal buffer clears and copies
2777 - radeonsi: don't sync before clear_buffer and copy_buffer if the buffer is idle
2778 - radeonsi: properly destroy buffers on failure
2779 - winsys/amdgpu: remove force_chaining parameter from cs_check_space
2780 - winsys/amdgpu: set max_ib_size and max_check_space_size later in cs_check_space
2781 - radeonsi: don't set inline_uniforms for viewperf because it's enabled by default
2782 - amd/addrlib: change how the license is formatted to match internal tree
2783 - amd/addrlib: cosmetic addrlib update
2784 - mesa: discard draws with count=0 to decrease overhead
2785 - st/mesa: don't crash when draw indirect buffer has no storage
2786 - mesa: remove USAGE_ELEMENT_ARRAY_BUFFER because it's unused and adding overhead
2787 - mesa: move setting USAGE_PIXEL_PACK_BUFFER out of BindBuffer to reduce overhead
2788 - mesa: remove redundant flagging USAGE_ARRAY_BUFFER
2789 - mesa: add a no_error path to _mesa_handle_bind_buffer_gen
2790 - glthread: don't execute display lists if they have no effect
2791 - glthread: don't sync for glIsEnabled with a few enums
2792 - glthread: add an option to make glCheckFramebufferStatus a no-op
2793 - glthread: add a trivial thread-safe way to skip display list execution
2794 - radeonsi: enable shader culling for indirect draws
2795 - radeonsi: remove unused parameters in si_emit_draw_packets
2796 - gallium/radeon: change the BO priority definitions to bits
2797 - gallium/radeon: remove/merge some BO priorities and remove holes
2798 - gallium/radeon: remove unused RADEON_DEPENDENCY_START_FENCE
2799 - gallium/radeon: merge BO read/write usage flags with priority flags
2800 - winsys/amdgpu: simplify parameter passing and derefs in cs_add_buffer
2801 - winsys/amdgpu: remove an amdgpu_cs dereference from amdgpu_cs_add_buffer
2802 - winsys/amdgpu: don't clear RADEON_USAGE_SYNCHRONIZED for last_added_bo_usage
2803 - winsys/amdgpu: increase the BO hash list size
2804 - winsys/amdgpu: don't use ip_instance and ring fields of fence and IB structures
2805 - winsys/amdgpu: move BO fence array updates to the CS thread
2806 - winsys/amdgpu: optimize looping inefficiencies in add_bo_fence_dependencies
2807 - radeonsi: don't invoke si_decompress_depth if textures are not dirty at binding
2808 - mesa: fix locking when destroying/overwriting/adding display lists
2809 - mesa: remove display list OPCODE_NOP
2810 - mesa: remove PADDING_64BIT by adding the dlist header into vbo_save_vertex_list
2811 - vbo: return a GL error earlier in vbo_save_playback_vertex_list_gallium
2812 - vbo: use int16_t for vbo_save_vertex_list::gallium::private_refcount
2813 - vbo: restructure vbo_save_vertex_list to get more cache hits
2814 - mesa: use alloca in search_resource_hash
2815 - glsl: add gl_resource_name to precompute "name" properties later
2816 - mesa: don't compute the same strlen up to 3x in _mesa_program_resource_find_name
2817 - mesa: precompute strlen in gl_resource_name::length and use it
2818 - mesa: rename locals in _mesa_program_resource_find_name for clarity
2819 - mesa: preparse [ and [0] in gl_resource_name and use it in shader_query.cpp
2820 - mesa: handle hash collisions in program resource lookups (e.g. uniforms)
2821 - mesa: add separate hash tables for each GLSL resource type
2822 - mesa: skip strlen when hashing strings for ProgramResourceHash
2823 - radeonsi: print the border color error message only once
2824 - util: add a util_bitcount variant that selects POPCNT through C++ template arg
2825 - st/mesa: change st_atom_array.c to cpp
2826 - st/mesa: use POPCNT in st_update_array if the CPU supports it
2827 - mesa: change gl_vertex_array_object::NewArrays to bool
2828 - mesa: add NewVertexBuffers/NewVertexElements flags to indicate state changes
2829 - cso: add missing parameters into cso_set_vertex_buffers
2830 - st/mesa: don't update vertex elements when GL doesn't change them
2831 - driconf: set vblank_mode=0 for viewperf2020
2832 - gallium/util: fix util_can_blit_via_copy_region with unbound render condition
2833 - gallium/u_blitter: disable sample shading for all blits
2834 - gallium/u_blitter: do MSAA copies in 1 pass using sample shading
2835 - gallium/u_blitter: work around broken sample shading in llvmpipe and zink
2836 - radeonsi: fix 2 issues with depth_cleared_level_mask
2837 - radeonsi: add a faster clear path for glClearTexImage
2838 - radeonsi: rename stencil_cleared_level_mask -> stencil_cleared_level_mask_once
2839 - radeonsi: allow and finish TC-compatible MSAA HTILE
2840 - radeonsi: fix a typo preventing a fast depth-stencil clear
2841 - radeonsi: increase tc_max_cpu_storage_size
2842 - vbo: utilize structure padding to optimize indirection cold->prims[0].begin
2843 - driconf: disallow 10-bit pbuffers for viewperf2020/maya due to X errors
2844 - gallium: rename PIPE_CAPF_MAX_POINT_WIDTH -> MAX_POINT_SIZE
2845 - gallium: add missing point and line CAPs
2846 - radeonsi: set correct point and line limits
2847 - st/mesa: use new point and line CAPs
2848 - nir: add new SSA instruction scheduler grouping loads into indirection groups
2849 - radeonsi: enable nir_group_loads for better performance
2850 - radeonsi: fix shader culling with integer pixel centers
2851 - radeonsi: fix view culling for wide lines
2852 - radeonsi: use ac_build_load_to_sgpr in gfx10_emit_ngg_culling_epilogue
2853 - radeonsi: make si_get_small_prim_cull_info static
2854 - radeonsi: set PERPENDICULAR_ENDCAP_ENA for wide AA lines
2855 - radeonsi: set EXTRA_DX_DY_PRECISION for lines where it's supported
2856 - radeonsi: add si_state_rasterizer::ngg_cull_flags_lines and rename the others
2857 - ac,radeonsi: cull small lines in the shader using the diamond exit rule
2858 - radeonsi: unify GFX9_VSGS_NUM_USER_SGPR and GFX9_TESGS_NUM_USER_SGPR
2859 - radeonsi: add dcc_msaa option to enable DCC for MSAA
2860 - radeonsi: improve memory instruction tracking
2861 - radeonsi: remove an incorrect comment at lds_byte0_accept_flag
2862 - radeonsi: make si_llvm_emit_clipvertex non-static
2863 - radeonsi: unify how ngg_cull_flags are set
2864 - radeonsi: cull against clip planes, clipvertex, clip/cull distances in shader
2865 - radeonsi: inline declare_vs_specific_input_sgprs
2866 - radeonsi: don't use GS SGPR6 for the small prim cull info
2867 - glx: add a workaround to glXDestroyWindow for Viewperf2020/Sw
2868 - radeonsi: separate culling code from VS/TES (to be reused by GS)
2869 - radeonsi: restructure code that declares merged VS-GS and TES-GS SGPRs
2870 - radeonsi: add is_gs parameter into si_vs_needs_prolog
2871 - radeonsi: simplify si_get_vs_key_outputs for GS
2872 - radeonsi: don't use ctx.stage outside of si_llvm_translate_nir
2873 - radeonsi: implement shader culling in GS
2874 - radeonsi: deduplicate min_esverts code in gfx10_ngg_calculate_subgroup_info
2875 - mesa: don't add attenuation constants if ffvp doesn't use them
2876 - mesa: add allow_glsl_compat_shaders for shader-db
2877 - ac/gpu_info: don't fail on amdgpu_query_video_caps_info failures
2878 - ac/surface: allow gfx6-8 to enter the gfx9 DCC codepath for SI_FORCE_FAMILY
2879 - mesa: add a more straightforward callback for replacing shaders
2880 - driconf: enable glthread for all Unigine benchmarks
2881 - driconf: enable glthread for Minecraft
2882 - driconf: enable glthread for Basemark GPU
2883 - radeonsi: don't print uninitialized inlined_uniform_values
2884 - radeonsi: add wave32 flag into prolog/epilog keys
2885 - radeonsi: add si_shader::wave_size because it will vary
2886 - radeonsi: use si_shader::wave_size
2887 - radeonsi: don't use si_get_wave_size in si_get_ir_cache_key
2888 - radeonsi: clean up compute_wave_size use in si_compute_blit.c
2889 - radeonsi: propagate si_shader::wave_size to VGT_SHADER_STAGES
2890 - radeonsi: don't use compute_wave_size directly
2891 - radeonsi: centralize wave size computation in si_get_shader_wave_size
2892 - radeonsi: rename si_get_shader_wave_size and make it non-inline
2893 - mesa: remove SourceChecksum from shader structures
2894 - mesa: rename gl_shader::sha1 to disk_cache_sha1
2895 - mesa: add shader source SHA1s that are propagated up to glCompileShader
2896 - mesa: add gl_linked_shader::linked_source_sha1
2897 - mesa: don't compute the same SHA1 twice in glShaderSource
2898 - util: add SHA1 printing and comparison functions
2899 - nir: add shader_info::source_sha1, its initialization and printing
2900 - radeonsi: print source_sha1 as part of shader dumps
2901 - radeonsi: print the shader stage for shader-db dumps
2902 - glthread: don't sync for more glGetIntegerv enums for glretrace
2903 - mesa: use simple_mtx_t for TexMutex (v2)
2904 - nir: handle more intrinsics in divergence analysis
2905 - nir: disable a NIR test due to undebuggable & locally unreproducible CI failures
2906 - nir: serialize divergent fields
2907 - nir: add nir_has_divergent_loop function
2908 - glsl: fix setting compiled_source_sha1 without a shader cache
2909 - radeonsi: add Wave32 heuristics and shader profiles
2910 - radeonsi: print more stats for shader-db
2911 - radeonsi: add shader profiles that disable binning
2912 - radeonsi: unroll loops of up to 128 iterations
2913 - mesa: include less stuff in dlist.c
2914 - mesa: inline vbo_initialize_save_dispatch and rename the functions
2915 - mesa: inline _mesa_install_arrayelt_vtxfmt
2916 - mesa: inline _mesa_install_eval_vtxfmt
2917 - mesa: inline _mesa_install_dlist_vtxfmt
2918 - mesa: move _mesa_initialize_vbo_vtxfmt calls to a common place and inline
2919 - mesa: inline _vbo_install_exec_vtxfmt
2920 - mesa: move the ES2 check from vbo_init_tmp.h to install_vtxfmt
2921 - glapi: rename exec="dynamic" to exec "vtxfmt" to make it self-explanatory
2922 - mesa: don't set CallList* redundantly in _mesa_initialize_save_table
2923 - mesa: rename dlist functions to match dispatch function names
2924 - glapi: rename gl_genexec.py to api_exec_init.py, api_exec.c to api_exec_init.c
2925 - glapi: move apiexec API condition determination to common code
2926 - glapi: replace dispatch.h inline functions with macros for faster compilation
2927 - mesa: add EXT suffix to VertexAttribI*EXT to match glapi name
2928 - vbo: rename vertex functions to match GL dispatch names
2929 - vbo: rename ES vertex functions to match GL dispatch names
2930 - glapi: move reusable glapi printing code to apiexec.py
2931 - glapi: autogenerate _mesa_initialize_save_table with python
2932 - glapi: autogenerate api_save.h with save_* function declarations
2933 - mesa: remove api_exec.h and move its contents into context.h
2934 - glapi: autogenerate all _mesa_* forward declarations in api_exec_decl.h
2935 - glapi: autogenerate install_vtxfmt with python
2936 - mesa: remove GLvertexformat
2937 - vbo: expose all exec entrypoints for glthread and match api_exec_decl.h names
2938 - glthread: add nop dispatch
2939 - glthread: set marshal functions in dispatch only if they exist in the API
2940 - glthread: inline _mesa_glthread_restore_dispatch and merge disable & destroy
2941 - glthread: fix restoring the dispatch in destroy when the context is not current
2942 - glthread: disable glthread if the context is lost
2943 - mesa: use ctx->GLThread.enabled now that it's correct
2944 - mesa: rename _ae_ArrayElement -> _mesa_ArrayElement to match glapi
2945 - mesa: remove COPY_DISPATCH code that doesn't do anything
2946 - glapi: add missing no_error settings for implemented functions
2947 - mesa,vbo: make ES wrapper functions static
2948 - mesa: remove all GL func forward declarations because they are autogenerated
2949 - mesa: use nop dispatch for ColorTable/Convolution/Histogram
2950 - radeonsi: add a debug option that disables DCC for all exported buffers
2951 - driconf: enable glthread for Minecraft-FTB, Stellaris, Battletech
2952 - ac/gpu_info: set cu_mask correctly for Arcturus
2953 - ac/gpu_info: add AMD_CU_MASK environment variable to set CU_EN
2954 - radeonsi: set COMPUTE_DESTINATION_EN_SEn to spi_cu_en
2955 - radeonsi: program COMPUTE_STATIC_THREAD_MGMT_SE4..7 on Arcturus
2956 - radeonsi: apply spi_cu_en to CU_EN
2957 - radv: set COMPUTE_DESTINATION_EN_SEn to spi_cu_en
2958 - radv: apply spi_cu_en to CU_EN
2959 - radeonsi: pack si_pm4_state
2960 - radeonsi: replace SI_PM4_MAX_DW with a max_dw field
2961 - radeonsi: decrease the size of si_pm4_state::pm4 except for cs_preamble_state
2962 - amd/registers: work around an assertion in parse_kernel_headers.py
2963 - ac/llvm: add vindex into ac_build_buffer_store_dword
2964 - ac/llvm: remove the num_channels parameter from ac_build_buffer_store_dword
2965 - radeonsi: modifiers can't disable DCC
2966 - radeonsi: make get_thread_id_in_tg non-static
2967 - radeonsi: clean up si_export_mrt_color
2968 - radeonsi: set done=1 for PS exports at the end of si_llvm_build_ps_epilog
2969 - radeonsi: remove unnecessary code that was used to find the last export
2970 - radeonsi: export mrtz before color exports
2971 - radeonsi: simplify compacted_mrt_index in si_export_mrt_color
2972 - radeonsi: reorder slots for internal buffers, reuse a slot for GS_QUERY_BUF
2973 - radeonsi: don't bind the ESGS ring twice, handle the difference in the shader
2974 - radeonsi: remove unused si_shader::prolog2
2975 - radeonsi: pass sample_coverage VGPR index to the PS prolog instead of guessing
2976 - radeonsi: move smoothing to the main shader part to remove 1 live VGPR
2977 - radeonsi: do opt_large_constants & lower_indirect_derefs after uniform inlining
2978 - radeonsi: use nir->scratch_size instead of ac_count_scratch_private_memory
2979 - radeonsi: change si_shader_output_values::vertex_stream to a bitmask
2980 - radeonsi: move si_nir_scan_shader into si_shader_info.c
2981 - radeonsi: add into the disk cache key whether cached shaders contain LLVM IR
2982 - radeonsi: move the GS copy shader into shader variants
2983 - ac: move ac_exp_param.h to ac_nir.h
2984 - ac/nir: move ac_are_tessfactors_def_in_all_invocs into radeonsi
2985 - radeonsi: print all streamout info
2986 - radeonsi: print the number of param exports for shader-db
2987 - ac/surface: allow displayable DCC with any resolution (e.g. 8K)
2988 - radeonsi: rename uses_vmem_* flags
2989 - radeonsi: apply fbfetch/indirect_descriptor to uses_vmem_load_other earlier
2990 - radeonsi: determine MEM_ORDERED after generating a shader variant
2991 - winsys/radeon: fix a hang due to introducing spi_cu_en
2992 - amd: add a workaround for an SQ perf counter bug
2993 - ac/surface: add more elements to meta equations because HTILE can use them
2994 - radeonsi: fix register shadowing after the pm4 state size was decreased
2995 - radeonsi: fix an assertion failure with register shadowing
2999 - util/cpu_detect: Add flag for IBM Z (s390x)
3000 - llvmpipe: Use lp_build_round_arch on IBM Z (s390x)
3001 - util/cpu_detect, gallium: use cpu_family CPU_S390X instead of separate flag
3005 - Fixed you're to your
3009 - radv/ci: mark the dEQP fails related to a missing VKCTS 1.3 as expected
3013 - util/format: Add PIPE_FORMAT_Y8_UNORM as an "other" layout format
3014 - tu: Expose required VK_FORMAT_FEATURE bits for planar YUV formats
3015 - ir3: Add support for (dis)assembling flat.b
3016 - freedreno/ir3: Add infrastructure for flat.b
3017 - freedreno/ir3: Use flat.b to load flat varyings on a6xx
3018 - freedreno/ir3: Use immediate for flat.b's src1
3019 - intel/genxml: capitalize decoder mode select properly
3023 - aux/trace: print enum names instead of integer values in gallium traces
3024 - aux/trace: implement missing trace calls
3025 - aux/trace: cosmetic cleanup
3029 - android: define cpp_rtti=false because libLLVM is built w/o RTTI
3030 - Revert "android: define cpp_rtti=false because libLLVM is built w/o RTTI"
3031 - android: define cpp_rtti=false because libLLVM is built w/o RTTI (v2)
3035 - v3dv: drop unused variable on handle_set_event_cpu_job
3036 - v3dv: wrap wait semaphores info in v3dv_submit_info_semaphores
3037 - v3dv: store wait semaphores in event_wait_cpu_job_info
3038 - drm-uapi/v3d: extend interface for multiple semaphores support
3039 - v3dv: check multiple semaphores capability
3040 - v3dv: enable multiple semaphores on cl submission
3041 - v3dv: enable multiple semaphores for tfu job
3042 - v3dv: enable multiple semaphores for csd job
3043 - v3dv: enable GPU jobs to signal multiple semaphores
3044 - v3dv: track submitted jobs by GPU queue type
3045 - v3dv: handle wait semaphores in the first job by queue
3046 - v3dv: process signal semaphores in the very last job
3047 - v3dv: signal fence when all submitted jobs complete execution
3048 - v3dv: move sems_info from event_wait job to wait_thread info
3052 - microsoft/spirv_to_dxil: turn sysvals into input varyings
3053 - microsoft/compiler: Use memcpy instead of a union to write dxil_features
3057 - Revert "wsi/x11: Avoid a class of deadlocks in the WSI queue thread"
3061 - vulkan: fix uninitialized variables
3062 - lavapipe: fix unused variable
3063 - llvmpipe: Fix Wpointer-to-int-cast
3064 - meson: check -mtls if has_exe_wrapper
3065 - meson: correctly detect linker arguments
3066 - zink: fix -Warray-bounds warning
3067 - zink: fix unused variable warning
3068 - vulkan/wsi: drop unused wsi_create_win32_image
3072 - nir/print: print bindless info as applicable
3073 - aux/pb: add a tolerance for reclaim failure
3074 - virgl: remove unused pipebuffer include
3075 - aux/pb: more correctly check number of reclaims
3076 - zink: use static array for detecting VK_TIME_DOMAIN_DEVICE_EXT
3077 - zink: add a read barrier for indirect dispatch
3078 - zink: fully zero surface creation struct
3079 - zink: clear descriptor refs on buffer replacement
3080 - zink: assert compute descriptor key is valid before hashing it
3081 - zink: don't update lazy descriptor states in hybrid mode
3082 - zink: move push descriptor updating into lazy-only codepath
3083 - zink: add an early return for zink_descriptors_update_lazy_masked()
3084 - zink: move last of lazy descriptor state updating back to lazy-only code
3085 - lavapipe: stop reading renderpass during pipeline creation
3086 - lavapipe: remove some unused struct members
3087 - lavapipe: use framebuffer attachment_count member instead of renderpass
3088 - lavapipe: remove lvp_subpass_attachment::in_render_loop
3089 - lavapipe: remove lvp_subpass_attachment::layout
3090 - lavapipe: add attachment index to lvp_render_pass_attachment
3091 - lavapipe: remove lvp_subpass::max_sample_count
3092 - lavapipe: remove lvp_subpass_attachment and use lvp_render_pass_attachment refs
3093 - lavapipe: remove last VK_ATTACHMENT_UNUSED check
3094 - lavapipe: store subpass directly to rendering_state
3095 - lavapipe: simplify some attachment derefs
3096 - lavapipe: remove lvp_subpass::has_color_att
3097 - lavapipe: pull layer count from render state during resolve
3098 - lavapipe: clamp attachment clear rect sizes
3099 - zink: don't ralloc zink_resource structs
3100 - zink: align pipe_resource and sampler_view allocations to cachelines
3101 - zink: use ctx params for program ref/destroy functions
3102 - zink: fix descriptor interface param for program_deinit
3103 - zink: remove descriptor pools from hash table on deletion
3104 - zink: unref descriptor pools in hybrid mode when they explode
3105 - zink: always invalidate descriptor sets on pool free
3106 - zink: don't clear descriptor pool cache on context destroy
3107 - zink: stop leaking descriptor pool references
3108 - zink: rescue surfaces/bufferviews for cache hits during deletion
3109 - zink: rename zink_descriptor_layout_key::num_descriptors -> num_bindings
3110 - zink: eliminate a hole in zink_descriptor_layout_key
3111 - zink: reduce hashed region of zink_descriptor_layout_key
3112 - zink: split out descriptor pool sizing into separate struct
3113 - zink: detect prim type more accurately for tess/gs lines
3114 - zink: don't break early when applying fb clears
3115 - zink: only reset zink_resource::so_valid on buffer rebind
3116 - zink: don't check rebind count outside of buffer/image rebind function
3117 - zink: add notes about binding points which aren't counted in util funcs
3118 - zink: stop exporting PIPE_SHADER_CAP_FP16_DERIVATIVES
3119 - zink: don't add dynamic vertex pipeline states if no attribs are used
3120 - zink: fix gl_SampleMaskIn spirv generation
3121 - zink: more accurately update samplemask for fs shader keys
3122 - nir/lower_samplers_as_deref: rewrite more image intrinsics
3123 - nir/lower_io_to_scalar: add support for bo and shared io
3124 - zink: run optimize_nir() only once during compile
3125 - zink: move bo load offset adjustment to compiler passes
3126 - zink: run lower_io_to_scalar before rewriting bo access
3127 - zink: move all 64-32bit shader load rewriting to nir pass
3128 - zink: move ssbo store offset adjustment to compiler passes
3129 - zink: move shared intrinsic offset adjustments to compiler passes
3130 - zink: move all 64-32bit shader store rewriting to nir pass
3131 - zink: add more glsl base types to get_glsl_basetype()
3132 - zink: move all shader bo/sharedmem access to compiler passes
3133 - zink: add better handling for CUBE_COMPATIBLE bit
3134 - zink: cache bo SpvId array types
3135 - zink: use align64 for allocation sizes
3136 - zink: set aspectMask for renderpass2 VkAttachmentReference2 structs
3137 - zink: always use explicit lod for texture() when legal in non-fragment stages
3138 - zink: be more permissive for injecting LOD into texture() instructions
3139 - zink: inject LOD for sampler version of OpImageQuerySize
3140 - zink: flag renderpass change when toggling fbfetch
3141 - zink: be more spec-compliant for unnormalizedCoordinates samplers
3142 - zink: don't clamp cube array surfacess to cubes
3143 - zink: don't clamp 2D_ARRAY surfaces to 2D
3144 - zink: error when trying to allocate a bo larger than heap size
3145 - zink: clamp max buffer sizes to smallest buffer heap size
3146 - zink: explicitly enable VK_EXT_shader_subgroup_ballot
3147 - zink: add more int/float types to cast switching in ntv
3148 - zink: force float dest types on some alu results
3149 - zink: stop double printing validation messages
3150 - zink: add SpvCapabilityStorageImageMultisample for multisampled storage images
3151 - zink: reject all storage multisampling if the feature is unsupported
3152 - gallium: add PIPE_CAP_VERTEX_ATTRIB_ELEMENT_ALIGNED_ONLY
3153 - zink: set PIPE_CAP_VERTEX_ATTRIB_ELEMENT_ALIGNED_ONLY
3154 - zink: add queue locking
3155 - build: add sha1_h to llvmpipe build
3156 - mesa/st: rework psiz lowering
3157 - mesa/st: lower psiz for shader precompile
3158 - anv: disable debug logging spam
3159 - zink: set fbfetch state on lazy batch data when enabling it
3160 - zink: always use lazy (non-push) updating for fbfetch descriptors
3161 - vk: update headers for 1.2.197
3162 - lavapipe: fix cmd queuing for dynamic render
3163 - lavapipe: VK_KHR_dynamic_rendering
3164 - features: add dynamic render for lavapipe
3165 - zink: do a better job conserving locations for packed xfb outputs
3166 - zink: clamp PIPE_SHADER_CAP_MAX_INPUTS for xfb
3167 - zink: remove lazy ci job
3168 - zink: add khr46 to ci
3169 - aux/primconvert: handle singular incomplete restarts
3170 - zink: rework cached fbfetch descriptor fallback
3171 - lavapipe: add some asserts for descriptor dynamic offsets
3172 - gallivm: handle TGSI SampleId sysval
3173 - Revert "gallium/u_blitter: work around broken sample shading in llvmpipe and zink"
3174 - aux/trace: fix PIPE_QUERY_PIPELINE_STATISTICS_SINGLE tracing
3175 - aux/trace: support pipe_context::get_query_result_resource
3176 - aux/trace: trace pipe_screen::is_format_supported better
3177 - aux/trace: fix vertex state tracing
3178 - aux/trace: add pipe_context::render_condition_mem
3179 - zink: set new point/line caps
3180 - zink: update radv ci passes
3181 - zink: flatten out draw templates a bit
3182 - zink: declare int/float size caps inline with type usage
3183 - zink: simplify 64bit vertex attrib lowering
3184 - zink: add another compiler pass to convert 64bit vertex attribs
3185 - zink: add 1DShadow sampler handling for drivers (radv) that don't support it
3186 - zink: radv ci updates for 1dshadow stuff
3187 - zink: implement cs uniform inlining
3188 - zink: always inline uniforms when running on a cpu driver
3189 - zink: ci updates
3190 - zink: be more consistent about applying module hash for gfx pipeline
3191 - zink: update gfx pipeline shader module pointer even if the program is unchanged
3192 - aux/primconvert: break out primconvert internals into util function
3193 - aux/primconvert: support pipe_context::draw_vertex_state
3194 - zink: stop running discard_if in generated tcs
3195 - zink: always add VK_IMAGE_CREATE_2D_ARRAY_COMPATIBLE_BIT for 3D images
3196 - move util/indices to core util
3197 - mesa: convert unsupported primtypes during display list compilation
3198 - zink: block suballocator caching for swapchain/dmabuf images
3199 - zink: set suballocator bo size to aligned allocation size
3200 - zink: implement pipe_context::draw_vertex_state
3201 - zink: stop using VK_IMAGE_LAYOUT_PREINITIALIZED for dmabuf
3202 - zink: always set matching resource export type for dmabuf creation
3203 - dri2: set dimensions on dmabuf import planes
3204 - zink: implement multiplanar modifier handling
3205 - zink: add error for bo allocation failure
3206 - zink: always use slab allocation placement for domains
3207 - zink: set zink_resource_object::host_visible based on actual bo placement
3208 - zink: demote BAR allocations to device-local on oom
3209 - zink: use IMMUTABLE for dummy xfb buffer
3210 - zink: fix memory availability reporting
3211 - zink: fail context creation more gracefully
3212 - zink: clamp to 500 max batch states on nvidia
3213 - gallium: add pipe_screen::is_compute_copy_faster hook
3214 - gallium: rename PIPE_CAP_PREFER_BLIT_BASED_TEXTURE_TRANSFER
3215 - mesa/st: make sampler_type_for_target public
3216 - mesa/st: make some pbo functions public
3217 - gallium: implement compute pbo download
3218 - aux/pb: add a new slab alloc function for reclaiming all bo objects
3219 - zink: use pb_slab_alloc_reclaimed(reclaim_all) for BAR heap sometimes
3220 - zink: enable PIPE_TEXTURE_TRANSFER_COMPUTE on non-cpu drivers
3221 - zink: explicitly init glsl
3222 - zink: add a compiler pass to scan for shader image use
3223 - zink: set batch state queue on creation
3224 - zink: be consistent about waiting on context queue on context destroy
3225 - mesa/vbo: be more comprehensive for degenerate primitive conversion in dlists
3226 - radv: fix xfb query copy param ordering
3227 - zink: add missing assert for 8bit vertex decompose
3228 - zink: add some wsi instance extensions
3229 - zink: always unset vertex shader variant key data when changing last vertex stage
3230 - zink: add extra synchronization for buffer descriptor binds
3231 - zink: use device-local heap for sparse backing allocations
3232 - zink: add a better threshold for clamping query pool resets on suspend
3233 - zink: always set number of timestamp results to 1 for internal qbo
3234 - zink: fix availability buffer sizing/copying for xfb queries
3235 - zink: skip readback of qbos with no results
3236 - Revert "zink: when performing an implicit reset, sync qbos"
3237 - zink: use even more accurate stride values for query result copies
3238 - zink: ci updates
3239 - zink: add flake
3240 - aux/trace: copy over stream_output_target_offset method from context
3241 - util/vbuf: fix buffer translation sizing
3242 - zink: remove SpvMemorySemanticsMakeVisibleMask from nir_intrinsic_memory_barrier
3243 - zink: add some nv ci results
3244 - aux/trace: add pipe_context::fence_server_signal tracing
3245 - zink: update gfx_pipeline_state.vertex_strides when necessary
3246 - zink: check EXT_image_drm_format_modifier for dmabuf support
3247 - Revert "zink: update gfx_pipeline_state.vertex_strides when necessary"
3248 - zink: add VK_KHR_external_semaphore_fd to device exts
3249 - zink: add VK_KHR_external_memory_capabilities to instance exts
3250 - zink: add driver/device uuid screen hooks
3251 - zink: implement GL semaphores
3252 - zink: implement external memory object resource handling
3253 - zink: enable EXT_external_objects pipe caps
3254 - util: add ptralloc
3255 - lavapipe: replace hard pointer calcs in dynamic render with ptralloc
3256 - lavapipe: replace hard pointer calcs in push descriptors with ptralloc
3257 - zink: update nv ci baseline
3258 - zink: add deqp ci baseline for nv
3259 - lavapipe: fix ptralloc typo
3260 - zink: add get_sparse_texture_virtual_page_size hook
3261 - zink: set up image create bits for sparse textures
3262 - zink: support sparse texture range commits
3263 - zink: allow sparse buffers to be suballocated
3264 - zink: stop allocating such massive staging regions for buffer maps
3265 - zink: fake sparse R9G9B9E5 support as needed
3266 - zink: implement sparse shader instructions in ntv
3267 - zink: always set actual_dest_type for ntv tex instruction emission
3268 - zink: handle residency return value from sparse texture instructions
3269 - zink: rename zink_so_info -> zink_shader_info
3270 - zink: always pass shader info to ntv
3271 - zink: add nir_intrinsic_image_deref_sparse_load to image scanning in compiler
3272 - zink: only allocate ntv residency info if it will be used
3273 - zink: emit sparse residency cap in ntv
3274 - zink: handle min_lod texture operands
3275 - zink: batch sparse texture binds
3276 - zink: handle sparse texture miptail commits
3277 - zink: enable ARB_sparse_texture pipe caps
3278 - zink: ARB_sparse_texture2
3279 - docs: add features/relnotes for zink sparse texture support
3280 - zink: fix non-modifer dmabuf usage
3281 - zink: link with vulkan utils
3282 - zink: hook up planar image format creation
3283 - vulkan/wsi: add VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT for swapchain image caps
3284 - lavapipe: remove unused struct member
3285 - zink: update nv fails
3286 - zink: update radv fails list
3287 - zink: never use SpvOpImageQuerySizeLod for texel buffers
3288 - zink: add anv (icl) fails
3289 - zink: handle bogus xfb draws
3290 - zink: reorder fbfetch flag-setting to avoid null deref
3291 - zink: return 256 for PIPE_CAP_MIN_MAP_BUFFER_ALIGNMENT
3292 - anv: silence wsi debug logging
3293 - zink: unify some context casts in zink_create_sampler_view
3294 - zink: emit same number of timeline signals as semaphore signals
3295 - zink: flag has_work when a GL semaphore is signalled
3296 - zink: allow resource creation without VK_FORMAT_FEATURE_COLOR_ATTACHMENT_BIT
3297 - zink: use VkImageViewUsageCreateInfo to remove attachment bits
3298 - zink: always create a separate VkBuffer for storage use
3299 - zink: use the storage buffer for bufferview creation when format allows
3300 - zink: flag all buffer resources with PIPE_BIND_SHADER_IMAGE
3301 - zink: simplify buffer case for zink_resource_object_init_storage()
3302 - zink: remove tmp buffer rebinds
3303 - aux/tc: add tc_buffer_write to replace pipe_buffer_write usage
3304 - zink: fix vertex buffer mask computation for null buffers
3305 - zink: replace qbo pipe_buffer_write usage with tc_buffer_write
3306 - zink: replace other pipe_buffer_write usage with pipe_buffer_write_nooverlap
3307 - zink: make pipe_buffer_write usage trigger compiler errors
3308 - zink: ci updates
3309 - zink: clamp tbo creation to maxTexelBufferElements
3310 - zink: add vertex shader pipeline bit for generated barrier construction
3311 - zink: print an error when the device is lost
3312 - zink: fix waiting on current batch id
3313 - zink: handle swizzled offset/count values for shader bitfield ops
3314 - zink: enable VK_EXT_shader_atomic_float
3315 - zink: add warning printf for drivers missing VK_EXT_shader_atomic_float
3316 - zink: cast image atomic op params/results based on image type
3317 - zink: use SpvScopeDevice over SpvScopeWorkgroup for atomic shader ops
3318 - aux/vbuf: use local var for modifying unaligned_vb_mask during update
3319 - aux/vbuf: move mask-clearing for vbuf updates after buffer scanning
3320 - aux/vbuf: add fastpath for skipping identical vbuf updates
3321 - mesa: stop truncating MESA_GLSL=dump
3322 - zink: ARB_sparse_texture_clamp
3323 - docs: update features/relnotes for zink sparse texture clamp
3324 - zink: disable PIPE_SHADER_CAP_FP16_CONST_BUFFERS
3325 - llvmpipe: disable PIPE_SHADER_CAP_FP16_CONST_BUFFERS
3326 - zink: add VK_BUFFER_USAGE_CONDITIONAL_RENDERING_BIT_EXT for query binds
3327 - zink: use scanout obj when returning resource param info
3328 - zink: fix PIPE_CAP_TGSI_BALLOT export conditional
3329 - zink: reject invalid draws
3330 - zink: min/max blit region in coverage functions
3331 - aux/draw: fix llvm tcs lane vec generation
3332 - lavapipe: use util_pack_color_union() for generating clear colors
3333 - zink: ci updates
3334 - lavapipe: ci updates
3335 - zink: always set VkPipelineMultisampleStateCreateInfo::pSampleMask
3336 - zink: always invalidate streamout counter buffer if not resuming
3337 - zink: ci updates
3338 - zink: fix cached descriptor set invalidation for array bindings
3339 - zink: free push descriptor pools on deinit
3340 - zink: don't free non-fbfetch dsl structs when switching to fbfetch
3341 - gallivm: avoid division by zero when computing cube face
3342 - zink: always update shader variants when rebinding a gfx program
3343 - zink: use a fence for pipeline cache update jobs
3344 - zink: wait on program cache fences before destroying programs
3345 - zink: fix descriptor cache pointer array allocation
3346 - zink: mark fbfetch push sets as non-cached
3347 - zink: stop leaking descriptor sets
3348 - zink: invalidate non-punted recycled descriptor sets that are not valid
3349 - zink: set shader key size to 0 for non-generated tcs
3353 - meson: Fix dri.pc dridriverdir
3357 - iris: Add missed tile flush flag
3358 - Revert "iris: add tile cache flush to iris_copy_region"
3359 - nir: Fix read depth for predecessors
3363 - iris: Convert some mod_info checks to asserts
3364 - iris: Disable the MC_CCS modifier with norbc
3365 - iris: Set DISABLE_AUX_BIT for AUX_USAGE_NONE modifiers
3366 - iris: Refactor the assignment to possible_usages
3367 - intel/isl: Allow creating non-Y-tiled ASTC surfaces
3368 - iris: Allow GPU-based uploads of ASTC textures
3369 - anv: Require transfer features for transfer usages
3370 - anv: Allow transfer-only linear ASTC images
3371 - anv: Drop code from get_blorp_surf_for_anv_buffer
3372 - intel/isl: Unify fmt checks in isl_surf_supports_ccs
3373 - iris: Enable CCS_E on 32-bpc float formats on TGL+
3374 - iris: Drop redundant iris_resource_disable_aux call
3375 - intel/isl: Restore CCS_E support for YUYV and UYVY
3376 - iris: Drop the YCRCB cases in finish_aux_import
3377 - intel/isl: Drop extra devinfo checks for CCS support
3378 - intel/isl: Require aux map for some 64K alignment
3379 - intel/blorp: Modify the SKL+ CCS resolve rectangle
3380 - intel/blorp: Modify get_fast_clear_rect for XeHP
3381 - iris: Change a param of iris_resource_init_aux_buf
3382 - iris: Use the aux BO and surf less during init
3383 - iris: Move some BO setup to iris_resource_init_aux_buf
3384 - iris: Simplify iris_get_aux_clear_color_state_size
3385 - iris: Don't allocate a clear color BO for some Z/S
3386 - iris: Drop row pitch param from iris_get_ccs_surf
3387 - iris: Don't allocate and initialize CCS on XeHP
3388 - iris: Don't assert a NULL aux BO during aux config
3389 - iris: Modify the comment about zeroing CCS
3390 - iris: Update the initial CCS state on XeHP
3391 - iris: Free the local cache bucket in bufmgr_destroy
3392 - iris: Replace bo->real.local with bo->real.heap
3393 - iris: Add and use flags_to_heap
3394 - iris: Use a num_buckets pointer in add_bucket
3395 - iris: Replace "local" with "heap" in bufmgr fn params
3396 - iris: Add and use BUCKET_ARRAY_SIZE
3397 - iris: Add and use bucket_info_for_heap
3398 - iris: Rework the DEVICE_LOCAL heap
3399 - iris: Disable the SMEM fallback for CCS on XeHP
3400 - iris: Use util packing fns in convert_clear_color
3401 - blorp: Drop multisampled code in blorp_can_hiz_clear_depth
3402 - intel/isl: Rework HiZ image align calculations
3403 - intel/isl: Update comment for the XeHP HiZ block
3404 - intel/isl: Use a new HiZ format on XeHP+
3405 - intel/isl: Require Y-tiling for depth on gfx4-5
3406 - intel/isl: Allow HiZ with Tile4/64 surfaces
3407 - intel/isl: Return false more in isl_surf_get_hiz_surf
3408 - anv,iris: Flush tile cache after color fast clears
3409 - anv,iris: Depth stall around color fast clears
3410 - intel: Rename the PSD bit in PIPE_CONTROL for XeHP
3411 - anv,iris: PSS Stall Sync around color fast clears
3412 - anv,iris: Flush HDC before color fast clears
3413 - anv: Don't fill lowered_storage_image_param on SKL+
3414 - intel/isl: Don't check pitch in isl_surf_get_mcs_surf
3415 - intel/isl: Strengthen MCS SINT format restriction
3416 - Revert "intel/isl: Don't reconfigure aux surfaces for MCS"
3417 - intel/gen125.xml: Increase Auxiliary Surface Pitch
3418 - intel/isl: Allow creating MCS in Tile4 memory
3419 - anv: Drop assert against modifier with aux on gfx12
3420 - anv: Disable CCS_E for some 8/16bpp copies on TGL+
3421 - anv: Use ANV_FAST_CLEAR_DEFAULT_VALUE for CCS on TGL+
3422 - anv: Re-enable CCS_E on TGL+
3423 - anv: Drop redundant disabling of non-renderable CCS
3424 - anv: Disable the SMEM fallback for local memory
3425 - anv: Require the local heap for CCS on XeHP
3426 - anv: Don't allocate VMA for CCS on XeHP
3427 - isl: Enable compression with multisampled Tile64
3428 - iris: Explicitly rely on gallium fallbacks for YUV
3429 - intel/isl: Support YUV pipe-to-isl format mapping
3430 - iris: Drop stale media compression import code
3431 - iris: Use iris_format_for_usage in map_aux_addresses
3432 - intel: Rename a RenderCompressionFormat field
3433 - intel/isl: Support the XeHP media compression format
3434 - iris: Support the XeHP media compression format
3435 - iris: Refactor a ternary in iris_resource_get_param
3436 - iris: Pick the right BO in iris_resource_get_param
3437 - iris: Return non-zero stride for clear color plane
3438 - intel/isl: Simplify Z-buffer tiling config during emit
3439 - intel/isl: Fix depth buffer TiledSurface programming
3440 - intel/isl: Add more PRM text for HiZ/STC requirement
3441 - iris: Don't fast clear with the view format
3442 - Revert "anv: Require the local heap for CCS on XeHP"
3443 - anv: Refactor anv_image_init_from_create_info
3444 - anv: Disable aux if the explicit modifier lacks it
3448 - st: Fix 64-bit vertex attrib index for TGSI path
3449 - st: Fix comments in commit be6d584de43966e
3450 - svga: Add GL43 commands support
3451 - svga: Add utility to check for GL43 support
3452 - tgsi: Add hw_atomic_declared in tgsi_info
3453 - svga: Add support for compute shader, shader buffers and image views
3454 - svga: shader translation for compute, image views and shader buffers
3455 - svga: add GL43 resource validation at draw time
3456 - svga: enable GL43 on SVGA GL43 capable device
3457 - svga: enable PIPE_CAP_IMAGE_STORE_FORMATTED on gl43 capable device
3458 - svga: store shared_mem_size in svga_compute_shader instead of svga_context
3462 - v3d: Update prim_counts when prims generated query in flight without TF
3466 - mesa/get: allow NV_pixel_buffer_object constants in GLES2
3470 - wsi/x11: support depth 30 visuals
3474 - llvmpipe: page-align memory allocations
3478 - iris: also dump bo's imported and exported flags
3479 - iris: destroy our mutexes a little later
3480 - iris: fix off-by-one error when clearing stale syncobjs
3481 - iris: call brw_process_intel_debug_variable() earlier
3482 - iris: extract iris_hw_context_set_unrecoverable()
3483 - intel/fs: Assert the GPU supports 64bit ops if present at lower_scoreboard time.
3484 - iris: improve error checking in functions that call vma_alloc()
3485 - iris: sprinkle some assertions for bufmgr->lock
3486 - iris: save some iris_syncobj_reference() calls at update_bo_syncobjs()
3487 - iris: implement inter-context busy-tracking
3488 - iris: handle IRIS_MEMZONE_BINDER with a real vma_heap like the others
3489 - iris: fix register spilling on compute shaders on XeHP
3493 - llvmpipe: fix wrong assumption on FB fetch shader opacity
3494 - llvmpipe: fix gl_FragColor and gl_LastFragData[0] combination
3495 - llvmpipe: protect from doing FB fetch of missing buffers
3496 - llvmpipe: fix FB fetch with non 32-bit render target formats
3497 - llvmpipe: remove dead args from load_unswizzled_block
3498 - llvmpipe: enable PIPE_CAP_FBFETCH_COHERENT
3502 - r300: Document the RADEON_DEBUG options
3503 - r300: Replace RADEON_NO_TCL with RADEON_DEBUG=notcl
3504 - r300: Remove broken optimization in rc_transform_KILL
3505 - r300: use point sprite coordinates only when drawing points (v5)
3506 - r300: fix translate_LRP
3507 - r300: properly initialize new_vs in r300_draw_init_vertex_shader
3508 - r300: Disable integers and indirect temporary addressing with swctl
3509 - r300: Set consistent PIPE_SHADER_CAP_PREFERRED_IR
3510 - r300: fix transformation of abs modifiers with negate
3514 - etnaviv: fix emit_if in case the else block ends in a jump
3518 - nv50/nir: Switch to the common NIR options
3520 Pierre-Eric Pelloux-Prayer (57):
3522 - radeonsi: use viewport offset in quant_mode determination
3523 - gallium/dri: replace bool with flag parameter
3524 - gallium/dri: add createImageFromFds2
3525 - gallium/dri: let the driver know if the imported image is DRI_PRIME buffer
3526 - radeon_winsys.h: add a parameter to buffer_from_handle
3527 - winsys/amdgpu: add uncached flag to the imported DRI_PRIME buffer
3528 - radeonsi/gfx10.3: enable SDMA for DRI_PRIME copies
3529 - radeonsi: use PIPE_BIND_DRI_PRIME instead of is_dri_blit_image
3530 - Revert "gallium: add a is_dri_blit_image bool to pipe_blit_info"
3531 - glsl/drirc: add an option for gl_ClipVertex / gl_CullDistance checks
3532 - drirc: enable do_dce_before_clip_cull_analysis for ANSA
3533 - mesa: don't reset SamplersValidated if nothing changed
3534 - glsl/nir: mark samplers inside a block as bindless
3535 - radeonsi: treat nir_intrinsic_load_constant as a VMEM operation
3536 - ac/surface: use a less strict condition in is_dcc_supported_by_L2
3537 - radeonsi/sdma: fix bogus assert
3538 - radeonsi: fix ps_uses_fbfetch value
3539 - ac/surface: don't validate DCC settings if DCC isn't possible
3540 - vbo/dlist: free copied.buffer if no vertices were copied
3541 - glx/drirc: add a force_direct_glx_context option
3542 - mesa: enable force_direct_glx_context for DiscoveryStudio2020
3543 - llvmpipe: add missing NIR alu-op handling
3544 - mesa: print a warning when an extension can't be disabled
3545 - mesa: don't use dummy_true for some MESA extensions
3546 - mesa/init: replace call_once with manual implementation
3547 - drirc: add mesa_extension_override option
3548 - drirc: add options for BETA CAE Ansa application.
3549 - mesa: always call _mesa_update_pixel
3550 - mesa: remove NEW_COPY_TEX_STATE
3551 - radeonsi/sqtt: fix shader stage values
3552 - radeonsi/sqtt: fix FINISH_DONE / BUSY usage
3553 - radeonsi/sqtt: reserve a vmid when sqtt is enabled
3554 - radeonsi: unreference framebuffer state after use
3555 - radeonsi/sqtt: increase the default buffer size to 32MB
3556 - radeonsi: fix fast clear / depth decompression corruption
3557 - radeonsi: silence a warning
3558 - radeonsi: use max_zplanes after the last write
3559 - glx: fix querying GLX_FBCONFIG_ID for Window
3560 - radeonsi: don't use perp. end caps when line smoothing is on
3561 - glthread: only log glthread destroy reason when it's not NULL
3562 - radeonsi/gfx8: use the proper dcc clear size
3563 - vbo/dlist: fix loopback crash
3564 - vbo/dlist: add vertices to incomplete primitives
3565 - amd: add SDMA_NOP_PAD
3566 - radv: partial sdma support
3567 - radv: allocate the prime buffer as uncached
3568 - vulkan/wsi: add use_prime_blit param to wsi_swapchain_init
3569 - vulkan/wsi: add a private transfer pool to exec the DRI_PRIME blit
3570 - radv: implement wsi's private transfer queue using SDMA
3571 - radeonsi/gfx10: fix si_texture_get_offset for mipmapped tex
3572 - radeonsi,radv: fix usages of surf_pitch
3573 - radeonsi/tests: update expected results
3574 - radeonsi/tests: add expected results for vega20
3575 - driconf: enable vs_position_always_invariant for Dirt Rally
3576 - mesa: use less temporaries in build_lighting
3577 - radeonsi: limit loop unrolling for LLVM < 13
3578 - radeonsi: change rounding mode to round to even
3582 - driconf: add executable_regexp application attribute
3583 - drirc: add Mari application workaround
3584 - xmlconfig_test: add unit test for executable_regexp
3585 - radeonsi: add radeonsi_force_use_fma32 driconf option
3586 - driconf: support META application
3587 - glx/dri3: fix glXQueryContext does not return GLX_RENDER_TYPE value
3588 - loader/dri3: fix piglit egl-copy-buffer test
3589 - ci: remove egl-copy-buffers from fail list
3590 - glx: no need to create extra pixmap for pbuffer
3591 - glx: check drawable type before create drawble
3592 - glx: add drawable type argument when create drawable
3593 - loader/dri3: add drawable type set by GLX and EGL
3594 - loader/dri3: remove unused present capability query
3595 - loader/dri3: pack window present event setup into a function
3596 - loader/dri3: setup present event with drawable type info
3597 - loader/dri3: replace is_pixmap with drawable type
3598 - loader/dri3: rename dri3_fake_front_buffer
3599 - loader/dri3: stop doing anything in swap buffer for some drawable
3600 - loader/dri3: support glx pbuffer swap
3601 - glx: fix regression for drawable type detection
3602 - glapi: should not add alias function to static_data.py
3603 - gallium: add caps for sparse texture support
3604 - mesa: add ARB_sparse_texture extension
3605 - mesa: add ARB_sparse_texture constants
3606 - gallium: add screen get_sparse_texture_virtual_page_size callback
3607 - gallium: add get_sparse_texture_virtual_page_size for noop/rbug/trace
3608 - mesa: add ARB_sparse_texture query in glGetInternalformativ
3609 - mesa: add ARB_sparse_texture texture param set/get
3610 - mesa/st: add st_GetSparseTextureVirtualPageSize interface
3611 - mesa: glTexStorage* support sparse texture allocation
3612 - mesa/st: add st_TexturePageCommitment interface
3613 - mesa: implement glTexPageCommitmentARB/glTexturePageCommitmentEXT
3614 - mesa/st: update NumSparseLevels from pipe_resource
3615 - winsys/radeon: change surface_init flags to 64bit
3616 - ac/surface: fix prt_first_mip_tail calculation for gfx9+
3617 - ac/surface: add prt_tile_depth
3618 - radeonsi: support alloc a sparse texture
3619 - radeonsi: use staging buffer for sparse texture when transfer map
3620 - radeonsi: implement get_sparse_texture_virtual_page_size
3621 - radeonsi: support texture resource commit
3622 - radeonsi: enable ARB_sparse_texture
3623 - nir: fix nir_tex_instr hash not count is_sparse field
3624 - gallium/dd_debug: add get_sparse_texture_virtual_page_size
3625 - gallium: add PIPE_CAP_QUERY_SPARSE_TEXTURE_RESIDENCY
3626 - mesa: add ARB_sparse_texture2 extension
3627 - mesa/main: relax alignment check when ARB_sparse_texture2 available
3628 - glsl: add ARB_sparse_texture2 extension
3629 - glsl: ir_texture support sprase texture
3630 - glsl: add _texture related sparse texture builtin functions
3631 - glsl: add _textureCubeArrayShadow related sparse texture builtin func
3632 - glsl: add _texelFetch related sparse texture builtin function
3633 - glsl: add sparse texture image load builtin functions
3634 - glsl: add vec5 glsl types
3635 - glsl/nir: convert sparse ir_texture to nir
3636 - glsl/nir: convert sparse image load to nir
3637 - glsl/nir: adjust sparse texture nir_variable
3638 - glsl: add sparseTexelsResidentARB builtin function
3639 - glsl/nir: convert is_sparse_texels_resident to nir
3640 - radeonsi: lower nir_intrinsic_is_sparse_texels_resident
3641 - mesa/main: allow multi sample sparse texture
3642 - mesa/main: export _is_multisample_target for external usage
3643 - gallium: add multi_sample parameter to get_sparse_texture_virtual_page_size
3644 - radeonsi: enable multi sample sparse texture support
3645 - radeonsi: enable ARB_sparse_texture2
3646 - gallium: add PIPE_CAP_CLAMP_SPARSE_TEXTURE_LOD
3647 - mesa: add ARB_sparse_texture_clamp extension
3648 - glsl: add ARB_sparse_texture_clamp extension
3649 - glsl: ir_texture add clamp field
3650 - glsl: _texture support clamp parameter
3651 - glsl: _textureCubeArrayShadow support clamp
3652 - glsl: add ARB_sparse_texture_clamp builtin functions
3653 - glsl/nir: convert ir_texture->clamp to nir
3654 - radeonsi: enable ARB_sparse_texture_clamp
3655 - radeonsi: workaround Specviewperf13 Catia hang on GFX9
3656 - radeonsi: fix depth stencil multi sample texture blit
3657 - glx: fix pbuffer refcount init
3661 - intel/compiler: Assert that unsupported tg4 offsets were lowered for XeHP
3662 - intel: Emit 3DSTATE_BINDING_TABLE_POOL_ALLOC for XeHP
3666 - anv: Enable implicit CCS for external images
3667 - anv: add helper methods related to enabling CCS for external images
3671 - radv: Use nir_var_mem_image in meta shaders
3672 - aco: disable mul(cndmask(0, 1, b), a) optimization sometimes
3673 - aco: simplify emit_stream_output()
3674 - radv: don't use a separate cache entry for GS copy shaders
3675 - aco: use std::vector and IDSet in RA validator
3676 - docs: update radv extensions in features.txt
3677 - nir/algebraic: optimize Cyberpunk 2077's open-coded bitfieldReverse()
3678 - nir/lower_system_values: replace local_invocation_id components with zero
3679 - radv: lower load_local_invocation_index with 1D workgroups
3680 - aco: optimize load_local_invocation_index with single-wave workgroups
3681 - aco: consider pseudo-instructions reading exec in needs_exec_mask()
3682 - aco/spill: use spills_entry instead of spills_exit to kill linear VGPRs
3683 - radv: make RADV_FORCE_FAMILY case-insensitive
3684 - radv: fix max_render_backends for Sienna Cichlid null winsys
3685 - spirv: run nir_copy_prop before nir_rematerialize_derefs_in_use_blocks_impl
3686 - radv: stop running copy-propagation before nir_opt_deref
3687 - nir/lower_tex: don't calculate texture_mask for texture_index>=32
3688 - nir/lower_subgroups: fix left shift of -1
3689 - aco: properly update use counts if a extract is still used
3690 - nir/dce: fix DCE of loops with a halt or return instruction in the pre-header
3691 - nir/cf: fix insertion of loops/ifs after jumps
3692 - nir/tests: add DCE test for loops following a jump
3693 - aco: don't create DPP instructions with SGPR operands
3694 - nir/opt_offsets: fix try_extract_const_addition recursion
3695 - nir/opt_offsets: remove need to loop try_extract_const_addition
3696 - nir/algebraic: optimize a*#b & -4
3697 - radv: add radv_meta_init_shader
3698 - radv: include RT shaders in RADV_DEBUG=shaders,shaderstats
3699 - radv: clone shader in radv_shader_compile_to_nir
3700 - radv: initialize workgroup_size in radv_meta_init_shader
3701 - aco: don't create unnecessary addition in indirect get_sampler_desc()
3702 - aco: improve clrx disassembly
3703 - aco: use more predictable tiebreaker when forming MADs
3704 - aco: create v_fmamk_f32/v_fmaak_f32 from nir_op_ffma
3705 - aco: swap operands if necessary to create v_madak/v_fmaak
3706 - aco: swap multiplication operands if needed to create v_fmac_f32/etc
3707 - radv,aco: implement nir_op_ffma
3708 - radv,aco: don't lower some ffma instructions
3709 - radv: have the null winsys set more fields
3710 - aco: fix check_vop3_operands() for f16vec2 ffma fneg combine
3711 - nir/opt_load_store_vectorize: improve ssbo/global alias analysis
3712 - nir/algebraic: optimize more 64-bit imul with constant source
3713 - nir/opt_access: infer CAN_REORDER for global access
3714 - aco: disallow SMEM offsets that are not multiples of 4
3715 - aco: skip &-4 before SMEM
3716 - aco: remove SMEM constant/addition combining out of the loop
3717 - radv: increase maxTaskOutputCount to 65535
3718 - nir/lower_shader_calls: fix store_scratch write_mask
3719 - ac/nir: fix store_buffer_amd write_masks
3720 - nir/uniform_atomics: fix is_atomic_already_optimized without workgroups
3721 - aco: remove pack_half_2x16(a, 0) optimization
3722 - radv: add RADV_DEBUG=splitfma
3723 - radv: set radv_split_fma=true for Proton SotTR
3724 - nir/unsigned_upper_bound: don't follow 64-bit f2u32()
3725 - aco: use p_extract for SGPR nir_op_unpack_half_2x16_split_y
3726 - nir: add nir_op_fmulz and nir_op_ffmaz
3727 - radv,aco,ac/llvm: implement fmulz and ffmaz
3728 - aco: consider legacy multiplications in optimizer
3729 - aco: create v_mac_legacy_f32/v_fmac_legacy_f32
3730 - nir/algebraic: add ignore_exact() wrapper
3731 - nir/algebraic: optimize open-coded fmulz/ffmaz
3732 - nir: add some missing nir_alu_type_get_base_type
3733 - nir/algebraic: optimize expressions using fmulz/ffmaz
3734 - nir/builder: set write mask if not provided
3735 - nir/builder: assume scalar alignment if not provided
3736 - radv: avoid providing a write_mask to intrinsic builders
3737 - radv: avoid providing an align_mul to intrinsic builders
3738 - radv: avoid providing an align_offset to intrinsic builders
3739 - ac/nir: avoid providing a write_mask to intrinsic builders
3740 - ac/nir: avoid providing an align_mul to intrinsic builders
3741 - ac/nir: use shorter builder names
3742 - radv: fix raytracing with wave32
3743 - radv: use wave32 for raytracing
3744 - radv: use 8x4 workgroups for wave32 RT
3745 - radv: fix optimized MSAA copies with suballocated images
3746 - aco: fix neg(abs(mul(a, b))) if the mul is not VOP3
3747 - aco: optimize abs(mul(a, b))
3748 - aco: preserve pass_flags during format conversions
3749 - aco: fix v_mac_legacy_f32
3750 - aco: don't encode src2 for v_writelane_b32_e64
3751 - radv: fix R_02881C_PA_CL_VS_OUT_CNTL with mixed cull/clip distances
3752 - anv: Enable nir_opt_access
3753 - radv: include disable_aniso_single_level and adjust_frag_coord_z in key
3757 - freedreno: Fix for large epilogues
3758 - freedreno/ir3+isa: Cleanup bindless cat5 samp/tex encoding
3759 - freedreno/isa: Fixes for validation
3760 - freedreno/ir3/tests: Fix indentation
3761 - freedreno/ir3/tests: Add additional disasm test vectors
3762 - freedreno/ir3: Don't lower s2en if samp/tex is too large
3763 - isaspec: Fix derived field width
3764 - isaspec: Do not emit duplicate field encodes
3765 - isaspec: Add bitfield size assertions
3766 - freedreno/isa: Add immed reg accessors
3767 - freedreno/ir3/tests: Don't skip encode test if decode fails
3768 - freedreno/ir3/tests: Add some 8/16b ldg/stg tests
3769 - freedreno/isa: Fix ldg/stg "halfness"
3770 - freedreno/computerator: Fix mergedregs
3771 - freedreno/ir3: 64b intrinsic lowering
3772 - freedreno/ir3: Lower 64b phis
3773 - freedreno/ir3: Add load/store_global lowering
3774 - isaspec: Fix gpu_id for default_options
3775 - freedreno: Skip built-in shaders for clover
3776 - freedreno/ir3: Handle MESA_SHADER_KERNEL
3777 - freedreno: implement set_compute_state()
3778 - freedreno/ir3: vec8+vec16 support
3779 - freedreno/ir3: implement load_work_dim intrinsic
3780 - freedreno/ir3: Add support for load_kernel_input
3781 - freedreno/ir3: Move lower_idiv_options
3782 - freedreno: Fix set_global_binding
3783 - freedreno/ir3: Deal with zero-source instructions
3784 - freedreno/ir3: 16b bools
3785 - freedreno/ir3: 8bit fixes
3786 - freedreno/ir3: Fix load/store_global_ir3 type
3787 - freedreno/ir3: Fix reg size validation
3788 - freedreno/ir3: Add wide load/store lowering
3789 - freedreno/ir3: Add ihadd/uhadd
3790 - freedreno/ir3: Get req_local_mem from pipe_compute_state
3791 - freedreno/ir3: Fix validation of subgroup macros
3792 - freedreno/drm: Add some asserts
3793 - freedreno/drm: Move suballoc_bo to device
3794 - freedreno/ir3/print: Show end's outidxs
3795 - freedreno/ir3: xfb fix for duplicate outputs
3796 - freedreno: Add PIPE_SHADER_IR_NIR_SERIALIZED support
3797 - freedreno: caps for clover
3798 - isaspec: Add prototypes for expr evaluators
3799 - freedreno/crashdec: Split out mempool decoding
3800 - freedreno/crashdec: HFI queue decoding
3801 - freedreno/crashdec: Fallback to chip_id for GPU id
3802 - freedreno/crashdec: Basing GMU log decoding
3803 - freedreno/ir3: Handle instr->address when cloning
3804 - freedreno/computerator: Fix @buf header
3805 - freedreno/computerator: Mark shader bo for dumping
3806 - freedreno/ir3: Dump const state with shader disasm
3807 - clover: Move min image support check
3808 - freedreno: Report system memory as video memory
3809 - freedreno: Small dev_id_compare() cleanup
3810 - freedreno: Rearrange dev_id_compare() logic
3811 - freedreno: Handle wildcard fuse-id in device matching
3812 - freedreno: Update chip-ids
3813 - freedreno/decode: Handle chip-id
3814 - mesa/st: Lowered ucp should still mark rast state dirty
3815 - freedreno: Pass shader cache key instead of shader key
3816 - freedreno: Add FD_DIRTY_RASTERIZER_CLIP_PLANE_ENABLE
3817 - freedreno/a6xx: Fix clip_mask
3818 - freedreno/a5xx: Fix clip_mask
3819 - freedreno: Add missing generated header dependency
3823 - intel/fs: OpImageQueryLod does not support arrayed images as an operand
3824 - ci: Do not remove libgbm-dev
3825 - ci: Move common variables out into a separate file
3826 - ci: Do not remove wget
3827 - ci/piglit: Start vtest server if driver is set to virpipe
3828 - anv: Enable VK_VALVE_mutable_descriptor_type
3829 - docs: Update features and new_features for anv
3833 - llvmpipe: adjust rounding for viewport scissoring
3837 - vulkan/wsi/x11: document implementation
3841 - android.mk: Add missing variables to the make target
3842 - v3dv: Fix dEQP-VK.info#instance_extensions test
3843 - v3dv: Fix V3DV_HAS_SURFACE preprocessor condition
3844 - v3dv: Hotfix: Rename remaining V3DV_HAS_SURFACE->V3DV_USE_WSI_PLATFORM
3845 - v3d: Don't force SCANOUT for PIPE_BIND_SHARED requests
3846 - v3dv: add Android support
3850 - frontends/omx: preserve omx to keep current mode for avc decoding
3851 - frontends/va: preparing to disable h264 extension flag in vaapi dec path
3852 - radeon/vcn: enable dynamic dpb Tier2 support for h264 dec vaapi path
3853 - radeon/vcn: enable dynamic dpb Tier2 for hevc dec vaapi path
3854 - frontend/va: Keep surface buf addr before reallocation
3855 - radeon/vcn: Updating render_pic_list for correction
3859 - util/xmlconfig: Allow DT_UNKNOWN files
3863 - venus: ignore framebuffer for VkCommandBuffer executed outside of render pass
3867 - intel/compiler: Set correct cache policy for A64 byte scattered read
3868 - intel/genxml: Add new Primitive ID Not Required bit field to 3DSTATE_DS
3869 - intel/compiler: Track primitive id in domain/evaluation shader
3870 - anv: Drop hint if primitive id is required or not
3871 - iris: Drop hint if primitive id is required or not
3872 - anv: Pass correct aux usage while filling out surface state
3873 - anv: Enable CCS for storage image formats
3874 - intel/compiler: Set correct return format for brw_SAMPLE
3875 - intel/compiler: Don't hardcode padding source type to 32bit
3876 - intel/compiler: Fix instruction size written calculation
3877 - intel/compiler: Add helper to support half float payload with padding
3878 - intel/fs: Define and set correct sampler simd mode
3879 - intel/compiler: Prepare disasm for 16-bit sampler params
3880 - anv, iris: Implement Wa_14014890652 for DG2
3881 - intel/genxml: Add L1 Cache Control bit field
3882 - intel/genxml: Add Un-Typed Data-Port Cache Flush field to pipe control
3886 - frontends/va/av1: handle multiple slice params
3890 - aco: do not return an empty string when disassembly is not supported
3891 - radv: fix removing PSIZ when it's not emitted by the last VGT stage
3892 - radv: fix OpImageQuerySamples with non-zero descriptor set
3893 - radv: do not remove PSIZ for streamout shaders
3894 - aco: fix invalid IR generated for b2f64 when the dest is a VGPR
3895 - aco: fix emitting stream outputs when the first component isn't zero
3896 - radv: add an assertion to prevent GPU hangs when VRS isn't supported
3897 - ac/rgp: remove useless code related to GFX6-7
3898 - aco: fix loading 64-bit inputs with fragment shaders
3899 - radv,aco: decouple shader_info/options from radv_shader_args
3900 - radv: re-emit prolog inputs when the nontrivial divisors state changed
3901 - radv: remove old RADV_TRACE_FILE warning
3902 - radv: fix build errors with Android
3903 - radv: report error messages when the driver can't be initialized
3904 - aco: only load streamout buffers if streamout is enabled
3905 - radv: lower the viewport index to zero when the VGT stage doesn't export it
3906 - radv: fix invalid wait_dst_stage_mask type
3907 - ac/nir: remove bogus assertion about the position for culling
3908 - zink: add CI lists and deqp-suite configuration for RADV
3909 - radv: move freeing the trigger SQTT file at a better place
3910 - radv: remove useless checks about GFX7 for SQTT
3911 - radv: enable SQTT instruction timing by default
3912 - radv: stop reporting SQTT/RGP support as experimental
3913 - docs: document RADV_THREAD_TRACE_* envvars
3914 - radv: rename radv_shader_variant to radv_shader
3915 - radv: only enable VK_EXT_display_control for vrcompositor (SteamVR)
3916 - radv: optimize subpass barrier flushes for imageless framebuffers
3917 - Revert "radv: only enable VK_EXT_display_control for vrcompositor (SteamVR)"
3918 - radv: remove unused parameter in radv_emit_subpass_barrier()
3919 - radv/sqtt: reserve a VMID for better profiling
3920 - radv/sqtt: stop calling radv_cs_add_buffer() for the thread trace BO
3921 - radv: do not expose buffer features for depth/stencil formats
3922 - ac/rgp: fix alignment of code object records to follow the RGP spec
3923 - radv: only emit PGM_LO for the vertex prolog
3924 - ac/rgp: add support for queue event timings
3925 - ac/rgp: add support for clock calibration
3926 - radv/winsys: use same IBs padding as the kernel
3927 - zink: update the CI lists for RADV
3928 - radv/sqtt: fix GPU hangs when capturing from the compute queue
3929 - ac: add initial SPM support
3930 - ac/rgp: add support for dumping SPM data
3931 - radv: fix a sync issue on GFX9+ by clearing the upload BO fence
3932 - ac/spm: fix determing the SPM wire
3933 - ac/spm: fix determining the counter slot
3934 - zink: update the CI lists for RADV
3935 - zink: skip one GLES31 subset to avoid GPU hangs on Navi10
3936 - nir: fix constant expression of ibitfield_extract
3937 - radv: simplify re-using cache entries in radv_pipeline_cache_insert_shaders()
3938 - radv: disable HTILE for D32S8 format and mipmaps on GFX10
3939 - util/queue: fix a data race detected by TSAN when finishing the queue
3940 - radv: allow TC-compat CMASK with storage images on GFX10+
3941 - radv: ignore the descriptor set layout when creating descriptor template
3942 - radv: convert remaining enums/structs to 1.2 versions
3943 - ac: change family names to uppercase in ac_get_family_name()
3944 - radv/winsys: report the real family name instead of OVERRIDDEN
3945 - radv: fix emitting VBO when vertex input dynamic state is used
3946 - radv: make radv_break_on_count() a non-static function
3947 - radv: make radv_copy_buffer() a non-static function
3948 - radv: implement optimized MSAA copies using FMASK
3949 - ac/nir: fix left shift of 1 by 31 places detected by UBSAN
3950 - radv: fix OOB access for inline push constants detected by UBSAN
3951 - aco: fix right shift of exponent 32 detected by UBSAN
3952 - radv: implement vkCmdWriteTimestamp2KHR()
3953 - radv: implement vkCmd{Reset,Set}Event2KHR()
3954 - radv: implement vkCmdWriteBufferMarker2AMD()
3955 - radv: implement vkCmdWaitEvents2KHR()/vkCmdPipelineBarrier2KHR()
3956 - radv: implement vkQueueSubmit2KHR()
3957 - radv: add support for VkMemoryBarrier2KHR
3958 - radv: add support for creating device-only events
3959 - radv: add support for new pipeline stages and access masks
3960 - radv: add support for VK_IMAGE_LAYOUT_ATTACHMENT_OPTIMAL_KHR
3961 - radv: switch the remaining stages/access to VK_PIPELINE_STAGE_2/VK_ACCESS_2
3962 - radv: advertise VK_KHR_synchronization2
3963 - radv: fix accessing NULL pointers when destroy the VRS image
3964 - radv: add a workaround to fix a segfault with Metro Exodus (Linux native)
3965 - radv: configure the number of SGPRs/VGPRs directly from the arguments
3966 - radv: copy the user SGPRs locations outside of radv_declare_shader_args()
3967 - radv: constify radv_shader_info in radv_declare_shader_args()
3968 - radv/llvm: stop trying to eliminate VS outputs
3969 - radv/llvm: constify radv_shader_info
3970 - radv: fix resetting the entire vertex input dynamic state
3971 - radv: remove never reached free() when compiling shaders
3972 - radv: add a helper function to upload a shader binary
3973 - radv: upload shader binaries after they are all compiled
3974 - radv: pass a pointer to a pipeline for the create/insert cache functions
3975 - radv: upload shader binaries of a pipeline contiguously in memory
3976 - radv: move a comment at the right place in CmdBindVertexBuffers2EXT()
3977 - radv: constify radv_vertex_binding in CmdSetVertexInputEXT()
3978 - radv: constify radv_vs_input_state() in more places
3979 - radv: fix dynamic rendering with VRS
3980 - radv: avoid prefixing few VkXXX structures by struct
3981 - radv/sqtt: always dump pipelines and shaders ISA
3982 - radv: add few helpers for configuring performance counters
3983 - radv: add initial SPM support on GFX10+
3984 - radv: fix resume/suspend render pass with depth/stencil attachment
3985 - radv: mark GFX10.3 (aka RDNA2) as conformant products with CTS 1.2.7.1
3986 - Revert "radv: Add bufferDeviceAddressMultiDevice support."
3987 - radv: do not perform depth/stencil resolves for suspended render pass
3988 - nir/lower_tex: add lower_lod_zero_width
3989 - radv: enable lower_lod_zero_width
3990 - radv: fix dynamic rendering inheritance if the subpass index isn't 0
3991 - radv: ignore dynamic inheritance if the render pass isn't NULL
3992 - radv/winsys: remove useless has_sparse_vm_mappings checks
3993 - radv/winsys: clear the PRT VA range when destroying a virtual BO
3994 - radv/winsys: stop using reference counting for virtual BOs
3995 - radv/winsys: update sparse mappings with OP_REPLACE instead of OP_MAP/OP_UNMAP
3996 - ac/nir: add an option to disable anisotropic filtering for single level images
3997 - radv,aco: do not disable anisotropy filtering for non-mipmap images
3998 - radv: fix dynamic rendering global scissor
3999 - radv: remove unnecessary NULL checks in vkMapMemory()/vkUnmapMemory()
4000 - radv/winsys: stop zeroing few structs in buffer_from_fd()
4001 - radv/winsys: remove radv_amdgpu_winsys_bo::is_shared
4002 - radv: fix restoring subpass during hw/fs color resolves
4003 - radv: re-enable fast clears for images that support comp-to-single
4004 - radv: re-apply "Do not access set layout during vkCmdBindDescriptorSets."
4005 - radv: fix clears with value of "1" and different DCC signedness
4006 - radv: stop checking buffer size in vkCreateBuffer()
4007 - radv: print number of levels with RADV_DEBUG=img
4008 - radv: fix copying mutable descriptors to sampler descriptors
4009 - radv: add drirc radv_disable_htile_layers and enable it for F1 2021
4010 - aco: fix a dynamic-stack-buffer-overflow when printing instructions
4011 - aco: do not print prologs disassembly if no disassembler
4012 - radv,aco: keep track of the prolog disassembly if necessary
4013 - radv: save the vertex prolog to the trace BO for debugging
4014 - radv: dump the VS prolog disassembly to the hang report
4015 - radv: add UMR markers for the vertex prolog
4016 - radv: do not use the common entrypoint for the Metro Exodus layer
4017 - radv: remove remaining dead code related to the old sync code
4018 - radv/winsys: remove unused syncobj functions
4019 - radv: stop checking if dynamic states changed
4020 - radv: reset VRS if the current subpass doesn't have a VRS attachment
4021 - radv: fix computing the fb size in presence of dynamic VRS attachment
4022 - radv: only clear VRS_HTILE_ENCODING on GFX10.3+
4023 - radv: allow to disable anisotropic filtering for single level image with drirc
4024 - radv: enable radv_disable_aniso_single_level for Battlefield 1 & V
4025 - radv/winsys: fix zero submit if no timeline semaphore support
4026 - vulkan/runtime: fix accessing NULL pointers detected by UBSAN
4027 - radv: fix missing destroy for the overallocation mutex
4028 - radv: fix computing the number of color samples if no attachments
4029 - radv: optimize CPU overhead of si_cp_dma_prefetch() slightly
4030 - radv: disable attachmentFragmentShadingRate for RADV_DEBUG=nohiz
4031 - radv: fix copying VRS rates to HTILE if the depth/stencil is cleared
4032 - radv: fix copying VRS rates if the ds attachment uses mips
4033 - radv/winsys: set GTT_WC flag for CS IBs on GFX6
4034 - radv: do not restore NULL compute pipelines after meta operations
4035 - radv: stop checking if pipelines are NULL during draws/dispatches
4036 - vulkan: Update the XML and headers to 1.3.204
4037 - radv: implement 1.3 features/properties
4038 - radv: report textureCompressionASTC_HDR as not supported
4039 - radv: add a no-op version of vkGetPhysicalDeviceToolPropertiesEXT()
4040 - radv: switch a bunch of struct/enum to 1.3 versions
4041 - radv: bump conformance version to 1.3.0.0 for RDNA2
4042 - radv: advertise Vulkan 1.3
4043 - radv/ci: mark dEQP-VK.api.version_check.version as expected failure on Stoney
4044 - radv/winsys: fix missing buffer_make_resident() for the null winsys
4045 - Revert "radv: re-apply "Do not access set layout during vkCmdBindDescriptorSets.""
4046 - radv/winsys: fix initializing debug/perftest options if multiple instances
4047 - radv: enable radv_disable_aniso_single_level for The Evil Within 1&2
4051 - egl/wayland: Remove unused wayland enum
4055 - vulkan: Unlock before return.
4059 - docs/features: Add VK_KHR_acceleration_structure, VK_KHR_pipeline_library, VK_KHR_ray_query, VK_K…
4063 - egl: introduce a log level getter function
4064 - egl: add config debug printout
4068 - meson: Try to link all-targets module if Gallium OpenCL is enabled
4072 - renderonly: write down usage rules
4073 - vulkan/wsi/wayland: use enum wl_shm_format
4074 - vulkan/wsi/wayland: use DRM_FORMAT_INVALID
4075 - vulkan/wsi/wayland: remove format switch from wl_shm_format_for_vk_format
4079 - llvmpipe: Add get_{driver,device}_uuid implementations
4083 - iris: clear bos_written when resetting a batch
4084 - vulkan/wsi: provide api for drivers to setup syncobj fd
4085 - radv: setup syncobj fd via wsi_device_setup_syncobj_fd
4086 - anv: setup syncobj fd via wsi_device_setup_syncobj_fd
4087 - vulkan/wsi: implement missing wsi_register_device_event
4088 - anv: allow VK_IMAGE_LAYOUT_UNDEFINED as final layout
4089 - glsl: fix invariant qualifer usage and matching rule for GLSL 4.20
4090 - iris: unref syncobjs and free r/w dependencies array for slab entries
4091 - mesa: free idalloc storage for display lists
4092 - mesa: free vbo_save_vertex_list store prims
4093 - mesa: refactor GetProgramiv to use program resource list
4094 - mesa: move GetProgramInterfaceiv as a shader_query function
4095 - mesa: change GetProgramiv name length queries to use program resources
4096 - intel/genxml: add PIPE_CONTROL field for L3 read only cache invalidation
4097 - anv: invalidate L3 read only cache when VF cache is invalidated
4098 - iris: invalidate L3 read only cache when VF cache is invalidated
4099 - iris: fix a leak on surface states
4100 - mesa/st: always use DXT5 when transcoding ASTC format
4104 - aco: support DPP8
4105 - aco: lower masked swizzle to DPP8
4106 - radv/sqtt: Add and enable basic EXT_debug_utils support.
4110 - tegra: Use private reference count for sampler views
4111 - tegra: Use private reference count for resources
4115 - nine: remove dead code
4116 - svga: fix bitwise/logical and mixup
4117 - meson: drop compatability with < 0.48
4118 - meson: drop a comment relating to old meson version
4119 - docs: update the required meson version
4120 - meson: drop a temp formatting variable
4121 - ci: clean up debian-clang no-error list
4122 - gallivm: avoid a self-assign warning
4123 - ci: debian-clang: drop -Wno-error for self-assign
4124 - lavapipe: fix implicit-fallthrough warning
4125 - i915g: fix implicit-fallthrough warning
4126 - r300: remove a set but not used variable
4127 - r600: remove a set but not used variable
4128 - glx: remove a set but not used variable
4129 - gallium/u_threaded: drop unused function
4130 - gallium/tgsi_exec: drop unused function
4131 - draw: drop unused function
4132 - microsoft/compiler: dxil_nir_opt_alu_deref_srcs: return progress
4133 - lavapipe: fix string-plus-int warning
4134 - i915g: avoid left shifting a negative number
4135 - panvk: use FALLTHROUGH to stop a warning
4136 - panvk: cast negative value to unint8_t
4137 - ci: debian-clang: build more drivers
4138 - tu/clear_blit: use || when working with bools
4139 - r600/sb: silence a sometimes-uninitialized warning
4140 - ci: debian-clang: -Wno-error for sometimes-uninitialized
4141 - zink: malloc/sizeof mismatch
4142 - xa: fix compile warning for -Wabsolute-value
4143 - broadcom/compiler: fix compile warning -Wabsolute-value
4144 - ci: debian-clang: drop -Wno-error=absolute-value
4145 - replace 0 with NULL for NULL pointers
4146 - freedreno: silence sometimes-uninitialized warning
4147 - freedreno: drop dead assignment
4148 - meson: add check kwarg to run_command
4149 - ci: clean up debian-android no-error list
4150 - anv: drop unused label
4151 - ci: debian-android: drop -Wno-error=unused-label
4152 - vulkan/vk_extensions_gen: fix -Wextern-initializer warning
4153 - ci: debian-android: drop -Wno-error=extern-initializer
4154 - microsoft/compiler: fix -Wbitwise-instead-of-logical warning
4155 - anv: avoid warning about unused function
4156 - vc4: drop unused function
4157 - v3d: avoid warning about unused function
4158 - v3d: avoid warning about unused function
4159 - broadcom: drop unused functions
4160 - panfrost: mark two variables as unused
4161 - anv: drop a set but unused variable
4162 - anv: drop a set but unused variable
4163 - svga: silence -Wsometimes-uninitialized
4167 - util: use anonymous file for memory fd creation
4171 - frontends/va: disable packed header support for h264 encoder
4172 - frontends/va/enc: hardcode h265 encoder ref pic list size
4173 - radeon/vcn: increase encoder dpb size
4174 - frontends/va/enc: allow for frames to be marked as (not) referenced
4175 - radeon: hardcode uvd/vce encoder not_referenced value to false
4176 - radeon/vcn: implement encoder dpb management
4177 - frontends/va/enc: default motion estimation parameters for performance
4178 - radeonsi: add check for graphics to si_try_normal_clear
4182 - mesa: fix buffer overrun in SavedObj texture obj array
4183 - mesa: remove old tnl device driver header files
4184 - mesa: remove _mesa_ir_link_shader()
4185 - mesa: make _mesa_associate_uniform_storage() static
4186 - mesa: remove GLSL IR to Mesa IR code
4187 - mesa/st: move _mesa_generate_parameters_list_for_uniforms() code to st
4188 - mesa: move _mesa_ensure_and_associate_uniform_storage() to uniform_query.cpp
4189 - mesa: tidy up ir_to_mesa.{cpp,h} includes, comments, etc
4190 - mesa: rename ir_to_mesa.{cpp,h} -> link_program.{cpp,h}
4191 - mesa: update or remove out of date references to ir_to_mesa
4192 - doc: update source tree doc to reflect recent classic/swrast deletions
4193 - util: add dri config option force_compat_shaders
4194 - util: add workaround for SNK HEROINES Tag Team Frenzy
4195 - mesa: make struct in gl_program a union and remove FIXME
4196 - glsl/glcpp: make sure to expand new token after concatenation
4197 - glsl/st: move st_nir_opts() into gl compiler common code
4201 - radv: Use MESA_VULKAN_SHADER_STAGES to make room for mesh/task.
4202 - aco: Fix how p_is_helper interacts with optimizations.
4203 - nir: Rename nir_get_io_vertex_index_src and include per-primitive I/O.
4204 - nir: Print Mesh Shader specific info.
4205 - nir: Fix nir_lower_io with per primitive outputs.
4206 - nir, spirv: Don't mark NV_mesh_shader primitive indices as per-primitive.
4207 - nir: Add new option to lower invocation ID from invocation index.
4208 - nir: Lower cull and clip distance arrays for mesh shaders.
4209 - nir: Don't compact per-vertex and per-primitive outputs together.
4210 - nir: Group per-primitive outputs at the end for driver location assign.
4211 - nir: Fix sorting per-primitive outputs.
4212 - util: Add util_widen_mask function.
4213 - aco: Use util_widen_mask.
4214 - radv: Use util_widen_mask.
4215 - nir: Print task and mesh shader I/O variable names.
4216 - aco: Clean up and fix quad group instructions with WQM.
4217 - aco/optimizer_postRA: Fix combining DPP into VALU.
4218 - aco/optimizer_postRA: Fix applying VCC to branches.
4219 - spirv: Allow VRS with mesh shaders.
4220 - gitlab-ci: Disable radv-fossils again.
4221 - ac/nir/ngg: Lower NV mesh shaders to NGG semantics.
4222 - radv: Add radv_pipeline_has_mesh helper.
4223 - radv: Add mesh shader specific info.
4224 - radv: Cleanup PS input generation.
4225 - radv: Cleanup VS output param assignment.
4226 - radv: Add support for per-primitive mesh shader outputs.
4227 - radv: Setup shader arguments for mesh shaders.
4228 - radv: Compile mesh shaders and apply the necessary NIR lowerings.
4229 - radv: Set output driver locations for mesh shaders.
4230 - radv: Create mesh shading pipelines.
4231 - radv: Add support for mesh shading pipelines in the command buffer.
4232 - aco: Add Mesh and Task shader stages.
4233 - aco: Update README about NGG and mesh shaders.
4234 - aco: Add 1D workgroup_id support for mesh shaders.
4235 - aco: Use the correct outinfo for mesh shaders.
4236 - aco: Export per-primitive mesh shader output attributes.
4237 - radv: Implement NV_mesh_shader draw calls.
4238 - radv: Enable NV_mesh_shader with a perftest flag.
4239 - radv: Note when a mesh shader writes the primitive shading rate.
4240 - radv: Lower primitive shading rate for mesh shaders.
4241 - aco: Emit VRS rate when it's per-primitive.
4242 - radv: Support VRS for mesh shaders.
4243 - radv: Disable IB2 on compute queues.
4244 - ac/nir/ngg: Fix mixed up primitive ID after culling.
4248 - ci: Add support for lazor Chromebooks
4249 - ci: Let manual LAVA jobs have a longer timeout than others
4250 - freedreno/ci: Test Turnip on Adreno 618
4251 - ci: Create symlink to /install early
4252 - ci: Don't set GALLIVM_PERF in the scripts
4253 - virgl/ci: Set GALLIVM_PERF=nopt,no_quad_lod
4254 - ci: Remove syslogd
4255 - virgl/ci: Run each dEQP instance in its own VM
4256 - ci: Uprev Crosvm
4257 - lvp: Free the driver_data pointer for all commands
4258 - anv/ci: Test with deqp-vk on Tiger Lake
4259 - anv/tests: Free BO cache and device mutex
4260 - ci: Rebalance Iris jobs
4264 - intel/compiler: Handle new sampler descriptor fields for 16bit sampler
4265 - intel/compiler/fs: Add support for 16-bit sampler msg payload
4266 - intel/compiler: Demote sampler params to 16-bit for CMS/UMS/MCS
4267 - intel/compiler: Prepare ld2dms_w for 4 mcs components
4268 - intel/compiler: Add new variant for TXF_CMS_W
4269 - intel/compiler: Deprecate ld2dms and use ld2dms_w instead
4273 - iris: enable adaptive sync for IRIS
4274 - anv: enable adaptive sync for ANV
4278 - intel/fs: Fix a cmod prop bug when cmod is set to inst that doesn't support it
4279 - anv: Include viewport size in scissor rectangle
4280 - glthread: Check out of bounds for MultiDrawElementsBaseVertex cmd
4284 - gallium: add PIPE_CAP_PREFER_POT_ALIGNED_VARYINGS
4285 - lima: enable PIPE_CAP_PREFER_POT_ALIGNED_VARYINGS
4286 - lima: add native txp support
4287 - lima: add support for 3D textures
4288 - lima: handle 1D samplers
4289 - lima: add more wrap modes
4290 - lima/ppir: check if mul node is a source of add node before inserting
4291 - lima/ppir: implement gl_FragDepth support
4292 - lima: fix crash with sparse samplers
4293 - lima: disasm: use last argument as a filename
4294 - lima: disasm: call util_cpu_detect() to init CPU caps
4295 - lima: implement dual source blend
4296 - lima: use 1 as blend factor for dst_alpha for SRC_ALPHA_SATURATE
4297 - lima: fix blending with min/max ops
4301 - glsl: fix for unused variable in glsl_types.cpp
4305 - anv: Fix assertion.
4306 - radv: Fix memory leak on error path.
4307 - clover: Add constructor for image_rd_argument.
4308 - zink: Remove duplicate variable unsized.
4309 - intel/compiler: Change selected_simd return type to int.
4310 - intel/compiler: Initialize SIMDSelectionTest member error.
4311 - vulkan/wsi: Unlock before return on error path.
4312 - microsoft/spirv_to_dxil: Fix non-Windows build.
4313 - virgl: Allocate qdws after virgl_init_context to avoid leak.
4314 - ac/rgp: Initialize clock_calibration with memset.
4315 - ir3: Make shift operand 64-bit.
4316 - panfrost: Avoid double unlock.
4317 - isaspec: Sort field names to generate deterministic output.
4318 - radv: Fix memory leak on error path.
4319 - r600/sfn: Remove unused AluInstruction members.
4320 - isaspec: Remove duplicate return statement.
4321 - v3dv: Add missing unlocks on errors.
4325 - zink: Do not access just freed zink_batch_state
4326 - zink: Fully initialize VkBufferViewCreateInfo for hashing
4330 - iris: Mark a dirty update when vs_needs_sgvs_element value changed
4334 - dri_interface: remove obsolete interfaces
4335 - dri_interface: remove gl header
4336 - venus: refactor private descriptor_set helpers to be private
4337 - venus: assign valid memoryTypeIndex of exportable ahb memory for image
4338 - venus: release queues on device creation failure
4339 - venus: refactor to add vn_device_init
4340 - venus: refactor to add vn_buffer_init
4341 - venus: refactor the ahb buffer mem_type_bits query api
4342 - venus: add struct vn_buffer_memory_requirements
4343 - venus: add struct vn_image_memory_requirements
4344 - venus: add buffer cache init and usage flows
4345 - venus: implement vn_buffer_get_max_buffer_size
4346 - venus: implement vn_buffer_cache_get_memory_requirements
4347 - venus: implement vn_buffer_cache_entries_create
4348 - venus: refactor vn_device_memory_simple_alloc
4349 - venus: simplify device memory pool alloc and refcount
4350 - venus: defer roundtrip waiting to vkFreeMemory time
4351 - venus: track memory type property flags in vn_device_memory
4352 - venus: move bo allocation for mappable memory to vn_MapMemory
4353 - venus: subtract appended header size in vn_CreatePipelineCache
4354 - venus: VkExternalImageFormatProperties is optional
4355 - tu: VkExternalImageFormatProperties is optional
4356 - venus: update some obsolete assumptions described
4357 - venus: track whether a fence is external
4358 - Revert "venus: remove vn_ring_wait_all"
4359 - venus: make vn_QueueSubmit async for native submissions
4360 - venus: properly destroy deferred ahb image before real image creation
4364 - vulkan: Open registry XML files as UTF-8
4365 - win32: Fixes thread local on win32 with clang/mingw (!14062)
4369 - virgl/drm: New optimization for uploading textures
4373 - build: add sha1_h for lp_texture.c
4377 - Fix compilation on armv7l with gcc 11.2.0
4381 - radeon/vce: Limiting max supported refernce frames to 1  for h264 encoding