Lines Matching +full:lima +full:- +full:rules

1 Mesa 21.1.0 Release Notes / 2021-05-05
20 ---------------
24     0128f10e22970d3aed3d1034003731f94623015cd9797c07151417649c1b1ff8  mesa-21.1.0.tar.xz
28 ------------
30 - VK_KHR_workgroup_memory_explicit_layout on Intel, RADV
32 - DRM format modifiers for AMD.
34 - VK_KHR_zero_initialize_workgroup_memory on Intel, RADV
36 - Zink exposes GL 4.6 and ES 3.1
38 - GL_EXT_depth_bounds_test on softpipe, zink
40 - GL_EXT_texture_filter_minmax on nvc0 (gm200+)
42 - GL_ARB_texture_filter_minmax on nvc0 (gm200+)
44 - GL_ARB_post_depth_coverage on zink
46 - VK_KHR_copy_commands2 on lavapipe
48 - lavapipe exposes Vulkan 1.1
50 - VRS attachment on RADV
52 - None
56 ---------
58 - No sRGB capable visuals/fbconfigs reported in glx
59 - Graphics corruption and GPU hang with RADV/LLVM
60 - old kernels (4.19) support in radv
61 - Elite Dangerous: Odyssey alpha crashes GPU on launch
62 - CSGO: Some default variables can cause problems with trust mode
63 - mesa git started to break wine + UnrealTournament.exe (old dx6 game)
64 - SuperTuxKart artifacting on RK3399
65 - [amdgpu]: Golf With Your Friends (431240): ERROR Waiting for fences timed out
66 - Strange results when trying to read from VK_FORMAT_R64_SFLOAT in compute shader
67 - anv: dEQP-VK.binding_model.buffer_device_address.set3.depth3.basessbo.convertcheck* slow
68 - Iris doesn't support INTEL_performance_query anymore
69 - RADV: TRUNC_COORD breaks gather operations
70 - [RADV] corruption in avatar after dying in Heroes of the Storm
71 - Metro Exodus crashing due to memory overflow
72 - Sauerbraten shader rendering broken on RV530 (r300g)
73 - texture glitches on CS:GO on Tiger Lake
74 - Build fail due to "parameter name omitted" on Gallium Nine
75 - Non-DRI builds broken by recent cleanups in Mesa core
76 - Cinnamon core dump after installing latest oibaf mesa build (165a69d2)
77 - yuv sampler lowering regression
78 - anv: anv_descriptor_set_binding_layout::array_size overflows u16
79 - RADV - Vertex explosion in DIRT 5 on RDNA2
80 - ci: Use renderdoc from debian
81 - ci: Use debian apitrace in x86 images
82 - SIGSEV in v3d_emit_gl_shader_state
83 - Xorg crash due to assertion failure after GPU soft reset
84 - AMD hevc_vaapi ffmpeg encoding = wrong image width (48px black bar on the right)
85 - panfrost: Page fault in glamor when running GIMP with X11 on Mali T860
86 - gallium: python trace scripts need updating
87 - EGL context creation fails when EGL_KHR_create_context_no_error is mentioned for OpenGL ES 1.1.
88 - [spirv-fuzz] NIR validation failed after spirv_to_nir: error: nir_block_dominates
89 - [bisected][regression][i965,iris] dEQP-VK.clipping.user_defined.clip_cull_distance.* failing on m…
90 - No Mans Sky GPU hang on Radeon ACO
91 - radeonsi: prusa-slicer crashes on mesa 21
92 - anv: dEQP-VK.subgroups.ballot_broadcast.compute.subgroupbroadcast_i8vec3_requiredsubgroupsize16 f…
93 - [radeonsi] Rendering of Firefox UI and website content corrupts randomly and after window geometr…
94 - max_binding inconsistency in vulkan descriptor set drivers
95 - anv: conservative rasterization ext question mark
96 - Unigine Tropics MSAA failure
97 - warning: xnack 'Off' was requested for a processor that does not support it! [AMD VEGAM with LLVM…
98 - Compiling some ARB assembly shaders leads to memory corruption
99 - Specifying an explicit location for an array output messes up transform feedback
100 - Request for VK_EXT_conservative_rasterization support on Anvil Intel driver for newest DXVK..
101 - radv: dEQP regressions after addrlib update
102 - Up to 30% performance drop (GLBenchmark, GfxBench)
103 - DOTA 2 don't no longer starts since commit ad241b15a9e517dd4c4e8d7b1d5dab7c3a74b37c
104 - Clover doesn't work for kmsro drivers
105 - aco_tests isel.sparse.clause fails with llvm-12
106 - util cpu detection breaks on 128-core AMD machines
107 - util cpu detection breaks on 128-core AMD machines
108 - Default GL_MAX_TEXTURE_BUFFER_SIZE very small
109 - intel_nullhw.c:41:38: error: field ‘vtable’ has incomplete type
110 - ACO error with GCN 1 GPU
111 - kmsro advertises EGL_MESA_device_software
112 - d3d12: Use ID3D12Device9::CreateCommandQueue1 when available
113 - [RADV] Halo: The Master Chief Collection: Crash in Halo Reach Firefight
114 - freedreno: use SAMPLE_COUNT to autotune sysmem vs gmem
115 - freedreno: draw_vbo optimizations
116 - [Bisected][RadeonSI] Mesa crashes when rendering with Eevee in Blender
117 - subgroupBallotFindMSB() broken in RADV/ACO 20.3.4
118 - nir_print: util_cpu_detect() is not called prior to _mesa_half_to_float()
119 - turnip: buffer overflow read on dEQP-VK.ycbcr.query.levels.tess_eval.r8g8b8a8_unorm
120 - RuneScape crashes GLOn12
121 - d3d12: Surfaces need to use shareable descriptors
122 - [RADV][RDNA2] Red Dead Redemption 2 image glitches during menu/overlay menu transitions
123 - "unknown intrinsic" assertion triggered by multiview shader in non-multiview renderpass in Vulkan…
124 - [i965][g965,ilk,g33,g45][bisected] dEQP-GLES2.functional.fbo.completeness.attachment_combinations…
125 - radv: VK_FORMAT_FEATURE_SAMPLED_IMAGE_BIT exposed for VK_FORMAT_R64_SFLOAT
126 - anv: android building error after commit 4fb6c05
127 - Compiling mesa with -Dtools=all throws deprecation warnings for intel tools
128 - DXVK is broken in latest master
129 - OpenCLOn12: Affinity Photo CL kernels produce invalid DXIL phis
130 - nir -> tgsi conversion problem
131 - [i965][g965,ilk,g33][bisected] fp16 enablement causes deqp test failures
132 - mesa/st: Uniforms are not updated after lowering alpha test
133 - [i965][bisected][regression] piglit failing primitive-restart-vbo_combined_vertex_and_index on mu…
134 - [RADV] Nioh 2 - The Complete Edition: "Bloom" on lights
135 - [RADV] Oblivion: Poor Performance while MSAA Is Enabled
136 - lima context state bugs with shader compile
137 - [RADV][BISECTED] The Surge 2 (644830) - In-game assets do not render correctly since 20.3.4.
138 - amd clang cannot convert ‘llvm::AtomicOrdering’ to ‘llvm::MaybeAlign’ build failure
139 - [iris][icl,tgl][bisected][regression] failure on piglit.spec.arb_separate_shader_objects.programu…
140 - opencl build fail
141 - anv: dEQP-VK.glsl.builtin_var.fragcoord_msaa.* fails
142 - Request - depth format feature SAMPLED_IMAGE_FILTER_LINEAR
143 - "radeonsi: Check pitch and offset for validity." is a bad commit
144 - Add OpenCL information to docs/features.txt
145 - [regression] [bisected] piglit.spec.arb_framebuffer_object.fbo-drawbuffers-none gldrawpixels fails
146 - RADV: robustBufferAccessUpdateAfterBind is not exposed
147 - debug build compilation failed: inlining failed in call to ‘always_inline’ ‘_nir_visit_dest_indir…
148 - [RADV/DXVK] Shadow artifacts with different games
149 - glxgears segfaults with classic i915
150 - mesa_glthread=true Black Mesa
151 - freedreno: rendering corruption in dead cells
152 - ac/rgp: Android building error after commit 12515d6
153 - d3d12: Assert failures & crashes on latest master
154 - RADV/ACO - DCC causing garbled output on RX570
155 - draw.c:121: _mesa_set_draw_vao: Assertion \`vao->_EnabledWithMapMode == _mesa_vao_enable_to_vp_in…
156 - ANV: Weird jitter in Witcher 1
157 - RADV - Path of Exile: Shimmering outlines where water and other objects meet
158 - ANV: Weird jitter in Witcher 1
159 - ANV: Weird jitter in Witcher 1
160 - meson: meson-built libraries have inconsistent compatability / current versions compared to older…
161 - device select layer breaks other layers
162 - RADV: Extreme overhead in vkQueueSubmit
163 - Graphical glitch of popupping missing texture on Mesa version >18.0.5 (Padoka Stable + Unstable/O…
164 - [regression] [bisected] dEQP-GLES2.functional.fbo.render.stencil_clear.rbo_rgb5_a1_stencil_index8…
165 - occasional corruption issue with RADV in multiple games, disappears after using amdvlk
166 - panfrost T860 regression
167 - OpenGL on GMA4500MHD
168 - piglit-replay: JUnit file contains wrong links to the tracie dashboard
169 - R8 texture upload / corruption bug on Radeon RX 5700 XT
170 - Ambient Occlusion in Two Point Hospital shows black spot artifacts
171 - freedreno: async background shader compile
172 - AMD VAAPI encoding has ceased to work
173 - Rage 2: Visual corruption on in-game menu with ACO.
174 - ACO doesn't correctly render map in Borderlands 3 vs. LLVM on 5700 XT
175 - Invalid shader under panfrost/wayland
176 - Strange Brigade refuses to load correctly since some recent commits
177 - GLonD3D12: Crashes and suboptimal fallback
178 - GLonD3D12: Crashes and suboptimal fallback
179 - GLonD3D12: Crashes and suboptimal fallback
180 - [RADV][REGRESSION][BISECTED] radv_GetMemoryFdPropertiesKHR returns no valid memory types for vaap…
181 - anv: vkQueueSubmit with waitSemaphore value of 0 hangs CPU
182 - ttn: invalid base/range triggering nir_validate assertion
183 - Sampling with mipmapped HiZ behaves unexpectedly on Gen9
184 - zink: ARB_map_buffers issues on CI
185 - u_upload_mgr: assert failure for large uploads
186 - [RADV][ACO] Overwatch game crash: amd/compiler/aco_insert_exec_mask.cpp: Failed Assertion
187 - PRIME render offloading broken
188 - Use out encoding for float immediates
189 - [RADV] Severe performance drop when exceeding VRAM compared to AMDVLK
190 - LIBGL_ALWAYS_SOFTWARE=1 picks zink over actual software rasterizers
191 - crash/assert in fd_set_viewport_states
192 - RADV: Occlusion query hangs Big Navi GPU
193 - "mesa: don't allocate matrices with malloc" cause eglCreateContext problem on android 7.
194 - Metal Gear Solid V: The Phantom Pain: texture issues and vertex stretches
195 - [iris and Navi 10] piglit.spec.arb_multi_draw_indirect.arb_draw_elements_base_vertex-multidrawele…
196 - miscompiled compute shader loop on llvmpipe (and Iris)
197 - ci: minio caching of arm64 artifacts for bare-metal
198 - Graphics glitches after upgrade to mesa 20.3 on Khadas VIM3 Pro (Mali G52 GPU)
199 - glthread crash in _mesa_glthread_upload
200 - freedreno piglit flakes
201 - RADV: NonUniform OpArrayLength on SSBO ignores NonUniform.
202 - Iris driver causing graphics glitch in QEMU spice egl DMA-BUF
203 - [RADV/ACO] Death Stranding cause a GPU hung (\*ERROR* Waiting for fences timed out!)
204 - [TGL] Elder Scrolls Online misrenders
205 - [ANV] System hang with GRVK demos
206 - ci: Fractional deqp runs with valgrind enabled.
207 - Regression: Segfault in cso_destroy_context() regression in 20.2
208 - Rendering artifacts in Barn Finders specifically on Radeon Vega
209 - Graphics regression in Assassins Creed Odyssey
210 - [ANV] Compilation warnings
211 - regression in !8152
212 - [bdw][icl][iris] fails new test \`clearbuffer-depth-cs-probe`
213 - ci: new traces runner needs dashboard links in the job log and junit
214 - zink: car model corruption with game TORCS
215 - glGetInternalformati64v(GL_TEXTURE_2D, GL_SR8_EXT, GL_COLOR_ENCODING) returns GL_NONE
216 - Windows: 32-bit build is broken hard
217 - ANV: Not handling separate stencil layouts properly
218 - [Regression][Intel][OpenGL][Bisected] Copying whole 2D array texture failed on latest driver
219 - turnip: dEQP-VK.tessellation.invariance.outer_triangle_set.quads_fractional_odd_spacing failure
220 - i915 regressions bisected to "vbo/dlist: use a shared index buffer"
221 - intel: Chrome OS "hatch" (cometlake) fails on dEQP-GLES31.functional.geometry_shading.layered.ren…
222 - radv: dEQP-VK.sparse_resources.* failures on GFX9
223 - freedreno: rendering corruption in War Robots
224 - radv: dEQP-VK.sparse_resources.* failures on GFX9
225 - Mesa 20.3.x crashes pidgin on AMD RX480
226 - timespec_get used unconditionally / build fails when targeting macOS 10.14 or earlier
227 - libunwind not located / used on macOS
228 - libunwind not located / used on macOS
229 - meson fails to locate libexpat on macOS
230 - CarX Drift Racing 2 fails to start
231 - Some games using FNA framework show blank screen
232 - Intel Vulkan regression of angle_end2end_tests
233 - Defer lavapipe warning to queue / command / swapchain buffer creation
237 -------
241 - clover: implement CL_IMAGE_ELEMENT_SIZE
242 - clover: implement CL_IMAGE_NUM_MIP_LEVELS and CL_IMAGE_NUM_SAMPLES
246 - intel: change urb max shader geometry for CML GT1
247 - intel: change urb max shader geometry for KBL GT1
251 - docs: Mark some non-core zink extensions complete
252 - docs: Mark some ES3 zink features complete
253 - egl: Fix error string returned by eglQueryDeviceAttribEXT
254 - zink: Factor out instance setup a bit more
255 - osmesa: Pacify MSVC in the test code
256 - glx: Fix GLX_SGI_video_sync for the no-current-drawable case
257 - nir: Silence a warning at -Og
258 - softfloat: Silence a warning at -Og
259 - glsl: Silence a warning at -Og
260 - glsl: Silence some warnings at -Og
261 - glsl: Silence a warning at -Og
262 - loader: Silence a warning at -Og
263 - gallivm: Silence a warning at -Og
264 - nir/ttn: Silence some warnings at -Og
265 - vl: Silence a warning at -Og
266 - gallivm: Silence a warning at -Og
267 - nouveau: Silence some warnings at -Og
268 - nouveau: Silence a warning at -Og
269 - xlib: Fix build regression since 99e25d183d9
270 - gallium/xlib: Partial fix for glXCopySubBufferMESA
271 - mesa: Store depth bounds test bounds as GLclampd
272 - softpipe: Fix depth comparison with float Z formats
273 - softpipe: Implement GL_EXT_depth_bounds_test
274 - docs: Document GL_EXT_depth_bounds_test
275 - zink: Enable GL_EXT_depth_bounds_test
276 - zink: more and better debug printfs
277 - zink: Fix a thinko in instance setup
278 - zink: Wire up ARB_post_depth_coverage
279 - glx: Pull use_x_font out of the context vtable
280 - glx: Pull get_proc_address out of the context vtable
281 - glx: Remove windows' stub {bind,release}_text_image context hooks
282 - glx/drisw: Implement WaitX and WaitGL
283 - dri: Explicitly handle all the config attributes
284 - dri: Fold attribMap into the code
285 - mesa: Remove misc pbuffer attributes from struct gl_config
286 - mesa: Remove the texture-from-pixmap state from struct gl_config
287 - mesa: Remove transparency state from struct gl_config
288 - mesa: Remove unused gl_config::level
289 - mesa: Remove the pretense of aux buffer support
290 - mesa: Stop tracking visual rating in gl_config
291 - mesa: Remove redundant gl_config::sampleBuffers
292 - ci: Bump the llvmpipe test timeout to 240 seconds
293 - mesa/st: Remove unused ST_ATTACHMENT_SAMPLE
294 - mesa/st: Check for successful framebuffer allocation in st_api_make_current
295 - gallium: Remove curious st_visual::no_config
296 - radeon: Exchange one curious idiom for another in radeonMakeCurrent
297 - mesa: Remove unused _mesa_create_framebuffer
298 - mesa: Make _mesa_initialize_visual return void
299 - mesa: Remove unused gl_config::mutableRenderBuffer
300 - mesa: gl_config::rgbBits should count alphaBits too
301 - dri: Don't tie the accum buffer's alpha-ness to the color buffer's
302 - glx: Stop pretending to validate the pbuffer fbconfig attributes
303 - glx: Don't downgrade the visual caveat from the server
304 - glx: Downgrade aux-buffer-ful fbconfigs
305 - glx: Downgrade tfp mipmap-capable fbconfigs
306 - glx: Downgrade sRGB-ful fbconfigs
307 - dri: Use __DRI_BUFFER_COUNT consistently internally
308 - glx: Default sRGBCapable in the same place as the other config attribs
309 - glx: Clean up fbconfig attribute handling
310 - glx: Remove some #if 0'd DRI config attribute fetch
311 - glx: Don't pointlesly add -D_REENTRANT to libGL's cflags
312 - glx: Move {Bind,Release}TexImage from context to screen vtable
313 - glx: Be more robust against null fbconfigs
314 - glx: Lift sending the MakeCurrent request to top-level code
315 - Revert "glx: Lift sending the MakeCurrent request to top-level code"
316 - gallium/xlib: Fix for recent gl_config changes
320 - docs: docker: minor stale documentation fix
324 - v3dv/pipeline: enable lower_add_sat NIR option
325 - v3d/compiler: enable lower_add_sat NIR option
326 - v3dv/descriptor: assert CrateDescriptorPool receives valid count values
327 - v3dv: drop v3dv_resource definition
328 - v3dv: properly handle two different binding points for cmd_buffers
329 - v3dv: move to subclassing instance/physical device
330 - v3dv: remove reference to v3dv_instance on v3dv_physical_device
331 - v3dv: port to using common dispatch code.
332 - v3dv: support for depthBiasClamp
333 - v3dv/device: clarify that we can't expose textureCompressionBC
334 - v3dv/formats: expose support for BC1-3 compressed formats
335 - v3dv/meta_copy: get tlb compatible BC compressed formats for copies
336 - v3dv/descriptor_set: don't free individual set if not allowed
337 - v3dv: avoid some maybe-uninitialized warnings
338 - v3dv/pipeline_cache: add more details when dumping debug info
339 - v3dv/pipeline: remove pregenerate_variant
340 - v3dv/pipeline: remove pipeline->use_push_constants
341 - broadcom/compiler: add local_size in v3d_compute_prog_data
342 - broadcom/compiler: add driver_location_map at vs prog data
343 - v3dv/pipeline: use driver_location_map instead of nir utilities
344 - v3dv/pipeline: move topology to pipeline
345 - v3dv/pipeline: remove compiled_variant_count field
346 - v3dv/pipeline: remove v3d_key from shader_variant and pipeline stage
347 - v3dv: define broadcom shader stages
348 - v3dv/pipeline: use broadcom_shader_stage as pipeline/variant stage type
349 - v3dv/pipeline: try to get the shader variant directly from the cache
350 - v3dv/pipeline: don't create a variant if compilation failed
351 - v3dv/pipeline: compute sha1 for no-op fragment shaders correctly
352 - v3dv/device: avoid unused-result warning with asprintf
353 - v3dv: Add support for the on-disk shader cache
354 - v3dv/cmd_buffer: return early for draw commands if there is nothing to draw
355 - v3dv: define a default attribute values with float type
356 - vulkan: track number of bindings instead of max binding for CreateDescriptorSetLayout
357 - v3dv/device: do not compute per-pipeline limits multiplying per-stage
358 - v3dv/device: fix and cleanup v3dv limits
359 - v3dv/pipeline: reduce descriptor_map size
363 - frontends/va/image: Eliminate repetitive code on error paths
367 - mesa: texparam: Add a clamping macro to handle out-of-range floats returned as integers.
371 - egl/haiku: Fix ConfigID naming inline with mesa
375 - pan/bi: Fix assertion
376 - pan/bi: Pipe scratch_size in from NIR
377 - pan/bi: Fix 64-bit SSBO addresses
378 - pan/bi: Fix RA of node 0
379 - pan/bi: Fix printing of node 0
380 - panfrost: Fix TLS sizing if cores are missing
381 - panfrost: Allow waiting on slots 6/7 during preload
382 - pan/bi: Add internal debug flag
383 - pan/bi: Validate format 12 tuple count in disasm
384 - pan/bi: Print FAU index in verbose mode
385 - pan/bi: Refactor PC-relative printing
386 - pan/bi: Lint for infinite loops
387 - pan/bi: Print disasm/stats with DEBUG=internal
388 - pan/bi: Fix IDLE register mode packing
389 - pan/bi: Fix staging register packing
390 - pan/bi: Fix dependency wait calculation
391 - pan/bi: Fix M1/M2 decoding in disassembler
392 - pan/bi: Pull out bi_count_read_registers helper
393 - pan/bi: Move bi_next_clause to bir.c
394 - pan/bi: Pass through wait_{6, 7} flags
395 - pan/bi: Add dead branch elimination pass
396 - pan/bi: Add "soft" mode to DCE
397 - pan/bi: Add bi_{before,after}_clause cursors
398 - pan/bi: Add bi_foreach_clause_in_block_rev
399 - pan/bi: Add bi_foreach_instr_in_tuple helper
400 - pan/bi: Add bi_foreach_instr_in_clause iterators
401 - pan/bi: Add destination iterator macro
402 - pan/bi: Don't open code bi_foreach_dest
403 - pan/bi: Permit multiple destinations in RA
404 - pan/bi: Add interference per clause
405 - pan/bi: Implement spilling at the clause-level
406 - pan/bi: Don't fill garbage
407 - pan/bi: Add CUBEFACE pseudoinstruction
408 - pan/bi: Print multiple destinations if needed
409 - pan/bi: Move init_builder to common code
410 - pan/bi: Add "word equivalence" relation for index
411 - pan/bi: Stub out scheduler unit test
412 - pan/bi: Factor nir_function_impl out of the context
413 - pan/bi: Add bi_can_{fma, add} predicates
414 - pan/bi: Annotate ISA.xml with 'last' parameter
415 - pan/bi: Pipe last flag into opcode tables
416 - pan/bi: Add bi_must_last predicate
417 - pan/bi: Add bi_must_message predicate
418 - pan/bi: Label table instructions
419 - pan/bi: Emit branch and table bits in opcode table
420 - pan/bi: Add various read predicates
421 - pan/bi: Unit test bi_can_{fma, add}
422 - pan/bi: Test bi_must_last
423 - pan/bi: Test bi_must_message
424 - pan/bi: Test read predicates
425 - pan/bi: Move bi_constants to bifrost.h
426 - pan/bi: Use canonical terminology for tuple
427 - pan/bi: Use enum bifrost_message_type
428 - pan/bi: Clarify tuple comment
429 - pan/bi: Amend misleading comment
430 - pan/bi: Pack multiple tuples in-memory
431 - pan/bi: Add clause encodings as a table
432 - pan/bi: Move bi_packed_tuple to compiler.h
433 - pan/bi: Add bi_pack_literal
434 - pan/bi: Add bi_pack_upper
435 - pan/bi: Add bi_pack_tuple_bits
436 - pan/bi: Add bi_pack_sync
437 - pan/bi: Add tuple/embedded constant pack
438 - pan/bi: Add subword 5/6 pack
439 - pan/bi: Add subword 4 or 7 pack
440 - pan/bi: Add pack_format helper
441 - pan/bi: Calculate pos for constant packing
442 - pan/bi: Pack multiple tuples per clause
443 - pan/bi: Add packing unit test group
444 - pan/bi: Test pack_literal
445 - pan/bi: Test pack_upper
446 - pan/bi: Test pack_tuple_bits
447 - pan/bi: Test pack_sync
448 - pan/bi: Add packing format tests
449 - pan/decode: Be explicit when printing invocations
450 - pan/decode: Remove tiler size checks
451 - pan/decode: Remove dependency of decoder on the encoder
452 - pan/decode: Deduplicate SFBD blend printing
453 - pan/decode: Deduplicate shader property printing
454 - pan/decode: Remove unused MEMORY_PROP macro
455 - pan/decode: Simplify tiler printing
456 - pan/decode: Remove pandecode_prop
457 - pan/decode: Remove unused disasm stats
458 - pan/decode: Remove mesa header dependencies
459 - pan/mdg: Drop unused stage parameter to disassembler
460 - pan/decode: Remove tile range validation
461 - pan/decode: Prefer sizeof to ARRAY_SIZE for char
462 - nir/lower_io: Fix grammar errors
463 - pan/bi: Fix NULL deref with empty shader
464 - pan/bi: Add side_effects helper
465 - pan/bi: Respect side effects in DCE
466 - pan/bi: Implement AXCHG
467 - pan/bi: Implement ACMPXCHG
468 - pan/bi: Add bi_fmul_f32 convenience method
469 - pan/bi: Fix FLOG_TABLE modifier handling
470 - pan/bi: Lower frcp to Newton-Raphson
471 - pan/bi: Lower frsq to Newton-Raphson
472 - pan/bi: Lower FEXP2 with a table
473 - pan/bi: Lower flog2 to a table and polynomial
474 - pan/bi: Rename NO_FP32_TRANSCENDENTALS quirk
475 - pan/bi: Fix bi quirks detection
476 - pan/bi: Lower FP32 transcendentals where required
477 - pan/bi: Lower transcendentals on G71
478 - pan/bi: Print program size in shader-db
479 - pan/bi: Cleanup terminal block check
480 - pan/bi: Dead code eliminate per-channel
481 - pan/bi: Include ATEST datum in the instruction
482 - pan/bi: Add scheduler data structures
483 - pan/bi: Add cubeface lowering
484 - pan/bi: Flatten block lists
485 - pan/bi: Stub worklist routines
486 - pan/bi: Add constant count estimates to scheduler
487 - pan/bi: Add FAU update helper
488 - pan/bi: Add bi_tuple_is_new_src
489 - pan/bi: Add bi_count_succ_reads helper
490 - pan/bi: Validate reads_t
491 - pan/bi: Add T0/T1 constraint check
492 - pan/bi: Add writes_reg predicate
493 - pan/bi: Add bi_instr_schedulable predicate
494 - pan/bi: Choose instructions to schedule
495 - pan/bi: Destructively schedule a single instruction
496 - pan/bi: Add passthrough register rewriting helper
497 - pan/bi: Extract bi_ec0_packed helper
498 - pan/bi: Add bi_foreach_instr_and_src_in_tuple
499 - pan/bi: Move bi_constant_field to bifrost.h
500 - pan/bi: Add pcrel_idx to bi_clause
501 - pan/bi: Derive M0 from pcrel_idx while packing
502 - pan/bi: Add trivial rewrite helpers
503 - pan/bi: Add constant to passthrough rewrite
504 - pan/bi: Add constant state constructor
505 - pan/bi: Add constant merging routines
506 - pan/bi: Add constant modifier handling
507 - pan/bi: Schedule blocks
508 - pan/bi: Switch to new scheduler
509 - pan/bi: Remove old FAU assignment code
510 - pan/bi: Remove older cube map lowering
511 - pan/bi: Add nosched debug option
512 - pan/bi: Fix 'last tuple' for terminal-NOP clauses
513 - pan/bi: Fix 2-write pseudo op scheduling
514 - pan/mdg: Fix multithreaded shader-db
515 - pan/mdg: Add MIDGARD_MESA_DEBUG=inorder option
516 - pan/mdg: Optimize UBO offset calculations
517 - pan/mdg: Set lower_uniforms_to_ubo
518 - panfrost: Fix race condition in UBO mapping to CPU
519 - panfrost: Respect buffer_offset when mapping to CPU
520 - panfrost: Move sysvals to dedicated UBO
521 - panfrost: Don't truncate uniform_count
522 - panfrost: Add UBO push data structure
523 - panfrost: Push uniforms required by the program
524 - panfrost: Set FAU count based on program->push
525 - panfrost: Don't store uniform_count on Midgard
526 - pan/mdg: Update UBO promotion comment
527 - pan/mdg: Push uniforms based on UBO analysis
528 - pan/bi: Fix multithreaded shader-db
529 - pan/bi: Add bi_replace_index helper
530 - pan/bi: Add bi_is_ssa helper
531 - pan/bi: Print FAU uniforms in IR
532 - pan/bi: Generalize bi_update_fau with fast zero
533 - pan/bi: Handle modifiers in rewrite_fau_to_pass
534 - pan/bi: Rework FAU lowering
535 - pan/bi: Simplify derivative lowering
536 - pan/bi: Add SSA-based scalar copy propagation
537 - pan/bi: Push UBOs on Bifrost
538 - panfrost: Enable ES3 conformant floating-point
539 - compiler, nir: Add and set barrier metadata
540 - panfrost: Set barriers flag for compute shaders
541 - panfrost: Pass is_blit flag around
542 - pan/bi: Skip ATEST for colour blit shaders
543 - panfrost: Fake shader images for bifrost+deqp
544 - pan/bi: Fix jumps to terminal block again
545 - pan/bi: Fix empty shader handling
546 - nir: Add sample_positions_pan intrinsic
547 - pan/decode: Cleanup sample locations decode
548 - pan/decode: Only print local storage for vertex jobs
549 - panfrost: Preload sample mask if needed
550 - panfrost: Add sample positions sysval
551 - panfrost: Push sample positions sysval for Midgard
552 - panfrost: Refactor sample shading state
553 - panfrost: Respect info.fs.uses_sample_shading
554 - panfrost: Add panfrost_sample_pattern helper
555 - panfrost: Set tiler descriptor sampler pattern
556 - panfrost: Generalize MSAA handling
557 - panfrost: Don't set REQ_MSAA in pan_mfbd
558 - panfrost: Don't use PAN_REQ_MSAA in SFBD
559 - panfrost: Remove PAN_REQ_MSAA
560 - panfrost: Remove PAN_REQ_DEPTH_WRITE
561 - panfrost: Remove batch_is_scanout
562 - panfrost: Set sample count/pattern for tiler FBD
563 - panfrost: Upload sample positions on device init
564 - panfrost: Use sample location LUT
565 - panfrost: Ensure open_device has pandecode initialized
566 - panfrost: Advertise MSAA 8x and 16x
567 - panfrost: Implement get_sample_position
568 - panfrost: Overhaul sysval handling
569 - panfrost: Add MULTISAMPLED sysval
570 - pan/mdg: Assert on bad 64-bit swizzle in disassembly
571 - pan/mdg: Remove unused pack_unorm_4x8 lowering
572 - pan/mdg: Lower bitfield instructions
573 - pan/mdg: Rename bitcount8 to popcnt, fixing the unit
574 - pan/mdg: Implement uclz
575 - pan/mdg: Lower ufind_msb, poorly
576 - pan/mdg: Stub load_barycentric_sample
577 - pan/mdg: Lower stores from helpers
578 - pan/bi: Remove redundant TEXC opcode check
579 - pan/bi: Implement texture gathers
580 - pan/bi: Lower bitfield inserts/extracts
581 - pan/bi: Implement bitfield_reverse
582 - pan/bi: Implement ufind_msb
583 - pan/bi: Lower ifind_msb
584 - pan/bi: Fix gl_SampleID read
585 - pan/bi: Implement load_sample_mask_in
586 - pan/bi: Implement nir_intrinsic_load_sample_positions_pan
587 - pan/bi: Decouple sysval loading from NIR
588 - pan/bi: Implement coverage mask updates
589 - pan/{mdg, bi}: Lower load_helper_invocation
590 - pan/{mdg, bi}: Lower load_sample_pos
591 - panfrost: Simplify bind_compute_state
592 - panfrost: Remove stale TODOs
593 - panfrost: Assert on indirect compute shaders
594 - panfrost: Advertise SAMPLE_SHADING
595 - panfrost: Bump max SSBO count
596 - panfrost: Bump advertised ESSL feature level
597 - panfrost: Advertise OES_standard_derivatives
598 - docs/features: Mark more TBO exts done on panfrost
599 - docs/features: Mark some ES3.1 done on Panfrost
600 - docs/features: Mark sample shading done on Panfrost
601 - docs/features: gl_HelperInvocation on Panfrost
602 - nir: Fix grammar error
603 - panfrost: Fix uniform_count on Midgard
604 - pan/bi: Stub scoreboarding
605 - pan/bi: Implement barriers
606 - pan/decode: Fix tiler printing on Bifrost
607 - pan/decode: Pretty print 22-bit pixel formats
608 - pan/decode: Disambiguate border colours
609 - panfrost: Label groups in GenXML
610 - panfrost: Track coverage, early fragment tests
611 - panfrost: Flesh out pixel kill / zs update
612 - panfrost: Handle PIPE_FORMAT_NONE as 'constant'
613 - panfrost: Flesh out allow_forward_pixel_to_kill check
614 - panfrost: Export bifrost_get_blend_desc with type size
615 - panfrost: Add RT conversion sysval
616 - panfrost: Fix NULL deref for an empty frag shader
617 - panfrost: Spell fix
618 - panfrost: Take panfrost_dev for AFBC selection
619 - panfrost: Set border colour on Bifrost sampler
620 - panfrost: Remove useless check
621 - pan/bi: Fix RA of node 0 again
622 - pan/bi: Don't inline 64-bit constants
623 - pan/bi: Fix LD_GCLK staging count
624 - pan/bi: Don't read alpha out of bounds
625 - pan/bi: Allow @rNULL with tied operands
626 - pan/bi: Add SEG_ADD.i64 pseudoinstruction
627 - pan/bi: Add 32-bit atomic pseudoinstruction
628 - pan/bi: Lower atomic pseudo-instructions
629 - pan/bi: Extract bi_atom_opc from NIR intrinsic
630 - pan/bi: Add ATOM_C1 promotion check
631 - pan/bi: Handle computational atomics
632 - pan/bi: Materialize \*DTSEL_IMM in the scheduler
633 - pan/bi: Implement image_atomic_exchange
634 - pan/bi: Implement image_atomic_comp_swap
635 - pan/bi: Implement shader_clock intrinsic
636 - pan/bi: Fix blend shaders using LD_TILE with MRT
637 - pan/bi: Flesh out LD_TILE emit
638 - pan/bi: Elucidate slot 6/7 operation
639 - pan/bi: Preload sample ID for sample shading
640 - pan/bi: Implement interpolateAtSample
641 - pan/bi: Add imm_f16 helper
642 - pan/bi: Implement interpolateAtOffset
643 - pan/bi: Allow dynamically uniform tex indices
644 - pan/bi: Use explicit move even for RT#0 of MRT
645 - panfrost: Comment on state of ARB_shader_clock
646 - panfrost: Advertise FRAMEBUFFER_NO_ATTACHMENTS
647 - docs/features: Mark atomics/images done on Bifrost
648 - panfrost/ci: Mark flaky test
649 - gallium/tessellator: Remove XBOX 360 code
650 - gallium/tessellator: Remove unused includes
651 - gallium/tessellator: Rename D3D11 defines
652 - pan/bi: Fix more jumps to terminal blocks
653 - pan/bi: Optimize out redundant jumps to #0x0
654 - pan/bi: Fix elimination of repeated branches
655 - panfrost: Fix infinite loop spilling
656 - panfrost: Fix NULL dereference adding cbuf to batch
657 - panfrost: Remove redundant NULL check
658 - panfrost: Fix NULL deref in pan_mfbd.c
659 - panfrost: Fix NULL derefs in pan_cmdstream.c
660 - panfrost: Fix NULL deref in pan_sfbd
661 - panfrost: Raise TEXTURE_BUFFER_OFFSET_ALIGNMENT
662 - panfrost: Hide MSAA 8x/16x support
663 - panfrost: Fix UNORM 16 rendering
664 - panfrost: Reinterpret format for reload blits
665 - panfrost: Fix typo in midgard.xml
666 - panfrost: Don't advertise OES_copy_image
667 - pan/bi: Use nir_opt_sink/move for constants
668 - pan/bi: Reduce liveness calculations in DCE
669 - pan/bi: Inline \`bytemask of read components`
670 - pan/bi: Mark branches as having side effects
671 - pan/bi: Mark DISCARD as having side effects
672 - pan/bi: Make bi_writemask take a destination
673 - pan/bi: Allow spilling with multiple destinations
674 - pan/bi: Annotate instructions by destination count
675 - pan/bi: Adapt builder to dest count
676 - pan/bi: Remove unused definitions
677 - pan/bi: Do copyprop in linear-time
678 - panfrost/lcra: Fix constraint counting
679 - pan/bi: Use replace_index in more places
680 - pan/bi: Allow negating constants
681 - pan/bi: Implement fsin/fcos
682 - pan/bi: Mark message-passing sources/dests live
683 - pan/bi: Set clause_state.message conservatively
684 - pan/bi: Treat +DISCARD.f32 as message-passing
685 - nir/lower_viewport_transform: Allow geom/tess
686 - pan/bi: Implement u{add, sub}_sat
687 - nir: Unify memory atomics
688 - meson: Remove kmsro from gallium-drivers
689 - pan/bi: Document register conventions
690 - pan/bi: Add bi_foreach_block_rev
691 - pan/bi: Handle 16-bit blend sr_count
692 - pan/bi: Only run copyprop once
693 - pan/bi: Only run DCE once
694 - pan/bi: Remove TODO: RA warnings
695 - pan/bi: Remove stale todo/assert
696 - panfrost: Deduplicate UBO count assignment
697 - panfrost: Deduplicate Bifrost fau_count
698 - panfrost: Only check blend work count on midgard
699 - vulkan: Deduplicate mesa stage conversion
700 - pan/bi: Enable all nir_opt_move/sink optimizations
701 - pan/bi: Split writemasks for memory stores
702 - pan/bi: Lower large arrays to scratch
703 - pan/bi: Add bi_swz_16 helper
704 - pan/bi: Optimize MKVEC.v2i16 generation
705 - pan/bi: Lower swizzles
706 - pan/bi: Fill in some more conversions
707 - pan/bi: Generalize f2i16, f2u16
708 - pan/bi: Remove conversion lowering
709 - nir: Add nir_type_convert
710 - nir: Add {i2f, u2f, f2i, f2u} helpers
711 - nir/lower_idiv: Convert to lower_instructions
712 - nir/lower_idiv: Factor out numer/denom load
713 - nir/lower_idiv: Add 8-bit and 16-bit lowering path
714 - pan/bi: Determine block successors correctly
715 - panfrost: Fix AFBC body_size for shared resources
716 - panfrost: Enable AFBC buffer sharing
717 - nir: Add varying precision linking helper (v2)
718 - docs: Add some notes on building for macOS
719 - panfrost: Fix formats converting uninit from AFBC
723 - v3dv: Output a message if file open fails in physical_device_init
727 - ci: recover tracie dashboard URLs for failing traces
728 - ci: correct the trace image URLs in the piglit summary
729 - ci: make piglit runner less noisy and show a better failure message
730 - ci: clean paths used in the piglit runner
731 - ci: correct piglit's HTML summary location for artifacts upload
732 - ci: make sure piglit's artifacts are not overwritten
733 - ci: correct artifacts location for piglit's runner messages
734 - ci: tracie dashboard URLs only in the failure after the testcase
735 - ci: piglit runner colors diff output on failures
736 - ci: remove pytest since we don't need it any more
737 - ci: only install piglit dependencies when installing piglit
738 - ci: build gfxreconstruct v0.9.5
739 - ci: add libdrm to the x86_test-vk container
740 - .mailmap: colapse duplicates for Timothy Arceri
741 - .mailmap: resolve duplicates for Icecream95
742 - .mailmap: resolve duplicates for Christopher Li
743 - .mailmap: resolve duplicates for Emmanuel Vadot
744 - .mailmap: resolve duplicates for Indrajit Das
745 - .mailmap: resolve duplicates for James Xiong
746 - .mailmap: resolve duplicates for Jan Zielinski
747 - .mailmap: resolve duplicates for Lin Johnson
748 - .mailmap: resolve duplicates for Mark Menzynski
749 - .mailmap: resolve duplicates for Matthias Hopf
750 - .mailmap: resolve duplicates for Matthias Lorenz
751 - .mailmap: resolve duplicates for Maya Rashish
752 - .mailmap: resolve duplicates for Mun Gwan-gyeong
753 - .mailmap: resolve duplicates for Satyeshwar Singh
754 - .mailmap: resolve duplicates for Yogesh Mohan Marimuthu
755 - .mailmap: add an alias for Eleni Maria Stea
759 - util: add mesa_glthread for Half Life 2 and Black Mesa.
763 - st/mesa: fix pbo upload/download for arrays of textures with only 1 layer
764 - iris: don't emit IRIS_DIRTY_VF depending on trash in restart_index
765 - mesa: ensure parameter list capacity before associating uniform storage
766 - glsl/linker: Fix xfb stride alignment for buffers containing 64bit types
767 - gitlab-ci: remove fixed tests
768 - spirv: repair ssa defs for switchs with only default case
769 - nir/spirv: remove unused fields from \`vtn_builder`
773 - zink: check shader stencil output
774 - zink: support stencil-export
775 - zink: fix destroy batch
776 - ci: Use lock file to build deqp-runner
780 - intel/anv: Fix condition to set MipModeFilter for YUV surface
781 - intel/anv: Fix condition for planar yuv surface
782 - intel: Rename files with gen\_ prefix in common code to intel\_
783 - intel: Rename "gen\_" prefix used in common code to "intel\_"
784 - intel: Fix broken alignment due to gen\_ prefix renaming
785 - intel: Rename "GEN\_" prefix used in common code to "INTEL\_"
786 - i965: Remove blank line at EOF
787 - i965: Rename files with "intel\_" prefix to "brw\_"
788 - intel/isl: Drop intel\_ prefix in function names
789 - anv: Remove redundant #if checks
790 - intel: Remove GEN_IS_HASWELL macro
791 - intel: Simplify version checks involving haswell
792 - intel: Remove GEN_IS_G4X macro
793 - intel: Simplify few version checks involving G4X
794 - intel: Rename GEN_VERSIONx10 macro to GFX_VERx10
795 - intel: Rename GEN_GEN macro to GFX_VER
796 - intel: Rename ISL_DEV_GEN to ISL_GFX_VER
797 - intel: Rename genx10 field in gen_device_info struct to verx10
798 - intel: Rename gen field in gen_device_info struct to ver
799 - intel: Rename genx keyword in filenames to gfxx
800 - intel: Rename GENx prefix in macros to GFXx in build files
801 - intel: Rename GENx prefix in macros to GFXx in source files
802 - intel: Rename genx keyword to gfxx in build files
803 - intel: Rename genx keyword to gfxx in source files
804 - intel: Rename Genx keyword to Gfxx
805 - intel: Rename GENx keyword to GFXx
806 - intel: Rename IS_GEN* macros to IS_GFX_VER*
807 - intel: Make line wrapping changes due to IS_GFX_VER_BETWEEN
808 - intel: Remove unused MAKE_GEN macro
809 - intel: Rename GEN_{ALL, LT, ..} macros to GFX_{ALL, LT, ..}
810 - intel: Rename GEN:BUG:### to Wa_###
811 - intel: Rename WA_### to Wa_###
813 Arcady Goldmints-Orlov (14):
815 - v3dv: Fix uninitialized variable warnings
816 - nir: add more intrinsics to divergence analysis
817 - nir: handle v3d intrinsics in divergence analysis
818 - nir: store the results of divergence analysis on loops
819 - broadcom/compiler: Use ANYA for branches in uniform ifs
820 - broadcom/compiler: Emit uniform loops using uniform control flow
821 - broadcom/compiler: Enable PER_QUAD TMU access only in uniform control flow
822 - v3dv: Only lower local arrays of size up to 2 to if-chains
823 - broadcom/compiler: improve generation of if conditions
824 - Revert "broadcom/compiler: improve generation of if conditions"
825 - v3dv: initialize render_fd at the top of physical_device_init
826 - broadcom/compiler: Add a v3d_compile argument to vir_set_[pu]f
827 - broadcom/compiler: Skip bool_to_cond where possible
828 - broadcom/compiler: Merge instructions more efficiently
832 - lima/ppir: increase usage of pipeline regs
836 - st/nine: Reduce system memory allocated by D3DUSAGE_AUTOGENMIPMAP
837 - st/nine: Do not allow depth buffer render targets
838 - st/nine: Clamp GetAvailableTextureMem
839 - st/nine: Unmap buffers after full unlock
840 - st/nine: Track formats compatible with FETCH4
841 - st/nine: Implement experimental FETCH4
842 - st/nine: Enable DF24 support
843 - st/nine: Add new debug and error checks
844 - st/nine: Refactor ht_guid_delete
845 - st/nine: Protect \*PrivateData also for Volumes
846 - st/nine: Fix leak at device destruction
847 - driconf: Rename csmt_int back to csmt_force
848 - st/nine: Simplify checks for driconf options
849 - st/nine: Add new function to know if we are the worker
850 - st/nine: Add RAM memory manager for textures
851 - st/nine: Use the texture memory helper
852 - st/nine: Control the memfd virtual limit
853 - st/nine: Add driconf option to limit texture memory
854 - st/nine: Set default dynamic_texture_workaround to true
855 - st/nine: Check memfd_create support
856 - st/nine: Fix compilation issue in nine_debug
857 - st/nine: Optimize EndScene
858 - st/nine: Implement SYSTEMMEM buffers same as MANAGED
859 - st/nine: Refactor DrawPrimitiveUp
860 - st/nine: Optimize DrawPrimitiveUp
861 - st/nine: Use correct bind flag at buffer creation
862 - gallium/util: Add new u_box helpers
863 - st/nine: Track pending MANAGED buffer uploads
864 - st/nine: Optimize dynamic systemmem buffers
865 - st/nine: Force DYNAMIC SYSTEMMEM for sw vertex processing
866 - st/nine: Always use DYNAMIC with SYSTEMMEM
867 - st/nine: Use stream_uploader for bad cases of systemmem
868 - st/nine: detect worker threads syncs for systemmem
869 - radeonsi: Limit the size of the in-memory shader cache
870 - radeonsi: fix leak when the in-memory cache is full
871 - st/nine: Disable fpu exceptions during init
872 - st/nine: Fix crash on texture creation failure
873 - st/nine: Fix cubetexture early destruction
874 - st/nine: Add missing breaks
875 - st/nine: Fix invalid NULL check
876 - st/nine: Prevent use after free on dtor
877 - st/nine: Fix reading invalid pointer
878 - st/nine: Fix compilation warnings
879 - st/nine: Fix read outside bounds for some textures
880 - st/nine: Fix value of pipe_draw_info's max_index vertex
881 - st/nine: Prevent negative reference count
882 - st/nine: Improve Surface GetContainer
883 - st/nine: Fix alpha to coverage states
884 - st/nine: Enable multisampling also without depth buffer
885 - st/nine: Handle D3DFMT_NULL multisampling
886 - st/nine: Remove errors on unsupported lock flags
887 - st/nine: Increase number of constants of vs1_sw
888 - st/nine: Ignore swizzle on samplers
889 - st/nine: Clamp max_anisotropy
890 - st/nine: Refuse depth buffers as rendertargets
891 - st/nine: Fix ps ff BLENDTEXTUREALPHA
892 - st/nine: Fix ff has_aNrm computation
893 - st/nine: Catch redundant scissor and viewport settings
894 - st/nine: Pseudo implement set/getClipstatus
895 - st/nine: Improve Reset on Ex devices
896 - st/nine: Pseudo implement Create*Ex functions
897 - st/nine: Complete \*Ex stubs
898 - st/nine: Add logging to Ex function
899 - st/nine: Have NOOVERWRITE win over DISCARD
900 - st/nine: Do not memset buffers twice
901 - st/nine: Add fallback for YUV formats
902 - st/nine: Use PIPE_MAP_ONCE for persistent buffers
903 - st/nine: Disable buffer_upload when csmt is off
904 - st/nine: Allow to override the vram size
905 - st/nine: Make it optional to use a sw renderer
906 - st/nine: Lower texture_memory_limit default
907 - st/nine: Bump num of backbuffers for tearfree thread_submit
908 - st/nine: Improve performance with thread_submit
909 - st/nine: Default thread_submit to true
910 - st/nine: Default tearfree_discard to true
911 - st/nine: Fix compilation error on non-x86 platforms
915 - ac/surface: Fix GFX9 sparse mip info.
916 - radv: Do not use a pipe offset for aliased sparse images.
917 - radv: Add a trivial implementation of VK_KHR_deferred_host_operation
918 - radv: Use stricter HW resolve swizzle compat check.
919 - radv: Expose VK_KHR_workgroup_memory_explicit_layout.
920 - radv: Do not hash vk_object_base in descriptor set layout.
921 - amd/common: Add modifier size helper.
922 - radv: Extract DCC format support handling.
923 - radv: Use the surface offset from ac_surface instead of a plane offset.
924 - radv: Don't relayout images with modifiers.
925 - radv: Add format modifier format queries.
926 - radv: Add drm format modifier queries.
927 - radv: Add image layout with drm format modifiers.
928 - radv: Enable DRM format modifiers on GFX9+.
929 - radv: Enable modifiers with the WSI.
930 - radv: Add modifier fails for CTS bug.
931 - radv: Fix assert.
932 - radv: Implement VK_KHR_zero_initialize_workgroup_memory.
933 - radv: Improve spilling on discrete GPUs.
934 - radv: Fix vram override with fully visible VRAM.
935 - radv: Remove custom icd json generation.
936 - radv: Define supported extensions in C.
937 - radv: Ignore WC flags for VRAM.
938 - radv: Determine swizzles correctly.
939 - radv: Add plane width/height helpers.
940 - radv: Use u_format helpers when possible.
941 - radv: Remove VK_SWIZZLE_*.
942 - radv: Do no use vk_format for getting divisors.
943 - radv: Do not use generated table for plane formats.
944 - radv: Stop checking for MULTIPLANE layout.
945 - radv: Stop using plane_count.
946 - radv: Only support format with a PIPE_FORMAT.
947 - radv: Start using util_format_description for everything.
948 - radv: Remove the format table.
949 - radv: Remove vk_format_has_stencil/depth helpers.
950 - radv: Properly handle modifier import failure.
951 - radv: Do pipe misalignment check per plane.
952 - radv: Don't use dedicated memory info to indicate sharing.
953 - vulkan/device_select: Stop using device properties 2.
954 - amd/common: constify ac_surface_set_umd_metata.
955 - radv: Handle UMD metadata on import.
956 - radv: Use shared code for setting opaque metadata.
957 - amd/common: Add retile map size helper.
958 - radv: Implement initialization of displayable DCC.
959 - radv: Implement displayable DCC retiling.
960 - radv: Add DCC info to the metadata.
961 - radv: Use ac_surface DCC settings for shareable images.
962 - radv: Enable displayable DCC.
963 - radv: Disable displayable DCC for GFX8 properly.
964 - ac/rgp: Only report double the prims per clock on GFX10.
965 - radv: Expose robustBufferAccessUpdateAfterBind correctly.
966 - frontends/va: Use correct size for secondary planes.
967 - radv: Enable linear sampling for depth textures.
968 - radv: Add sam option.
969 - radv: Add nodisplaydcc option.
970 - radv: Use correct DCC compressed block size for sampling.
971 - radv: Dedupe winsyses per device.
972 - radv: Allow extra planes for DCC.
973 - radv: Enable sharing with DCC with modifiers.
974 - radv: Ensure we never decompress or FCE read-only textures.
975 - radv: Allow DCC for images with modifiers that are read-only.
976 - radv: Use 8x8 meta compute workgroups.
977 - radv: Enable DCC for image stores on GFX10.
978 - radv: Only set WRITE_COMPRESS_ENABLE on supported HW.
979 - vulkan: Fix descriptor set creation with zero bindings.
980 - lavapipe: Free sorted descriptor array.
981 - zink: Remove initialization of some arrays
982 - zink: Only set the needed number of scissors.
983 - radv: Flush caches for shader read operations.
984 - nir: Fix shader calls with nir_opt_dead_write_vars.
985 - nir: Extract shader_info->cs.shared_size out of union.
986 - nir: Remove nir_shader->shared_size.
987 - nir: Do not reset shared_size in nir_lower_io.
988 - radv: Support DCC without a fast clear value.
989 - radv: Support DCC without DCC/FCE predicates.
990 - radv: Add retiling for foreign queues.
991 - radv: Support DCC modifiers fully.
992 - radv: Add clang-format for AMD code.
993 - radv: Format.
994 - radv: Update editorconfig.
995 - radv: Re-enable retiling.
996 - radv: Refactor cs_domain to be a winsys function.
997 - radv: Use VRAM cmdbuffers in more situations.
998 - radv/winsys: Remove use_local_bos
999 - radv: Fix memory leak on descriptor pool reset with  layout_size=0.
1000 - amd/common: Use cap to test kernel modifier support.
1001 - radv: Only require DRM 3.23.
1005 - glx: Assign unique serial number to GLXBadFBConfig error
1009 - util: When building 'ARM64EC', don't use x64 intrinsics which need to be emulated
1013 - CI: windows: augment the timeout of building the windows container
1014 - CI: windows: split the layers to meet new registry requirements
1015 - CI: windows: Force using LLVM 12
1019 - d3d12: fix for upside-down multisample stencil blit
1023 - panfrost: Don't skip the test with a 4k shader
1024 - panfrost: Fix tiler job injection (again)
1025 - panfrost: Get rid of IS_BIFROST
1026 - panfrost: Don't memset the last attribute buffer entry twice
1027 - panfrost: Only allocate the extra attribute buffer entry on Bifrost
1028 - panfrost: Set attribs and attrib_bufs to NULL when attrib_count = 0
1029 - panfrost: Rename and move pan_render_condition_check()
1030 - panfrost: Use dev->arch where appropriate
1031 - panfrost: Add a panfrost_compile_shader() helper
1032 - panfrost: Update ctx->batch when a fresh batch is requested
1033 - panfrost: Fix a polygon list corruption in the multi-context case
1034 - panfrost: Don't add the tiler BO when it's not accessed
1035 - pan/bi: Add an is_terminal_block() helper
1036 - pan/bi: Make sure we never branch to an non-existing clause
1037 - pan/bi: Add uclz() support
1038 - pan/bi: Support bit_count()
1039 - panfrost: Use panfrost_get_shader_options() in panfrost_build_blit_shader()
1040 - panfrost: Hide backend compiler internals
1041 - panfrost: Prefix shader related helpers with pan_shader\_
1042 - panfrost: Move sysval_to_id out of panfrost_sysvals
1043 - panfrost: Keep the compiler inputs in the context
1044 - panfrost: Move the shader compilation logic out of the gallium driver
1045 - panfrost: Provide a helper to prepare the shader related parts of an RSD
1046 - panfrost: Use the pan_shader_prepare_rsd() helper
1047 - panfrost: Rename pan_blend.h into pan_blend_cso.h
1048 - panfrost: Move the blend lowering code out of the gallium driver
1049 - panfrost: Move the blend logic out of the gallium driver
1050 - Revert "pan/bi: Optimize out redundant jumps to #0x0"
1051 - pan/bi: Move int64 lowering before idiv lowering
1052 - panfrost: Split the direct and indirect draw logic
1053 - panfrost: Add a parameter to suppress next job prefetching
1054 - panfrost: Allow passing an explicit global dependency when queuing a job
1055 - panfrost: Add a pan_section_offset() helper
1056 - panfrost: Move pan_special_varying definition to pan_encoder.h
1057 - pan/bi: Extend the bi_builder to support type variants correctly
1058 - panfrost: Add a knob to disable the UBO -> push constants optimization
1059 - panfrost: Allow passing an explicit UBO index for the sysval UBO
1060 - panfrost: Print the correct UBO size when dumping UBO information
1061 - panfrost: Don't count the special vertex/instance ID attributes on Bifrost
1062 - panfrost: Split the sampler and texture count
1063 - panfrost: Expose panfrost_modifier_to_layout()
1064 - pan/gen_pack: Parse alignment requirements
1065 - panfrost: Specify descriptor alignment requirements
1066 - panfrost: Provide various helpers to simplify descriptor allocation
1067 - panfrost: Define the Surface and Surface-with-stride descriptors
1068 - panfrost: Emit surface descriptors with pan_pack()
1069 - panfrost: Use the descriptor allocators where appropriate
1070 - panfrost: Get rid of panfrost_pool_alloc()
1071 - panfrost: Move the blend shader cache at the device level
1072 - panfrost: Use the blend shader cache attached to the device
1073 - panfrost: Don't reserve space in the color buffer for disabled RTs
1074 - panfrost: Skip disabled RTs when selecting a RT for transaction elimination
1075 - panfrost: Stop including pan_device.h from pan_bo.h
1076 - panfrost: Add helpers to support indirect draws
1077 - panfrost: Prepare things for indirect draws
1078 - panfrost: Hook up indirect draw support
1079 - panfrost: s/panfrost_slice/pan_image_slice_layout/
1080 - panfrost: Move image states out of pan_image_layout
1081 - panfrost: Add a format field to pan_image_layout
1082 - panfrost: Stop passing a depth > 1 when creating 2D textures
1083 - panfrost: Add extra info to the pan_image_layout struct
1084 - panfrost: Split pan_image in two
1085 - panfrost: Add an offset field so we can attach a sub-buffer to an image
1086 - panfrost: Move out-of-band CRC info to pan_image
1087 - panfrost: Move special Z32_S8X24 case out of panfrost_setup_layout()
1088 - panfrost: Add a pan_image_layout_init() helper
1089 - panfrost: Patch the gallium driver to use pan_image_layout_init()
1090 - panfrost: Pass an image view to panfrost_new_texture()
1091 - panfrost: Provide a helper to calculate the polygon list size
1092 - panfrost: Provide a helper to retrieve image surface pointers
1093 - panfrost: Pass a const device to panfrost_sample_positions()
1094 - pan/midg: Use the sampler index passed to the texture instruction
1095 - panfrost: Add various helpers to simplify FB desc emission
1096 - panfrost: Add an helper to emit fragment jobs
1097 - panfrost: Add align info to the draw and draw padding definitions
1098 - panfrost: Add the early ZS pre frame mode
1099 - panfrost: s/pandecode_vertex_tiler_postfix_pre/pandecode_dcd/
1100 - panfrost: Decode pre/post frame DCDs
1101 - panfrost: Extend pan_fb_info to allow passing a tile enable map
1102 - panfrost: Extend pan_fb_info to allow passing pre/post frame DCDs
1103 - panfrost: Always pass a non-NULL screen to set_damage_region()
1104 - panfrost: Create a blitter library to replace the existing preload helpers
1105 - panfrost: Fix partial update
1106 - panfrost: Use the generic preload and FB helpers in the gallium driver
1107 - panfrost: Kill the old tile-buffer preload logic
1108 - panfrost: Pass a tile enable map to avoid reloading untouched tiles
1109 - panfrost: Fix pan_blitter_get_blit_shader()
1110 - panfrost: Don't advertise AFBC mods when the format is not supported
1111 - panfrost: Reserve thread storage descriptor in panfrost_launch_grid()
1112 - panfrost: Fix indirect draws
1113 - panfrost: Fix ZS reloading on Bifrost v6
1117 - frontend/va/image: add pipe flush for vlVaPutImage
1118 - frontends/omx/h265: search entire dpb list
1122 - intel/fs: Separate SLM size calculation from encoding
1123 - nir: Add a data pointer to the callback in nir_remove_dead_variables
1124 - spirv: Don't remove variables used by resource indexing intrinsics
1125 - nir/linking: Remove system_value handling from helper
1126 - compiler: Use util/bitset.h for system_values_read
1127 - ci: Add nouveau chipset 162 to shader-db runs
1128 - vulkan: Update XML and headers to 1.2.168
1129 - spirv: Update headers and metadata from latest Khronos commit
1130 - nir: Two shared memory \*blocks* may alias each other
1131 - spirv: Implement SPV_KHR_workgroup_memory_explicit_layout
1132 - anv: Implement VK_KHR_workgroup_memory_explicit_layout
1133 - spirv: Don't bother counting num_images/num_textures
1134 - spirv: Don't remove dead variables in \`create_library` mode
1135 - spirv: Store SPIR-V version of the module
1136 - spirv: Refactor variable initializer code
1137 - spirv: Recognize zero initializers in Workgroup variables
1138 - nir: Add nir_zero_initialize_shared_memory
1139 - anv: Implement VK_KHR_zero_initialize_workgroup_memory
1140 - spirv: Fail when parsing invalid Initializers
1141 - spirv: Use OpEntryPoint to identify valid I/O variables
1142 - spirv: Count variables \*after* unused ones are removed
1143 - spirv: Skip creating unused variables in SPIR-V >= 1.4
1144 - spirv: Allow variable pointers pointing to an array of blocks
1145 - intel/compiler: Use gl_varying_slot_name_for_stage()
1146 - freedreno/ir3: Use gl_varying_slot_name_for_stage()
1147 - etnaviv: Use gl_varying_slot_name_for_stage()
1148 - st/atifs: Use gl_varying_slot_name_for_stage()
1149 - compiler: Drop now unused gl_varying_slot_name()
1150 - spirv: Reuse nir_is_per_vertex_io()
1151 - spirv: Explicitly break when finished handling SpvDecorationBuiltIn
1152 - spirv: Update a couple of comments in variable handling
1153 - anv: Lower ViewIndex to zero when multiview is disabled
1154 - spirv: Update headers and metadata from latest Khronos commit
1155 - nir: Handle deref_atomic_fadd in a couple of passes
1156 - intel/compiler: Make vue_map parameter const for brw_compile_fs
1157 - intel/compiler: Use a struct for brw_compile_fs parameters
1158 - intel/compiler: Use a struct for brw_compile_vs parameters
1159 - intel/compiler: Refactor the shader INTEL_DEBUG checks
1160 - intel/compiler: Make brw_postprocess_nir take debug_enabled as a parameter
1161 - intel/compiler: Make vec4 generator take debug_enabled as a parameter
1162 - intel/compiler: Make visitors take debug_enabled as a parameter
1163 - intel/compiler: Use INTEL_DEBUG=blorp to dump blorp shaders
1164 - intel/compiler: Use a struct for brw_compile_cs parameters
1168 - anv/image: Replace bo_is_owned with from_gralloc (v2)
1169 - anv/image: Rename anv_image_plane::surface -> primary_surface
1170 - anv/image: Move vkGetImageMemoryRequirements
1171 - anv/image: Drop duplicate 'format' in anv_image_create()
1172 - anv/image: Fix interpretation of 'disjoint'
1173 - anv/android: Fix size check for imported gralloc bo
1174 - anv: Add anv_surface_is_valid()
1175 - anv/image: Clean up anv_GetImageMemoryRequirements2
1176 - anv: Refactor anv_image_get_compression_state_addr
1177 - anv/image: Add anv_image_address()
1178 - blorp/gen12: Don't use aux address if implicit CCS
1179 - anv/image: Make memory layout more explicit
1180 - vulkan: Track dependencies of Python imports
1181 - anv/image: Simplify assertions in anv_image_from_swapchain()
1182 - anv/image: Fix tiling if VkImageSwapchainCreateInfoKHR
1183 - anv/image: In vkCreateDmaBufImageINTEL use modifiers
1184 - anv/image: Check that anv_image is compatible with its modifier
1185 - anv/image: Refactor check_memory_bindings()
1186 - anv/image: Fix cleanup of failed image creation
1187 - anv/image: Add ANV_IMAGE_MEMORY_BINDING_PRIVATE
1188 - anv/image: Fix Vk*ImagePlaneMemory*Info for modifier images
1189 - anv: Move assert in vkGetImageSubresourceLayout
1190 - anv/image: Fix vkGetImageSubresourceLayout for modifier images
1191 - anv: Implement image acquire/release of modifier images
1192 - anv: Declare anv_layout_to_* as pure functions
1193 - anv/image: Add 'offset' param to add_surface()
1194 - anv/image: Support VkImageDrmFormatModifierExplicitCreateInfoEXT
1195 - anv: Enable VK_EXT_image_drm_format_modifier
1196 - anv: Remove vkCreateDmaBufINTEL (v4)
1197 - anv: Drop unused anv_image_create_info::stride
1201 - gallivm: increase size of texture target enum bitfield
1203 Chia-I Wu (38):
1205 - virgl: update headers
1206 - virgl: add support for VIRGL_CAP_V2_UNTYPED_RESOURCE
1207 - targets/libgl-xlib: add support for virgl
1208 - virgl: update headers from virglrenderer
1209 - venus: add driver skeleton
1210 - venus: add generated venus-protocol headers
1211 - venus: add experimental renderers
1212 - venus: add a CS encoder/decoder
1213 - venus: add a ring buffer
1214 - venus: initial support for vkCreateInstance
1215 - venus: initial support for VkPhysicalDevice commands
1216 - venus: initial support for VkDevice commands
1217 - venus: initial support for queue/fence/semaphore
1218 - venus: initial support for VkDeviceMemory commands
1219 - venus: initial support for buffers/images/samplers
1220 - venus: initial support for descriptor sets
1221 - venus: initial support for render pass and fb
1222 - venus: initial support for events and queries
1223 - venus: initial support for module and pipelines
1224 - venus: initial support for command buffers
1225 - venus: advertise extensions promoted to 1.1
1226 - venus: advertise extensions promoted to 1.2
1227 - venus: initial support for transform feedback
1228 - venus: initial support for WSI
1229 - venus: update venus-protocol headers
1230 - venus: prepare for splitting vn_device.[ch]
1231 - venus: split out vn_command_buffer.[ch]
1232 - venus: split out vn_pipeline.[ch]
1233 - venus: split out vn_query_pool.[ch]
1234 - venus: split out vn_render_pass.[ch]
1235 - venus: split out vn_descriptor_set.[ch]
1236 - venus: split out vn_buffer.[ch]
1237 - venus: split out vn_image.[ch]
1238 - venus: split out vn_device_memory.[ch]
1239 - venus: split out vn_queue.[ch]
1240 - venus: include individual venus-protcol headers
1241 - ci: enable venus in some meson build jobs
1242 - venus: check vn_renderer_info::vk_xml_version
1246 - etnaviv: handle NULL views in set_sampler_views
1247 - vc4: add drm-shim
1248 - ci: Update baremetal kernel to 5.11 plus patches
1249 - nir: add load_texture_rect_scaling
1250 - nir: add has_txs flag
1251 - nir/lower_tex: 'txs free' tex_rect lowering
1252 - nir/lower_tex: wider usage of nir_tex_instr_src_index(..)
1253 - gallium: add PIPE_CAP_TEXRECT
1254 - gallium/st: lower rectangle textures if not supported
1255 - ttn: lower rectangle textures if not supported
1256 - etnaviv: nir: support nir_intrinsic_load_texture_rect_scaling
1257 - etnaviv: let st lower rect tex
1258 - vc4: let st lower rect tex
1259 - etnaviv: nir: add ubo lowering pass
1260 - etnaviv: use nir_lower_uniforms_to_ubo(..)
1261 - etnaviv: fix etna_nir_lower_ubo_to_uniform pass
1262 - etnaviv: extend lower ubo tests
1263 - gallium: call util_cpu_detect()
1264 - etnaviv: use nir_lower_idiv(..) before opt loop
1265 - ci/bare-metal: fix fastboot
1266 - etnaviv: etnaviv: put sampler limit determination into own function
1267 - etnaviv: factor out TS state emitting
1268 - etnaviv: add support for NTE
1269 - etnaviv: rename struct members
1270 - ci/bare-metal: no need to use tee
1271 - etnaviv: tell the truth if alpha-test is supported
1275 - nir/lower_tex: Handle sized tex destination types
1276 - freedreno/ir3: Handle sized tex destination types
1277 - ntt: Handle sized tex destination types
1278 - nir/lower_bool: Rewrite dest_type for boolean destinations
1279 - brw/vec4: Don't convert tex dest type to glsl_type
1280 - radv/meta: Use sized types for nir_tex_instr::dest_type
1281 - v3dv/meta: Use sized types for nir_tex_instr::dest_type
1282 - intel/blorp: Use sized types for nir_tex_instr::dest_type
1283 - anv: Use sized types for nir_tex_instr::dest_type
1284 - dxil: Use sized types for nir_tex_instr::dest_type
1285 - panfrost/blit: Use sized types for nir_tex_instr::dest_type
1286 - d3d12/blit: Use sized types for nir_tex_instr::dest_type
1287 - nir: Use sized types for nir_tex_instr::dest_type
1288 - st/mesa: Use sized types for nir_tex_instr::dest_type
1289 - gallium/nir: Use sized types for nir_tex_instr::dest_type
1290 - ttn: Use sized types for nir_tex_instr::dest_type
1291 - st/atifs: Use sized types for nir_tex_instr::dest_type
1292 - glsl/nir: Use sized types for nir_tex_instr::dest_type
1293 - vtn: Use sized types for nir_tex_instr::dest_type
1294 - ptn: Use sized types for nir_tex_instr::dest_type
1295 - nir: Validate nir_tex_instr::dest_type bitsize
1296 - nir/lower_tex: Assume that nir_tex_instr::dest_type is sized
1297 - panfrost: Assume that nir_tex_instr::dest_type is sized
1298 - ir3: Assume that nir_tex_instr::dest_type is sized
1299 - ntt: Assume that nir_tex_instr::dest_type is sized
1300 - freedreno/a6xx: Document threadsize-related fields
1301 - freedreno/cffdec: Use rb trees for tracking buffers
1302 - ir3/parser: Fix parsing of "0.0" in @const line
1303 - freedreno/computerator: Fix example assembly
1304 - ir3/parser: Support labels
1305 - ir3/parser: Add ability to specify branchstack
1306 - freedreno/computerator: Add branching example
1307 - freedreno/computerator: Fix thrsz type
1308 - freedreno/a6xx: Fix compute threadsize type
1309 - freedreno/registers: Handle typed registers with fields
1310 - freedreno/a6xx: Cleanup SP_XS_CTRL_REG0 definitions
1311 - freedreno: Add local_size to ir3_shader_variant
1312 - ir3: Calcuate max_waves and threadsize
1313 - turnip: Use threadsize calculated by ir3
1314 - freedreno: Use threadsize calculated by ir3
1315 - freedreno/computerator: Use threadsize calculated by ir3
1316 - freedreno: Report max_waves in shaderdb output
1317 - freedreno/computerator: Add script for finding reg file size
1318 - util/bitset: Avoid out-of-bounds reads
1319 - freedreno/a3xx: Fix SP_FS_CTRL_REG1_INITIALOUTSTANDING
1320 - ir3/legalize: Fix last input (ss) insertion
1321 - ir3: Fix valid flags for STIB
1322 - ir3/cp_postsched: Set address of uses for relative mov's
1323 - ir3: Don't copy propagate arrays in ir3_cp
1324 - ir3/postsched: Make sure to schedule inputs before kill
1325 - vtn: Handle ZeroExtend/SignExtend image operands
1326 - tu: Expose VK_KHR_spirv_1_4 and VK_EXT_scalar_block_layout
1327 - tu: Expose VK_KHR_relaxed_block_layout
1328 - ir3/sched: Don't penalize uses of already-waited tex/SFU
1329 - ir3/sched: Don't schedule too many tex/SFU instructions
1330 - ir3: Fix list corruption in legalize_block()
1331 - tu: Correctly preserve old push descriptor contents
1332 - ir3: Prevent oob writes to inputs/outputs array
1333 - nir/lower_clip_disable: Fix store writemask
1334 - tu: Fix SP_GS_PRIM_SIZE for large sizes
1335 - ir3/postsched: Fix dependencies for a0.x/p0.x
1339 - ci: Use lavacli from master
1343 - aco: fix VOP3P assembly, VN and validation
1344 - aco/RA: fix subdword operands on VOP3P instructions
1345 - aco: allow constants/literals on every src position for VOP3P
1346 - aco: allow SGPRs on every src position for VOP3P
1347 - aco: change usesModifiers() considering opsel_hi on packed instructions
1348 - aco: create helpers to emit vop3p instructions
1349 - aco: emit packed 16bit instructions
1350 - radv: vectorize 16bit instructions
1351 - aco: simplify multiply-add combining
1352 - aco: optimize packed mul+add to v_pk_fma_f16
1353 - aco: optimize packed clamp
1354 - aco: optimize packed fneg
1355 - aco: optimize v_pk_fma_f16 -> v_pk_fmac_f16 on GFX10
1356 - aco: propagate swizzles when optimizing packed clamp & fma
1357 - aco: remove divergent branches which only jump over very few instructions
1358 - aco/optimizer: don't copy-prop logical phis
1359 - aco/optimizer: don't propagate subdword temps of different size
1360 - aco: generalize subdword constant copy lowering
1361 - aco/validate: validate that p_create_vector operands are aligned unless they are subdword operands
1362 - aco/validate: ensure that Operand and Definition size matches for parallelcopies
1363 - aco/validate: relax subdword restrictions
1364 - aco: propagate temporaries into PSEUDO instructions if it can take it
1365 - aco/optimizer: expand subdword vectors with SGPRs on all generations
1366 - aco/optimizer: convert extract_vector with index 0 into parallelcopies if possible
1367 - radv: don't vectorize shift operations
1368 - aco: fix VCC hint on boolean subgroup operations
1369 - aco: fix nir_intrinsic_ballot with wave32
1370 - aco: fix shared VGPR allocation on RDNA2
1371 - aco: change gpr_alloc_granule to full alignment
1372 - aco: refactor GPR limit calculation
1373 - aco: don't decrease the vgpr_limit when encountering bpermute
1374 - aco: also consider VCC in get_reg_specified()
1375 - aco: check get_reg_specified() on register hints
1376 - aco: don't abort() if disassembly fails
1377 - aco: use VCC as regular SGPR pair on GFX10
1378 - aco: don't create unnecessary exec phi on merge blocks
1379 - aco: handle non-temp phi definitions and operands
1380 - aco: make all exec accesses non-temporaries
1381 - aco: remove dead code for the handling of exec temporaries
1382 - aco: fix assertion in insert_exec_mask pass
1383 - nir: lower load_helper to is_helper if the shader uses demote()
1384 - nir: lower is/load_helper to zero if no helper lanes are needed
1385 - aco: remove special handling of load_helper_invocation
1386 - aco: don't rematerialize exec
1387 - aco: value number VOPC instructions with different exec masks
1388 - aco/value_numbering: use can_eliminate() function to avoid unnecessary hashmap lookups
1389 - aco/optimizer: set VCC hint on new v_cmp_* definitions
1390 - aco/ra: allow VCC on SMEM sbase operand on GFX10+
1391 - .mailmap: fix email for Daniel Schürmann
1392 - aco/ra: split affinity creation into separate function
1393 - aco/ra: split register_file initialization into separate function
1394 - aco/ra: refactor SSA repairing during register allocation
1395 - aco/ra: iterate backwards when coalescing phis
1396 - aco/ra: allow m0 in get_reg_specified()
1397 - aco/ra: remove exec handling for phis
1398 - aco/spill: refactor spill decision taking
1399 - aco/spill: reload spilled exec masks directly to exec
1400 - aco/spill: spill phi constants and exec directly to VGPR
1401 - aco/spill: don't count phis as variable access
1402 - aco/spill: refactor some more spill decision taking
1403 - aco/spill: refactor live-in registerDemand calculation
1404 - aco/spill: use correct next_use_distances at loop header
1405 - aco: lower p_spill with constants correctly
1406 - aco: fix kill flags on phi operands
1407 - aco: add new reindex_ssa() pass
1408 - aco/cssa: rewrite lower_to_cssa pass
1409 - aco/cssa: don't create parallelcopies for constants and exec
1413 - CI: Try really hard to get updated Windows TLS certs
1414 - CI: Trigger Windows builds for llvmpipe & Vulkan changes
1415 - CI: Change LAVA job visibility
1419 - turnip/ir3: handle image load/stores produced by AtomicLoad/Store
1420 - turnip: make GS use correct varyings size from previous stage
1421 - ir3: add debug option to override shader assembly
1422 - freedreno/ir3/parser: add cat7 support
1423 - turnip: don't emit tess consts if they are not used
1424 - freedreno: clamp scissor bounds
1425 - freedreno/a2xx: fix scissors clamp bounds
1426 - turnip: enable inheritedQueries
1427 - turnip: consider HW limit on number of views when apply multipos opt
1428 - turnip: consider tile_max_h when calculating tiling config
1429 - turnip,freedreno/a6xx: tell hw the size of shared mem used by CS
1430 - turnip/ir3: check for bindless IBOs in atomic dests fixup
1431 - turnip: fix leak of tu_shader object during compute pipeline creation
1432 - ir3: prevent duplication of instruction's dependencies
1433 - ir3: make mark_kill_path exit early if instr is already seen
1434 - ir3: disallow moving memory writes over discard
1435 - freedreno/hw: fix populating branch targets in isa_decode pre-pass
1436 - turnip: fix SP_HS_WAVE_INPUT_SIZE value
1437 - freedreno/a5xx: port handling of PIPE_BUFFER textures from a6xx
1438 - ir3: use OPC_GETBUF to get size of sampler buffers
1439 - turnip: lower device index to zero
1440 - turnip: fill VkMemoryDedicatedRequirements
1441 - turnip: set zmode to A6XX_EARLY_Z if FS forces early fragment test
1442 - turnip: implement intrinsic_vulkan_resource_reindex
1443 - ci/freedreno: run freedreno jobs on any change in src/freedreno/
1444 - ir3: fix oob access to regs array for getbuf,getinfo,rgetinfo
1445 - ir3/isa,parser: fix encoding and parsing of bindless s2en SAM
1446 - ir3: match mova1 mnemonic when writing to A1
1447 - freedreno/isa: assert if field's range is out of bitset's range
1448 - ir3: disallow .sat on SEL instructions
1449 - ir3: update info about applicability of saturation modifier
1450 - turnip: expose several already implemented extensions
1451 - nir: add nir_shader_as_str function
1452 - turnip: implement VK_KHR_pipeline_executable_properties
1453 - turnip: clamp to zero negative upper left corner of viewport
1454 - turnip,ir3: account for dispatch group offsets
1455 - freedreno/a6xx: copy full 64bit of primitive counter
1456 - freedreno/a6xx: fix primitive counters debug output
1457 - ir3/isa: account for randomly set by blob lowest bit of ibo atomics
1458 - glsl/linker: Fix attempts to split up 64bit varyings between slots
1459 - glsl/linker: Fix xfb with explicit locations and 64bit types
1460 - ir3: nir_op_f2f16 should round to even
1461 - ir3: convert shift amount to 16b for 16b shifts
1462 - turnip: enable infinities for f16 math and document the register
1463 - turnip: enable VK_KHR_16bit_storage on A650
1464 - turnip: handle format list for compressed formats
1465 - docs: mark float_controls,float16_int8,16bit_storage as done on Turnip
1466 - turnip: fix alignment of non-32b types in workgroup memory
1467 - turnip: implement variableMultisampleRate
1468 - turnip: support copying both aspects of D32_SFLOAT_S8_UINT
1472 - device-select-layer: update for vulkan 1.2
1473 - lavapipe: fix missing piece of VK_KHR_get_physical_device_properties2
1474 - vk-device-select: add device group support
1475 - lavapipe: refactor image surface creation
1476 - lavapipe: rewrite attachment clearing for conditional rendering.
1477 - gallium: add a cond rendering hook for vulkan.
1478 - llvmpipe: handle vulkan conditional rendering
1479 - lavapipe: add VK_EXT_conditional_rendering support.
1480 - CI: add lavapipe to llvmpipe rules.
1481 - lavapipe: add support for external memory/fd/sempahore extensions
1482 - llvmpipe: handle firstvertex for vulkan draw parameters
1483 - lavapipe: handle shader draw parameters
1484 - lavapipe: add missing loader interface negoitation
1485 - lavapipe: move to subclassing instance/physical device.
1486 - lavapipe: add missing wsi entrypoint.
1487 - lavapipe: sort extensions in proper order.
1488 - lavapipe: use common dispatch layer.
1489 - radv: move queue object to a common base object
1490 - radv: remove all entrypoint enabled debug option
1491 - radv: move to subclassed instance/physical_device structs
1492 - radv: port to using common dispatch code.
1493 - zink: don't pick a cpu device ever.
1494 - llvmpipe: add a mutex around debug resource tracking
1495 - llvmpipe: fix use after free with fs variant cleanup
1496 - lavapipe: reset shader constant buffers after execution
1497 - glsl: fix leak in gl_nir_link_uniform_blocks
1498 - llvmpipe: enable GL spir-v support
1499 - util/format: add helper to check if a format is scaled.
1500 - llvmpipe: don't support scaled formats outside vertex buffers
1501 - lavapipe: add support for 2/10/10/10 scaled formats.
1502 - lavapipe: add support for missing 10/10/10/2 formats.
1503 - lavapipe: add reference counting to descriptor set layout
1504 - lavapipe: avoid pointer to pipeline layout in execution
1505 - lavapipe: set viewport state dirty on first execute
1506 - lavapipe: implement physical device group enumeration
1507 - lavapipe/meson: drop megadrivers build req
1508 - lavapipe: fix some void ptr arithmetic
1509 - lavapipe: use msvc compatible 0 init
1510 - lavepipe: some misc msvc fixes
1511 - lavapipe: make OPT macro MSVC compatible
1512 - lavapipe: use os_time for timing related things
1513 - vulkan/util: add api to reset object magic + private data.
1514 - radv: reset object base on recycled command buffers
1515 - tu: reset object base on recycled command buffers
1516 - lavapipe: reset object base on recycled command buffers
1517 - util: add optimised memset64
1518 - u_surface: use optimised memset64
1519 - llvmpipe: zs clear use 64-bit memset
1520 - lavapipe: use clear interface for renderpass clears
1521 - glx: proposed fix for setSwapInterval
1522 - zink: use extensioned draw indirect functions.
1523 - zink/ci: update results now that we are testing zink/lavapipe
1524 - lavapipe: add calibrated timestamp support
1525 - zink/ci: update results for GL 3.3 testing enables
1526 - zink/ci: disable arb_timer_query tests
1527 - lavapipe: use the common icd generator
1528 - lavapipe: fix msvc initialiser
1529 - lavapipe: add dll definition file instead of using PUBLIC
1530 - lavapipe: fix icd generation for windows
1531 - meson/llvm: add native for gallium swrast
1532 - lavapipe: handle tessellation domain winding
1533 - lavapipe: enable KHR_maintenance2
1534 - lavapipe: enable KHR_maintenace3
1535 - lavapipe: fix descriptor set layout freeing.
1536 - lavapipe: fix depth texturing swizzle
1537 - lavapipe: use null probe path on win32
1538 - ci: try building lavapipe on windows
1539 - zink/instance: work with vulkan 1.0 and later loader.
1540 - lavapipe: expose a 1.0 vulkan API for now.
1541 - lavapipe: Define supported extensions in C
1542 - lavapipe: VK_EXT_extended_dynamic_state support
1543 - lavapipe: reorder descriptor set stages to get correct binding
1544 - lavapipe: sort bindings before creating descriptor set
1545 - clover: fix array images view creation
1546 - lavapipe: fix pipeline vp/scissor mixup.
1547 - lavapipe: fix dynamic viewport/scissor pipeline emission
1548 - draw: fix uses viewport index for tess eval shader
1549 - draw/prim_assembler: write correct decomposed primitive lengths
1550 - llvmpipe: add support for shader viewport layer
1551 - lavapipe: enable EXT_shader_viewport_index_layer
1552 - zink/ci: update results after layer extensions enabled in lavapipe
1553 - util/panfrost/glsl: rename BITSET_LAST_BIT to BITSET_LAST_BIT_SIZED
1554 - util/bitset: add a new last bit api
1555 - shader_info: convert textures_used to a bitset.
1556 - gallium: add a sampler reduction cap + settings
1557 - gallium: add a view mask to the draw command
1558 - gallivm: mark subpass input attachments as 2d arrays
1559 - gallivm: add support for load_view_index intrinsic
1560 - draw: add interface to notify renderer of the current view index
1561 - draw: refactor out the instances drawing code
1562 - draw: add view_mask rendering support
1563 - draw: pass the view index to the render driver
1564 - draw/vs: pass the view index to the vertex shader
1565 - draw: add tess/gs support for multiview index
1566 - llvmpipe: add the view index callback from draw
1567 - llvmpipe: add view index support to rasterizer
1568 - lavapipe: add clear support for multiview
1569 - lavapipe: add draw support for multiview
1570 - lavapipe: add input attachment support for multiview
1571 - lavapipe: add render pass support for multiview
1572 - lavapipe: enable KHR_multiview
1573 - llvmpipe: add reduction mode support
1574 - lavapipe: add EXT_sampler_filter_minmax support
1575 - lavapipe: add support for VK_KHR_create_renderpass2
1576 - lavapipe: move queue to base object
1577 - lavapipe: move to the common casting interfaces
1578 - lavapipe: move to common create render pass code
1579 - lavapipe: add single ssbo variable pointer support.
1580 - docs: update lavapipe features.txt
1581 - lavapipe: enable KHR_uniform_buffer_standard_layout
1582 - lavapipe: enable EXT_scalar_block_layout
1583 - lavapipe: add missing break
1584 - lavapipe: fix writing availability for queries.
1585 - lavapipe: add host query reset
1586 - gallivm: convert packing to uint64 not double
1587 - lavapipe: only init immutable samplers for correct types.
1588 - lavapipe: add support for KHR_buffer_device_address.
1589 - lavapipe: bump maxMemoryAllocationCount
1590 - lavapipe: fix image format properties
1591 - lavapipe: add missing sampler minmax properties
1592 - lavapipe: add missing device group api
1593 - lavapipe: drop unused vk_format in image struct
1594 - lavapipe: fix templated descriptor updates
1595 - gallivm: fix non-32bit ubo loads
1596 - gallivm/nir: handle bool registers.
1597 - nir: port fp16 casting code from dxil
1598 - nir: lower 64-bit floats to 32-bit first.
1599 - gallivm: use fp16 casts lowering
1600 - lavapipe: enable 8/16-bit storage extensions
1601 - llvmpipe: fix cube image size query
1602 - st/glthread: allow for invalid L3 cache id.
1603 - util: rework AMD cpu L3 cache affinity code.
1604 - gallivm: add 64-bit atomic support for ssbo/shared.
1605 - gallivm: add 64-bit atomic global support
1606 - lavapipe: enable KHR_shader_atomic_int64
1607 - lavapipe: only reference pCounterBuffers if non-NULL
1608 - lavapipe: fail out if spirv->nir fails
1609 - lavapipe: fix only clearing depth or stencil paths.
1610 - zink/ci: update results after lavapipe clear fixes
1611 - lavapipe: add support for KHR_imageless_framebuffer
1612 - drisw: move zink down the list below the sw drivers.
1613 - zink/ci: handle getting correct drisw driver.
1614 - llvmpipe: when depth clamp is disable clamp to 0.0/1.0
1615 - llvmpipe: always take depth clamping from state tracker
1616 - ci: update zink/virgl results for depth clamping fixes
1617 - lavapipe: add vulkan 1.1 properties/features apis
1618 - lavapipe: fix missing protected memory properties
1619 - gallivm: add subgroup vote 64-bit and feq support.
1620 - gallivm: move get_flt_bld to header.
1621 - gallivm: add subgroup system values support
1622 - gallivm: add subgroup elect intrinsic support.
1623 - gallivm: add subgroup reduction + in/ex scan support
1624 - gallivm: add subgroup ballot support
1625 - gallivm: add subgroup read invocation support
1626 - gallivm: add subgroup lowering support
1627 - gallivm: add compute shader subgroup system values support
1628 - lavapipe: enable subgroups features
1629 - lavapipe: enable correct workgroup sizing
1630 - lavapipe: enable Vulkan 1.1 support
1631 - docs: update lavapipe bits for 1.1
1632 - lavapipe: add vk1.1 image swapchain support
1633 - lavapipe: add dummy sampler ycbcr conversion
1634 - lavapipe: fix mipmapped resolves.
1638 - radv: fix divide by zero with no tesselation params
1642 - gallium/indices: Use "__restrict" to help the compiler
1646 - radeonsi: Report multi-plane formats as unsupported
1650 - VERSION: bump for 21.1.0 cycle
1651 - docs: add release notes for 20.3.3
1652 - docs: Add sha256sum for 20.3.3
1653 - docs: update calendar and link releases notes for 20.3.3
1654 - docs: update calendar for 21.0.0-rc1
1655 - bin/post_version: convert the csv.reader into a concrete list
1656 - docs: add release notes for 20.3.4
1657 - docs: Add sha256sum for 20.3.4
1658 - docs: update calendar and link releases notes for 20.3.4
1659 - docs: update calendar for 21.0.0-rc2
1660 - docs: update calendar for 21.0.0-rc3
1661 - Scons: check for timespec_get on windows as well as unices
1662 - docs: Remove 21.0 features from features_new.txt
1663 - docs: add release notes for 21.0.0
1664 - docs: update calendar and link releases notes for 21.0.0
1665 - docs: Add calendar entries for 21.0 release.
1666 - docs: Extend calendar entries for 21.0 by 1 releases.
1667 - docs: Add calendar entries for 21.1 release candidates.
1668 - docs: add release notes for 20.3.5
1669 - docs: Add hashes for 20.3.5
1670 - docs: update calendar and link releases notes for 20.3.5
1671 - docs: add release notes for 21.0.1
1672 - docs: Add 21.0.1 hashes
1673 - docs: update calendar and link releases notes for 21.0.1
1674 - docs: add release notes for 21.0.2
1675 - relnotes: Add sha256sum for 21.0.2
1676 - docs: update calendar and link releases notes for 21.0.2
1677 - meson: OpenMP is supposed to be optional
1678 - .pick_status.json: Update to ee9b744cb5d1466960e78b1de44ad345590e348c
1679 - VERSION: bump for 21.1.0-rc3
1680 - .pick_status.json: Update to cbd6e5f2e592a9834a03004a473537f25aea4336
1681 - .pick_status.json: Update to ede0b3c643279f4126fb10552a2f1d00be27f16d
1682 - .pick_status.json: Update to b80720acb13e1014aea89e6bd25f22d43df85356
1686 - clover: Implement CL_MEM_OBJECT_IMAGE1D
1690 - anv: Added the VK_EXT_sample_locations extension to the anv_extensions list
1691 - anv: Implement physical device properties for VK_EXT_sample_locations
1692 - anv/state: Take explicit sample locations in emit helpers
1693 - anv: Add support for sample locations
1694 - anv: Removed unused header file
1695 - anv: Enabled the VK_EXT_sample_locations extension
1696 - iris: fix in fences backend for ext_external_objects edge case
1698 Ella-0 (1):
1700 - glsl: build without bison
1704 - microsoft/spirv_to_dxil: Add support for load_vulkan_descriptor
1705 - microsoft/spirv_to_dxil: Use non-zero exit code on failed compilations
1709 - gallium/ttn: Add support for TGSI_OPCODE_I64NEG/ABS.
1710 - gallium/ntt: Stop lowering integer source mods.
1711 - gallium/tgsi: Assert that we don't see integer abs modifiers.
1712 - gallium/tgsi: Remove support for f64 src modifiers.
1713 - gallium/tgsi: Rewrite the docs on source modifiers.
1714 - gallium/tgsi: garbage collect unused TGSI_UTIL_SIGN_MODE.
1715 - mesa/st: Make a single helper for the NIR-to-TGSI transfer.
1716 - mesa/st: Lower shader images before handing off to NIR-to-TGSI.
1717 - mesa/st: Dump nir-to-tgsi output when ST_DEBUG=tgsi or nir is set.
1718 - gallium/ntt: Don't vectorize IBFE/UBFE/BFI.
1719 - gallium/ntt: Add support for store_per_vertex_output.
1720 - gallium/ntt: Avoid referencing undefined channels of system values.
1721 - ci/freedreno: Mark some a5xx separate_shader tests as xfails.
1722 - ci/freedreno: Fix up the xfail/flake handling of a3xx texture functions.
1723 - ci/freedreno: Remove a bunch of stale flakes from a3xx.
1724 - ci/freedreno: Drop some long-unseen a5xx flakes.
1725 - ci/freedreno: Drop skip list stuff from a5xx flakes.
1726 - ci/freedreno: Remove some long-unseen a6xx known flakes.
1727 - util/format: Fix pack/unpack of A1R5G5B5_UINT.
1728 - mesa: Add some little unit tests showing format unpack behavior.
1729 - mesa: Drop incorrect statement about Z unpack behavior.
1730 - mesa: Replace the float[4] unpack code with util/format's.
1731 - mesa: Make _mesa_unpack_rgba_block() use the u_format pack/unpack.
1732 - util: Move most of src/mesa/main/format_utils.h to util/format/
1733 - util: Fix UBSan failure on _mesa_unorm_to_unorm.
1734 - util: Fix rounding of unpack_unorm8 from small unorm formats.
1735 - mesa: Reuse util_format's unpack_8unorm.
1736 - mesa: Reuse util_format's unpack functions for pure integer formats.
1737 - util: Give a reasonable answer when unpacking z32unorm from floats.
1738 - mesa: Use a bunch of util functions for Z/S unpacking.
1739 - mesa: Move the rest of format_unpack.py out of code generation.
1740 - util/format: Simplify the generated unpack code.
1741 - swrast: Use util_format_write_4/4ub for the scattered pixel writes.
1742 - mesa/main: Replace float pack function with util_format_pack_rgba().
1743 - mesa/main: Replace the uint format packing code with util/format's.
1744 - ci/piglit: Upgrade to a newer piglit in our containers.
1745 - ci/freedreno: Fix xfail setup for sampler3d_float_vertex.
1746 - nir_to_tgsi: Store directly to TGSI outputs when possible.
1747 - r300,i915g: Report no shader buffers or images on non-TCL HW.
1748 - nir_to_tgsi: Fix buffer overflow in atomic image compswap.
1749 - swr: Don't report support for shader images.
1750 - panfrost: Stub out set_shader_images().
1751 - gallium: Fix leak of shader images on context destruction.
1752 - util/bitset: Avoid dereferencing the bitset for size == 0.
1753 - ci: Add a fractional deqp run of softpipe with asan enabled.
1754 - freedreno/a6xx: Skip the body of emit_state if we're clean.
1755 - freedreno: Move blend gmem checks to a blend dirty state check.
1756 - freedreno: Move framebuffer state checks under a ctx->dirty flag.
1757 - freedreno: Skip some batch dependency tracking if !ctx->dirty.
1758 - ci/freedreno: Detect cheza HFI errors and restart the run.
1759 - ci/freedreno: Ban vs-clip-vertex-enables which flakes in CI.
1760 - ci/freedreno: Ban more flaky clip-enables tests.
1761 - ci/freedreno: Make a630 piglit_shader run a manual run, too.
1762 - freedreno: Use a real type instead of void * for the fd_batch->key.
1763 - freedreno: Early-out from the resource write path when we're the writer.
1764 - freedreno: Remove duplicate bc invalidate on flush_write_batch().
1765 - ci: Update baremetal kernel to 5.11-rc5 plus patches.
1766 - mesa/st: Allocate the gl_context with 16-byte alignment.
1767 - ci/freedreno: Drop pointless GIT_STRATEGY setting for a630.
1768 - ci/freedreno: Use the new nginx cache for trace downloads.
1769 - ci/freedreno: Use the http cache for artifacts downloads, too.
1770 - ci/docs: Update CI farm requirements suggestions.
1771 - docs/ci: Document setting up the http cache for traces.
1772 - ci/lava+baremetal: Add an xserver to the root fs.
1773 - ci/freedreno: Do our piglit runs against Xorg.
1774 - ci/freedreno: Add Valve games and other traces now that we have GLX.
1775 - freedreno: Make sure that queries are disabled during shadow blits.
1776 - freedreno: rename batch->active_providers to query_providers_used.
1777 - freedreno: Backport a5xx/a6xx fix for active query handling.
1778 - freedreno: Drop pointless clear of used providers.
1779 - freedreno/a6xx: Skip guessing VSC size with indirect TF draw counts.
1780 - docs: Document PIPE_CAP_STREAM_OUTPUT_PAUSE_RESUME.
1781 - freedreno/a6xx: Add support for glDrawTransformFeedback().
1782 - ci/deqp: Bump runner to 0.5.1 for recent runtime perf improvements.
1783 - ci/freedreno: bump VK coverage to 1/4 of the CTS.
1784 - ci/freedreno: Run a3xx gles3 in parallel and increase coverage.
1785 - ci/virgl: Fix GLES31 testing on desktop GL.
1786 - freedreno: Force updating active queries on batch reordering.
1787 - freedreno: Remove FD_STAGE_* in favor of a "disable_all" flag.
1788 - freedreno/a5xx: Don't forget to count our custom blits against queries.
1789 - mesa/st: Always precompile the first shader variant.
1790 - mesa/st: Assume that the default variant is always first in the list.
1791 - vc4: Remove vestiges of alpha test lowering.
1792 - vc4: Stop advertising support for PIPE_CAP_TWO_SIDED_COLOR.
1793 - vc4: Stop advertising support for VS color clamping.
1794 - v3d: Clean up vestiges of alpha test lowering.
1795 - v3d: Stop advertising support for PIPE_CAP_TWO_SIDED_COLOR.
1796 - v3d: Stop advertising support for PIPE_CAP_*_COLOR_CLAMPED.
1797 - v3d: Stop advertising support for flat shading.
1798 - gallium: Document behavior of more lowering pipe caps.
1799 - freedreno: Add missing dep on freedreno tracepoints.
1800 - gallium: Flip the default value of PIPE_CAP_SHAREABLE_SHADERS.
1801 - virgl: Drop a context dependency from part of the shader compile path.
1802 - v3d/qpu: Avoid leaking memory in the QPU disasm test.
1803 - mesa/st: Make sure to unbind cb0 on transition away from gs/tess shaders.
1804 - ci: Allow better customization of the name of the artifacts for minio.
1805 - ci/freedreno: Add a fractional gles31 run with asan enabled.
1806 - ci/freedreno: Drop the "arm64" in front of job names.
1807 - ci: Move specific driver testing to separate files in separate dirs.
1808 - ci/freedreno: Fix a5xx piglit runs.
1809 - ci/freedreno: Remove stray BM_DTB definition.
1810 - ci/bare-metal: Use an upstream kernel for db820c.
1811 - ci/a5xx: Update the piglit expectations.
1812 - ci/a5xx: Increase our dEQP GLES3 fraction by 4x.
1813 - ci: Move the dEQP and traces expectations to the per-driver CI dirs.
1814 - ci: Move the piglit expectations lists to the per-driver CI dirs.
1815 - ci/zink: Add tests of gles2, gles3, and gl33 on lavapipe.
1816 - zink: Use mesa_loge() for should-never-be-reached initialization errors.
1817 - zink: Remove NULL checks after GET_PROC_ADDR_INSTANCE().
1818 - softpipe: Fix the const buffer overflow check.
1819 - mesa: Get the FXT1 compressor/decompressor off of GL types.
1820 - mesa: Move the FXT1 compressor/decompressor to util/
1821 - llvmpipe: Enable FXT1 texture decompression.
1822 - v3d: Replace driver lowering of GL_CLAMP with mesa/st's.
1823 - ci/piglit: Stop including the test counts at the end of expectations.
1824 - ci/iris: Move the traces yml file to the driver-specific dir.
1825 - mesa: Always make sure uniform storage doesn't get reallocated.
1826 - freedreno: Remove uniform variables after finalizing NIR.
1827 - freedreno: Drop custom driver lowering of two-sided color.
1828 - freedreno: Drop custom driver lowering of GL's color clamping.
1829 - freedreno: Use the mesa/st frontend lowering of GL_CLAMP.
1830 - freedreno/a5xx+: Stop recompiling on texture samples changes.
1831 - freedreno/a5xx+: Drop the unused no_decode_srgb flag.
1832 - freedreno/a5xx: Fix cube image load/stores.
1833 - nir: Add a nir_src_is_undef() helper, like nir_src_is_const().
1834 - nir/vec_to_movs: Don't generate MOVs for undef channels.
1835 - ci: Move deqp-default-skips.txt back to .gitlab-ci/
1836 - ci/lava: Move the per-driver gitlab-ci.yml to each driver.
1837 - ci/lava: Move the driver expectation files to the per-driver CI dir.
1838 - tgsi_exec: Roll the loops for condmask handling.
1839 - tgsi_exec: Jump over entirely non-taken THEN or ELSE branches.
1840 - ci/freedreno: Also retest when only CI configuration changes.
1841 - ci/freedreno: Switch the fastboot boards to using nfsroot.
1842 - ci/a5xx: Run all of gles2 in one job.
1843 - ci/a3xx: Run all of GLES3 dEQP.
1844 - ci/a5xx: Increase the gles3/31 coverage.
1845 - ci/a5xx: Update piglit expectations.
1846 - ci/zink: Add another primitive restart flake.
1847 - ci/turnip: Mark a flaky WSI test.
1848 - lima: stop encoding the texture format in the shader key
1849 - lima: don't look at dirty bits for setup of FS key
1850 - lima: upload the shader to a BO at shader creation
1851 - lima: avoid stomping over bound shader state when creating new shaders
1852 - nir-to-tgsi: Fix handling of partial writemasks on SSA/REG decls.
1853 - docs: Add some documentation of game GL buffer object mapping behavior.
1854 - freedreno/a5xx: Introduce an event write helper like a6xx has.
1855 - freedreno/a5xx: Flush depth at the end of sysmem, like a6xx does.
1856 - ci/freedreno: Mark another a5xx TF flake.
1857 - u_format: Mark the generated pack/unpack src/dst args as restrict.
1858 - mesa/st: Unify st_get_vp_variant() and st_get_common_variant().
1859 - mesa/st: Add perf debug for draw-time variant compiles.
1860 - mesa/st: Fix precompile misses on compat GL VSes writing to color outputs.
1861 - virgl: Update GLES expectations.
1862 - ci/freedreno: Add three more a5xx flakes from the last day.
1863 - freedreno/a5xx: Fix the texel buffer alignment requirement.
1864 - freedreno/a5xx: Fix the max texture buffer size.
1865 - ci/panfrost: Disable t860/radeonsi testing while the runners are struggling.
1866 - ci: Bump deqp-runner to v0.6.3.
1867 - ci/freedreno: Switch the piglit testing to the new piglit runner.
1868 - ci/bare-metal: Restart a run on intermittent kernel lockups.
1869 - ci/freedreno: Mark an a630 piglit flake from async shader compiling.
1870 - ci/freedreno: Mark the rest of the glx_arb_sync_control\@timing as flakes.
1871 - nir_to_tgsi: Respect PIPE_SHADER_CAP_TGSI_SQRT_SUPPORTED.
1872 - freedreno/a5xx: Fix stream-output binning handling.
1873 - freedreno/ir3: Demote centroid usage to pixel on non-msaa.
1874 - ci/freedreno: Mark all of dEQP TF as flaky.
1875 - ci/bare-metal: Move the db820c lockup detect to the right boot script.
1876 - ci/freedreno: Mark glx-swap-copy as a flake on a630.
1877 - freedreno/a6xx: Rename the RB_BLIT_INFO.INTEGER field to SAMPLE_0.
1878 - freedreno/a6xx: Disable sample averaging on depth/stencil resolves.
1879 - freedreno: Move the ir3 linked shader cache to the context.
1880 - freedreno/a3xx: Switch to using ir3_cache for looking up our VS/FS.
1881 - freedreno/a4xx: Switch to using ir3_cache for looking up our VS/FS
1882 - freedreno/a5xx: Switch to using ir3_cache for looking up our VS/FS
1883 - turnip: Fix KGSL build since common dispatch rework.
1884 - broadcom: Disbale CLIF dumping when libexpat isn't available.
1885 - ci/android: Make sure we don't detect system libexpat.
1886 - ci/android: Build the v3dv driver.
1887 - iris: Flag for resolves when stencil enable changes, too.
1888 - freedreno: Assert that TF prims generated can ignore active_queries.
1889 - freedreno/ir3: Move max-tf-vtx calculation to a .c file
1890 - freedreno: Move max-tf-vtx calculation to just the HW that needs it.
1891 - freedreno: Move the primitives generated/written updates after the draw.
1892 - freedreno: Don't count SW TF queries on a6xx.
1893 - freedreno: Clamp TF prims written to buffer size pre-a6xx.
1894 - ci/freedreno: Mark a630 as flaky on arb_draw_indirect-transform-feedback
1895 - nir: Update clip_distance_array_size in clip lowering.
1896 - freedreno/a6xx: Use the frontend userclip lowering.
1897 - freedreno/a5xx: Add support for clip distances and use them for userclip.
1898 - freedreno/a5xx: Use VALIDREG/CONDREG like a6xx do.
1899 - ci/freedreno: Demote a630-asan to a manual test for now.
1900 - ci: Drop the custom db820c kernel/dtb from the kernel+rootfs.
1901 - ci/freedreno: Add more new traces for a630 (minetest, TDM, pioneer, glyphy).
1902 - ci/freedreno: Rename a306-test and a530-test to drop "arm64" from the name.
1903 - ci/freedreno: Add trace testing on a3xx, a5xx.
1904 - freedreno/a5xx: Fix alpha test vs early Z bugs.
1905 - freedreno/a6xx: Fix alpha tests.
1906 - ci/freedreno: Switch to the trimmed glxgears trace.
1907 - ci/freedreno: Fix up the a5xx border color flake annotation.
1908 - ci: Uprev deqp runner to 0.6.5.
1909 - ci: Uprev piglit to 6a4be9e9946d ("piglit: NOTE! Default branch is now main")
1910 - ci: bump bare-metal kernel to bring in an a530 stability fix
1911 - freedreno: Fix YUV sampler regression.
1912 - nir_to_tgsi: Use ARL instead of UARL in the !native_integers case.
1913 - nir: Generate load_ubo_vec4 directly for !PIPE_CAP_NATIVE_INTEGERS
1914 - freedreno/a6xx: Don't try to do Z-as-RGBA blits for mismatched formats.
1915 - ci/virgl: Mark a couple of new Crash tests as flakes.
1919 - VERSION: bump to 21.1.0-rc1
1920 - .pick_status.json: Update to c74d93cf0187e07cdfacc448a947a8cae485eb41
1921 - .pick_status.json: Update to 95d9d811c91076d50385b2fbd330335b68688c69
1922 - .pick_status.json: Update to fcb5ba58165cd407408f8dd9a102f0c5e16a9956
1923 - VERSION: bump for 21.1.0-rc2
1924 - .pick_status.json: Mark 8acf361db4190aa5f7c788019d1e42d1df031b81 as denominated
1925 - .pick_status.json: Update to 35a28e038107410bb6a733c51cbd267aa79a4b20
1926 - .pick_status.json: Update to 7e905bd00f32b4fa48689a8e6266b145662cfc48
1927 - .pick_status.json: Update to 72eca47c660b6c6051be5a5a80660ae765ecbaa5
1928 - .pick_status.json: Update to f3d2fade82c168a7ffffa4bd7bf22585c45c711b
1929 - .pick_status.json: Update to f5d6a1b916fb163ee72e6a6f356937b1fbac53e0
1930 - .pick_status.json: Update to 1d418e79b8a0f4270775277b7115b88ac4c77113
1934 - lima: introduce fs and vs shader cache
1935 - lima/ppir: fix creation of mov node for non-ssa tex dest
1936 - lima: set yuv formats as external_only
1937 - lima: enable r and rg pixel formats again
1938 - lima: always set stride in texture descriptor
1939 - lima: implement GL_EXT_texture_swizzle
1940 - docs/features: add lima features
1941 - lima: fix max sampler views
1942 - lima: run nir dce after nir_lower_vec_to_movs
1943 - lima/ppir: remove liveness info from blocks
1944 - lima/ppir: remove use of live_out
1945 - lima/ppir: rework liveness data structures to bitset
1946 - lima: fix half float render
1947 - lima: enable rg formats for fp16 render
1948 - lima: increase epsilon for depthrange near == far
1950 Erik Faye-Lund (158):
1952 - zink: handle NULL views in zink_set_sampler_views
1953 - zink: fix vertex-stride wrangling
1954 - docs: fix sphinx-warnings due to lacking escaping
1955 - docs: fix broken link
1956 - docs: turn non-code into comment
1957 - docs/features: add missing features for zink
1958 - docs/features: remove a few redundant zink mentions
1959 - zink: always expose linear float textures
1960 - zink: respect feature-cap for robust buffer access
1961 - zink: respect feature-cap for independent blending
1962 - zink: respect feature-cap for sample-shading
1963 - zink: respect feature-cap for multi-draw indirect
1964 - zink: check for extension instead of function
1965 - zink: require vulkan memory model for tesselation
1966 - zink: make all xfb caps depend on extension
1967 - zink: respect fragment-shader depth-layout
1968 - zink: clone shader before lowering clip_halfz
1969 - docs/zink: add missing colon
1970 - docs/zink: add two missing required features
1971 - docs/zink: document the independentBlend requirement for GL3
1972 - docs/zink: fix phrasing of GL 3.3 requirements
1973 - docs/zink: add GL 4.0 requirements
1974 - docs/zink: add GL 4.1 requirements
1975 - docs/zink: add GL 4.2 requirements
1976 - docs/features: mark off two more extensions for zink
1977 - docs/zink: correct vk version for GL 4.2
1978 - mesa/main: remove leftover bumpmap code
1979 - compiler/nir: add texcoord replace lowering pass
1980 - gallium/st: lower point-sprites if not supported
1981 - zink: request texcoord replace lowering
1982 - docs/features: mark ssbos as done for zink
1983 - zink: remove stale TODO
1984 - zink: be more careful about limits when unsupported
1985 - zink: correct return-type for function
1986 - zink: only emit SpvCapabilityDerivativeControl when needed
1987 - zink: only emit cap when needed
1988 - zik: correct spir-v caps for textures and images
1989 - zink: do not insist shaders come from glsl
1990 - zink: add a get_primitive_mode-helper
1991 - zink: add a get_spacing-helper
1992 - zink: refactor vertex-order emitting
1993 - zink: wrap some long lines
1994 - docs: fix invalid rst syntax
1995 - zink: check for error when calling vkEnumeratePhysicalDevices
1996 - zink: explicitly check for VK_NULL_HANDLE
1997 - zink: support using lavapipe
1998 - CI: always expose docs artifacts
1999 - ci: make sure all lava-builders have libvulkan
2000 - ci: run piglit on zink with lavapipe
2001 - lavapipe: report correct value for minMemoryMapAlignment
2002 - ci: document arm oddity in build-rules
2003 - zink: correctly handle 64 valid timestamp bits
2004 - zink: enable excluded test
2005 - ci: enable max texture size tests for zink
2006 - lavapipe: handle null-buffers for xfb
2007 - ci: disable sporadically failing test
2008 - zink: drop extra set of parens
2009 - zink: do not use extra staging resource unless needed
2010 - zink: don't always require linear display-targets
2011 - zink: limit host-visible bind-flags
2012 - zink: ignore irrelevant bind-flags
2013 - zink: use gallium api to copy to display-target
2014 - zink: add X32_S8X24 format
2015 - zink: correct inaccurate comment
2016 - lavapipe: fix primitive-restart for uint8 indices
2017 - zink: fix emulation of no mipfilter
2018 - zink: fix free of ralloced pointer
2019 - gallium/st: fix shader_has_one_variant
2020 - gallium/st: fix shader_has_one_variant
2021 - gallium/st: reserve space in default uniform block for lowered constants
2022 - docs: remove stray newline
2023 - docs: remove excessive wrapping
2024 - docs: remove excessive quoting
2025 - docs: document zink GL 4.3 requirements
2026 - docs: document zink GL 4.4 requirements
2027 - docs: document zink GL 4.5 requirements
2028 - docs: document zink GL 4.6 requirements
2029 - docs: simplify format requirements
2030 - zink: factor out interpolation to helper
2031 - zink: emit all interpolation modes
2032 - zink: check for pipeline statistics feature
2033 - zink: check for depth-bias-clamp feature
2034 - zink: check for stores and atomics features
2035 - zink: add missing required feature
2036 - zink: check for mirror-clamp extension
2037 - zink: fix vector comparison
2038 - zink: drop bool attempt in float vector compares
2039 - zink: do not open-code vector-compares
2040 - zink: follow spir-v 1.0 spec
2041 - docs: Add 21.0.0 hashes
2042 - zink: tighten emitted image spir-v caps
2043 - zink: remove no-longer-needed clipdist1 patching
2044 - frontends/va: correct check for invalid format
2045 - zink: handle errors in nir_to_spirv
2046 - zink: pre-populate locations in variables
2047 - zink: do not depend on shader_slots_reserved for xfb
2048 - zink: use pre-populated shader-locations
2049 - lavapipe: report correct value for maxTexelBufferElements
2050 - docs: do not try to copy missing file
2051 - compiler/glsl: avoid null-pointer deref
2052 - docs: remove bogus zink-requirement
2053 - docs: remove zink incorrect requirement
2054 - zink: do not enable unused extension
2055 - docs: clarify VK_KHR_external_memory requirement
2056 - zink: check base-requirements
2057 - zink: assert that pstage is within range
2058 - zink: simplify shader-removal
2059 - zink: document why we're calling pipe_shader_type_from_mesa
2060 - docs: appling -> applying
2061 - docs: sytem -> system
2062 - docs: ie. -> i.e.
2063 - docs: vulkan -> Vulkan
2064 - zink: do not request scoped memory barriers
2065 - docs: optimisation -> optimization
2066 - docs: opencl -> OpenCL
2067 - docs: Xorg -> X.Org
2068 - docs: nops -> NOPs
2069 - docs: lod -> LOD
2070 - docs: lex / yacc -> Lex / Yacc
2071 - docs: dfsm -> DFSM
2072 - docs: fix incorrect possessive form
2073 - docs: fix invalid rst
2074 - docs: fix rst-quoting issues in release-notes
2075 - docs: spell out full name of gitlab instance
2076 - docs: spell out development
2077 - docs: spell out environment
2078 - docs: spell out freedesktop.org
2079 - docs: no-op'd -> disabled
2080 - docs: fix release notes for 20.3.5
2081 - ci: turn sphinx-build warnings into errors
2082 - bin/gen_release_notes.py: more robust rST escaping
2083 - compiler/glsl: correct the number of string-arguments
2084 - compiler/glsl: fix volatile string
2085 - compiler/glsl: clean up output
2086 - glsl: fix is_integer_16_32
2087 - glsl: fix int16 type
2088 - glsl: tolerate int16 loop counters
2089 - gallium/st: correct range for float16
2090 - gallium/st: correct range for int16
2091 - zink: document scalarBlockLayout requirement
2092 - zink: fix typo in function name
2093 - compiler/glsl: drop rogue argument to _mesa_glsl_error
2094 - compiler/glsl: do not cast struct to string
2095 - lavapipe: do not subtract 8 from enum
2096 - lavapipe: check all vertex-stages
2097 - lavapipe: check all graphics stages
2098 - lavapipe: ask pipe-driver for int16 support
2099 - zink: do not clear on cpu
2100 - zink: fall back from cached to non-cached memory
2101 - zink: do not dereference NULL pointer
2102 - zink: verify that src/dst support blitting
2103 - zink: verify that source-format support linear-filter
2104 - zink: fix stencil-export cap emission
2105 - gallivm: handle 16-bit input in i2b32
2106 - zink: do not read outside of array
2107 - zink: do not require vulkan memory model for shader-images
2108 - zink: correct image cap checks
2109 - zink: fix shader-image requirements
2113 - intel/tools: Show unknown instructions in decoded state.
2117 - intel/gen12: Fix memory corruption issues in fused Gen12 parts.
2118 - intel/genxml: Fix pixel hashing 3DSTATE_3D_MODE field definitions for Gen12 and Gen12.5.
2119 - intel/genxml: Define 3DSTATE_SUBSLICE_HASH_TABLE command for Gen12 and Gen12.5.
2120 - intel/dev: Implement pixel pipe subslice counting for Gen12+.
2121 - iris/gen11+: Calculate pixel hashing tables instead of hardcoding.
2122 - iris/gen12: Implement programming of pixel pipe hashing tables.
2123 - anv/gen11+: Calculate pixel hashing tables instead of hardcoding.
2124 - anv/gen12: Implement programming of pixel pipe hashing tables.
2125 - iris/gen12: Work around push constant corruption on context switch.
2129 - vulkan/device_select: Only call vkGetPhysicalDeviceProperties2 if the device supports it.
2133 - r600/nir: clone shader before first query to shader key
2134 - r600/sfn: fix use of b32all/and
2135 - r600: Add flags to INTERP_X and INTERP_Z two-slot ops
2136 - r600/sb: Add support for INTERP_X and INTERP_Z ops
2137 - r600/nir: pass array info to r600_shader for sb
2138 - r600/sfn: update shader array info
2139 - r600/sfn: Keep array registers alive for the whole shader
2140 - r600/sb: fix boundary assert for mem-instruction decoding
2141 - r600/sb: fall back to un-optimized byte code when ra_init fails
2142 - r600: Enable sb also for NIR
2143 - compiler/nir: Add support for lowering stores with nir_lower_instruction
2144 - r600/sfn: Fix use of cnde_int for bcsel
2145 - r600/sfn: Set unnormalized flag for z-coordinate when fetching from array
2146 - r600/sfn: Add the position input as varying
2147 - r600/sfn: Fix FS inputs when reading from the same position
2148 - r600/sfn: Fix dual source blend lowered to FRAG_DATA
2149 - r600/sfn: Use the constant buffer ID when given
2150 - gallium/tgsi_to_nir: Handle SAMPLE_MASK output in FS
2151 - gallium/tgsi-to-nir: Take property NUM_CLIPDIST_ENABLED into account
2152 - r600/sfn: Handle memory_barrier_atomic_counters
2153 - r600/sfn: Fix indirect_file flag for IMAGES
2154 - r600/sfn: remove duplicate barriers
2155 - r600/sfn: Base instr lowering class on nir_lower_instruction code
2156 - nir: Add flag to tex instruction to indicate lowering cube to array
2157 - nir: Add r600 specific CUBE opcode to evaluate cube texture coords and face
2158 - r600/sfn: Add support for cube_r600 instruction
2159 - r600/sfn: add lowering pass for cube textures
2160 - r600/sfn: fix gather with cube lowering
2161 - r600/sfn: use lower bool to int32 and lower int_tg4 only on shader clone
2162 - r600/sfn: use lowering pass for cube textures
2163 - r600/sfn: remove old cube texturing code
2164 - r600/sfn: Lower FS inputs to temps late and, and lower interpolate at
2165 - r600/sfn: set info about using helper_invocation to skip sb
2166 - r600/sfn: lower isign and iabs in nir
2167 - r600/sfn: Allow any channel for the helper invocation evaluation
2168 - r600: unify nir shader options evaluation
2169 - r600/sfn: remove code for nir_op_fsign since it is lowered
2170 - r600/sfn: remove unused emit_alu_op2_split_src_mods
2171 - r600/sfn: remove some old debug output
2172 - r600/sfn: encode component in address for local IO
2173 - nir: disaallow reordering for r600 shared load and remove component field
2174 - r600/sfn: handle querying the number of layers in cube arrays
2175 - r600/sfn: Fix loading TES gl_PatchVerticesIn
2176 - r600: Don't optimize using source modifiers on literals
2177 - r600: Enable GLSL 450 for nir shaders.
2178 - r600/sfn: Update status
2179 - nir: Add r600 specific intrinsic for loading the tesselation coords
2180 - r600/sfn: lower intrinsic_load_tess_coord to driver version
2181 - r600/sfn: eliminate loading unused component loads from shared memory
2182 - virgl: implement support for  PIPE_CAP_STRING_MARKER
2183 - r600/sfn: sort alu opcodes in switch statememt
2184 - r600/sfn: remove unused code
2185 - r600/sfn: fix buffer offset for ssbo writes
2186 - r600/sfn: Fix including/not including c++ parts of header
2187 - r600/sfn: lower bitfield_extract and bitfield_insert in NIR
2188 - r600/sfn: lower idiv, imod, etc in nir
2189 - r600/sfn remove some leftover debug output
2190 - nir: add opcodes for \*find_msb_rev and lowering
2191 - nir: Add opcodes for fused comp + csel and optimizations
2192 - nir: Add r600 specific sin and cos variants
2193 - r600/sfn: Add algebraic lowering for fsin and  fcos
2194 - r600/sfn: optimize comp+csel using fused ops
2195 - r600/sfn: lower find_msb variants to find_msb_rev
2196 - r60/sfn: don't lower scomp
2197 - r600: Handle negate of second operator in TGSI_OPCODE_UADD
2198 - r600/sfn: Make some value pool functions private
2199 - r600/sfn: Add skelton for visitor pattern
2200 - r600/sfn: remove extra parameter from alu assemebly emission
2201 - r600/sfn: fix some formatting
2202 - r600/sfn: switch assembler creation to use visitor
2203 - r600/sfn: drop the local register mep
2204 - r600/sfn: lower VS IO and drop old deref code
2205 - r600/sfn: lowered FS output IO
2206 - r600/sfn: remove old deref code
2207 - r600/sfn: force dual source blend output handling in some cases
2208 - r600/sfn: remove find_msb lowering in driver
2209 - r600/sfn: clean up multi-sample texture load
2210 - r600/sfn: clean up value pool interface usage in emit_stream
2211 - r600/sfn: use get_temp_vec4 directly when fetching
2212 - r600/sfn: make allocate_temp_register private to valuepool
2213 - r600: Fix texture buffer offset alignment
2214 - r600: don't set an index_bias for indirect draw calls
2215 - r600: Don't advertise support for scaled int16 vertex formats
2216 - r600/sfn: allocate number of VS outputs based on max driver_location
2217 - r600/sfn: Only fetch the constants that are needed in fdd*
2218 - r600: Enable sb for nir only on specific request
2219 - Revert "r600: don't set an index_bias for indirect draw calls"
2220 - Revert "r600: Don't advertise support for scaled int16 vertex formats"
2221 - r600: don't set an index_bias for indirect draw calls
2225 - disk_cache: Fail creation when cannot inizialize queue.
2226 - anv: Allow null handle in DestroyDescriptorUpdateTemplate.
2230 - meson: Add missing lavapipe dep
2234 - gitlab-ci: extend x86_64 kernel config to suport Intel devices
2235 - gitlab-ci: build the iris gallium driver as well
2236 - gitlab-ci: add intel APL and GLK devices with manual triggers
2237 - traces-iris: fix expectation for Intel GLK
2239 Hans-Kristian Arntzen (2):
2241 - radv: Take image alignment into account when allocating MUTABLE pool.
2242 - radv: Allocate buffer list for MUTABLE descriptor types as well.
2246 - zink/codegen: add some new attributes to Extension
2247 - zink/codegen: make 'struct' optional in Version
2248 - zink/codegen: enable instance extension unconditionally if promoted
2249 - zink/codegen: codegen-ize load_instance_extensions()
2250 - zink/codegen: adding/fixing comments and copyright notice
2251 - zink/codegen: find promotion version using vulkan registry
2252 - zink: remove excessive checks for loader version
2253 - zink: expose PIPE_CAP_ANISOTROPIC_FILTER
2254 - zink: VK_KHR_draw_indirect_count is a device extension
2255 - zink/codegen: introduce ExtensionRegistry
2256 - zink/codegen: more validation in zink_instance
2257 - zink/codegen: introduce notion of non-standard extensions
2258 - zink/codegen: make zink_device_info accept vk.xml
2259 - zink/codegen: perform basic validation in zink_device_info
2260 - zink/codegen: validate has_properties and has_features
2261 - zink/codegen: fix type annotations
2262 - zink/codegen: do not enable extensions that are now core
2263 - zink: enable KHR_shader_draw_parameters on Vulkan <1.2
2264 - zink: fix detection of KHR_maintenance1/2
2268 - turnip: Return correct value of tu6_load_state_size
2269 - nir: Set access at lower_ubo_vec4
2270 - ir3: fix has_src() to return correctly in ir3_nir_lower_tex_prefetch
2271 - ir3: Add nonuniform encodings to ir3 encoder and parser
2272 - turnip: Enable nonuniform descriptor indexing
2276 - v3dv: only update uniforms for dirty descriptors if stage has descriptors
2277 - v3dv: disable early Z writes if Z writes are disabled
2278 - v3dv: don't wait for idle on occlusion query pool resets
2279 - v3dv: use PIPE_TIMEOUT_INFINITE
2280 - v3dv: refactor checks for subpass attachment clears
2281 - v3dv: refactor checks for subpass attachment loading
2282 - v3dv: refactor checks for subpass attachment stores
2283 - v3dv: do not emit full tile buffers clears to handle Z/S clears
2284 - v3dv: enable early Z/S clears
2285 - v3dv: fix disabling Early Z for the whole frame
2286 - broadcom/compiler: fix indentation with TABs
2287 - broadcom/compiler: prepare TMU spilling code to account for TMU pipelining
2288 - broadcom/compiler: implement pipelining for general TMU operations
2289 - broadcom/compiler: support pipelining of tex instructions
2290 - broadcom/compiler: refactor image load/store TMU emission code
2291 - broadcom/compiler: support pipelining of image load/store instructions
2292 - broadcom/compiler: disable TMU pipelining if we fail to register allocate
2293 - broadcom/compiler: disallow spilling if TMU pipelining was enabled
2294 - broadcom/compiler: log spilling shaders to perf output
2295 - broadcom/compiler: let QPUs stall on TMU input/config overflows
2296 - v3dv: handle D/S buffer to image copies with the texel buffer path
2297 - v3dv: batch copies in the copy_buffer_to_image_blit path
2298 - v3dv: allow a component swizzle in copy_buffer_to_image_shader
2299 - v3d/common: use spaces instead of TABs
2300 - v3dv: serialize pipeline compilation when debugging shaders
2301 - v3dv: add a perf trace when a device is created with robust buffer access
2302 - v3d/compiler: fix QPU scheduler TMU sequence shuffling
2303 - broadcom/compiler: add V3D_QPU_WADDR_UNIFA
2304 - broadcom/compiler: pass a devinfo to check if an instruction writes to TMU
2305 - broadcom/compiler: name registers correctly based on V3D version
2306 - broadcom/compiler: don't check for GFXH-1633 on V3D 4.2.x
2307 - broadcom/compiler: add a helper to check if an instruction writes unifa
2308 - broadcom/compiler: disallow unifa overlap with thread switch/end
2309 - broadcom/compiler: preserve ordering of unifa/ldunifa sequences
2310 - broadcom/compiler: ensure 3-slot delay between unifa and ldunifa
2311 - broadcom/compiler: disallow reading two uniforms in the same instruction
2312 - broadcom/compiler: do not DCE ldunifa
2313 - broadcom/compiler: emit ldunifarf when needed
2314 - broadcom/compiler: use unifa for UBO loads from uniform addresses
2315 - broadcom/compiler: don't emit redundant ldunif
2316 - broadcom/compiler: use a helper function to decide on TMU spilling
2317 - broadcom/compiler: don't dump shader-db stats for failed shaders
2318 - broadcom/compiler: fix ldunif optimization
2319 - broadcom/compiler: allow dead code elimination of unused trailing ldunifa
2320 - broadcom/compiler: remove unused leading ldunifa
2321 - broadcom/compiler: add a constant alu optimization pass
2322 - broadcom/compiler: skip unnecessary unifa writes
2323 - broadcom/compiler: use nir_opt_sink
2324 - v3dv: fix branching to large secondaries with more than one BCL buffer.
2325 - broadcom/compiler: fix DAG pre-remove for merged instructions
2326 - broadcom/compiler: fix indentation style
2327 - broadcom/compiler: track pipelineable ldvary sequences
2328 - broadcom/compiler: pipeline smooth ldvary sequences
2329 - broadcom/compiler: allow pipelining of flat and noperspective varyings
2330 - broadcom/compiler: ldvary pipelining tracking and documentation clean-ups
2331 - broadcom/compiler: drop the destination for unused ldunifa
2332 - broadcom/compiler: be more aggressive skipping unifa writes
2333 - broadcom/compiler: always restart ldvary pipelining when scheduling ldvary
2334 - broadcom/compiler: ldvary doesn't implicitly write to r3 since V3D 4.1
2335 - broadcom/compiler: fix flags check for ldvary merge
2336 - broadcom/compiler: add an additional sanity check assert to the ldvary fixup
2337 - broadcom/compiler: move code block around
2338 - broadcom/compiler: simplify ldvary pipelining
2339 - broadcom/compiler: disallow ldunif during ldvary sequences if possible
2340 - v3dv: call util_cpu_detect() when initializing the instance
2341 - broadcom/compiler: flag wrtmuc with a read dependency on last_tmu_config
2342 - broadcom/compiler: be more flexible scheduling TMU writes
2343 - vulkan/util: call glsl_type_singleton_init_or_ref from vk_instance_init
2344 - compiler/glsl: call util_cpu_detect from glsl_type_singleton_init_or_ref
2345 - broadcom/compiler: fix end of tmu sequence detection
2346 - broadcom/compiler: use nir_opt_load_store_vectorize
2347 - broadcom/compiler: use nir_lower_wrmasks to simplify TMU general stores
2348 - broadcom/compiler: handle implicit uniform loads when optimizing constant alu
2349 - broadcom/compiler: optimize constant vfpack
2350 - broadcom/compiler: use nir_lower_undef_to_zero
2351 - v3dv/pipeline_cache: fix assert
2352 - broadcom/compiler: convert add to mul when possible to allow merge
2353 - broadcom/compiler: add a v3d_qpu_writes_accum helper
2354 - broadcom/compiler: try to fill up delay slots after a thrsw
2355 - broadcom/compiler: flag TMU read dependencies against last TMU config
2356 - broadcom/compiler: flag TMU reads with a read dependency on last TMU config
2357 - broadcom/compiler: dump instruction index when failing to pack instructions
2358 - broadcom/compiler: add a NOP count stat to shader-db
2359 - broadcom/compiler: try to fill up delay slots after a branch instruction
2360 - broadcom/compiler: try to fill up delay slots after unconditional branch
2361 - broadcom/compiler: implement restriction for branch after setmsf
2362 - broadcom/compiler: optimize branch emission for uniform break/continue
2363 - v3dv: fix index buffer binding
2364 - broadcom/compiler: add a definition for the unifa skip distance
2365 - broadcom/compiler: allow compilation strategies to limit minimum thread count
2366 - broadcom/compiler: sort constant UBO loads by index and offset
2367 - broadcom/compiler: rename unifa tracking fields
2368 - v3dv: fix descriptor set limits
2372 - i965: Don't advertise OpenGL 3.3+ if driconf disables GL_ARB_blend_func_extended
2373 - i965: Use allow_higher_compat_version option during screen initialization
2374 - i965: Don't parse driconf again
2375 - nir/algebraic: Fix a >> #b << #b for sizes other than 32-bit
2376 - nir/algebraic: add patterns for a >> #b << #b and a << #b >> #b
2377 - nir/algebraic: Partially revert 3f782cdd2591
2378 - intel/eu/validate: Add some checks for CMP and CMPN
2379 - intel/compiler: Enable the ability to emit CMPN instructions
2380 - intel/compiler: Make the CMPN builder work like the CMP builder
2381 - intel/compiler: Use CMPN for min / max on Gen4 and Gen5
2382 - nir/algebraic: Fix some min/max of b2f replacements
2383 - nir/algebraic: Remove some redundant b2f logic-op reduction patterns
2384 - nir/algebraic: Add some max/min optimizations with 3 variables
2385 - nir/range-analysis: C++ linkage
2386 - nir/range_analysis: Handle vectors better in ssa_def_bits_used
2387 - intel/compiler: Silence unused parameter warnings in files that include brw_eu.h
2388 - intel: Silence unused parameter warnings in files that include gen_device_info.h
2389 - intel: Silence unused parameter warnings in files that include genX_pack.h
2390 - intel/compiler: Relax some conditions in try_copy_propagate
2391 - gallium/dri: Remove dri2_format_mapping::cpp
2392 - nir/search: Constify instruction parameter to search helpers
2393 - nir/algebraic: Apply addition property of equality more conservatively
2394 - nir/algebraic: Apply addition property of equality to the other ordering too
2395 - nir/range_analysis: Refactor fsat handling
2396 - nir/range_analysis: Add "is finite" range analysis tracking
2397 - nir/range_analysis: Add "is a number" range analysis tracking
2398 - nir/range_analysis: Fix analysis of fmin, fmax, or fsat with NaN source
2399 - nir/search: Use range analysis for is_finite
2400 - nir/range_analysis: Simplify analysis of bcsel
2401 - mesa: Add anything dynamically indexed before any non-dynamically indexed
2402 - mesa: Clean up _mesa_layout_parameters after previous commit
2403 - tgsi_exec: Fix NaN behavior of saturate
2404 - tgsi_exec: Fix NaN behavior of min and max
2408 - pan/bi: Lower 64-bit integers
2409 - pan/bi: Handle 64-bit pack and unpack operations
2410 - pan/bi: Add some compute intrinsic loads
2411 - pan/bi: Set compute lowering options
2412 - pan/bi: Improve interoperability of the command-line disassembler
2413 - pan/bi: Implement load/store intrinsics
2414 - pan/bi: Implement load_kernel_input
2415 - panfrost: Set bifrost_props for compute shaders
2416 - pan/bi: Improve unknown intrinsic error
2417 - panfrost: Use the correct NIR options for OpenCL on Bifrost
2418 - pan/bi: Use pan_nir_lower_64bit_intrin
2419 - panfrost: Add a sysval for local_group_size
2420 - panfrost: Add a sysval for local_work_dim
2421 - panfrost: Assert on sysval overflow
2422 - pan/mdg: Limit int64 vectorization
2423 - pan/mdg: Don't reorder loads/stores past each other
2424 - pan/mdg: Allow 64-bit src_bitsize for comparison operations
2425 - pan/bi: Add w0 to the 'h01' swizzle bucket
2426 - pan/bi: Lower umul_high
2427 - panfrost: Set TLS for compute jobs
2428 - pan/bi: Implement saturated add/sub operations
2429 - pan/bi: Implement ihadd/irhadd operations
2430 - pan/bi: Implement packing ops between 32-bit vec1 and 16-bit vec2
2431 - pan/mdg: Fix spilling when scratch memory is used
2432 - pan/bi: Iterate from zero when setting RA interference
2433 - panfrost: Add a function to determine if a resource is 2D
2434 - panfrost: Only checksum resources when it makes sense to
2435 - panfrost: Add a debug flag to disable checksumming
2436 - panfrost: Transaction elimination support
2437 - panfrost: Fix the tile size assertion
2438 - pan/decode: Free mapped memory objects on BO unreference
2439 - panfrost: Add support for INTEL_blackhole_render
2440 - panfrost: Use normal malloc/free instead of ralloc for surfaces
2441 - panfrost: Add the tiler heap to fragment jobs
2442 - pan/bi: Return the size of the last clause from bi_pack
2443 - pan/bi: Fix shader prefetch size
2444 - panfrost: Fix clear color packing for 12-byte formats
2445 - pan/bi: Don't check liveness unless the index is valid
2446 - pan/bi: Use the correct size for UBO loads
2447 - pan/bi: Remove check for first_ubo_is_default_ubo
2448 - pan/bi: Implement image load/store
2449 - pan/bi: nir_intrinsic_image_size support
2450 - st/mesa: Update constants on alpha test change if it's lowered
2451 - panfrost: Disable early-z when alpha test is used
2452 - pan/mdg: Rename load/store operations
2453 - pan/mdg: Use appropriate sizes for global loads/stores
2454 - pipe-loader,gallium/drm: Fix the kmsro pipe_loader target
2455 - pipe-loader: Stop trying to use kmsro for vgem
2456 - panfrost: Implement panfrost_set_global_binding
2457 - panfrost: Flush output after disassembling shaders
2458 - panfrost: Only do point coord replacement for PIPE_PRIM_POINTS
2459 - panfrost: Only add resource checksum BOs to the batch once
2460 - panfrost: Align BO size to 4096 bytes
2461 - panfrost: Add fast path for graphics work group computation
2462 - panfrost: Unset shared/scanout binding flags for staging resources
2463 - pan/bi: Skip nir_opt_move/sink for blend shaders
2464 - panfrost: Fix shader texture count
2465 - pan/decode: Allow frame shader DCDs to be in another BO than the FBD
2466 - pan/mdg: Fix calculation of available work registers
2467 - panfrost: Fix viewport scissor for preload draws
2471 - nv50/ir: ignore FS_BLEND_EQUATION_ADVANCED
2472 - nv50,nvc0: explicitly list recently-added caps
2473 - st/mesa: fix broken moves for u2i64 and related ops
2474 - nv50/ir: clear dnz flag when converting mul/mad to simpler ops
2475 - glsl: only expose int64 atomics when extension is enabled
2476 - cso: set index_bounds_valid = true for arrays draws
2477 - nvc0: index_bias is now only set for indexed draws
2478 - nvc0/ir: add fixup to deal with interpolateAtSample with non-MSAA
2479 - nv50,nvc0: clear internal vbo masks based on the trailing slots
2480 - ci: remove nouveau from shader-db runs
2481 - nouveau: reinstate fencing on screen destroy
2482 - nv50: add PIPE_CAP_NIR_IMAGES_AS_DEREF to unsupported list
2483 - nv50,nvc0: add scissored clear support
2484 - st/mesa: do scissored clears on depth/stencil as well when supported
2485 - i965: support GL_EXT_color_buffer_half_float
2486 - mesa: fix conditions for fp16 render format eligibility
2487 - mesa: fix fbo attachment size check for RBs, make it trigger in ES2
2488 - mesa: add tracking of reduction mode
2489 - st/mesa: add EXT_texture_filter_minmax support
2490 - nvc0: enable minmax reductions on gm200+
2491 - docs: add notes about nvc0 support of ARB/EXT_texture_filter_minmax
2492 - mesa: only report INCOMPLETE_FORMATS for GLES1 / desktop
2493 - gallium,st: add missing viewport swizzles
2494 - nv50: initialize target for blit source surfaces
2495 - nv50,nvc0: remove explicit target argument from view creation
2496 - nv50: add appropriate space check before adding new pushbuffer
2497 - nvc0: ensure sufficient push space for indirect data
2498 - nvc0: fix reported driver queries for Pascal and later GPUs
2499 - mesa: fix restoring of texture attributes for msaa binding points
2500 - nv50: adapt texture and constbuf paths for compute shaders
2501 - nv50: add resource tracking for shader images and buffers
2502 - nv50: implement memory barrier handling
2503 - nv50: add texture, constbuf, image, buffer validation
2504 - nv50: pass in third axis via user param
2505 - nv50/ir: retrieve (n)ctaid.z from first user param
2506 - nv50/ir: force shared memory indirect to be an address
2507 - nv50/ir: do not use inline offsets for global, ensure indirect access
2508 - nv50/ir: fix emission of RED
2509 - nv50/ir: lower buffer to global
2510 - nv50/ir: fix emitting movs from imm to short registers
2511 - nv50/ir: fix emission of cvt with half-reg destinations
2512 - nv50/ir: fix emission of logic ops on half-regs
2513 - nv50/ir: fix emission of shifts on half-regs
2514 - nv50/ir: logic ops on half-regs can't take an immediate
2515 - nv50/ir: add support for 16-bit immediates
2516 - nv50/ir: fix emission of 16-bit add
2517 - nv50/ir: fix emission of cas without a destination
2518 - nv50: fix expression for ucp offset
2519 - nv50/ir: avoid inlining results of a locked load
2520 - nv50/ir: fix emission of ld/st lock/unlock
2521 - st/mesa: adapt for the case where buffers are not supported in frag
2522 - nv50/ir: fix texture size for msaa textures
2523 - nv50: emulate indirect draws
2524 - nv50/ir: fake SV_THREAD_KILL support
2525 - nv50: enable ARB_framebuffer_no_attachments
2529 - panfrost: fix attribute continuation decoding
2530 - panfrost: add 3d attribute buffer continuation to XML
2531 - panfrost: decode 3d attribute continuation
2532 - panfrost: add resource modifier conversion
2533 - panfrost: implement gallium->set_shader_images
2534 - panfrost: emit shader image attribute descriptors
2535 - panfrost: implement image_size sysval
2536 - pan/mdg: create nir pass to lower image coord bitsize
2537 - pan/mdg: enable image bitsize lowering pass
2538 - pan/mdg: add ld_image opcodes
2539 - pan/mdg: rename st_image opcodes and add float16 versions
2540 - pan/mdg: implement shader image instructions
2541 - pan/mdg: implement nir_intrinsic_image_size
2542 - panfrost: advertise images for midgard
2543 - pan/mdg: prevent csel_v from being scheduled alongside writeout
2547 - anv: don't advertise mipmaps for linear 3D surfaces on BDW
2548 - anv: move buffer size alignment into helper function
2549 - anv: use helper function to get the buffer size
2550 - intel, anv: propagate robustness setting to nir_opt_load_store_vectorize
2554 - nouveau: Stash supported sector layout in screen
2555 - nouveau: Use DRM_FORMAT_MOD_NVIDIA_BLOCK_LINEAR_2D
2556 - nouveau: no modifier != the invalid modifier
2557 - nouveau: Use format modifiers in buffer allocation
2561 - radv: Use linker pragma to simulate weak functions
2562 - radv: Remove unnecessary headers
2563 - radv: Stub sections that don't have _WIN32 support
2564 - radv: Modify radv_descriptor_set for MSVC legality
2565 - radv: Pointer arithmetic on char/uint8_t, not void
2566 - radv: Disable calibrated timestamps on Windows
2567 - radv: Use typed outarray API
2568 - radv: Fix struct initialization for MSVC
2569 - gallium/tessellator: Fix warning suppression
2570 - microsoft: Fix comma in variadic macro for MSVC
2571 - ac: Remove unnecessary header
2572 - radv: Use typed outarray API
2573 - amd: Disable surface modifier test on Windows
2574 - radv: Pointer arithmetic without void*
2575 - radv: Update JSON generator if Windows
2576 - vulkan: Use typed outarray API
2577 - vulkan: Fix source list for vulkan_wsi on Windows
2578 - vulkan: Update dispatch table gen for Windows
2579 - vulkan/util: Use util_bitcount
2580 - ac/rgp: BSD elf library compatibility
2581 - amd: Hide amdgpu_drm.h on Windows
2582 - amd: Hide drm_fourcc.h on Windows
2586 - amd: add Aldebaran chip enum
2590 - ac: prefer system EM_AMDGPU definition
2591 - ac/rgp: define EM_AMDGPU if missing for compatibility
2595 - intel/tools: Decode COMPUTE_WALKER
2596 - intel/fs: Allow compute dispatch without a pushed subgroup ID on Gen12-HP
2597 - anv: Add a general state pool
2598 - intel/fs: Emit code for Gen12-HP indirect compute data
2599 - anv: Enable push constants on gen12-hp
2600 - intel/genxml,anv,iris: Drop the legacy compute path from gen125.xml
2601 - anv: Add a trivial implementation of VK_KHR_deferred_host_operation
2602 - anv: Exit early from cmd_buffer_apply_pipe_flushes
2603 - anv: Take the set of stages to flush in flush_descriptor_sets
2604 - anv: Only flush descriptors used by the pipeline
2605 - anv: Early-exit from cmd_buffer_flush_state
2606 - nir/lower_int64: Add a level of wrapper functions
2607 - nir/lower_int64: Fix lowering of f2[ui]64 for 16-bit float
2608 - nir/lower_int64: Add lowering for some 64-bit subgroup ops
2609 - nir/lower_int64: Add lowering for 64-bit iadd shuffle/reduce
2610 - nir/lower_int64: Lower 64-bit vote_ieq
2611 - intel/compiler: Return 1 for immediates in regs_read
2612 - intel/compiler: Move brw_reg_type_for_bit_size to brw_reg_type.h
2613 - intel/reg,fs: Handle immediates properly in subscript()
2614 - intel/fs: QUAD_SWIZZLE requires packed data
2615 - intel/fs: Support 64-bit SEL_EXEC on Gen11+
2616 - intel/fs: Support 64-bit SHUFFLE on Gen11+
2617 - intel/fs: Support 64-bit CLUSTER_BROADCAST on Gen11+
2618 - intel/fs: Properly lower 64-bit MUL on 64-bit-incapable platforms
2619 - intel/fs: Refactor our shuffle emit code
2620 - intel/fs: Implement umin/umax shuffle
2621 - anv: Advertise shaderInt64 on Gen11+
2622 - anv: Break SAMPLE_PATTERN and MULTISAMPLE emit into helpers
2623 - intel/fs: Add an ex_desc field to fs_inst for SHADER_OPCODE_SEND
2624 - anv: Drop anv_dump
2625 - anv: Fix an old parameter name in GetDeviceQueue
2626 - anv: Refactor anv_queue_finish()
2627 - anv: Add an anv_queue_family struct
2628 - nir/from_ssa: Respect and populate divergence information
2629 - vulkan/meson: Add missing dependencise on vk_extensions_gen.py
2630 - anv: Clean up the vk_device on the CreateDevice error path
2631 - radv: Properly clean up vk_device
2632 - turnip: Properly clean up vk_device
2633 - v3dv: Properly clean up vk_device
2634 - lavapipe: Properly clean up vk_device
2635 - vulkan: Move vk_device to its own file
2636 - vulkan: Add a return code to vk_device_init
2637 - vulkan: Add common extension tables
2638 - anv: Use the common extension table struct
2639 - vulkan: Add common dispatch table generation
2640 - vulkan: Add dispatch table loading helpers
2641 - vulkan-overlay-layer: Use the new dispatch tables
2642 - vulkan: Add dispatch table lookup helpers
2643 - vulkan: Add common instance and physical device structs
2644 - vulkan: Add generators for instance trampoline functions
2645 - vulkan: Add entrypoint tables and related helpers
2646 - vulkan: Add common Get*ProcAddr implementations
2647 - vulkan: Add a common entrypoint table generator
2648 - anv: Add a single anv_genX switch macro
2649 - anv: Use the common dispatch framework
2650 - vulkan: Add framework for common entrypoints
2651 - vulkan,anv: Move GetDeviceProcAddr to common code
2652 - vulkan,anv: Add common entrypoints for VK_EXT_private_data
2653 - anv: Make Get*MemoryRequirements a wrapper
2654 - vulkan,anv: Move a bunch of trivial wrappers to common code
2655 - vulkan,anv: Move VK_KHR_copy_commands2 wrappers to common code
2656 - vulkan: Add a truly common VK_EXT_debug_report implementation
2657 - anv: Switch to the common VK_EXT_debug_report
2658 - turnip: Use the common dispatch framework
2659 - turnip: Use common entrypoints for VK_EXT_private_data
2660 - turnip: Drop some legacy wrappers in favor of common code
2661 - turnip: Switch to the common VK_EXT_debug_report
2662 - lavapipe: Drop some wrappers in favor of common code
2663 - v3dv: Drop v3dv_instance::app_info
2664 - v3dv: Use common entrypoints for VK_EXT_private_data
2665 - v3dv: Switch to the common VK_EXT_debug_report
2666 - radv: Use common entrypoints for VK_EXT_private_data
2667 - radv: Switch to the common VK_EXT_debug_report
2668 - vulkan: Make vk_debug_report_callback derive from vk_object_base
2669 - anv: Use vk_object_base::type for debug_report
2670 - vulkan: Use vk_object_base::type for debug_report
2671 - vulkan: Make the debug_report implementation internal
2672 - anv,radv: Use common entrypoints for VK_KHR_deferred_operation
2673 - vulkan: Rework vk_device_init and friends
2674 - vulkan: Drop the type_prefix parameter from gen_extensions
2675 - nir: Add some ssa-only fast-paths for nir_src rewrite
2676 - nir: Drop the lower_mem_constant_vars declaration
2677 - vulkan: Add a common helper for enumerating instance extension properties
2678 - vulkan: Rework extension disabling on Android
2679 - anv: Pull the patch version from the XML
2680 - anv: Make anv_icd.py more generic and independent
2681 - anv,vulkan: Move anv_icd.py to a common location
2682 - anv: Move extension tables to anv_device.c
2683 - anv: Add fake graphics-only and compute-only queue families
2684 - nir: Add a couple helpers for phis and cursors
2685 - nir/lower_bit_size: Support phi instructions
2686 - intel/nir: Lower 8-bit phis on Gen11+
2687 - nir: Add some range analysis for used bits
2688 - nir/algebraic: Clean up up-cast of down-cast when we can
2689 - nir/algebraic: Covert up-cast of down-cast to extract on Intel
2690 - spirv: Store the nir_function in vtn_function
2691 - spirv: Delete the impl for prototype-only functions
2692 - nir: Don't optimize bcsel-of-shuffle across blocks
2693 - nir: Fix parameter order in the bcsel-of-shuffle optimization
2694 - nir/opt_large_constants: Handle generic pointers
2695 - intel/fs: Shuffle can't handle source modifiers
2696 - anv/formats: Advertise linear sampling on depth formats
2697 - anv/android: Re-implement AcquireImageANDROID
2698 - intel/mi_builder: Create a context in the tests
2699 - intel/mi_builder: Delete a bogus comment
2700 - intel/mi_builder: Fix a misleading comment
2701 - intel/mi_builder: Short-circuit shifts in more cases
2702 - intel/mi_builder: Add constant folding
2703 - intel/mi_builder: Rewrite unit tests in terms of constant folding
2704 - intel/mi_builder: Add tests for gen_mi_z and gen_mi_nz
2705 - intel: Rename gen_mi_builder.h to mi_builder.h
2706 - intel/mi_builder: Drop the gen\_ prefix
2707 - intel/mi_builder: Use AddCSMMIOStartOffset for LRI
2708 - intel/mi_builder: Add ieq/ine helpers
2709 - intel/mi_builder: Support inverted values in mi_store
2710 - intel/mi_builder: Add load/store_offest on GFX 12.5+
2711 - genxml: Clean up MI_SET_PREDICATE
2712 - intel/batch_decoder: Don't follow predicated MI_BATCH_BUFFER_START
2713 - intel/mi_builder: Use softpin for tests on gen8+
2714 - intel/mi_builder: Return an address from __gen_get_batch_address
2715 - intel/mi_builder: Add control-flow support
2716 - nir: Add and use a new nir_ssa_def_rewrite_uses_src helper
2717 - nir: Make nir_ssa_def_rewrite_uses take an SSA value
2718 - nir: Make nir_ssa_def_rewrite_uses_after take an SSA value
2719 - intel/mi_builder: Fix some indentation
2720 - intel/mi_builder: Fix a couple of #ifs
2721 - anv: Drop anv_extensions.py
2722 - turnip: Move the CreateRenderPass wrapper to common code
2723 - anv: Move multialloc to common code
2724 - vulkan: Use VK_MULTIALLOC in CreateRenderPass
2725 - anv: Move vk_format helpers to common code
2726 - vulkan: Use correct aspectMask in CreateRenderPass
2727 - vulkan: Add some asserts and checks for multiview in CreateRenderPass
2728 - vulkan: Preserve preserve attachments in CreateRenderPass
2729 - anv: Drop CreateRenderPass
2730 - radv/meta: Use CreateRenderPass2
2731 - radv: Drop CreateRenderPass
2732 - intel/fs: Use INTEL_MASK for pushish constant address masking
2733 - intel/fs: Handle payload node interference in destinations
2734 - vulkan: Use ALWAYS_INLINE for multialloc
2735 - vk/alloc: Handle zero sizes better in vk_multialloc_add
2736 - vulkan/alloc: Add VK_MULTIALLOC_DECL macros
2737 - vulkan/util: Add a type parameter to vk_multialloc_add
2738 - vulkan/alloc: Use char * for pointer arithmetic
2739 - anv,genxml: Handle L3SQCREG1_SQGHPCI in GenXML
2740 - anv: Add an anv_batch_write_reg macro
2741 - iris: Add an iris_write_reg macro
2742 - genxml: Make 1-bit L3$ config register fields bool on Gen7
2743 - intel/fs,rt: Add a predicate to load_global_const_block
2744 - anv: Use load_global_constant for shader constants
2745 - anv: Use nir_shader_instructions_pass in apply_pipeline_layout
2746 - anv/apply_pipeline_layout: Refactor descriptor chasing code
2747 - anv/apply_pipeline_layout: Rework the early pass index/offset helpers
2748 - anv/apply_pipeline_layout: Lower UBO loads in the early pass
2749 - anv/apply_pipeline_layout: Run DCE between the early and late passes
2750 - anv/apply_pipeline_layout: Move bounds checking later for index/offset
2751 - anv/apply_pipeline_layout: Plumb through a UBO address format
2752 - anv/apply_pipeline_layout: Add some switch statements
2753 - nir: Add a new 64+32-bit address format
2754 - anv: Use 64bit_global_32bit_offset for SSBOs
2755 - anv: Rework the 64bit_bounded_global resource index format
2756 - anv: Zero out the last dword of UBO/SSBO descriptors in the shader
2757 - anv/apply_pipeline_layout: Apply dynamic offsets in load_ssbo_descriptor
2758 - anv/apply_pipeline_layout: Refactor all our descriptor address builders
2759 - anv/apply_pipeline_layout: Rework the desc_addr_format helper
2760 - anv/apply_pipeline_layout: Use the new helpers for early lowering
2761 - anv/apply_pipeline_layout: Use the new helpers for images
2762 - nir/lower_io: Support global addresses for UBOs in nir_lower_explicit_io
2763 - anv: Add a pass for lowering A64 UBO access
2764 - anv: Do UBO loads with global addresses for bindless
2765 - anv/apply_pipeline_layout: Add support for A64 descriptor access
2766 - nir: Add image atomic_fmin/fmax intrinsics
2767 - spirv: Add support for SPV_EXT_shader_atomic_float_min_max
2768 - intel/fs: Add support for 16-bit A64 float and integer atomics
2769 - intel/genxml: Binding table pointers are 15 bits on GFX version 12.5+
2770 - intel/tools: Handle milti-LRI in the batch decoder
2771 - intel/tools: Handle GT_MODE in the batch decoder
2772 - intel/genxml: Make BindingTablePoolEnable a bool
2773 - intel/tools: Handle BINDING_TABLE_POOL_ALLOC in batch decoding
2774 - anv: Align inline uniform data to ANV_UBO_ALIGNMENT
2775 - anv: Implement VK_EXT_conservative_rasterization
2776 - anv: Fix coverage masks for VK_EXT_conservative_rasterization
2777 - intel: Drop gen_device_info::has_resource_streamer
2778 - anv: Clean up anv_device_memory::base on failure
2779 - anv: Refactor framebuffer creation
2780 - anv: Clean up anv_descriptor_pool::base on the error path
2781 - anv: Clean up anv_semaphore::base on the error path
2782 - vulkan: Add a vk_object_multialloc helper
2783 - anv: Use vk_object_alloc/free
2784 - anv: Make memory type and queue family pointers const
2785 - intel: fix querying mip levels on null surfaces on SKL and prior
2786 - intel/compiler: Don't insert barriers for NULL sources
2787 - anv: Use the same re-order mode for streamout as for GS
2788 - intel/isl: Fix isl_color_value_unpack to match the prototype
2789 - intel/nir: Set lower txs with non-zero LOD
2793 - darwin: Use the system libexpat
2794 - util: Fix pointer to integer conversion error when using libunwind
2795 - darwin: Use the system libunwind
2796 - Fall back on clock_gettime when timespec_get() is unavailable
2797 - Adjust dylib compatibility versions to match what was set by mesa-18.3's autotools-based builds
2801 - nir: Work around MSVC x86 internal compiler error
2802 - main: Undefine MemoryBarrier for Windows
2803 - glapi: Undefine MemoryBarrier
2804 - mapi: Undefine MemoryBarrier
2805 - drisw: Disable automatic use of layered drivers with LIBGL_ALWAYS_SOFTWARE
2806 - wgl: Refactor screen creation to a function
2807 - wgl: Add a loop for screen creation with an ordered list of fallbacks
2808 - d3d12: Fail screen creation if a shader validator is needed and can't be created
2809 - wgl: Disable automatic use of layered drivers with LIBGL_ALWAYS_SOFTWARE
2810 - CI: Use a sha for the Windows SPIRV-LLVM-Translator dependency
2811 - microsoft/clc: Add -fgnu89-inline to clang args
2812 - microsoft/clc: Add test with inline function
2813 - clover: Add -fgnu89-inline to Clang command line
2814 - microsoft/clc: Only apply float scaling to 32bit fdiv
2815 - microsoft/clc: Let lower_vars_to_explicit_types fill kernel input driver_location
2816 - microsoft/clc: Fix wrap modes for inline samplers for integer textures
2817 - microsoft/clc: Move inline samplers to the end of the variable list
2818 - microsoft/clc: Use driver_location for metadata instead of re-computing offsets
2819 - microsoft/clc: Re-order dead variable removal after uniform vars_to_explicit_types
2820 - microsoft/clc: Add a test with an unused kernel arg
2821 - glapi: Support "ELF" TLS on Windows
2822 - docs: Document USE_ELF_TLS can work on Windows too
2823 - meson/gallium: Add an option to not use LLVM for gallium draw module
2824 - d3d12: Handle null constant buffers
2825 - nir: Add a nir_after_instr_and_phis helper
2826 - microsoft/compiler: Don't separate phis while inserting upcasts
2827 - d3d12: Move descriptor pools to screen, and add lock
2828 - d3d12: Handle is_new_style_shadow comparison filtering
2829 - d3d12: Really handle null constant buffers
2830 - u_format: Add restrict to fn pointer and manual format pack/unpack/fetch
2831 - panfrost: Add a Meson dependency on bi_opcodes.h for bifrost_compiler
2832 - meson, util: Make zlib optional again
2833 - nir: Temporarily disable optimizations for MSVC ARM64
2834 - wgl: Fix wglCreatePbufferARB pixel format lookup
2835 - d3d12: Use ID3D12Device9::CreateCommandQueue1 when available
2836 - d3d12: Use CreateDXGIFactory2 and use the debug flag when appropriate
2837 - wgl: Add unit test infrastructure for OpenGL32.dll on Windows
2838 - wgl: Add a context to framebuffer destruction
2839 - d3d12: Add a constant for num_buffers
2840 - d3d12: Clean up swapchains on framebuffer destruction
2841 - wgl, d3d12: Add a d3d12-specific test for swapchain leaks
2842 - microsoft/compiler: Move blob_init earlier to prevent crash on failure
2843 - microsoft/compiler: Add copy_prop_vars to optimization loop
2844 - microsoft/compiler: Add a lowering pass to split clip/cull distance compact arrays
2845 - microsoft/compiler: Enable dxil_nir.h to be included from C++
2846 - microsoft/compiler: Support compact arrays for clip/cull in nir_to_dxil
2847 - d3d12: Use compact arrays for clip/cull distance
2848 - microsoft/spirv_to_dxil: Implement TODO for removing dead functions
2849 - spirv_to_dxil: Handle clip/cull distance
2850 - microsoft/compiler: Fix barrier flag for shared memory
2851 - microsoft/spirv_to_dxil: Lower globals to function_temp
2852 - microsoft/spirv_to_dxil: Lower io arrays
2853 - microsoft/compiler: Support fp16 i/o vars
2854 - nir: Add a new opcode for [un]packing doubles
2855 - microsoft/compiler: Add a lowering pass to emit double [un]pack instructions
2856 - microsoft/compiler: Implement new double pack/unpack alu ops
2857 - microsoft/spirv_to_dxil: Support doubles
2858 - microsoft/compiler: Add some more float16 support
2859 - meson: Refuse to build lavapipe without llvmpipe
2860 - vtn: Don't warn about linkage capability if we're creating a NIR library
2861 - vtn: Add a cap for CL drivers to support read-write images
2862 - microsoft/clc: Update unit test to always use COMMON state for buffers
2863 - meson: For MSVC, suppress warnings generated by useless delayloads
2864 - driconf: Remove default values from string driconf entries
2865 - CI: Enable -werror for Windows
2866 - vtn: Support scoped control barriers for OpenCL too
2867 - nir_opt_deref: ptr_as_array(deref_cast<T*>(x))[0] isn't the same as x[0] if the cast has alignment
2868 - nir: Fix MSVC warning C4334 (32bit shift cast to 64bit)
2869 - d3d12: Fix MSVC warning C4334 (32bit shift cast to 64bit)
2870 - microsoft/clc: Fix MSVC unreferenced variable warnings
2871 - microsoft/clc: Fix undeclared function warning
2872 - microsoft/compiler: Fix MSVC warning C4334 (32bit shift cast to 64bit)
2873 - shader_enums: Fix MSVC warning C4334 (32bit shift cast to 64bit)
2874 - gallium/aux: Fix MSVC warning C4334 (32bit shift cast to 64bit)
2875 - llvmpipe: Fix MSVC warning C4334 (32bit shift cast to 64bit)
2876 - xmlconfig: Fix MSVC warning C4334 (32bit shift cast to 64bit)
2880 - glcpp: Fix undefined behaviour in glcpp
2884 - freedreno/a2xx: fix/add RBBM perfcounter
2885 - freedreno/a2xx: add RB perfcounter 1-3
2889 - turnip: fix logicOp
2890 - turnip: delete unused vk_format_parse.py file
2891 - turnip: use vk_format_is_int to disable COLOR_ATTACHMENT_BLEND_BIT
2892 - turnip: IMAGE_FILTER_{LINEAR,CUBIC}_BIT only for non-integer formats
2893 - turnip: don't always use 3d ops for blit_image
2894 - turnip: add missing register write to disable dithering
2895 - freedreno/registers: use macro instead of inline function for array regs
2896 - freedreno/a6xx: update perfcntr registers (declare as arrays)
2897 - freedreno/a6xx: always use reg64 for address registers (no LO/HI)
2898 - freedreno/a6xx: update some registers
2899 - freedreno/a6xx: set SP_PERFCTR_ENABLE in computerator
2900 - turnip: implement VK_KHR_shader_float_controls
2901 - turnip: enable VK_KHR_shader_float16_int8
2905 - intel/genxml/gen125: Add CFE_STATE and COMPUTE_WALKER
2906 - intel/compiler: Disable push constants on gen12-hp
2907 - anv: Emit CFE_STATE for gen12-hp
2908 - anv: Don't use MEDIA_INTERFACE_DESCRIPTOR_LOAD for gen12-hp
2909 - anv: Use COMPUTE_WALKER for gen12-hp
2910 - iris: Add support for COMPUTE_WALKER
2911 - iris: Fix android build due to missing link to libmesa_iris_gen125
2912 - anv: Add exec_flags to anv_queue
2913 - anv: Turn device->queue into an array
2914 - anv: Print queue number with INTEL_DEBUG=bat
2915 - anv: Support i915 query (DRM_IOCTL_I915_QUERY) from Linux v4.17
2916 - anv: Gather engine info from i915 if available
2917 - anv: Add anv_gem_count_engines
2918 - anv: Support multiple engines with DRM_IOCTL_I915_GEM_CONTEXT_CREATE_EXT
2919 - anv: Add ANV_QUEUE_OVERRIDE env-var to override advertised queues
2920 - i965: Call util_cpu_detect() early in screen creation
2921 - intel: Use GEN_VERSIONx10 in more places
2922 - intel/dev: Add devinfo genx10 field
2923 - intel: Use devinfo genx10 field
2924 - anv: Restructure mem heap/type init code
2925 - anv: Put cache memory type first on non-llc platforms
2926 - anv: Add mem heap/type support for local-mem
2927 - anv: Drop has_slm in emit_l3_config for gen11+
2928 - anv: Use fallback paths if DRM_I915_QUERY_ENGINE_INFO fails
2929 - i965/gen11: Fix must-be-ones bit positions in 3D_MODE
2930 - genxml/gen12: 3D_MODE bits 31:16 are no longer must-be-one
2931 - Revert "intel/compiler: Silence unused parameter warning in update_inst_scoreboard"
2932 - intel/compiler: Fix INTEL_DEBUG=hex
2936 - v3d: Enables DRM_FORMAT_MOD_BROADCOM_SAND128 support
2937 - v3d: YUV formats at is_dmabuf_modifier_supported are external_only
2938 - v3d: YUV formats at query_dmabuf_modifiers are external_only
2939 - v3d: DRM_FORMAT_MOD_BROADCOM_SAND128 only available for NV12 format.
2943 - lavapipe: handle NULL pStrides in CmdBindVertexBuffers2EXT
2944 - lavapipe: implement CmdBindVertexBuffers with CmdBindVertexBuffers2EXT
2945 - lavapipe: enable KHR_image_format_list
2949 - scons: Add u_format_pack.h include path.
2950 - wgl: Match opengl32.dll export ordinals.
2951 - util: Always use timespec_get on Windows.
2952 - appveyor: Remove integration.
2953 - scons: Remove.
2954 - gitlab-ci: Build meson-mingw32-x86_64 w/o zlib.
2958 - v3d: fix dest offset in TFU setup
2959 - v3d: use a compatible supported format for TFU-based blit
2960 - vc4/ci: Replace expect script by python script
2961 - ci/vc4: allow custom timeout values for activity
2962 - ci/vc4: rename stage to Broadcom
2963 - ci/vc4: Add piglit job
2964 - ci: Bump deqp to current vulkan-cts-1.2.5.1
2965 - ci: add option to overwrite CPU arch
2966 - ci/v3d: Add V3D and V3DV testing
2967 - ci/v3d: Update expected resuls for piglit
2968 - ci/piglit: allow parallel piglit jobs
2969 - ci/vc4/v3d: Parallelize piglit jobs
2970 - ci/piglit: fix parallel piglit results
2971 - ci/baremetal: highlight message errors
2972 - ci/broadcom: retry always when serial log timeout
2973 - ci: Bump deqp to vk-gl-cts 1.2.5.2
2974 - ci/broadcom: allow custom kernels
2975 - vc4: destroy renderonly object if present
2976 - ci/armXX: add libgl1-mesa-dev dependency
2977 - ci/v3dv: add flaky test in the skip list
2978 - ci/vc4/v3d: run piglit testsuite against Xorg
2979 - ci/broadcom: use new piglit runner
2980 - ci/broadcom: update piglit expected results
2981 - ci/v3d: run full GLES3 and GLES31 testsuite
2982 - broadcom/compiler: fix unused value
2983 - v3dv: fix unused value
2984 - ci/v3dv: update flaky tests
2985 - broadcom/cle: do not leak spec
2986 - ci/broadcom: update expected list
2987 - v3d: use uint type in _gen_unpack_uint
2988 - broadcom/compiler: fix first_component assertion
2989 - broadcom/compiler: use signed pointers for packed condition
2990 - ci/broadcom: use SNMP to turn on/off devices
2991 - broadcom/compiler: use VPM offsets in GS load_per_vertex input
2992 - v3d: use GS_BIN inputs as VS_BIN outputs
2993 - v3dv: fix assertion
2994 - ci: Update VK-GL-CTS to 1.2.6.0
2995 - v3d: do not emit attribute if has no resource
2996 - ci/v3dv: skip Vulkan waiver tests
2997 - util/hash_table: do not leak u64 struct key
3001 - gallium: add support for SVM (Share Virtual Memory) migrate
3002 - clover: implement clEnqueueSVMMigrateMem
3003 - nouveau: add support for SVM migrate
3007 - clover: track allocated svm pointers
3008 - clover/api: make use of validate_mem_migration_flags in clEnqueueMigrateMemObjects
3009 - nouveau: print warning about unhandled cap only once
3010 - clover: simplify image arguments
3011 - clover: rework quering image max sizes
3012 - clover: Fix build with llvm-12.
3013 - clover: Add missing include for llvm-12 build fix
3014 - tegra/context: fix regression in tegra_draw_vbo
3015 - tegra/context: unwrap indirect_draw_count as well
3019 - glx: Provide glvnd wrapper for glXSwapIntervalEXT
3020 - wsi/x11: Fix type of target_msc argument to x11_present_to_x11_dri3
3024 - vbo: Don't set node->min_index = max_index = indices_offset when merging
3025 - vbo: Only mark merged line strips as lines when actually converting them
3026 - tnl: Try not to botch index buffer munging when start > 0.
3027 - tnl: Respect \`start` when converting indices to GLuint
3028 - tnl: Reset nr_bos to 0 between map/unmap cycles.
3029 - Revert "mesa: allow half float textures based on ARB_half_float_pixel"
3030 - iris: Consider resolves after changing a resource's aux state
3031 - iris: Drop find_existing_assembly optimization from program cache
3032 - iris: Drop iris_print_program_cache().
3033 - iris: Refactor iris_debug_recompile interface to take a shader.
3034 - intel: Produce a "constrained" output from gen_get_urb_config()
3035 - iris: Reconfigure the URB only if it's necessary or possibly useful
3036 - iris: Move the URB size checks into iris_update_compiled_xs
3037 - iris: Properly handle new unbind_num_trailing_slots parameters
3038 - iris: Use shader_info rather than vs_prog_data for draw parameter checks
3039 - iris: Minor code restyling in iris_bind_vs_state
3040 - iris: Move VS draw parameter dirty flagging to iris_bind_vs_state
3041 - iris: Refcount shader variants
3042 - iris: Store a list of shader variants in the shader itself
3043 - iris: Enable PIPE_CAP_SHAREABLE_SHADERS.
3044 - iris: add hooks to call INTEL_MEASURE
3045 - iris: Fill out scratch base address dynamically
3046 - iris: Remove context from iris_debug_recompile
3047 - iris: Remove context from iris_upload_shader()
3048 - iris: Remove context from iris_compile_vs and friends
3049 - iris: Remove context from iris_create_uncompiled_shader
3050 - iris: Remove context from iris_disk_cache_retrieve
3051 - iris: Make a pin_scratch_space() helper
3052 - iris: Reference the shader variant for last_vue_map as well
3053 - iris: Pin surface state buffers after possibly updating the clear color
3054 - i965: Rename use_intel_mipree_map_blit to use_blitter_to_map
3055 - i965: Rename intel_batchbuffer_* to brw_batch_*.
3056 - i965: Rename intel_screen to brw_screen
3057 - i965: Rename intel_texture_{object,image} to brw_texture_{object,image}
3058 - i965: Rename intel_renderbuffer to brw_renderbuffer
3059 - i965: Rename intel_mip* to brw_mip*.
3060 - i965: Use __func__ in blorp perf_debug macros
3061 - i965: Rename intel_buffer_object to brw_buffer_object
3062 - i965: Rename intel_image_format and intel_buffer to brw_*
3063 - i965: Rename the rest of intel_* functions to brw_*
3064 - i965: Rename intelInit and brwInit camel-case functions to brw_*
3065 - i965: Rename some camel-case local variables
3066 - i965: Rename more camel-case functions to brw and underscore style
3067 - i965: Rename DRI extension structs to be "brw" instead of "intel"
3068 - i965: Eliminate all tabs except in brw_defines.h
3069 - tnl: Call _mesa_matrix_analyse to make sure the inverse MVP is updated
3070 - glsl/float64: Bump #version to 400
3071 - iris: Defer uploading of surface states
3072 - iris: Defer stream output target space allocation until set time
3073 - iris: Rework zeroing of stream output buffer offsets
3074 - iris: Support rebinding of stream output targets
3075 - iris: Use different shader uploaders for precompile vs. draw time
3076 - iris: Make various classes inherit from u_threaded_context base classes
3077 - iris: Use thread safe slab allocators in transfer_map handling
3078 - iris: Enable u_threaded_context
3079 - vbo: Fix vbo_sw_primitive_restart for start > 0
3080 - intel/genxml: Add a partial GT_MODE definition for Gen11+.
3081 - iris: Delete stale comment in iris_lost_context_state
3082 - intel: Fix release build breakage
3083 - Half-revert "gallium/dri2: Pass the resource that corresponds to the plane"
3084 - intel: Mark an otherwise unused variable in intel_dump_gpu as ASSERTED
3085 - ci: Enable iris testing in meson-release
3089 - macros: Add thread-safety annotation macros
3093 - radeon/vcn: clean the message buffers and their indexes logic
3094 - radeon/vcn: add dynamic dpb interface
3095 - radeon/vcn: add dynamic dpb buffer Tier1 support
3096 - radeon/vcn: enable dynamic dpb Tier1 support
3097 - radeon/vcn: add dynamic dpb Tier2 message buffer interface
3098 - radeon/vcn: implement dynamic dpb Tier2 support
3099 - radeon/vcn: enable dynamic dpb Tier2 support
3100 - meson: bump drm amdgpu version to 2.4.105
3101 - ci: Fix meson-i386 build failed after libdrm bump version
3102 - include/drm-uapi: bump AMDGPU headers
3103 - ac: add function for querying video capabilities
3104 - radeonsi: replace the hard coded video decode and encode caps
3108 - virgl: Don't destroy resource while it's in use.
3109 - virgl: Use atomic operation directly.
3110 - virgl: move new added field to the end.
3114 - anv: add transfer usage for color/depth/stencil attachments
3115 - anv: don't disable KHR_performance_query in debug mode
3116 - intel/mi_builder: optimize 64bit immediate register loads & memory stores
3117 - intel/mi_builder: fix self modifying batches
3118 - intel/perf: restructure i915 perf version checks
3119 - intel/perf: add definition for generic perf counters
3120 - intel/perf: link queries back to the gen_perf_config object
3121 - intel/perf: move gt_frequency to results
3122 - anv: Fix stencil layout in render passes
3123 - intel: silence unused var warnings in release builds
3124 - anv: fix invalid programming of BLEND_STATE
3125 - intel/common: store sample position in plain arrays
3126 - anv: pass context to reset stats helper
3127 - anv: store queue creation flags on anv_queue
3128 - genxml: PERFCNT registers are available since HSW
3129 - intel/perf: prep work to enable new perf counters
3130 - intel/perf: query register descriptions
3131 - intel/perf: add performance query layout using MI_SRM
3132 - intel/perf: switch query code to use query layout
3133 - anv: fix layout comment
3134 - anv: remove unused query pool field
3135 - intel/perf: rename lkf into ehl
3136 - intel/perf: add reorder script
3137 - intel/perf: reorder xml files
3138 - intel/perf: remove reordering script
3139 - intel/perf: update files from IGT
3140 - intel/perf: small ICL equation refactor
3141 - intel/perf: add async compute metrics
3142 - intel/dev: identify tigerlake
3143 - intel/perf: break TGL perf configs in GT1/2
3144 - intel/dev: identify rocketlake
3145 - intel/perf: add RKL support
3146 - intel/perf: add DG1 support
3147 - intel/perf: drop the special READ_REG operator
3148 - anv: compute commands required to implement perf queries
3149 - anv: switch khr perf query code to use query layout
3150 - anv: switch intel perf queries to query layout
3151 - anv: add a comment describing has_relocs field
3152 - anv: break up internal queueing function
3153 - anv: only signal wsi fence BO on last command buffer
3154 - drm-shim: report support for timeline semaphores
3155 - intel/stub: plug some gaps in our ioctl faking
3156 - anv: print out perf permission warning only once
3157 - anv: discard all timeline wait/signal value=0
3158 - vulkan: document flags choice for vkGetDeviceQueue
3159 - genxml: add MI_SET_APPID on Gen12+
3160 - genxml: Add PIPE_CONTROL protected memory bits
3161 - isl: add external parameter to isl_mocs()
3162 - anv: track command buffer pool flags
3163 - anv: track buffer creation flags
3164 - intel/dev: identify alderlake
3165 - intel/perf: Add Alderlake metrics
3166 - intel/perf: fix roll over PERF_CNT counter accumulation
3167 - anv: reset binary syncobj to be signaled before submission
3168 - anv: don't wait for completion of work on vkQueuePresent()
3169 - anv: Fix wait_count missing increment
3170 - anv: make use of new helper function directly in anv_QueueSubmit()
3171 - anv: track the end of the command buffers
3172 - anv: end command buffer with a potential jump
3173 - anv: allow multiple command buffers in anv_queue_submit
3174 - anv: group as many command buffers into a single execbuf
3175 - anv: fix missing general state pool in validation list
3176 - anv: implement INTEL_DEBUG=submit
3177 - anv: fix MI_PREDICATE_RESULT write
3178 - intel/tools: fix meson warning
3179 - intel/dev: add helpers to compute subslice/eu total
3180 - intel/dev: add warning on missing kernel uAPI for Gen8+
3181 - iris: use gen_device_info helper to get subslice total
3182 - i965: stop using get_param for things queried by gen_device_info
3183 - anv: stop using get_param for things queried by gen_device_info
3184 - intel/dev: switch over to mesa log infrastructure
3185 - anv: move L3 config emission to genX_state.c
3186 - anv: move L3 initialization to device init on Gen11+
3187 - intel: install intel_device_info
3188 - intel/fs/vec4: add missing dependency in write-on-write fixed GRFs
3189 - intel/dev: store size of CS prefetch
3190 - intel/mi_builder: use device info to use the right CS prefetch size
3191 - anv: use the device size of CS prefetch to pad secondary buffer calls
3192 - meson: switch vulkan layer to list of choices
3193 - intel: Add null hw layer
3194 - gitlab-ci: fix vulkan build layer enabling
3195 - intel/nullhw: fix build
3196 - etnaviv/drm: only print out fence error on non timeout
3197 - intel/fs/copy_prop: check stride constraints with actual final type
3198 - intel/fs: implement another copy propagation restriction
3199 - intel/compiler: lower bit sizes in NIR postprocessing
3200 - anv: put correct number of BT prefetch for compute on XeHP+
3201 - intel/fs: limit OW reads to 8 owords on XeHP+
3202 - microsoft: fixup clc_log() define
3203 - anv: bump internal descriptor index fields to 32bits
3204 - anv: fix 3DSTATE_MULTISAMPLE emission on gen8+
3205 - anv: disable baked in pipeline bits from dynamic emission path
3206 - spirv: fix uToAccelerationStructure handling
3207 - spirv: fixup pointer_to/from_ssa with acceleration structures
3208 - vulkan/wsi/display: don't report support if there is no drm fd
3209 - i965/bufmgr: fix invalid assertion
3213 - renderonly: remove layering violations
3214 - renderonly: close the gpu fd when destroying renderonly
3215 - etnaviv: don't try to copy PIPE_BUFFER with the 3D engine
3216 - etnaviv: remove stale comment in etna_resource_copy_region
3217 - Revert remaining half of "gallium/dri2: Pass the resource that corresponds to the plane"
3221 - v3dv: fix assertion in job_compute_frame_tiling
3222 - v3dv: fix stride in buffer copy
3226 - intel/perf: export information about units of performance counters
3227 - intel/compiler: cache computed register pressure benefit
3228 - intel/tools/aub: print better error message when mmap fails
3229 - intel/tools/aub: handle truncated input file
3230 - intel/tools/aub: remove superfluous new line from error messages
3231 - intel/dump_gpu: mark bo as unmapped if its address changes
3232 - anv: fix memory allocation error handling
3233 - iris: fix decode_get_bo
3234 - i965: fix decode_get_bo
3235 - intel/batch_decoder: catch invalid sampler state pointer
3236 - intel/batch_decoder: drop bogus check
3237 - intel/batch_decoder: fix decoding of sampler states
3238 - intel/batch_decoder: assert on invalid sampler pointer
3239 - intel/aub_viewer: catch invalid sampler state pointer
3240 - intel/aub_viewer: drop bogus check
3241 - intel/aub_viewer: fix decoding of sampler states
3242 - gallium: add PIPE_CAP_ALLOW_DYNAMIC_VAO_FASTPATH
3243 - iris: disable dynamic VAO fastpath on GFX version 9
3244 - gallium/u_threaded: implement INTEL_performance_query hooks
3245 - gallium/u_threaded: offload begin/end_intel_perf_query
3249 - mesa: always set valid index bounds for non-indexed draws for classic drivers
3250 - st/nine: stop using cso_set_sampler_views
3251 - st/xa: stop using cso_set_sampler_views
3252 - gallium/tests: stop using cso_set_sampler_views
3253 - gallium/api: add state invalidate interface as alternative to cso_save/restore
3254 - gallium/hud: don't use cso_context to restore VBs, constbuf 0 and sampler views
3255 - gallium/pp: don't use cso_context to restore VBs, constbuf 0 and sampler views
3256 - st/mesa: don't use cso_context to restore VBs, sampler views for glBitmap
3257 - st/mesa: don't use cso_context to restore VBs for glClear
3258 - st/mesa: don't use cso_context to restore VBs, sampler views for glDrawPixels
3259 - st/mesa: don't use cso_context to restore VBs, sampler views for glDrawTex*OES
3260 - st/mesa: don't use cso_context to restore VBs, etc. for PBO glReadPixels
3261 - st/mesa: don't use cso_context to restore VBs, etc. for PBO glTexSubImage
3262 - st/mesa: don't use cso_context to set const bufs, sampler views and images
3263 - st/mesa: replace st->pipe with pipe in a few places
3264 - cso_context: remove ability to restore VBs, const bufs, sampler views, images
3265 - st/mesa: unbind sampler views, images, and vertex buffers after meta ops
3266 - st/mesa: optimize binding and unbinding shader images
3267 - radeonsi: constant buffer cleanups
3268 - radeonsi: don't clear unaligned bits when unbinding vertex buffers
3269 - radeonsi: move emit_cache_flush functions into si_gfx_cs.c
3270 - radeonsi: don't pass pipe_draw_info into si_emit_vs_state
3271 - radeonsi: don't pass pipe_draw_info into si_emit_ia_multi_vgt_param
3272 - radeonsi: translate pipe_prim_type only when it changes
3273 - radeonsi: don't pass pipe_draw_info into si_emit_derived_tess_state
3274 - radeonsi: don't compute average vertex count in si_draw_vbo
3275 - radeonsi: fix si_num_prims_for_vertices for PIPE_PRIM_POLYGON
3276 - radeonsi: make cik_emit_prefetch_L2 templated and move it to si_state_draw.cpp
3277 - radeonsi: add a specialized function for CP DMA L2 prefetch
3278 - radeonsi: make sctx->vertex_elements always non-NULL
3279 - radeonsi: remove MRT-draw-calls, spill-draw-calls, spill-compute-calls
3280 - radeonsi: get out of si_emit_vs_state early for blit vertex shaders
3281 - radeonsi: rearrange condition for streamout workaround on gfx7 and gfx8
3282 - radeonsi: don't use si_get_vs_state in most places
3283 - radeonsi: trim the size of si_vgt_param_key and si_vgt_stages_key
3284 - mesa: fix alpha channel of ETC2_SRGB8 decompression for !bgra
3285 - radeonsi: unify uploaders on APUs too
3286 - radeonsi: don't pass pipe_draw_info into si_emit_draw_registers
3287 - radeonsi: don't set context_roll for non-gfx9 in templated functions
3288 - radeonsi: add si_get_user_data_base selecting user data registers
3289 - radeonsi: evaluate sh_base in si_emit_vs_state at compile time
3290 - radeonsi: inline the last use of si_get_vs_state
3291 - radeonsi: evaluate si_get_vs in si_draw_vbo at compile time
3292 - radeonsi: enable the GS tri strip adj workaround with primitive_restart
3293 - radeonsi: clear dirty_atoms and dirty_states only if we entered the emit loop
3294 - radeonsi: move variables closer to their use in most draw state functions
3295 - radeonsi: don't validate inlinable uniforms at draw time
3296 - radeonsi: allow instance_count == 0 on chips that handle it correctly
3297 - glthread: remove marshal="draw" because it doesn't do much
3298 - glthread: don't sync with NV_half_float vertex attrib functions
3299 - glthread: add specialized versions of unmarshal_Draw funcs without user buffers
3300 - glthread: track all matrix stack depths
3301 - glthread: implement glGetIntegerv for states that glthread tracks
3302 - glthread: rename inside_dlist to ListMode for future use
3303 - glthread: remove if (COMPAT) conditions from functions that are GL-compat-only
3304 - mesa: add _mesa_get_list helper
3305 - glthread: add display list support to fix state tracking with display lists
3306 - mesa: remove _mesa_initialize_exec_dispatch from draw.c by autogenerating it
3307 - mesa: remove redundant glRect functions for display lists
3308 - mesa: optimize glCallLists by using loops inside a switch
3309 - mesa: simplify handling OPCODE_CONTINUE for display lists
3310 - mesa: simplify terminating display list loops
3311 - mesa: remove STATE_INTERNAL
3312 - mesa: combine STATE_ENV, STATE_LOCAL enums with STATE_xxx_PROGRAM
3313 - mesa: flatten STATE_MATERIAL and STATE_LIGHTPROD tokens
3314 - mesa: eliminate the switch statement for STATE_TEXGEN
3315 - glsl: remove unused internal builtin gl_CurrentAttribVertMESA
3316 - glsl: split gl_CurrentAttribFragMESA into elements
3317 - mesa: skip memmove in optimize_state_parameters if it's no-op
3318 - mesa: rename STATE_LIGHT_ATTRIBS -> STATE_LIGHT_ARRAY for consistency
3319 - mesa: optimize get_local_param_pointer and program_local_parameters4fv
3320 - mesa: don't allocate local parameters in fetch_state
3321 - mesa: merge local and env program parameters for faster uploads
3322 - mesa: sort state vars with constant indexing for ARB programs
3323 - mesa: add upper bound to limit program state var iterations
3324 - mesa: compute gl_program_parameter_list::UniformBytes accurately
3325 - mesa: don't handle STATE_* enums in fetch_state that don't do anything
3326 - mesa: sort and tightly pack STATE_* enums to generate better switch code
3327 - mesa: merge equivalent switch cases in prog_statevars.c
3328 - st/mesa: enable state var merging to improve fetch_state performance
3329 - radeonsi: add new possibly faster command submission helpers
3330 - radeonsi: clear dirty_states if si_pm4_bind_state is unbinding or no-op
3331 - radeonsi: don't mark NULL states as dirty in si_pm4_reset_emitted
3332 - radeonsi: optimize translating index_size to index_type
3333 - radeonsi: don't use rasterizer_discard to validate draws, only check ps_shader
3334 - radeonsi: add internal blitter_running flag
3335 - radeonsi: simplify determining whether render condition is enabled at draw time
3336 - radeonsi: inline si_blend_color and si_clip_state structures
3337 - radeonsi: move y_inverted out of si_viewports
3338 - radeonsi: don't set vertex buffer dirty flags when they don't do anything
3339 - radeonsi: move if (sctx->vertex_buffers_dirty) into the upload function
3340 - radeonsi: rename SI_SGPR_RW_BUFFERS to SI_SGPR_INTERNAL_BINDINGS
3341 - radeonsi: skip some code for ALLOW_PRIM_DISCARD_CS if tess or GS is enabled
3342 - radeonsi: enable accidentally disabled fast launch with non-indexed tri strips
3343 - radeonsi: iterate from draw 1 for total/min_direct_count computation
3344 - st/mesa: don't enable smoothing if multisampling is enabled
3345 - Revert "gallium/u_upload_mgr: allow use of FLUSH_EXPLICIT with persistent mappings"
3346 - gallium: add take_ownership param into set_constant_buffer to eliminate atomics
3347 - gallium: add unbind_num_trailing_slots to set_vertex_buffers
3348 - gallium: add unbind_num_trailing_slots to set_shader_images
3349 - gallium: add unbind_num_trailing_slots to set_sampler_views
3350 - gallium: add take_ownership param into set_vertex_buffers to eliminate atomics
3351 - cso_context,u_vbuf: add take_ownership param into set_vertex_buffers
3352 - st/mesa: eliminate all atomic ops when setting vertex buffers
3353 - st/mesa: skip atomics when binding UBOs
3354 - gallium/u_upload_mgr: eliminate all atomics for the upload buffer
3355 - gallium/u_threaded: add a null constant buffer codepath
3356 - gallium/u_threaded: unify user and non-user codepaths in set_constant_buffer
3357 - gallium/util: optimize pipe_vertex_buffer_reference binding the same buffer
3358 - gallium,u_threaded: add pipe_draw_info::take_index_buffer_ownership
3359 - st/mesa: set take_index_buffer_ownership to skip an atomic in u_threaded
3360 - gallium/u_vbuf: skip draws with 0 vertices
3361 - radeonsi: fix centroid with VRS coarse shading
3362 - glthread: fix interpreting vertex size == GL_BGRA for vertex attribs
3363 - glthread: fix glVertexAttribDivisor calls not being tracked by non-VBO uploads
3364 - glapi: guard against invalid XML definitions for glthread
3365 - ac,radeonsi: track memory usage in KB to reduce types from uint64 to uint32
3366 - radeonsi: optimize no-op cases in si_upload_shader_descriptors
3367 - radeonsi: mark shader_pointers dirty once outside the upload descriptors loop
3368 - radeonsi: move si_pm4_delete_state logic into si_pm4_free_state
3369 - radeonsi: delete si_pm4_delete_state
3370 - radeonsi: don't check for redundant and NULL states in si_emit_all_states
3371 - radeonsi: optimize si_emit_prefetch_L2 when it's split
3372 - radeonsi: reorganize si_draw_vbo for lower register pressure (part 1)
3373 - radeonsi: reorganize si_draw_vbo for lower register pressure (part 2)
3374 - radeonsi: set VB user SGPRs in si_upload_vertex_buffer_descriptors
3375 - radeonsi: prefetch VB descriptors right after uploading
3376 - radeonsi: precompute NGG cull flags in si_create_rs_state
3377 - mesa: remove/replace FLUSH_VERTICES when it doesn't do anything
3378 - mesa: optimize most _mesa_ActiveTexture calls in glPopAttrib
3379 - mesa: optimize glPopAttrib(GL_VIEWPORT_BIT)
3380 - mesa: don't push/pop gl_texture_object::Target
3381 - mesa: fix glPopAttrib for many texture fields
3382 - mesa: flush glBegin/End before changing GL_DEPTH_STENCIL_TEXTURE_MODE
3383 - mesa: for every state change, remember states we changed for glPopAttrib
3384 - mesa: fix trivial bugs in glPopAttrib
3385 - mesa: optimize out _NEW_ALL in glPopAttrib(GL_ENABLE_BIT)
3386 - mesa: only pop states in glPopAttrib that have been changed since glPushAttrib
3387 - mesa: partially skip glPush/PopAttrib for MSAA textures and texture buffers
3388 - mesa: don't save gl_shared_state in glPushAttrib
3389 - mesa: don't push/pop default texture attributes redundantly
3390 - mesa: pop all textures up to NumCurrentTexUsed, not just MaxTextureUnits
3391 - mesa: don't count buffer references for the context that created them
3392 - radeonsi: set current_rast_prim at bind time for tess and GS
3393 - radeonsi: simplify the NGG culling condition in si_draw_vbo
3394 - radeonsi: tune NGG shader culling vertex threshold for each chip
3395 - radeon: decrease the size of radeon_cmdbuf by switching prev fields to uint16
3396 - Revert "gallium/u_vbuf: skip draws with 0 vertices"
3397 - gallium/u_vbuf: skip non-indirect draws with 0 vertices
3398 - winsys/amdgpu,radeonsi: add HUD counters for how much memory is wasted by slabs
3399 - winsys/amdgpu: clean up slab alignment code, handle small buffers better
3400 - winsys/amdgpu,pb_slab: add slabs with 3/4 of power of two sizes to save memory
3401 - winsys/amdgpu: expand the slab allocation range to [256 B, 1 MB]
3402 - winsys/amdgpu: optimize out conditionals in amdgpu_lookup_buffer
3403 - winsys/amdgpu: remove amdgpu_winsys_bo::num_cs_references to remove atomics
3404 - winsys/amdgpu: pack amdgpu_winsys_bo::is_shared and protect it by a mutex
3405 - winsys/amdgpu: move amdgpu_winsys_bo::cpu_ptr into the u.real union
3406 - winsys/amdgpu: move amdgpu_winsys_bo::is_shared to the u.real union
3407 - winsys/amdgpu: move amdgpu_winsys_bo::is_user_ptr to the u.real union
3408 - winsys/amdgpu: move amdgpu_winsys_bo::use_reusable_pool to the u.real union
3409 - winsys/amdgpu: don't inc/dec num_active_ioctls for backing BOs of sparse BOs
3410 - winsys/amdgpu: don't set unused usage for backing BOs of sparse BOs
3411 - winsys/amdgpu: try not to skip any code with RADEON_NOOP=1 to test CPU perf
3412 - tgsi_to_nir: translate SAMPLEID
3413 - tgsi_to_nir: translate FBFETCH
3414 - gallium/u_tests: test no-op fragment shader instead of NULL fragment shader
3415 - winsys/amdgpu: disallow pb_cache for backing buffers of sparse buffers
3416 - ac/gpu_info: print use_late_alloc
3417 - ac/gpu_info: rename num_tcc_blocks -> max_tcc_blocks
3418 - ac/gpu_info: add radeon_info::num_tcc_blocks
3419 - ac/gpu_info: remove redundant radeon_info::num_sdp_interfaces
3420 - ac/gpu_info: inline get_l2_cache_size and set cache sizes farther down
3421 - ac/gpu_info: conceal L2 cache sizes
3422 - amd: sort chip enums based on hw revision
3423 - radeonsi: skip s_sendmsg(gs_alloc_req) for NGG passthrough on new chips
3424 - radeonsi: add debug options nodisplaytiling and nodisplaydcc
3425 - amd: update addrlib
3426 - mesa: optimize draw index type checking
3427 - mesa: precompute all valid primitive types at context creation
3428 - mesa: precompute draw time prim validation during state changes
3429 - mesa: move check_valid_to_render call into _mesa_valid_prim_mode
3430 - mesa: fold most of check_valid_to_render into _mesa_update_valid_to_render_state
3431 - mesa: inline check_valid_to_render
3432 - mesa: add skeleton code for DrawPixels/CopyPixels/Bitmap precomputed validation
3433 - mesa: don't report 1 for GL_VALIDATE_STATUS if user didn't validate pipeline
3434 - mesa: move shader pipeline validation from draws to state changes
3435 - mesa: move sampler uniform validation from draws to state changes
3436 - mesa: move some uniform debug code from draws to state changes
3437 - mesa: move FBO completeness checking from draws to state changes
3438 - mesa: move ARB program and integer FBO validation from draws to state changes
3439 - mesa: move GL_FILL_RECTANGLE validation from draws to state changes
3440 - mesa: move blending validation from draws to state changes
3441 - mesa: inline _mesa_valid_to_render now that it doesn't do validation
3442 - mesa: optimize the dual source blend error checking using a bitmask
3443 - mesa: remove VERBOSE_DRAW
3444 - mesa: remove optional draw validation code to increase performance
3445 - mesa: call _mesa_update_state() before validation
3446 - mesa: remove an optional GL error about mapped buffers during execution
3447 - mesa: skip MultiDrawArrays with primcount == 0
3448 - mesa: don't skip draws with count == 0 or numInstances == 0
3449 - mesa: add a separate valid primitive mask just for glDrawElements
3450 - mesa: move disallowed TFB in DrawElements on GLES from draws to state changes
3451 - mesa: validate numInstances in common functions to unify code
3452 - mesa: optimize set_varying_vp_inputs by precomputing the conditions
3453 - mesa: move gl_context::varying_vp_inputs into ctx->VertexProgram._VaryingInputs
3454 - mesa: set _DrawVAOEnabledAttribs only when it changes
3455 - mesa: precompute _mesa_get_vao_vp_inputs
3456 - mesa: precompute draw time determination of enabled vertex arrays
3457 - mesa: gather errors and call _mesa_error only once in validate_Draw
3458 - mesa: inline _mesa_set_draw_vao and set_varying_vp_inputs for draw calls
3459 - mesa: inline draw validate functions
3460 - mesa: add debug code to catch missing _mesa_update_valid_to_render_state calls
3461 - ac/surface: use family_id so as not to crash with SI_FORCE_FAMILY in addrlib
3462 - radeonsi: for tess, determine the minimum num_patches before optimizing tg size
3463 - radeonsi: improve comments in si_emit_derived_tess_state
3464 - radeonsi: allocate filled_size for streamout targets in set_streamout_buffers
3465 - radeonsi: do late NIR optimizations after uniform inlining
3466 - radeonsi: fix the value of uses_bindless_samplers
3467 - radeonsi: gather info about bindless images and memory stores with strstr(intr)
3468 - radeonsi: gather shader info about indirect UBO/SSBO/samplers/images
3469 - radeonsi: gather shader info about VMEM usage for MEM_ORDERED
3470 - radeonsi: set MEM_ORDERED optimally
3471 - glthread: assume all parameters are fixed if marshal_sync is present
3472 - glthread: don't declare pointers with const in unmarshal functions
3473 - glthread: don't sync when using pixel buffer objects
3474 - glthread: ignore the return value of glUnmapBuffer, don't sync, and return true
3475 - i915: use align_calloc for the context to fix m32 crashes
3476 - radeon,r200: use align_calloc for the context to fix m32 crashes
3477 - nouveau_vieux: use align_calloc for the context to fix m32 crashes
3478 - mesa: remove unnecessary NewState flagging for glPopAttrib(GL_ENABLE_BIT)
3479 - mesa: move fixed-func-related _mesa_update_state code closer together
3480 - mesa: split _NEW_LIGHT into 3 flags: _NEW_LIGHT_(FF_PROGRAM|CONSTANTS|STATE)
3481 - mesa: rework _MESA_NEW_NEED_EYE_COORDS to reduce fixed-func program updates
3482 - mesa: don't compute the inverted projection matrix if not used
3483 - mesa: don't compute the ModelView * Projection matrix if not used
3484 - mesa: add _NEW_MATERIAL to reduce the weight of _NEW_LIGHT_CONSTANTS
3485 - mesa: don't update derived material values in _mesa_update_state and elsewhere
3486 - mesa: remove _NEW_VARYING_VP_INPUTS in favor of _NEW_FF_(VERT|FRAG)_PROGRAM
3487 - mesa: remove _NEW_LIGHT_FF_PROGRAM in favor of _NEW_FF_(VERT|FRAG)_PROGRAM
3488 - mesa: don't push/pop ctx->Texture._* derived states
3489 - mesa: remove the fixed-func vert prog dependency on all texture states
3490 - mesa: sort state parameters for ffvp to enable better parameter merging
3491 - mesa: merge STATE_LIGHTPROD parameters
3492 - mesa: merge STATE_LIGHT_ATTENUATION and STATE_LIGHT_POSITION_* parameters
3493 - vbo: optimize copy_to_current functions
3494 - vbo: don't call update_color_material in copy_to_current if it's a no-op
3495 - mesa: be precise about when to set _NEW_CURRENT_ATTRIB and _NEW_MATERIAL
3496 - mesa: move _mesa_update_pixel out of _mesa_update_state
3497 - mesa: only update fixed-func programs on texture matrix enablement changes
3498 - mesa: don't update fixed-func vert prog on irrelevant _NEW_TRANSFORM changes
3499 - mesa: don't update fixed-func programs on irrelevant _NEW_POINT changes
3500 - mesa: don't update fixed-func programs on irrelevant _NEW_FOG changes
3501 - mesa: don't update fixed-func programs on irrelevant _NEW_RENDER_MODE changes
3502 - mesa: don't update the fixed-func frag prog on irrelevant _NEW_COLOR changes
3503 - mesa: don't update tnl spaces on irrelevant _NEW_POINT/TEXTURE_STATE changes
3504 - mesa: skip a subset of _mesa_update_state if no relevant flags are set
3505 - radeonsi: don't index si_context::shaders with enum gl_shader_stage
3506 - ac/llvm: fix ac_build_atomic_rmw with LLVM 13
3507 - radeonsi: don't crash on NULL images in si_check_needs_implicit_sync
3508 - ac/llvm: add support for 16-bit source operands for samplers
3509 - ac/llvm: implement 16-bit and 64-bit fpow correctly
3510 - ac/llvm: fix visit_load_ubo_buffer to use SMEM for 16 bits instead of VMEM
3511 - ac/llvm: add type parameter into ac_build_buffer_load to fix 16-bit TES inputs
3512 - ac/llvm: open code fpow on LLVM 12 using fmul.legacy
3513 - driconf: add performance tweaks for viewperf
3514 - ac/surface: select best swizzle mode for 3D sampler performance
3515 - ac,radeonsi: add sampler changes for Aldebaran
3516 - ac: set the TCC line size for Aldebaran
3517 - ac/llvm: unpack thread IDs on Aldebaran
3518 - ac: handle bigger instruction prefetch for Aldebaran
3519 - ac,radeonsi: use correct VGPR granularity on Aldebaran
3520 - ac: remove switch cases for pc_lines for compute-only chips
3521 - radeonsi: enable RGP on gfx10.3
3522 - gallium/u_threaded: don't sync in create_stream_output_target
3523 - gallium: add pipe_screen::num_contexts for skipping mutex locking in util_range
3524 - radeonsi: update pipe_screen::num_contexts
3525 - ac/llvm: handle demote in LLVM 13 that just added support for it
3526 - ac/gpu_info: fix more non-coherent RB and GL2 combinations
3527 - radeonsi: use pipe_sampler_state::border_color_is_integer to simplify stuff
3528 - mesa: fix Blender crash due to optimizations in buffer reference counting
3529 - mesa: add assertions for buffer reference counts
3530 - mesa: fix a oldNum typo in reallocation in _mesa_reserve_parameter_storage
3531 - mesa: don't overallocate ParameterValues 4 times (v2)
3532 - mesa: clear reserved parameter storage because it's stored in the shader cache
3533 - mesa: fix parameter reservation size
3534 - st/mesa: add a driconf option to transcode ETC2 to DXTC
3535 - util: add most missing formats with reversed RGB channel order
3536 - util: fail assertion in util_format_linear if it can't translate SRGB format
3537 - util: add new helper util_format_rgb_to_bgr
3538 - radeonsi: select the optimal micro tile mode at clear regardless of fast clear
3539 - radeonsi: add a fast path for MSAA resolving with RGB -> BGR swizzling
3540 - amd/addrlib: add back the incorrect original DCC checking
3541 - amd/addrlib: prevent defining regparm differently
3542 - amd/addrlib: define endianess differently
3543 - amd: update addrlib
3544 - ac/llvm: don't set unsupported xnack options to fix LLVM crashes on gfx6-8
3545 - radeonsi: disable sparse buffers on gfx7-8
3546 - radeonsi: set the clear/copy cache policy based on L2 cache size
3547 - radeonsi: don't insert start/stop pipeline stat events if it has no effect
3548 - radeonsi: never set DISABLE_WR_CONFIRM for CP DMA clears and copies
3549 - radeonsi: rename internal compute sync flags
3550 - radeonsi: remove unused SI_CP_DMA_SKIP_* definitions
3551 - radeonsi: merge CP DMA flags with internal compute flags
3552 - radeonsi: inline clear_buffer in si_screen_clear_buffer
3553 - radeonsi: set compute/cpdma sync flags in the outermost caller
3554 - radeonsi: reduce syncing in si_dcc_decompress
3555 - radeonsi: reduce syncing for initializing new buffers
3556 - radeonsi: reduce syncing in si_compute_expand_fmask when it's already idle
3557 - radeonsi: don't do an L2 flush in compute_do_clear_or_copy if we're not syncing
3558 - radeonsi: rename and apply SI_OP_CPDMA_SKIP_CACHE_FLUSH to compute as well
3559 - radeonsi: use the optimal packet order before draw packets for VS flushes too
3560 - radeonsi: add SI_CONTEXT_PFP_SYNC_ME to skip syncing PFP for image operations
3561 - radeonsi: return false from si_is_format_supported instead of NULL
3562 - radeonsi: don't use constbuf and set cache policy for 12-byte clear shader
3563 - radeonsi: don't use a constant buffer for the copy_image compute shader
3564 - radeonsi: decrease the maximum variable block size
3565 - radeonsi: pack the variable block size in one SGPR, 10 bits per component
3566 - amd: fix parsing the last dword of DMA_DATA packets
3567 - ac/surface: add CMASK info for level 0
3568 - radeonsi: determine accurately whether the framebuffer state has DCC MSAA
3569 - radeonsi: remove si_screen::dcc_msaa_allowed
3570 - radeonsi: parallelize CMASK and DCC clears
3571 - radeonsi: return success/failure from si_alloc_separate_cmask
3572 - radeonsi: add num_layers variable into si_do_fast_color_clear
3573 - radeonsi: group and parallelize all clears in si_texture_create_object
3574 - radeonsi: set better default depth clear value
3575 - radeonsi: enable HTILE with mipmapping on gfx9+
3576 - radeonsi: unset PIPE_CLEAR_* flags for non-existent buffers
3577 - radeonsi: turn the loops over color buffers into while loops in si_clear
3578 - radeonsi: don't use CP DMA for clears/copies except for very small ones
3579 - ac/surface/tests: move shareable code into ac_surface_test_common.h
3580 - radeonsi: fix si_compute_copy_image if DCC decompression happens before a copy
3581 - gallium/pb: pass an optional winsys pointer to the buffer destroy function
3582 - winsys/radeon: rename radeon_bo_reference -> radeon_ws_bo_reference
3583 - radeon_winsys.h: add a new function radeon_bo_reference that takes a winsys
3584 - radeon_winsys.h: add a winsys parameter to most winsys buffer functions
3585 - winsys/amdgpu: remove amdgpu_winsys_bo::ws
3586 - winsys/amdgpu: add amdgpu_cs::ws to reduce dereferences
3587 - gallium/pb: change pb_buffer::alignment to alignment_log2
3588 - gallium/pb: remove 8 bytes from pb_buffer by packing variables
3589 - winsys/amdgpu: remove another 8 bytes from amdgpu_winsys_bo by packing better
3590 - ac/surface: split dcc level info from surface_info to save space
3591 - ac/surface: overlap color and Z/S fields using a union in legacy_surf_layout
3592 - ac/surface: change legacy_surf_level::offset to 32-bit offset_256B shifted by 8
3593 - ac/surface: inline and reorder gfx9_surf_flags for better packing
3594 - ac/surface: pack gfx9_surf_meta_flags better
3595 - ac/surface: pack gfx9_surf_layout:resource_type better to save 8 bytes
3596 - ac/surface: pack radeon_surf::num_htile_levels better
3597 - ac/surface: pack alignments by storing log2 in radeon_surf
3598 - ac/surface: overlap color and Z/S fields using a union in gfx9_surf_layout
3599 - ac/surface: pack radeon_surf better
3600 - ac/surface: unify htile_levels and dcc_levels as meta_levels
3601 - ac/surface: unify htile_* and dcc_* fields as meta_* fields
3602 - ac/surface: use named "color and "zs" structures in unions
3603 - radeonsi: don't cache FMASK transactions from CB in L2
3604 - radeonsi: restructure DCC disablement into a switch
3605 - radeonsi: allow trivial DCC clears for shared textures with DCC constant encode
3606 - radeonsi: implement per-level DCC and CMASK fast clears for gfx10+
3607 - radeonsi: implement Z/S fast clear for non-zero mipmap levels
3608 - radeonsi: parallelize Z/S conversion into TC-compatible with fast color clears
3609 - radeonsi: clean up some mess around htile_stencil_disabled
3610 - radeonsi: add si_can_fast_clear_depth/stencil helpers
3611 - radeonsi: indent the code for TC-compatibility HTILE transition
3612 - radeonsi: implement fast Z/S clears using clear_buffer on HTILE
3613 - radeonsi: enable DCC fast clears for non-zero mipmap levels and 0/1 clear values
3614 - radeonsi: when transitioning to TC-compat HTILE, try to do a proper clear
3615 - radeonsi: do Z-only or S-only HTILE clear using a compute shader doing RMW
3616 - radeonsi: refine fast clears for small buffers, always use them for large HTILE
3617 - radeonsi: try to fix DCC coherency issues with DCC decompression
3618 - radeonsi: allow DCC_DECOMPRESS via CB with MSAA textures
3619 - ac/surface: only apply the 3D swizzle mode tuning to gfx10+
3620 - ac/surface/tests: test Sienna Cichlid and Navy Flounder
3621 - ac/surface/tests: fix a random segfault in the modifier test
3622 - amd/addrlib: expose DCC address equations to drivers
3623 - meson: add an optional OpenMP dependency for AMD tests
3624 - ac/surface: add a test of DccAddrFromCoord prototype outside of addrlib
3625 - ac/surface: limit the number of swizzle modes that can have displayable DCC
3626 - ac,radeonsi: rewrite DCC retiling without the DCC retile map
3627 - radeonsi: fix and enable full DCC with MSAA 2x on gfx9
3628 - radeonsi: implement DCC MSAA 4x/8x fast clear using DCC equations on gfx9
3629 - radeonsi: enable DCC for MSAA 4x and 8x on gfx9
3630 - radeonsi: move binding the internal compute shader into si_launch_grid_internal
3631 - radeonsi: unify internal compute with SSBOs in si_launch_grid_internal_ssbos
3632 - compiler: move TEXTURE_COORD/VERTEX_GENERIC_ATTRIB limits into shader_enums.h
3633 - nir: add src and dest types to all IO loads and stores for mediump
3634 - nir: add new VARYING_SLOTs and shader info for packed 16-bit varyings
3635 - nir: add many passes that lower and optimize 16-bit input/outputs and samplers
3636 - glsl: pack 16-bit uniforms in the NIR linker
3637 - mesa: implement glUniform for packed FP16 uniforms
3638 - mesa: implement glGetUniform for FP16 uniforms
3639 - mesa: implement glGetActiveUniform for FP16 uniforms
3640 - glsl: lower mediump uniforms to FP16 based on an option
3641 - gallium: add PIPE_SHADER_CAP_FP16_CONST_BUFFERS for FP16 uniforms
3642 - st/mesa: fix nir_lower_io if it's done right after IO vectorization
3643 - ac/llvm: implement 16-bit packed VS outputs and FS inputs
3644 - radeonsi: implement 16-bit VS->PS varyings
3645 - radeonsi: implement 16-bit VS inputs
3646 - radeonsi: optimize and legalize 16-bit samplers
3647 - radeonsi: kill 16-bit VS outputs if PS doesn't use them or doing Z-only draw
3648 - radeonsi: enable FP16 for mediump on gfx9+ if radeonsi_fp16=true
3649 - nir: handle mediump varyings in varying compaction helpers
3650 - radeonsi: don't decompress DCC for float formats in si_compute_copy_image
3651 - radeonsi: fix automatic DCC retiling after DCC clear and DCC decompression
3652 - radeonsi: fix automatic DCC retiling after compute image stores
3653 - radeonsi: make the gfx9 DCC MSAA clear shader depend on the number of samples
3654 - util: fix (re-enable) L3 cache pinning
3658 - compiler/nir: Increment shader input count and mark as used when adding new gl_PointCoord
3659 - etnaviv: Fix point sprite Z,W coordinate replacement
3663 - intel: Print GPU timing data based on INTEL_MEASURE
3664 - anv: enable timestamp for INTEL_MEASURE
3665 - anv: implement anv layer of INTEL_MEASURE
3666 - blorp: add hook for INTEL_MEASURE
3667 - anv: add hooks to call INTEL_MEASURE
3668 - iris: implement iris layer of INTEL_MEASURE
3669 - iris: add a iris_context reference to iris_batch
3670 - intel: stop tracking submission state in INTEL_MEASURE
3671 - intel: support secondary command buffers in INTEL_MEASURE
3672 - intel: combine common gather routines in INTEL_MEASURE
3673 - intel: check setuid before writing output file in INTEL_MEASURE
3674 - Revert "blorp/gen12: Don't use aux address if implicit CCS"
3678 - docs/freedreno: Fix a few typos
3679 - turnip: Remove unused TU_DEBUG_IR3 flag
3680 - docs: Mark VK_KHR_maintenance1 as done on turnip
3681 - ci: Use CI_PROJECT_ROOT_NAMESPACE
3682 - tu: Skip tu_tiling_config_update_tile_layout() if not using gmem
3683 - ci: Disable panfrost g52
3684 - Remove Scons leftovers
3685 - ir3: Don't count (nopX) towards the wrong category
3689 - gallium: Fix broken trace XML output
3690 - gallium/tools: update trace scripts to Python 3
3694 - android: r600/sfn: add sfn_nir_lower_64bit.cpp to Makefile.sources
3695 - android: freedreno/hw/isa: Add description of ir3 ISA
3696 - android: freedreno/ir3: Switch over to new encoder/decoder
3697 - android: pan/mdg: create nir pass to lower image coord bitsize
3698 - android: intel: Print GPU timing data based on INTEL_MEASURE
3699 - android: anv: implement anv layer of INTEL_MEASURE
3700 - android: iris: implement iris layer of INTEL_MEASURE
3701 - android: radv: port to using common dispatch code.
3702 - android: radv: fix building error in radv_android.c
3703 - android: util/fossilize_db: add missing sources to Makefile.sources
3704 - android: ac/rgp: fix building error
3705 - android: mesa: Move the FXT1 compressor/decompressor to util/
3706 - android: pan/bi: reorder static dependencies in gallium/dri
3707 - driconf: avoid Non-ASCII character error in driconf_static.py
3708 - android: driconf: Generate a static table when no xmlconfig
3709 - android: i965: Rename files with "intel\_" prefix to "brw\_"
3710 - android: util: create some standalone compression helpers
3711 - android: anv: add libcutils shared dependency
3712 - android: r600/sfn: fix sfn_nir_algebraic.c gen rules
3713 - android: vulkan/util: add vk_descriptors.{c,h} to Makefile.sources
3714 - android: amd/addrlib: define endianess to build
3715 - android: panfrost: Use the blend shader cache attached to the device
3716 - vulkan/util: Fix implicit declaration of ffs for Android build
3717 - android: anv: Remove anv_intel.c from Makefile.sources
3718 - android: anv: fix build error in anv_android.c
3719 - compiler/glsl: fix include for Android build
3720 - android: panfrost/lib: add pan_cs.c to Makefile.sources
3721 - android: gallium/radeonsi: add nir include path
3722 - android: amd/common: add nir include path
3726 - microsoft/compiler: Make resource_state_manager only build with_gallium_d3d12
3727 - util: Make os_read_file use O_BINARY on Windows
3728 - microsoft/spirv_to_dxil: Fix spirv2dxil I/O to use binary mode
3729 - microsoft/spirv_to_dxil: Add lowering pass to handle gl_PerVertex
3730 - microsoft/spirv_to_dxil: Add extra lowering functions according to the docs on nir_inline_functio…
3734 - ci: Remove .gitlab-ci/meson-build.bat
3735 - ci: Use meson test directly instead of ninja test
3736 - wsi/x11: Use get_screen_resources_current in wsi_x11_detect_xwayland
3737 - ci: Enable process isolation for softpipe & freedreno piglit jobs
3738 - ci: Use GNU time as meson test wrapper
3739 - ci: Run 'time' in the background and propagate signals to test process
3740 - ci: Fix MESA_TEMPLATES_COMMIT value
3741 - ci: Update to newer ci-fairy
3742 - ci: Set GALLIVM_PERF=no_filter_hacks for llvmpipe-piglit-quick_shader
3743 - ci: Set GALLIVM_PERF=no_filter_hacks for llvmpipe-piglit-quick_gl
3744 - ci: Set GALLIVM_PERF=nopt,no_filter_hacks for llvmpipe-gles2
3745 - ci: Use MESA\_ namespace for image variables in Windows jobs
3746 - ci: Use MESA_IMAGE_TAG everywhere
3747 - ci: Move FDO_DISTRIBUTION_TAG assignment to template
3748 - ci: Add and use .set-image template to construct docker image name
3749 - ci: Incorporate base image tag into dependent image tags
3750 - ci: Append build image tag to LAVA tag used for minio path
3751 - ci: Add trailing slash to path for documentation preview
3752 - ci: Restrict meson-gallium job to gstreamer runners
3753 - ci: Disable scons-win64 job
3754 - ci: Move meson-build.sh to meson/build.sh
3755 - ci: Drop SIGINT handling from meson test wrapper script
3756 - ci: Move /usr/bin/time check from meson test wrapper to build script
3757 - aco/tests: Use _exit in child process
3758 - ci: Add strace to the x86_build docker image
3759 - ci: Run meson tests in strace if it's available and can be used
3760 - ci: Don't run meson tests in strace for meson-mingw32-x86_64 job
3761 - intel/tools: Use subprocess.Popen to read output directly from a pipe
3762 - Revert "ci: Restrict meson-gallium job to gstreamer runners"
3763 - glcpp: Fully initialize struct gl_context
3764 - ci: Disable valgrind in some build jobs
3765 - glsl/tests: Bump glcpp valgrind test timeout to 240 seconds
3766 - glsl/tests: Don't use tempfiles
3767 - glsl/tests: Use exit code 126 to detect valgrind errors
3768 - Revert "ci: disable glcpp tests for now"
3769 - Revert "meson: add enable-glcpp-tests option"
3770 - Revert "glsl/test: Don't run whitespace tests in parallel"
3771 - ci: Remove INCLUDE_PIGLIT
3772 - ci: Build ARM baremetal rootfs in native container
3773 - ci: Merge ARM testing docker images to a single arm_test one
3774 - wsi/x11: Wait for fences with IMMEDIATE on Xwayland
3775 - ci: Fix HTML summary path for piglit OpenCL job artifacts
3776 - intel/blorp: Initialize texture_data[0]
3777 - ci: Do not install armhf LLVM packages
3778 - ci: Bump LLVM/clang from 10 to 11
3779 - ci: Move docker images from Debian buster to bullseye
3780 - ci: Install librenderdoc from Debian bullseye
3781 - ci: Install spirv-tools from Debian bullseye
3782 - ci: Install llvm-spirv from Debian bullseye
3783 - ci: Install GLVND from Debian bullseye
3784 - ci: Install Rust & cargo from Debian for x86_test* images
3785 - ci: Do not append ci-templates commit hash to Windows docker image tag
3786 - ci: Update to latest ci-templates
3790 - vulkan/lavapipe: add missing VKAPI_ATTR/CALL
3791 - vulkan: Fix windows api conflict
3792 - zink: Fix win32 build
3793 - vulkan: Fix windows api conflict
3794 - meson: invalid keyword argument dependencies
3795 - zink: fix win32 build
3796 - util: fix gcc vsnprintf overflow
3797 - glapi: keep declspec(thread) msvc-specific
3798 - vulkan: implement wsi_win32 backend
3799 - lavapipe: add mingw32 def file
3800 - lavapipe: set empty dll prefix
3801 - gallium: remove DROP_PIPE_LOADER_MISC
3802 - meson/xmlconfig: win32 regex fallback
3803 - meson: detect winflex/bison only on native win32
3804 - turnip: update features.txt
3805 - lavapipe: update features.txt
3806 - vulkan: fix CreateRenderPass prototype
3807 - swr: extern declaration for win32 intrinsics
3808 - swr: fix win32 intrinsics
3809 - swr: Fix SWR_CONTEXT pre-declaration
3810 - swr: fix unused SplitString warning
3811 - swr: fix deprecated llvm 11 declaration warning
3812 - swr: fix array-bounds warning
3813 - lavapipe: Fix type narrowing
3814 - docs: missing lvp win32surface ext in features.txt
3818 - zink: clamp sampler+samplerview limits
3819 - util/hash_table: optimize rehash for empty table and no-func clears
3820 - util/set: optimize rehash for empty table and no-func clears
3821 - util/set: add the found param to search_or_add
3822 - util/set: split off create() into an init() function
3823 - zink: optimize renderpass hash table
3824 - nir/lower_uniforms_to_ubo: set explicit_binding on uniform_0
3825 - zink: add spirv builder function for runtime array type
3826 - zink: add util function for emitting ntv atomic ops
3827 - zink: add set_shader_buffers pipe_context method
3828 - zink: hook up ssbo shader bindings
3829 - zink: emit ssbo variables in ntv
3830 - zink: modify ubo loading in ntv to work for ssbos
3831 - zink: start supporting atomic shader ops
3832 - zink: split UBOs and samplers into 'read' batch references during draw
3833 - zink: flag ssbo buffer resources as having pending writes on batch
3834 - zink: add more usage bits for buffer types
3835 - zink: partially enable SSBO pipe cap
3836 - zink: only emit streamout targets during draw if we have them
3837 - zink: rework framebuffer state
3838 - zink: add batch flag for checking renderpass state
3839 - zink: remove renderpass refcounting
3840 - zink: ralloc zink_framebuffer structs
3841 - zink: rename param in zink_create_framebuffer
3842 - zink: use 'fb' variable name for zink_framebuffer objects in zink_framebuffer.c
3843 - zink: decouple renderpass from framebuffer state
3844 - zink: move zink_clear to zink_clear.c
3845 - zink: start to refactor clearing
3846 - zink: handle clears with scissor regions
3847 - zink: break out scissor region testing for clear functions
3848 - zink: break out color/zs no_rp clear into separate functions
3849 - zink: break out some of the u_blitter setup into util function
3850 - zink: add a pipe_context::clear_texture hook
3851 - zink: enable PIPE_CAP_CLEAR_TEXTURE
3852 - zink: reduce blendfactor when alpha_to_one is set
3853 - zink: tweak xfb slot mapping in ntv
3854 - zink: process ubos with location values set as long as they're actually ubos
3855 - zink: add VK_KHR_driver_properties
3856 - zink: enable WSI-faking for RADV too
3857 - zink: rename zink_context::\*image_views -> sampler_views
3858 - zink: add ntv util function for getting image type
3859 - zink: rewrite image/sampler glsl -> vk type functions for robustness
3860 - zink: add spirv_builder function for hexops
3861 - zink: add spirv builder functions for image ops
3862 - zink: add ntv function for emitting variable access decorations
3863 - zink: verify format caps and add storage image usage when possible in creation
3864 - zink: add 'has_draw' flag to batch struct
3865 - zink: add a pipe_context::memory_barrier hook
3866 - zink: add shader image support to zink_binding()
3867 - zink: add new 'sampler_types' variable to ntv_context struct
3868 - zink: handle image variable types in ntv
3869 - zink: handle more atomic ops in ntv
3870 - zink: handle nir_intrinsic_memory_barrier in ntv
3871 - zink: add nir_var_uniform case to get_storage_class()
3872 - zink: expand ntv array derefs to track image derefs
3873 - zink: add handling for all basic image ops in ntv
3874 - zink: enable early frag test execution in ntv when necessary
3875 - zink: enable image caps in ntv when a shader has images
3876 - zink: handle image descriptors during zink_shader creation
3877 - zink: break out bufferview creation into separate function
3878 - zink: add a pipe_context::set_shader_images hook
3879 - zink: handle shader image descriptor updates during draw
3880 - zink: check if multisample support exists for shader image formats
3881 - zink: export shader image caps using features
3882 - zink: GLSL 420
3883 - docs/features: mark off GL 4.2 for zink
3884 - zink: set PIPE_CAP_COPY_BETWEEN_COMPRESSED_AND_PLAIN_FORMATS
3885 - zink: force per-sample interpolation
3886 - zink: set PIPE_CAP_FRAMEBUFFER_NO_ATTACHMENT
3887 - zink: set PIPE_CAP_ROBUST_BUFFER_ACCESS_BEHAVIOR
3888 - zink: support VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL usage
3889 - zink: add layout member to barrier setup in draw
3890 - zink: support PIPE_FORMAT_X24S8_UINT
3891 - zink: check correct caps for PIPE_CAP_IMAGE_LOAD_FORMATTED
3892 - zink: enable PIPE_CAP_SAMPLER_VIEW_TARGET
3893 - gallium/u_inlines: add helper for simplifying pipe_context::resource_copy_region
3894 - zink: add function for waiting on a specific batch's fence
3895 - zink: don't force a renderpass start when setting framebuffer state
3896 - spirv: handle NoContraction in GLSL450 alu ops
3897 - zink: fix streamout for clipdistance
3898 - zink: add a VkExternalMemoryImageCreateInfo for PIPE_BIND_SHARED images
3899 - zink: set lower_mul_2x32_64 when 64bit int support is available
3900 - zink: enable PIPE_CAP_TEXTURE_MIRROR_CLAMP_TO_EDGE
3901 - zink: flag gfx pipeline dirty using newer mechanism
3902 - zink: guarantee surface lifetimes for shader images
3903 - st/bitmap: use GL_CLAMP_TO_EDGE for bitmap samplers
3904 - st/drawpixels: use GL_CLAMP_TO_EDGE instead of GL_CLAMP for samplers
3905 - zink: don't export PIPE_CAP_MAX_COMBINED_SHADER_BUFFERS value
3906 - zink: add spirv builder function for OpAtomicStore
3907 - zink: flag ssbo buffer resources as having pending writes per stage
3908 - zink: handle null ssbo attachments without crashing
3909 - zink: handle more ssbo ops in ntv
3910 - zink: rework ssbo indexing and binding
3911 - zink: support nir_intrinsic_store_ssbo
3912 - zink: implement get_ssbo_size nir intrinsic
3913 - zink: flatten out ssbo/ubo variable decls in ntv
3914 - zink: export ssbo caps
3915 - Revert "glcpp: disable 'windows' tests"
3916 - meson: add enable-glcpp-tests option
3917 - ci: disable glcpp tests for now
3918 - zink: add barrier helper for buffer resources
3919 - zink: add a stage param for buffer resource barriers
3920 - zink: add helper function for checking if access flags include write access
3921 - zink: improve barrier helper for buffer resources and add check for barrier need
3922 - zink: flag previous vertex stages as dirty when toggling a later stage
3923 - zink: add shader key for vs shaders
3924 - zink: flag shaders as needing update when clip_halfz changes
3925 - zink: move tess/geom shader info to vs shader key
3926 - glsl: support 64bit integer loop iterators
3927 - radv: print image array size in debug mode
3928 - zink: move maintenance2 extension to right file
3929 - zink: unify shader image unbind codepath
3930 - zink: be a little more precise about query types in one conditional
3931 - radv: null bo list pointer for null descriptors on update
3932 - radv: zero the bo descriptor array when allocating a new set
3933 - zink: force 4 component formats for samplerview/render textures
3934 - zink: support nir_intrinsic_memory_barrier_buffer
3935 - zink: add defines for compute batch and gfx batch count
3936 - zink: bump resource usage flags to allow 5 batches
3937 - zink: make get_resource_usage() public
3938 - zink: make zink_batch_reference_resource_rw return usage info
3939 - zink: wait on compute batch when necessary during transfer map
3940 - zink: add spirv_builder function for emitting a 3word literal exec mode
3941 - zink: handle COMPUTE bindings in compiler/ntv
3942 - zink: handle COMPUTE setup in ntv
3943 - zink: handle COMPUTE glsl variables
3944 - zink: implement shared load/store nir ops in ntv
3945 - zink: add handling for shared atomic ops in ntv
3946 - zink: handle nir_intrinsic_memory_barrier_shared in ntv
3947 - zink: ignore compute batch when starting/ending batches
3948 - zink: take a pipe_reference param in zink_batch_reference_program
3949 - zink: refactor batch creation
3950 - zink: make allocate_descriptor_set() take more params instead of a gfx_program
3951 - zink: explicitly get shader stage from shader during binding setup in draw
3952 - zink: rename pipeline_cache_entry -> gfx_pipeline_cache_entry
3953 - zink: add compute programs and pipelines
3954 - zink: break out descriptor updating into separate function
3955 - zink: setup compute batch and add handling
3956 - zink: handle memory barriers for compute batch
3957 - zink: handle descriptor set updates for compute operations
3958 - zink: flush gfx/compute batches when the other pipeline needs resource sync
3959 - zink: add launch_grid pipe_context hook for compute handling
3960 - zink: export compute-specific shader/compute caps
3961 - zink: enable compute
3962 - zink: GLSL 430
3963 - features: mark off GL 4.3 for zink
3964 - zink: add spirv_builder wrapper for vote intrinsics
3965 - zink: handle vote intrinsics in ntv
3966 - zink: rework viewport handling
3967 - zink: handle nir_texop_texture_samples
3968 - zink: add a texture barrier hook
3969 - zink: use = and not \|= for VkMemoryPropertyFlags during resource creation
3970 - zink: set HOST_COHERENT bit for coherent resource creation
3971 - zink: track persistent, non-coherent, writable transfer map count for resources
3972 - zink: slightly refactor batch resource referencing in update_descriptors()
3973 - zink: flush all resources with persistent maps on work batch before draw/compute
3974 - zink: enable PIPE_CAP_BUFFER_MAP_PERSISTENT_COHERENT
3975 - zink: rename zink_context::dummy_buffer -> dummy_vertex_buffer
3976 - zink: create dummy xfb buffer
3977 - zink: handle null xfb buffers
3978 - zink: use better mapping for PIPE_FORMAT_X24S8_UINT
3979 - zink: handle nir_intrinsic_load_helper_invocation
3980 - zink: fix xfb buffer refcounting
3981 - zink: add PIPE_BIND_QUERY_BUFFER to the all-purpose resource creation path
3982 - zink: add a get_query_result_resource hook
3983 - zink: enable PIPE_CAP_TGSI_ARRAY_COMPONENTS
3984 - zink: enable PIPE_CAP_QUERY_BUFFER_OBJECT
3985 - zink: GLSL 440
3986 - zink: enable PIPE_CAP_CONDITIONAL_RENDER_INVERTED
3987 - zink: enable PIPE_CAP_CLIP_HALFZ
3988 - zink: enable PIPE_CAP_TGSI_TXQS
3989 - zink: enable PIPE_CAP_TEXTURE_BARRIER
3990 - zink: GLSL 450
3991 - features: mark off GL 4.5 for zink
3992 - zink: add spirv interfaces for bo and image/sampler/push variables
3993 - zink: lower flrp64 and ffma64 when in softfp64 mode
3994 - zink: always use query->type for starting/stopping xfb queries
3995 - zink: make the xfb_query_pool into an array
3996 - zink: break out cpu query reading for qbos into separate function
3997 - zink: put SO_OVERFLOW queries on the primgen list
3998 - zink: support SO_OVERFLOW pipe query types
3999 - zink: fix streamout for tess stage
4000 - zink: flag exact alu op results in ntv with NoContraction
4001 - zink: unset generated TCS if its parent TESS is unset
4002 - zink: hook up cs invocation queries to the compute batch
4003 - zink: add support for pipeline statistics queries
4004 - zink: fix slot mapping for legacy gl io with tess stages
4005 - zink: handle 1bit undef values in ntv
4006 - zink: add handling for ARB_shader_draw_parameters variables in ntv
4007 - zink: create a struct for tracking push constant layout
4008 - zink: rework tcs injection to be more compatible with new push const struct
4009 - zink: add push constant value to indicate whether the current draw is indexed
4010 - zink: wrap shader gl_BaseVertex access with a bcsel based on push constant state
4011 - zink: add a draw_id param to vs push constants
4012 - zink: add a vs shader key for rewriting gl_DrawID
4013 - zink: break out push constant creation in compiler and add drawid value
4014 - zink: rewrite drawid based on shader key value
4015 - zink: add util function for submitting the compute batch
4016 - zink: enable PIPE_CAP_TGSI_VOTE
4017 - zink: enable PIPE_CAP_DRAW_PARAMETERS
4018 - zink: enable PIPE_CAP_POLYGON_OFFSET_CLAMP
4019 - zink: enable PIPE_CAP_QUERY_SO_OVERFLOW
4020 - zink: enable pipeline statistics cap
4021 - zink: PIPE_CAP_GL_SPIRV
4022 - zink: GLSL 460
4023 - features: mark off GL 4.6 and ES 3.1 for zink
4024 - zink: support nir_intrinsic_group_memory_barrier
4025 - zink: fix device codegen extension detection
4026 - zink: add nir_intrinsic_memory_barrier_image handling
4027 - zink: use nir_shader_instructions_pass for draw params pass
4028 - zink: add flag for no-oping fence finish
4029 - zink: hook up valid_buffer_range for buffer resources using util_range
4030 - zink: create a VkPipelineCache object on the screen and use it
4031 - zink: add a disk cache for pipeline objects
4032 - gallium/trace: add a pipe_screen::get_compiler_options method
4033 - zink: handle dual blending override from driconf
4034 - zink: move command pool to the batch
4035 - nir/lower_tex: rewrite tex/txb -> txd/txl before saturating srcs
4036 - mesa/st: add pipe_sampler_state::border_color_is_integer
4037 - mesa/st: add PIPE_CAP_GL_CLAMP
4038 - zink: enable GL_CLAMP cap
4039 - gallium/trace: remove transfer_map assert
4040 - zink: add helper function for getting pipeline stage from shader stage
4041 - zink: set buffer resource barriers for descriptor resources in update_descriptors()
4042 - zink: rework xfb counter resource barriers
4043 - zink: rework xfb barrier transitions when reusing as vertex inputs
4044 - zink: remove aspect param from zink_resource_barrier
4045 - zink: add a VkPipelineStageFlags param to zink_resource_barrier()
4046 - zink: add helper for image resource barriers and avoid unnecessary barriers
4047 - zink: use define for max descriptor array size
4048 - zink: add generic wrapper for checking whether a resource needs a barrier
4049 - zink: avoid emitting unnecessary pipeline barriers during update_descriptors
4050 - zink: break out barrier transitioning in update_descriptors
4051 - zink: combine resource barriers where possible during update_descriptors
4052 - zink: take struct zink_batch param instead of direct cmdbuf in barrier helpers
4053 - zink: assert batch is not in a renderpass when emitting pipeline barrier
4054 - zink: add barriers for index and draw param buffers
4055 - zink: add access param for image resource barriers
4056 - zink: add access info for update_descriptor image barriers
4057 - zink: add batch references for resources in clear functions
4058 - zink: improve barrier usage for clear functions
4059 - zink: zink_resource_barrier -> zink_resource_image_barrier
4060 - zink: add general zink_resource_barrier() wrapper
4061 - zink: be more explicit with image barriers for copy operations
4062 - zink: fix surface creation for cube slices
4063 - zink: tag some missing ES features
4064 - zink: update relnotes
4065 - zink: just call context destructor on creation fail
4066 - zink: add buffer barriers for resource_copy_region
4067 - zink: break out buffer copying into util function with batch param
4068 - zink: just end the current renderpass in zink_batch_no_rp()
4069 - zink: break out even more of zink_blit state saving
4070 - zink: use vkGetFenceStatus when we're obviously checking for status
4071 - zink: fix buffer resource usage flags
4072 - zink: break out query result buffer copying into util function
4073 - zink: simplify some of the qbo direct buffer write code
4074 - zink: better handling for availability queries on qbos when query/resource is busy
4075 - zink: improve batch flushing for queries when compute batches are involved
4076 - zink: always use 64bit flag for query results
4077 - zink: handle scissor+viewport states dynamically if extension is available
4078 - zink: remove 'scissors' member of viewport state
4079 - zink: always set VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT for non-staging resources
4080 - zink: add available|visible masks to all barriers in ntv
4081 - zink: set conformant ubo/ssbo size limits
4082 - zink: destroy renderpass objects on context destroy
4083 - zink: rename 'has_draw' flag on batches and set it when the batch is used
4084 - zink: move gfx pipeline creation closer to the bind point
4085 - zink: only reset pipeline hash conditionally when updating fb state
4086 - zink: simplify barrier usage
4087 - zink: beef up zink_transfer_flush_region
4088 - zink: only wait on last write-batch for resources during transfer_map
4089 - zink: change some transfer_map cases of waiting on cs batch to flushing cs
4090 - zink: handle PIPE_TRANSFER_DISCARD_WHOLE_RESOURCE in transfer_map for buffers
4091 - zink: update ci results
4092 - zink: disable some builtin-gl-sample-mask sample shading tests on ci
4093 - zink: actually disable sample mask tests on ci
4094 - lavapipe: fix color-only renderpass clears
4095 - zink: ralloc the main context
4096 - zink: create framebuffer and renderpass objects just before vkCmdBeginRenderPass()
4097 - zink: defer pipe_context::clear calls when not currently in a renderpass
4098 - zink: also defer fb clears when conditional render is active
4099 - zink: break out region overlap testing function into helper
4100 - zink: add helper for converting pipe_box -> u_rect
4101 - zink: add another helper for checking whether one rect covers another
4102 - zink: break out fb clear apply into helper function
4103 - zink: add helper for applying/discarding clears based on a rect
4104 - zink: discard pending clears during blit/copy if we'll overwrite the data
4105 - zink: add yet another clear helper, this time for applying overlap regions
4106 - zink: optimize the remaining read cases of applying pending clear calls
4107 - zink: move all the clear stuff to zink_clear.h
4108 - zink: always do full-fb clears in renderpass begin when possible
4109 - zink: ci changes
4110 - zink: improve descriptor set oom handling
4111 - zink: ci updates
4112 - zink: set PIPE_CAP_TEXTURE_BORDER_COLOR_QUIRK
4113 - zink: force 128 fs input components on intel drivers
4114 - zink: add some spirv builder functions for spec constants
4115 - zink: support gl_LocalGroupSize
4116 - zink: add more ci flakes
4117 - util/bitscan: add u_foreach_bit macros
4118 - v3dv: remove for_each_bit() macro
4119 - radv: for_each_bit -> foreach_bit
4120 - freedreno/vulkan: for_each_bit -> foreach_bit
4121 - anv: for_each_bit -> foreach_bit
4122 - zink: use 0 as default for spec constants
4123 - zink: no-op descriptor updating for draws without descriptors
4124 - nir/texcoord_replace: add a yinvert param
4125 - zink: store prim mode to context during draw
4126 - zink: handle point sprite
4127 - zink: ci updates
4128 - zink: avoid memset during update_descriptors() for resources refs
4129 - zink: move samplerview referencing around in update_descriptors()
4130 - zink: reorder zink_bind_vertex_buffers()
4131 - zink: create a single fence per batch on startup and then reuse
4132 - zink: only flush batches in pipe_context::flush if they actually have work
4133 - zink: add a define for compute batch count
4134 - zink: add util function for returning previous batch
4135 - zink: handle PIPE_FLUSH_DEFERRED
4136 - zink: handle VK_IMAGE_LAYOUT_PRESENT_SRC_KHR barriers
4137 - zink: set VK_IMAGE_LAYOUT_PRESENT_SRC_KHR on fb resources at eof flush
4138 - zink: setup CmdBindVertexBuffers2EXT member in screen for dynamic state
4139 - zink: make dynamic state usage in pipeline creation more explicit/flexible
4140 - zink: use dynamic vertex buffer strides
4141 - zink: rename zink_context::buffers -> vertex_buffers (and usage mask)
4142 - zink: add zink_program struct as a base class for compute/gfx structs
4143 - zink: use zink_program in zink_batch_reference_program()
4144 - zink: ralloc zink program structs
4145 - zink: unref programs last in batch reset
4146 - zink: properly size descriptorset layout binding stack array
4147 - zink: increment batch->descs_used during update_descriptors flushing
4148 - zink: do batch-program tracking after possibly cycling batch in update_descriptors()
4149 - zink: add spirv builder methods for OpImageQueryLevels
4150 - zink: hook up nir_texop_query_levels
4151 - zink: relax tessellation shader reqs
4152 - zink: ci updates
4153 - zink: fix dynamic bo lowering for ssbo stores
4154 - zink: pre-fetch all format properties during screen init
4155 - zink: use pre-fetched format properties everywhere
4156 - zink: don't start renderpasses during descriptor update
4157 - zink: add more usage bits for buffer resource creation
4158 - zink: handle null src for fb refs
4159 - zink: track all framebuffers per batch
4160 - zink: store total memory size on zink_screen
4161 - zink: track resource mem usage per batch
4162 - zink: force batch flush if batches are using more than 1/10 total system memory
4163 - mesa/st: clamp scissored clear regions to fb size
4164 - mesa/st: no-op scissored clear calls with size zero
4165 - zink: handle GLSL_SAMPLER_DIM_EXTERNAL in ntv
4166 - zink: ci updates
4167 - mesa/st: even better no-oping for clears
4168 - zink: apply only the pending zs clear bits during deferred clears
4169 - zink: enable PIPE_CAP_CLEAR_SCISSORED
4170 - zink: export PIPE_CAP_TGSI_VS_LAYER_VIEWPORT
4171 - zink: use staging resource for write transfer_map in order to not stall
4172 - zink: ci updates
4173 - zink: rewrite macro for getting KHR device functions
4174 - zink: add vk/spirv caps/extension for shader LAYER variable
4175 - zink: remove ntv streamout assert
4176 - zink: fix streamout emission for super-enhanced layouts
4177 - zink: fix slot mapping for fat io variables
4178 - zink: fix location usage for explicit xfb outputs
4179 - zink: run more nir passes for tess shaders
4180 - zink: stop allocating xfb slot map
4181 - zink: handle direct xfb output from output variables
4182 - zink: evaluate existing slot map during program init and force new map as needed
4183 - zink: rename variable in update_so_info()
4184 - zink: use info.has_transform_feedback_varyings to determine xfb enablement
4185 - zink: pass so_info directly to update_so_info()
4186 - zink: use slightly stricter check for update_so_info() callsite
4187 - zink: only export necessary xfb outputs to ntv
4188 - zink: don't pass so_info to ntv at all unless it's necessary
4189 - zink: unref ctx->framebuffer on context destroy
4190 - zink: fix instance/device versioning (for real this time)
4191 - zink: simplify some update_descriptor code
4192 - zink: move descriptor sets/pools from batches to programs
4193 - zink: store and reuse descriptorsets after batch completion
4194 - zink: move descriptor set alloc function to zink_program.c
4195 - zink: use more precise sizing for descriptor pools
4196 - zink: add helper function for cycling a batch
4197 - zink: even better handling for descriptor oom
4198 - zink: remove flushes for batch descriptor use
4199 - zink: add bucket allocating for descriptor sets
4200 - zink: add scaling factor for descriptor set bucket allocations
4201 - zink: add caching for descriptor sets
4202 - zink: add second level cache for descriptor sets
4203 - zink: move streamout to draw_vbo
4204 - zink: reorder descriptor barrier applying during updating
4205 - zink: move surface refs to the end of descriptor updating
4206 - zink: split descriptor sets based on usage
4207 - zink: use dynamic offsets for first ubo
4208 - zink: introduce descriptor states
4209 - zink: add a null sampler view descriptor hash to the screen
4210 - zink: pre-hash sampler views and states
4211 - zink: store last-used descriptor set for each type of set for quick reuse
4212 - zink: actually flag all used resources as used during update_descriptors
4213 - zink: add program pointer to desc set struct
4214 - zink: move descriptor set allocation near the top of update_descriptors
4215 - zink: only batch-reference the program in use once per descriptor update
4216 - zink: improve descriptor cache invalidation
4217 - zink: add flag for recycled descriptor sets
4218 - zink: don't double iterate all the per-batch sets on reset
4219 - zink: add VkPipelineLayout to zink_program meta struct
4220 - zink: split out ubo descriptor updating
4221 - zink: break out ssbo descriptor updating
4222 - zink: break out sampler descriptor updating
4223 - zink: break out image descriptor updating
4224 - zink: deduplicate VkWriteDescriptorSet setup
4225 - zink: break out descriptor stuff into new files
4226 - zink: break out all the descriptor pool/layout stuff into a new struct
4227 - zink: change program pointer on struct zink_descriptor_set to pool pointer
4228 - zink: track number of sets currently allocated per descriptor pool
4229 - zink: move descriptor type to pool object from set
4230 - zink: allow reuse of zink_descriptor_pools between programs
4231 - zink: remove intermediate func for descriptor set getting
4232 - zink: simplify check for knowing whether descriptor updating is needed
4233 - zink: pre-size descriptor transition hash table
4234 - zink: move descriptor binding out of the update codepath
4235 - zink: reuse descriptor barriers across draws
4236 - zink: track resource count on descriptor pool object
4237 - zink: directly use resource count from pool instead of accumulating every time
4238 - zink: remove struct zink_descriptor_resource from descriptor updating
4239 - zink: don't create descriptor barrier hash tables for cached descriptor set
4240 - zink: always use VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL for sampler bindings
4241 - zink: remove image layouts from descriptor states
4242 - zink: avoid hashing states without descriptors
4243 - zink: fix arrays of texel buffer descriptors
4244 - zink: ci updates
4245 - zink: move vertex_buffers_enabled_mask to non-hashed part of pipeline state
4246 - zink: optimize pipeline hashing
4247 - zink: implement an extremely dumb resource memory cache
4248 - zink: ci updates
4249 - zink: remove debug printf
4250 - lavapipe: fix push descriptor set indexing
4251 - lavapipe: set more resource bind flags using image/buffer usage bits
4252 - zink: move buffer<->image copying to pipe_context::resource_copy_region hook
4253 - zink: remove extraneous flush in transfer_map_region_flush
4254 - zink: ci updates
4255 - zink: optimize resource usage tracking
4256 - zink: use _mesa_set_search_and_add() for set management
4257 - zink: ralloc screen objects
4258 - zink: implement a surface cache
4259 - zink: use a safer iteration for fb surfaces during rp init
4260 - zink: determine whether the vulkan driver requires mesa flush wsi
4261 - zink: force mesa flush implicit fencing on ANV
4262 - zink: force explicit fence only on first frame flush
4263 - zink: use internal api for first-frame fence
4264 - zink: return true from program ref functions upon free
4265 - zink: unset ctx->program pointers when an unref destroys the object
4266 - zink: stop leaking programs
4267 - vk/util: add unified shader module struct/functions
4268 - vk/util: add a util macro for initializing stack vk_shader_module structs
4269 - lavapipe: use common interfaces for shader modules
4270 - radv: use common interfaces for shader modules
4271 - v3dv: use common interfaces for shader modules
4272 - tu: use common interfaces for shader modules
4273 - anv: use common interfaces for shader modules
4274 - zink: add batch usage flags for sampler views/states and desc sets
4275 - zink: avoid looping for non-ubo descriptor updates based on set usage
4276 - zink: break out batch id finding for resource usage into util function
4277 - zink: move resource internals to a separate struct
4278 - zink: split out backing resource object create/destroy
4279 - zink: track resource_object usage instead of resource usage
4280 - zink: handle cached descriptor set punting
4281 - zink: add some asserts for pipeline barriers to check renderpass state
4282 - zink: add util function for checking whether a shader descriptor is a buffer
4283 - zink: add util function for checking whether shader descriptor is buffer from program
4284 - zink: use an explicit zink_buffer_view struct
4285 - zink: explicitly use zink_surface objects for sampler/image view objects
4286 - zink: store bufferview hash to bufferview struct
4287 - zink: simplify bufferview and imageview descriptor state hashing
4288 - zink: add extra batch tracking for sampler views
4289 - zink: massively beef up batch tracking for shader images
4290 - zink: add helper function for getting a resource for a descriptor
4291 - zink: update null descriptor hashes to reflect current descriptor states
4292 - zink: improve debug asserts for samplers/images during descriptor updates
4293 - zink: properly handle null bufferview descriptor states
4294 - zink: cache bufferviews
4295 - zink: add missing null check
4296 - zink: unset last_vertex_stage_dirty after applying it
4297 - zink: run nir_convert_from_ssa last during compile
4298 - zink: use intermediate var for glsl non-array type during shader create
4299 - zink: break out bo array type construction into ntv util function
4300 - zink: also break out whole ntv bo struct pointer construction
4301 - zink: add unsized array type to get_glsl_type_element() handling
4302 - zink: add debug info about missing atomic ops
4303 - zink: add ntv util function for checking if a glsl type is an atomic counter
4304 - zink: break out sized uint array construction into util function
4305 - zink: flatten binding numbers a bit
4306 - zink: directly set nir variable bindings for reuse during ntv
4307 - zink: move zink_binding() to compiler.c
4308 - zink: apply Delete All The Code methodology to the ubo/ssbo variables
4309 - zink: set ntv variable descriptor sets during compile phase
4310 - zink: ci updates
4311 - ci/panfrost: disable the rest of these jobs temporarily
4312 - zink: hook up resource bind history
4313 - zink: remove direct samplerview batch-tracking
4314 - zink: add a pipe_context::invalidate_resource hook
4315 - zink: set valid region for streamout buffers on bind
4316 - zink: handle streamout buffer rebinds
4317 - zink: invalidate resources on map when discarding range
4318 - zink: enable PIPE_CAP_INVALIDATE_BUFFER
4319 - zink: switch to deqp-runner for piglit jobs
4320 - zink: always use requested format for sampler view creation
4321 - zink: ci updates
4322 - zink: more consolidation for null sampler/image view hashing
4323 - zink: add a pipe_context::fence_server_sync hook
4324 - zink: add enum for different queues
4325 - zink: refactor resource_sync_writes_from_batch_usage() to manage batch id internally
4326 - zink: convert ZINK_RESOURCE_ACCESS defines to enum
4327 - zink: abstract zink_get_resource_usage() and move it to be internal
4328 - zink: return enum zink_queue from zink_batch_reference_resource_rw()
4329 - zink: split out batch resource-set clearing into separate function
4330 - zink: move active query pruning to batch reset
4331 - zink: move batch init into zink_batch.c
4332 - zink: also move batch destructor into zink_batch.c
4333 - zink: move other batch-tracking implementations to unified codepath
4334 - zink: use macro to streamline batch struct member init
4335 - zink: remove query batch-tracking init from begin_query()
4336 - zink: move fence reset to zink_fence_init()
4337 - zink: clear framebuffer state on context destroy
4338 - zink: enable spirv extension for post depth coverage
4339 - compiler/spirv: fix image sample queries
4340 - zink: handle nir_intrinsic_image_deref_samples
4341 - zink: flatten 2d_array surfaces when necessary
4342 - lavapipe: support VK_KHR_copy_commands2
4343 - lavapipe: rewrite cmdbufs to always do descriptor binds/pushes first
4344 - lavapipe: force state updates when beginning queries
4345 - llvmpipe/setup: force fs constant updating upon beginning queries
4346 - zink: break out surface viewtype clamping into util function
4347 - zink: improve surface viewtype clamping
4348 - zink: correctly clamp samplerview surface types
4349 - ci/lavapipe: split out lavapipe ci into lavapipe dir
4350 - llvmpipe/setup: use bigger hammer to force fs constant updating correctly
4351 - zink: split off a bunch of batch struct members to new batch state struct
4352 - zink: rewrite queue dispatch to use monotonic batch ids instead of hardcoded ones
4353 - zink: more accurately check samplecount caps for shader images
4354 - zink: make fb ref func return bool on free
4355 - zink: add explicit surface/bufferview batch-tracking functions
4356 - zink: use surface references for fb attachments
4357 - zink: break out surface destroy function into a screen function
4358 - zink: use a custom surface referencing function whenever unrefing a surface
4359 - zink: implement a global framebuffer cache
4360 - vk: consolidate dynamic descriptor binding sorting
4361 - ci: update xfails for ppc64le and s390x
4362 - zink: break out buffer mapping part of zink_transfer_map
4363 - zink: cache transfer maps
4364 - zink: unify clear color conversion code
4365 - nir: add nir_lower_indirect_builtin_uniform_derefs()
4366 - st/glsl_to_nir: lower indirect derefs of builtins in non-packed uniform case
4367 - softpipe: ci updates
4368 - zink: move 'batch_id' and 'is_compute' members to fence
4369 - zink: make batch usage unsetting function public
4370 - zink: always reset batch states when finding a new one
4371 - zink: move batch-tracked resources to fence object
4372 - zink: fix spirv image operand ordering
4373 - zink: fix multisampled shader image load/store
4374 - zink: force PIPE_SWIZZLE_1 for X channels in samplerviews
4375 - zink: handle blitting of color formats with ignored alpha channels
4376 - zink: emulate PIPE_FORMAT_R8G8B8X8_UNORM
4377 - zink: ci updates
4378 - zink: relax unreachable() to debug_printf when waiting on batch
4379 - zink: rework public batch flush function to be useful again
4380 - zink: move zink_flush_compute() users to zink_flush_queue()
4381 - zink: always flag xfb barrier on gfx flush when appropriate
4382 - zink: simplify some queue-related query code
4383 - zink: refactor clears a little to track a bitfield of enabled clears on the context
4384 - zink: trigger pending clears during flush
4385 - zink: ci updates
4386 - zink: add wrapper to reset batch state structs
4387 - zink: call clear() instead of reset() for batch states on context destroy
4388 - zink: unify gfx and compute batches
4389 - zink: isolate gfx stage bits when updating shader modules
4390 - zink: store conditional render predicate to query and split out start/stop
4391 - zink: only update conditional render buffer when it needs to be updated
4392 - zink: toggle conditional render when beginning/ending a renderpass
4393 - zink: ci updates
4394 - zink: handle gallium multi draws more effectively
4395 - zink: create separate upload mgr for constants
4396 - zink: explicitly use stream uploader for staging buffers
4397 - zink: add buffer_subdata hook
4398 - zink: avoid unnecessary resource refs during descriptor update
4399 - zink: remove handling for resource flushing between compute/gfx batches
4400 - zink: remove unnecessary flush during image maps
4401 - zink: add more rp cache asserts
4402 - compiler/spirv: use undefs when extending image coords
4403 - zink: don't generate sampled image type for non-sampled images
4404 - util/set: stop leaking u32 key sets which pass a mem ctx
4405 - lavapipe: fix CmdCopyQueryPoolResults for partial pipeline statistics queries
4406 - lavapipe: use the passed offset for CmdCopyQueryPoolResults
4407 - lavapipe: stop tracking draw start/count on rendering state
4408 - zink: ci updates
4409 - lavapipe: ignore templateType when descriptor template isn't for push descriptors
4410 - lavapipe: remove lvp_descriptor_update_template::descriptor_set_layout
4411 - zink: fix handling for image types in resource_copy_region hook
4412 - zink: also fix image buffer layer copying
4413 - lavapipe: fix array texture region copies
4414 - zink: only do shader updates when relevant stages are dirty
4415 - zink: use correct surface ref function for context destroy
4416 - zink: stall when we start getting a lot of uncompleted batches
4417 - zink: reset all fences when waiting on batch state
4418 - zink: fix format support detection for storage texel buffers and shader images
4419 - zink: break out image/buffer create info structs into helper funcs
4420 - zink: make descriptor state invalidate public
4421 - zink: reorder barrier util functions to set up barrier struct before batch
4422 - zink: break out barrier struct initializing into helper funcs
4423 - zink: create separate vk image/buffer objects for shader image use
4424 - zink: incrementally add image usage flags based on device caps
4425 - zink: add color output bit and/or use linear tiling for sampled images
4426 - zink: check image format props before creating image
4427 - zink: toggle between linear/optimal tiling during image creation
4428 - zink: flatten out buffer creation usage flags codepath
4429 - zink: ralloc shader cache and keys
4430 - zink: rework border color handling
4431 - zink: clean up query creation failure paths
4432 - zink: create result buffers for all query streams
4433 - zink: remove flush from query buffer copy
4434 - zink: manually handle more bool query types for copying
4435 - zink: remove special casing for occlusion qbos
4436 - zink: rewrite query internals
4437 - zink: bump pools up to 5k queries each
4438 - zink: don't use PARTIAL bit for query results with time queries
4439 - zink: reorder availability handling for (user) qbos
4440 - zink: remove explicit fencing for query results
4441 - zink: ci updates
4442 - lavapipe: refactor base draw dispatch to handle multidraws
4443 - lavapipe: refactor indexed draw dispatch to handle multidraws
4444 - aux/draw: stop copying draw params unnecessarily
4445 - aux/draw: rewrite PRIM_RESTART_LOOP macro as a function
4446 - aux/draw: pass the full draw params through to draw_instances()
4447 - aux/draw: pass the full draw params through to draw_pt_arrays_restart()
4448 - aux/draw: move draw param sanitization to end of function
4449 - aux/draw: track increment_draw_id value from draw info
4450 - aux/draw: pass full draw params to draw_pt_arrays()
4451 - llvmpipe: stop flattening multidraws
4452 - lavapipe: ignore unused clearvalues when beginning renderpass
4453 - zink: rework texture_barrier hook
4454 - zink: move update_descriptors & related funcs to zink_descriptors.c
4455 - zink: move descriptor barrier handling to main update function
4456 - zink: simplify some descriptor update function parameters
4457 - zink: use GENERAL layout for sampler images that are also bound as shader images
4458 - zink: rework some includes
4459 - zink: rework memory_barrier hook
4460 - zink: add locking for descriptor pools
4461 - zink: add locking for resource maps
4462 - zink: manually invoke cpu detection during screen init
4463 - zink: add locking for batch states
4464 - zink: add function for checking whether a batch is done
4465 - zink: split fence finish func
4466 - zink: add locking for fence resources
4467 - zink: explicitly reset a couple more batch state members
4468 - zink: assume fence has already completed if a batch state isn't found
4469 - zink: rename init_batch_state to get_batch_state
4470 - zink: store context to batch state
4471 - zink: make a local screen pointer in zink_flush
4472 - zink: remove zink_fence_init()
4473 - zink: move VkQueue to batch object
4474 - zink: break out queue submit into separate functions
4475 - zink: also check for device lost reset on flush
4476 - zink: remove zink_create_fence()
4477 - zink: track coherent resource objects
4478 - zink: use cached memory for all resources when possible
4479 - radv: stop zeroing radv_draw_info during draw
4480 - radv: refactor draw dispatch
4481 - radv: track whether gl_BaseInstance is used
4482 - radv: simplify vs draw param counting during setup
4483 - radv: set gfx pipeline vtx_emit_num to the number of sgprs
4484 - radv: track whether drawid is used on the pipeline struct
4485 - radv: track whether baseinstance is used on the pipeline struct
4486 - radv: break out vertex shader param emission into separate function
4487 - radv: make vertex param sgpr count more explicit
4488 - radv: reorder vertex shader params
4489 - radv: don't emit baseinstance and drawid if neither is used
4490 - radv: don't reset vertex state params on pipeline bind if reg layout matches
4491 - zink: implement threaded context
4492 - zink: ci updates
4493 - zink: handle PIPE_MAP_DONTBLOCK for buffer read maps
4494 - zink: add set_context_param hook
4495 - zink: add batch tracking id for program struct
4496 - zink: track last completed batch id to optimize checking states
4497 - zink: handle expired deferred fences more reasonably
4498 - zink: hook up timeline semaphore signalling during batch submission
4499 - zink: add timeline semaphore fastpath for checking/triggering batch completion
4500 - zink: optimize batch states for timeline use
4501 - zink: enforce device lost status
4502 - zink: be more explicit about blit layer/depth usage
4503 - zink: use VkSubresourceLayout::depthPitch as layer_stride when mapping 3D imgs
4504 - zink: zink_push_constant -> zink_gfx_push_constant
4505 - zink: use max_rt to determine number of blend state attachments
4506 - zink: emit ImageCubeArray cap when accessing arrayed cube dimension images
4507 - zink: fix layercount for array texture blits
4508 - zink: add some asserts to avoid zero-sized blit regions
4509 - features: mark off ARB_compute_variable_group_size for zink
4510 - features: mark off GL_OES_viewport_array for zink
4511 - zink: store shader_info to ntv_context struct
4512 - zink: only emit SpvCapabilitySampleMaskPostDepthCoverage if the mode is set
4513 - zink: enable PIPE_CAP_TGSI_TES_LAYER_VIEWPORT
4514 - features: mark off ARB_shader_viewport_layer_array for zink
4515 - zink: avoid cached memory allocations when not requested
4516 - util/threaded_context: support pipe_context::set_sample_locations
4517 - zink: hook up cs push constant for nir_intrinsic_load_work_dim
4518 - zink: use better usage flags for staging resources
4519 - zink: use vkGetPhysicalDeviceFormatProperties2 when available
4520 - zink: use 2 variant to check image format props during create
4521 - zink: only use host mem for staging resources with linear tiling
4522 - zink: move cmdpool reset to batch state reset
4523 - zink: split total_mem off to total_video_mem, use total_mem for tc
4524 - zink: relax maybe_flush mem threshold
4525 - zink: relax maybe_flush batch count threshold
4526 - zink: check last_finished first in fence_finish early out case
4527 - zink: defer timestamp query pool resets to end_query
4528 - zink: reset queries when suspending if >50% of total pool is used
4529 - zink: don't use cached mem for staging resources
4530 - zink: flag DYNAMIC resources as coherent
4531 - zink: drop VK_PIPELINE_CREATE_DISABLE_OPTIMIZATION_BIT from compute path
4532 - aux/trace: add a set_inlinable_constants hook
4533 - intel: avoid dumping null cs sampler/binding states
4534 - zink: emit WorkgroupSize when not using ExecutionModeLocalSize
4535 - lavapipe: add some asserts for blit region extents
4536 - zink: export PIPE_CAP_PREFER_BLIT_BASED_TEXTURE_TRANSFER
4537 - aux/trace: add screen deduplication for zink+lavapipe tracing
4538 - aux/trace: add a bunch of methods for lavapipe
4539 - util/set: add macro for destructively iterating set entries
4540 - util/hash_table: add macro for destructively iterating entries
4541 - aux/trace: add GALLIUM_TRACE_TRIGGER mode
4542 - zink: add a pipe_screen::finalize_nir hook
4543 - zink: implement uniform inlining
4544 - zink: add env var to force uniform inlining
4545 - zink: remove atomic usage from batch tracking comparisons
4546 - zink: bypass separate stencil path in resource_reference_rw when not a zs image
4547 - zink: fix conditional when assigning tess variable io
4548 - zink: stop unmapping resources
4549 - zink: simplify clear-apply on fb state change
4550 - zink: use set_foreach_remove()
4551 - zink: use explicit subpass deps
4552 - zink: hook up EXT_fragment_shader_interlock
4553 - zink: support ARB_fragment_shader_interlock
4554 - aux/trace: dump all the blend state members
4555 - features: mark off ARB_fragment_shader_interlock for zink
4556 - gallium/threaded_context: add another rule for buffer mapping
4557 - zink: fix CI flakiness in glx-multithread-clearbuffer
4558 - zink: make timeline semaphores per-screen
4559 - zink: handle checking batch completion from other contexts without timelines
4560 - zink: only unmap PIPE_MAP_ONCE in synchronous mode
4561 - zink: don't lose existing pNext when using wsi_image_create_info in image creation
4562 - anv: fix debugoptimized build compile
4563 - zink: move descriptor state management to descriptors.c
4564 - zink: make a bunch of descriptor functions static
4565 - zink: create separate linear tiling image for scanout
4566 - zink: flag anv for mesa image create wsi
4567 - zink: disable mutable formats for zs formats and scanout images
4568 - aux/trace: enhance trigger mode to dump context states during bind
4569 - aux/trace: dump current fb state on trigger-mode draw if it hasn't been seen yet
4570 - aux/trace: do deep dumps of fb state for triggered traces
4571 - aux/trace: use ralloc_free for ralloc()ed state pointers
4572 - zink: compare against screen batch id when determining which semaphore to use
4573 - zink: always copy the nir shader before compiling
4574 - zink: fix tcs slot map eval for user vars
4575 - zink: fix tcs input reservation for user vars
4576 - zink: merge copy-to-scanout path into non-deferred flush path
4577 - zink: force scanout sync when mapping scanout resource
4578 - zink: use undefined layout for first scanout obj transition
4579 - zink: move scanout sync to end of batch
4580 - zink: add a flag indicating whether scanout object needs updating
4581 - zink: move wsi flush info conditional to queue submission
4582 - zink: directly set batch->state->flush_res from flush_resource hook
4583 - zink: add clear-on-flush mechanic deeper into flush codepath
4584 - Revert "zink: force scanout sync when mapping scanout resource"
4585 - softpipe: fix render condition checking
4586 - softpipe: fix streamout queries
4587 - softpipe: ci updates
4588 - zink: track persistent resource objects, not resources
4589 - zink: restore previous semaphore (prev_sem) handling
4590 - zink: use cached memory for staging resources
4591 - zink: only reset query on suspend if the query has previously been stopped
4592 - zink: when performing an implicit reset, sync qbos
4596 - gallium: Map _DRI_IMAGE_FORMAT_NONE to NULL
4597 - gallium: Flush GL API resources in eglCreateImage
4598 - iris: Disable aux as needed in iris_flush_resource
4599 - blorp: Assert 8x4 alignment for a HiZ op on Gen8-9
4600 - i965,iris: Delete misleading HiZ sampling comments
4601 - iris: Drop an XXX comment about sampling HiZ arrays
4602 - iris: Drop a stale comment about HiZ sampling
4603 - iris: Delete redundant assertion in iris_hiz_exec
4604 - iris: Drop batch param from iris_resource_prepare_render
4605 - iris: Fix the depth aspect aux usage in iris_blit
4606 - iris: Keep aux_usage in iris_blorp_surf_for_resource
4607 - iris: Fix aux usage of depth buffer prepare/finish
4608 - iris: Loosen aux state getter/setter assert on HiZ
4609 - iris: Don't avoid aux state getter/setter with HiZ
4610 - iris: Drop iris_resource::aux::has_hiz
4611 - iris: Call iris_sample_with_depth_aux earlier
4612 - iris: Set BO maps to NULL in bo_free
4613 - drm-uapi: Update drm_fourcc.h for new TGL modifier
4614 - isl: Describe I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS_CC
4615 - iris: Support clear color plane imports for RC_CCS_CC
4616 - iris: Support RC_CCS_CC modifier in plane queries
4617 - iris: Support I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS_CC
4621 - mesa: set states in fast path for restoring light attributes
4622 - gallium/u_vbuf: use updated pipe_draw_start_count while using draw_vbo
4623 - nir_to_tgsi: Fix indices for CMP in nir_to_tgsi for nir_op_fcsel
4627 - meson: Fix missing xcb-xrandr dependency for Vulkan X11 WSI
4631 - docs/features: Add OpenCL status
4632 - spirv: Ignore WorkgroupSize in non-compute stages
4633 - nv50: Replace hardcoded texture/constbuf count with define
4634 - nv50: Update texture indices to match stage indices
4636 Pierre-Eric Pelloux-Prayer (55):
4638 - ac: add ifdef __cplusplus guard to header
4639 - radeonsi: invalidate compute sgprs in si_rebind_buffer
4640 - radeonsi: inhibit clockgating when using SQTT
4641 - ci: split src/mesa/\**/* matching rule
4642 - radeonsi/sqtt: use more event identifier
4643 - radeonsi/sqtt: fix SQTT bo size overflow
4644 - radeonsi/sqtt: allow AMD_THREAD_TRACE_TRIGGER to be a frame number
4645 - radeonsi/sqtt: forward string markers to sqtt
4646 - radeonsi: don't use cp_dma prefetch on GFX6
4647 - gallium/u_upload_mgr: lower risk of hitting an assert
4648 - radeonsi: fix indentation issue in si_texture.c
4649 - radeonsi: store si_context::xxx_shader members in union
4650 - radeonsi: fix read from compute / write from draw sync
4651 - radeonsi: fix si_check_render_feedback
4652 - radeonsi: replace force_cp_dma arg of si_clear_buffer by enum
4653 - radeonsi: enable dcc image stores on gfx10+
4654 - radeonsi: force dcc clear to use compute clear
4655 - mesa: update vao _EnabledWithMapMode in copy_array_object
4656 - radeonsi: properly set SPI_SHADER_PGM_HI_ES
4657 - ac/rgp: make the max gap between shader code a warning
4658 - ac/rtld: make ac_rtld_upload returns the code size
4659 - ac/rgp: move radv/sqtt functions to ac
4660 - radeonsi/sqtt: keep a copy of the uploaded shader code
4661 - radeonsi/sqtt: remove duplicate token
4662 - radeonsi/sqtt: don't always use WGP 0
4663 - radeonsi/sqtt: export shader code to RGP
4664 - radeonsi/sqtt: fix user event max size
4665 - frontends/va: fix protected slice data buffer read size
4666 - mesa/st: fix lower_tex_src_plane in multiple samplers scenario
4667 - dlist: remove ListExt feature
4668 - mesa: remove 2 recursive lock usages of _mesa_HashTable
4669 - mesa/hash: make the mtx non-recursive
4670 - mesa/hash: switch to simple_mtx
4671 - mesa: make _mesa_HashTable InDeleteAll debug only
4672 - vbo/dlist: use DrawGallium(Complex)
4673 - nir/lower_tex: ignore texture_index if tex_instr has deref src
4674 - mesa/st: fix st_nir_lower_tex_src_plane arguments
4675 - mesa/st: ignore texture_index if tex_instr has deref src
4676 - gallium/u_threaded: split draws that don't fit in a batch
4677 - st/draw: remove st_draw_vbo
4678 - vbo: inline vbo_primitive_restart in brw_primitive_restart
4679 - radeonsi/rgp: export barriers
4680 - radeonsi/rgp: export compute shader programs
4681 - gallium/u_threaded: skip refcounting only once
4682 - driconf: add workarounds for Teardown
4683 - amdgpu,radeon: add needs_reset param to ctx_query_reset_status
4684 - radeonsi: submit cs to failed context instead of skipping them
4685 - radeonsi: use SI_CONTEXT_FLAG_AUX when recreating the aux context
4686 - radeonsi: do not recreate the aux context from the aux context
4687 - radeonsi: only recreate the aux_context when soft recovery failed
4688 - radeonsi: re-create the aux context in si_create_context
4689 - amdgpu,radeon: add full_reset_only param to ctx_query_reset_status
4690 - radeonsi: avoid querying gpu state if possible
4691 - r600/sb: Use assignments for resetting struct r600_sb::literal
4692 - driconf: add workaround for Golf With Friends
4696 - lima: fix xserver page flip fail for full screen client
4700 - nir/loop_unroll: unroll more aggressively if it can improve load scheduling
4701 - aco: fix convert_to_SDWA() check in add_subdword_definition()
4702 - aco: add test for incorrect convert_to_SDWA() check
4703 - radv: fix max_waves estimation on GFX10.3
4704 - aco: fix num_waves on GFX10+
4705 - aco: have emit_wqm() take Builder instead of isel_context
4706 - aco: add emit_mimg() helper
4707 - aco: move VADDR to the end of the operand list
4708 - aco: use non-sequential addressing
4709 - aco: only require texture coordinates to be in WQM if NSA is used
4710 - aco: add affinity for non-sequential MIMG operands
4711 - radv,aco: don't use MUBUF for multi-channel loads on GFX8 with robustness2
4712 - nir/lower_io: fix array_length lowering if buffer is smaller than offset
4713 - radv,aco: use deref_buffer_array_length
4714 - radv: use nir_opt_access
4715 - nir/sink,nir/move: sink/move reorderable load_ssbo
4716 - radv: sink load_ssbo
4717 - aco: don't consider a phi trivial if same's register doesn't match the def
4718 - aco: remove Format::{VOP3A,VOP3B}
4719 - aco: add instruction cast and format-check methods
4720 - aco: use instruction cast methods
4721 - aco: use format-check methods
4722 - aco: return references in instruction cast methods
4723 - aco: fix WQM for texture instructions with args before the coordinates
4724 - nir/opt_uniform_atomics: recognize more complicated invocation comparisons
4725 - nir/opt_uniform_atomics: fix elect detection
4726 - aco: disable a*1.0 optimization if the instruction is precise
4727 - nir/algebraic: optimize out exact a*1.0 if it's used only as a float
4728 - aco: optimize a*0.0
4729 - aco: optimize out a*1.0 if it's used as a float
4730 - nir/algebraic: optimize out exact a+0.0 if it's used only as a float
4731 - nir/algebraic: eliminate exact a*0.0 if float execution mode allow it
4732 - aco: don't affect isPrecise() after applying output modifiers
4733 - nir,spirv: allow non-uniform OpArrayLength
4734 - radv,ac/nir: implement non-uniform get_ssbo_size
4735 - aco: implement non-uniform get_ssbo_size
4736 - radv: round-up num_records division in radv_flush_vertex_descriptors
4737 - radv: correctly enable WGP_MODE for NGG and GS
4738 - radv: correctly enable WGP_MODE for tessellation control
4739 - aco: add fallback algorithm in get_reg()
4740 - aco: always set exec_live=false
4741 - aco: optimize AC_FETCH_FORMAT_SNORM alpha adjust
4742 - aco: do not flag all blocks WQM to ensure we enter all nested loops in WQM
4743 - aco: rewrite setting of Exact_Branch
4744 - aco: remove loop to flag loop blocks as WQM
4745 - aco: fix adjust_vertex_fetch_alpha
4746 - radv: use a more relaxed alignment for upload buffer allocations
4747 - radv: fix max_lds_per_simd on GFX10
4748 - radv: switch MaxWaves statistic to wave32 waves
4749 - ac: split lds_granularity into encode and allocation granularities
4750 - radv: use lds_{encode,alloc}_granularity
4751 - radv: round up max_lds_per_simd / lds_per_wave
4752 - aco: fix waves calculation for wave32
4753 - aco: add Program::wgp_mode
4754 - radv,aco: add radv_nir_compiler_options::wgp_mode
4755 - aco: consider that GFX10.3 allocates LDS in 1024 byte blocks
4756 - aco: add DeviceInfo
4757 - aco: fix transition_to_{WQM,Exact} if exec.back() is not in exec
4758 - radv: relax shared alignment requirements in mem_vectorize_callback
4759 - radv,aco: allow unaligned LDS access on GFX9+
4760 - aco/lower_phis: fix all_preds_uniform with continue_or_break
4761 - nir/dce: replace instruction worklist with ssa def bitset
4762 - nir: inline nir_foreach_{src,dest}
4763 - nir/dce: perform DCE for unlooped instructions in a single pass
4764 - aco: calculate all p_as_uniform and v_readfirstlane_b32 sources in WQM
4765 - aco: use p_as_uniform for get_sampler_desc and convert_pointer_to_64_bit
4766 - nir: fix build at -O1
4767 - nir: add nir_ssa_def_is_unused()
4768 - nir/copy_prop: remove unused copies
4769 - nir/copy_prop: visit copies instead of sources
4770 - nir/copy_prop: use nir_{instr,if}_rewrite_{src,condition}_ssa
4771 - Revert "radv,aco: allow unaligned LDS access on GFX9+"
4772 - aco: add missing usable_read2 check
4773 - nir/opt_shrink_vectors: add option to skip shrinking image stores
4774 - radv: don't shrink image stores for The Surge 2
4775 - radv: don't set sx_blend_opt_epsilon for V_028C70_COLOR_10_11_11
4776 - radv,aco: remove aco_compiler_statistics
4777 - radv: cache pipeline statistics
4778 - aco: set compr for fp16 exports
4779 - radv/llvm: fix enabled_channels for compressed exports
4780 - aco: simplify loop_nest_depth tracking in isel
4781 - aco: track divergent and uniform branch depth
4782 - aco: move wait_imm to aco_ir.h
4783 - aco: lower p_constaddr into separate instructions earlier
4784 - aco: add instruction classes
4785 - aco: add latency and inverse throughput statistics
4786 - aco: add print option to print program without temporary IDs
4787 - aco: add ACO_DEBUG=perfinfo
4788 - aco: remove vmem/smem score statistics
4789 - aco: fix NSA MIMG followed by MUBUF/MTBUF
4790 - aco/tests: add test for NSAToVMEMBug
4791 - aco: fix NSA following writelane
4792 - aco/tests: add test for waNsaCannotFollowWritelane
4793 - nir: Don't update base in vectorize_loads()
4794 - aco: implement 64-bit VGPR {u,i}find_msb
4795 - aco: use uadd32_sat() helper for nir_op_uadd_sat
4796 - aco: use a single instruction for uadd32_sat() on GFX8
4797 - aco: implement image_deref_samples
4798 - aco: add aco_print_program() flag to print kill flags
4799 - aco: add aco_print_program() flags to print live_out and register demand
4800 - docs: document ACO_DEBUG=perfinfo
4801 - aco: add ACO_DEBUG=liveinfo
4802 - radv: lower variables to ssa before nir_propagate_invariant
4803 - radv: lower view_index to zero if multiview is disabled
4804 - ci: add expected fail for RADV
4805 - aco: don't optimize min(a*1.0, ...) to min(a, ...) on GFX8
4806 - aco: use -1.0*x and 1.0*|x| for fneg/fabs
4807 - aco/tests: add tests for denormal-aware propagation
4808 - ac: invalidate metadata after hs_emit_write_tess_factors()
4809 - aco/tests: fix isel.sparse.clause for LLVM 12+
4810 - lavapipe: fix initialization of pipe_stream_output with unwritten outputs
4811 - nir/gather_info: implement partial masking of struct and compact I/O
4812 - nir/lower_tex: handle deref casts
4813 - nir_to_tgsi: run constant folding after nir_opt_algebraic
4814 - aco: fix integer tg4 workaround with unnormalized coordinates
4815 - draw: fix pstipple, aaline and aapoint without LLVM
4816 - aco: ensure loops nested in a WQM loop are in WQM
4817 - nir/gather_info: fix partial masking of compact I/O with location_frac!=0
4818 - radv: remove second nir_lower_idiv
4819 - nir/lower_idiv: add options to use fp32 for 8-bit division lowering
4820 - nir/lower_idiv: make lowered divisions exact
4821 - aco: fix 16-bit u2f32
4822 - aco: fix 16-bit f2{u8,i8} on GFX6/7
4823 - radv: don't use fp16 for 8-bit division lowering before GFX9
4824 - nir: add nir_block_get_predecessors_sorted() helper
4825 - nir/lcssa: fix nondeterminism in predecessor iteration
4826 - nir/loop_unroll: fix is_indirect_load() with load_global
4827 - radv: fix conditions for running nir_opt_vectorize
4828 - aco/ra: use original names when renaming loop carried phi operands
4829 - aco/ra: remove live-in temporary from live_out_per_block when moving it
4830 - radv: fix barrier in radv_decompress_dcc_compute shader
4831 - radv: fix clearing DCC-compressed e5b9g9r9 images
4832 - aco: set TRUNC_COORD=0 for nir_texop_tg4
4833 - ac/nir: set TRUNC_COORD=0 for nir_texop_tg4
4834 - Revert "radeonsi: set TRUNC_COORD=0 for Total War: WARHAMMER to fix it"
4835 - aco: don't update register demand during RA validation
4836 - aco: allow SDWA sels smaller than the operand size
4837 - radv: disable VK_FORMAT_R64_SFLOAT
4838 - vulkan: fix use-after-free in vk_common_DestroyDebugReportCallbackEXT
4839 - radv: fix use-after-free upon GS copy shader cache hits
4840 - radv,ac/llvm: use a dword alignment for descriptor loads
4844 - freedreno/ir3: Fix ldg decoding/parsing
4845 - freedreno/ir3: Decouple ir3_info collection from assembler
4846 - freedreno/ir3: Add some new "logical" opcodes
4847 - freedreno/hw: Add isaspec mechanism for documenting/defining an ISA
4848 - freedreno/hw/isa: Add description of ir3 ISA
4849 - freedreno/hw/isa: Add expression caching
4850 - freedreno/ir3/tests: Switch disasm test over to new decoder
4851 - freedreno/ir3: Switch over to new encoder/decoder
4852 - freedreno/ir3: Small resinfo disasm tweak
4853 - freedreno/ir3: Better sstall estimation
4854 - freedreno/ir3: Realign disasm shader stats
4855 - freedreno/ir3/decode: Switch over to new disasm
4856 - freedreno/ir3: Remove legacy packed-struct encoding
4857 - frontend/dri: Expose RGB[AX]_SRGB as well
4858 - freedreno/isa: Fix branch/jump offset encoding
4859 - freedreno/a6xx: Add r2d support for GMEM resolves
4860 - gallium/util: Add helpers to determine if z/s is written
4861 - freedreno/a6xx: Don't early-z if there are stencil writes
4862 - r300: Use util_writes_depth_stencil() helper
4863 - radeonsi: Use util_writes_stencil() helper
4864 - freedreno: Add perf_warn() trace helper
4865 - freedreno: Add fmt/args macros for pipe_resource
4866 - freedreno/a6xx: Add helper to check if UBWC is supported
4867 - freedreno: Add perf_warn() for missed UBWC opportunities
4868 - ci/freedreno/a6xx: Skip vs-output-array-vec2-index-wr-before-gs
4869 - freedreno/a6xx: Fix 3dmark misrendering with unwritten MRTs
4870 - mesa: Remove _mesa_destroy_context()
4871 - freedreno/decode: Fix overflow
4872 - freedreno: Put an upper limit on VSC size
4873 - freedreno: Misc cleanup
4874 - freedreno/a5xx: Drop fd5_compute_stateobj
4875 - freedreno/a6xx: Drop fd6_compute_stateobj
4876 - freedreno/ir3+a5xx+a6xx: De-duplicate create_compute_state()
4877 - freedreno/ir3: Add ir3_shader_state
4878 - freedreno/ir3: Move ir3_compiler_create()
4879 - freedreno/ir3: Add ir3_screen_fini()
4880 - freedreno/ir3: Reshuffle ir3_shader_create()
4881 - freedreno/ir3: Reshuffle compute state creation
4882 - freedreno/ir3: Async shader compile
4883 - freedreno/ir3: Add missing shader prog cache invalidation
4884 - freedreno: Quiet fallthrough warnings
4885 - freedreno: Split batch_flush_reset_dependencies()
4886 - freedreno: driver-thread annotations
4887 - freedreno/ir3/print: More sane ssa src/dst display
4888 - freedreno/ir3/print: Improve branch printing
4889 - util/fossilize_db: Fix compile error with clang
4890 - freedreno: Handle InvalidateBufferData() case
4891 - freedreno: Add perf_debug logging for bo stalls
4892 - freedreno: Workaround for UNSYNC+DISCARD_RANGE
4893 - driconf: Generate a static table when no xmlconfig
4894 - xmlconfig: Reshuffle to keep attr processing
4895 - xmlconfig: Add static driconfig support
4896 - freedreno/ir3: Drop foreach_bit() macro
4897 - freedreno: Drop foreach_bit() macro
4898 - etnaviv: Drop foreach_bit() macro
4899 - v3d: Drop foreach_bit() macro
4900 - freedreno: Fix think-o in fd_resource_wait()
4901 - freedreno/ir3: Fix initial_variants_synchronous() condition
4902 - freedreno: Add FD_DBG() macro
4903 - freedreno: Slight perf_debug rework
4904 - freedreno: Add macro for duration based warns
4905 - util/u_queue: Ensure num_cpu_mask_bits is valid
4906 - util: Add accessor for util_cpu_caps
4907 - freedreno/a6xx: Always pass ctx to fd6_emit_textures()
4908 - freedreno/a6xx: Fix uncompressed resource vs stale CSO
4909 - freedreno/ir3: Add comments about shader key/gen
4910 - freedreno: Deduplicate fixup_shader_state()
4911 - freedreno/a6xx: Fix compile warning
4912 - driconf: Add ignore_map_unsynchronized option
4913 - freedreno: Remove dead-cells MBR workaround
4914 - util: Extract thread-id helpers from u_current
4915 - gallium/u_threaded: Add helper to assert driver thread
4916 - gallium/u_threaded: use mesa_log for debug msgs
4917 - freedreno: Fix u_blitter constant-buffer leak
4918 - freedreno: Factor out common fd_resource init
4919 - freedreno: Split out batch/resource tracking
4920 - freedreno: Restructure transfer_map()
4921 - freedreno: Extend threaded_resource
4922 - freedreno: Extend threaded_transfer
4923 - freedreno: Extract out helper for transfer-map flag munging
4924 - freedreno: Add fd_replace_buffer_storage()
4925 - freedreno: Add transfer_pool_unsync
4926 - freedreno/a6xx: Move UBWC demotion to first sampler view bind
4927 - freedreno: Check cb0 in rebind_resource()
4928 - freedreno: threaded_context support
4929 - freedreno: threaded_context async flush support
4930 - freedreno: Fix fd_fence_finish()
4931 - freedreno/drm: Avoid unitialized timestamp in submit fail
4932 - freedreno/drm: Split softpin "reloc" functions
4933 - freedreno/drm: Split 64b vs 32b paths
4934 - freedreno/drm: Move emit_reloc_tail to head
4935 - freedreno/drm: Inline iova calculation
4936 - freedreno/ir3: Precompute whether we need driver-params
4937 - freedreno: Add helpers to mark dirty state
4938 - freedreno: Add mapping to generation specific dirty state
4939 - freedreno/a6xx: Convert to dirty_groups
4940 - freedreno: Small dirty flag re-org
4941 - freedreno: Add dirty bit for state that needs rsc tracking
4942 - freedreno: Don't ignore geom/tess stage resources
4943 - freedreno: Split out helper for updating sw stats
4944 - freedreno: Only collect sw stats when required
4945 - freedreno/a6xx/vsc: Be more tolerate of degenerate prims
4946 - freedreno: Drop u_trim_pipe_prim() from fast-paths
4947 - u_draw: Add helper to emultate multi-draw
4948 - freedreno: Use multi-draw helper
4949 - freedreno: Handle multi-draw edge cases
4950 - freedreno: Push multi-draw closer to backend
4951 - freedreno/a6xx: Emit streamout state on every draw
4952 - freedreno: Add draw cost estimation
4953 - freedreno/batch: Export key/hash fxns
4954 - freedreno/batch: Add a way to clone a batch key
4955 - freedreno: Add gmem_reason_mask
4956 - freedreno/a6xx: Fix sRGB/snorm vs sysmem clear path
4957 - freedreno: Autotune bypass vs GMEM rendering decision
4958 - freedreno/a6xx: Fix typo
4959 - freedreno: Make headers C++ happy
4960 - freedreno/fdperf: Use os_read_file()
4961 - freedreno: Split out devicetree helpers
4962 - ci: Disable panfrost t760
4963 - freedreno/a6xx: Fix indirect+patches draws
4964 - freedreno/a6xx: Fix obsolete comment
4965 - d3d12: Use util_draw_multi() helper
4966 - etnaviv: Use util_draw_multi() helper
4967 - i915: Use util_draw_multi() helper
4968 - iris: Use util_draw_multi() helper
4969 - lima: Use util_draw_multi() helper
4970 - llvmpipe: Use util_draw_multi() helper
4971 - nouveau: Use util_draw_multi() helper
4972 - r300: Use util_draw_multi() helper
4973 - r600: Use util_draw_multi() helper
4974 - softpipe: Use util_draw_multi() helper
4975 - svga: Use util_draw_multi() helper
4976 - tegra: Use util_draw_multi() helper
4977 - vc4: Use util_draw_multi() helper
4978 - v3d: Use util_draw_multi() helper
4979 - virgl: Use util_draw_multi() helper
4980 - freedreno: Don't handle multi-draw in indirect case
4981 - util/primconvert: Handle indirect and multi-draw
4982 - freedreno: Add .clang-format
4983 - freedreno: Some manual reformatting
4984 - freedreno: Re-indent
4985 - freedreno: Manual fixups
4986 - freedreno: Add missing foreach macros and update indentation
4990 - virgl: Cache depth and stencil buffers
4991 - ci: Ensure that jobs inherting the ci-deqp jobs artifact meson logs
4992 - intel/genxml: Free resource before exiting
4993 - intel/compiler: Free resources on test teardown
4994 - virgl: update headers
4995 - virgl: Return total video memory if available
4996 - virgl: Add support for querying detailed memory info
4997 - virgl: Support the ETC1_RGB8 format as virglrenderer supports it
5001 - egl: android: use num_planes param in createImageFromDmaBufs()
5005 - radeon/vcn: release si buffer for encoding at the end.
5009 - Revert "virgl: fix BGRA emulation artifacts during window resize"
5013 - anv: Invalidate the correct AUX-TT entry
5014 - anv: Skip CCS ambiguate which preceed fast-clears
5015 - intel/mi_builder: Added support for command streamer shift operations
5016 - anv: Add anv_memregion structure
5017 - Revert "Revert "blorp/gen12: Don't use aux address if implicit CCS""
5018 - intel/blorp: Fix condition to figure out aux_address
5019 - anv: Set correct binding table entry count
5023 - turnip: disable UBWC on Z24_S8 MSAA images on A630
5024 - turnip: set sparseAddressSpaceSize to zero
5025 - turnip: fix UINT64_MAX size wrapping in tu_GetBufferMemoryRequirements()
5026 - turnip: fix resolve MSAA D24_UNORM_S8_UINT image to S8_UINT
5027 - turnip: fix resolve MSAA D32_SFLOAT_S8_UINT image to S8_UINT
5028 - util: fix parsing of /proc/meminfo MemAvailable value
5029 - turnip: keep track of memory heap usage, size and flags
5030 - turnip: VK_EXT_memory_budget implementation
5031 - turnip: set depth plane control zmode to A6XX_LATE_Z when sample mask is written
5035 - radv: do not invalidate the L2 metadata cache on compute queues
5036 - ci: mark some sparse CTS as expected failures on RAVEN
5037 - radv: flush L2 metadata as part of CB/DB flush instead of CS_DONE on GFX9
5038 - radv: add a comment explaining the micro tile mode resolve
5039 - radv: enable TC-compat HTILE with D32S8 and MSAA on GFX9+
5040 - radv: enable TC-compat HTILE for D16S8 on GFX9+
5041 - radv: restore invalidating the vector cache for internal meta operations
5042 - radv: flush L2 for images affected by the pipe misaligned issue on GFX10+
5043 - ci: exclude one CTS test that timeout most of the time for RADV CI
5044 - radv: remove redundant check in radv_process_depth_stencil()
5045 - radv: remove unnecessary radv_image::tc_compatible_htile
5046 - radv: remove redundant check in depth_view_can_fast_clear()
5047 - radv: fix a sync issue with geometry shader primitives query on GFX10+
5048 - radv: fix overflow when computing the SQTT buffer size
5049 - radv: inhibit clock gating when tracing with SQTT
5050 - ac/rgp: add support for GFX10.3
5051 - ac,radv: add SQTT support on GFX10.3
5052 - radv: enable SQTT support on GFX10.3
5053 - radv: fix separate depth/stencil layout in render pass
5054 - radv: add multi-layer support to FMASK color expand
5055 - radv: use the range aspect mask in FMASK color expand
5056 - radv: use a workgroup size of 8x8 for FMASK color expand
5057 - radv: only decompress the depth/stencil aspect that needs to be resolved
5058 - radv: enable sparseImageInt64Atomics/sparseImageFloat32Atomics
5059 - radv,aco: fix shifting input VGPRs for the LS VGPR init bug on GFX9
5060 - radv: synchronize Cmd{Set,Write}Event() using PS_DONE/CS_DONE events
5061 - radv: add support for emitting PS_DONE/CS_DONE on GFX6-8
5062 - radv: remove radv_util.h
5063 - radv: remove stub() macros
5064 - radv: remove unused EMPTY constant in radv_descript_set.c
5065 - nir/algebraic: mark more optimization with fsat(NaN) as inexact
5066 - ac/surface: store HTILE mip info into the surface
5067 - radv: use the image view range when fast clearing depth
5068 - radv: check if HTILE is enabled per-level instead of the entire image
5069 - radv: do not decompress/resummarize levels without HTILE
5070 - radv: remove mipmaps related assertions when initializing HTILE
5071 - radv: add support for fast clearing levels of the HTILE buffer
5072 - radv: teach radv_htile_enabled() about the number of HTILE levels
5073 - radv: enable TC-compat HTILE for mipmaps on GFX10+
5074 - radv: re-disable TC-compat HTILE for D32S8 on all generations
5075 - radv: fix centroid with VRS coarse shading
5076 - radv/winsys: move the initial BO domain to radeon_winsys_bo
5077 - radv: prefer CP DMA for GTT buffer copies/clears on dGPUs due to slow PCIe
5078 - radv: fix waiting on the last enabled RB for occlusion queries
5079 - radv/winsys: use an array for the global BO list instead of a list
5080 - radv/winsys: remove the radv_amdgpu_winsys_bo::ws indirection
5081 - radv/winsys: remove useless continue preamble CS for IBs path
5082 - radv/winsys: remove useless is_local check in radv_amdgpu_cs_add_buffer()
5083 - radv/winsys: remove unused radeon_bo_usage enum
5084 - radv/winsys: simplify the user fence logic for submission
5085 - radv/winsys: remove unused fields in radv_amdgpu_cs_request
5086 - radv/winsys: stop zeroing radv_amdgpu_cs_request
5087 - radv: use less AMDGPU contexts by creating only one per queue priority
5088 - radv: add radeon_winsys_bo::use_global_list
5089 - radv: stop using VM_ALWAYS_VALID on APUs
5090 - radv/winsys: move the debug_all_bos check outside of the add/del helpers
5091 - radv/winsys: set use_global_list to avoid adding a BO twice
5092 - radv/winsys: add buffer_make_resident() to the API
5093 - radv/winsys: add the resident BOs to the list of BOs at submit time
5094 - radv/winsys: enable the global BO list unconditionally
5095 - radv: use the global BO list from the winsys
5096 - radv: fix printing the debug option names
5097 - radv: fix double free when creating a fence failed
5098 - radv: stop allocating useless ESGS scratch BO on GFX10+
5099 - radv: fix memory leaks if a submission fails
5100 - radv: do not overallocate the SQTT buffer
5101 - radv: adjust an error message related to the SQTT buffer size
5102 - radv: add support for resizing the SQTT buffer automatically
5103 - ac/rgp: append the number of seconds to the generated RGP file
5104 - radv: emit pipeline bind markers for SQTT
5105 - radv: only make the WSI images resident if the global BO list is used
5106 - radv/winsys: set use_global_list inside the critical section
5107 - radv: only apply the MRT output NaN fixup to non-meta shaders
5108 - radv: create the start/stop CS for SQTT dynamically
5109 - radv: move SQTT parameters initialization to radv_thread_trace_init()
5110 - radv: remove an outdated TODO about SQTT cache flushes
5111 - radv: make sure to allocate enough space when emitting SQTT userdata
5112 - radv: stop emitting pipeline bind markers
5113 - radv: do not allow to capture SQTT on the compute queue
5114 - radv: add support for user event markers with SQTT
5115 - radv: only emit pipeline bind markers for application pipelines
5116 - radv: use the pipeline key as hash for pipeline bind markers
5117 - radv: set correct value for OFFCHIP_BUFFERING on GFX10+
5118 - radv: make the border color BO a resident buffer
5119 - radv: make the trace BO a resident buffer
5120 - radv: make the TMA/TBA BOs resident buffers
5121 - radv: emit the trap handler registers earlier
5122 - radv: rework radv_cmd_buffer_resolve_subpass() a bit
5123 - radv: emit missing subpass resolve marker for SQTT
5124 - ac/rgp: fill CPU info by parsing /proc/cpuinfo
5125 - radv: store a pointer to the code in radv_shader_variant
5126 - radv: add support for exporting pipelines with RGP
5127 - radv: add support for instruction timing with RGP
5128 - radv: do not scale the depth bias for D16_UNORM depth surfaces
5129 - include/drm-uapi: bump AMDGPU headers
5130 - ac/rgp: recognize more memory types
5131 - ac/rgp: report LDS size in CU mode on GFX10+
5132 - ac/rgp: report the number of memory operations per clock
5133 - ac/rgp: report the number of primitives per clock
5134 - radv: remove duplicate REG_INCLUDE_CONTEXT setting for SQTT
5135 - radv: always select the first active CU when profiling with SQTT
5136 - radv: fix exporting SQTT pipelines with LLVM
5137 - radv: exclude perf counters for SQTT also on GFX10.3
5138 - Revert "radv: do not overallocate the SQTT buffer"
5139 - radeonsi,radv: do not overallocate the SQTT buffer size
5140 - radv: remove useless decompression of the DS resolve attachment
5141 - radv: do not trace inactive shader engines with SQTT
5142 - ac/sqtt: fix determining if the trace is complete on GFX10+
5143 - radv: double the SQTT buffer size when it is resized
5144 - radv: trigger a new SQTT capture automatically after resizing the buffer
5145 - radv: bump the initial SQTT buffer size to 32MB per SE
5146 - radv: fix RGP barrier layout transition for TC-compatible CMASK images
5147 - Revert "radv: stop using VM_ALWAYS_VALID on APUs"
5148 - radv: cleanup enabling TC-compat HTILE for depth surfaces
5149 - radv: remove useless check about mips+layers for TC-compat HTILE images
5150 - radv: skip useless FCE when fast-clearing MSAA images with DCC enabled
5151 - radv: re-enable TC-compat HTILE for MSAA D32S8 images on GFX9+
5152 - radv: do not declare push constants for DCC decompress on compute
5153 - radv: check if dynamic VRS state changed
5154 - radv: check if dynamic line stipple state changed
5155 - radv: disable sampling with VK_FORMAT_R64_SFLOAT
5156 - radv: fix meta save/restore state with non renderable images
5157 - radv: fix potential clears with non renderable images on GFX9+
5158 - radv: fix initialization of disable_compression when clearing color image
5159 - radv: add missing SQTT events for copy_commands2/create_renderpass2
5160 - radv: remove useless DCC disable check for 3D images on GFX10+
5161 - radv: rework radv_use_dcc_for_image() a bit
5162 - vulkan: add missing vk_shader_module.c/h includes to Makefile
5163 - radv: use common entrypoints for VK_KHR_copy_commands2
5164 - radv: do not enable TC-compat CMASK if the image isn't readable by a shader
5165 - radv: remove redundant check when enabling TC-compat CMASK
5166 - radv: make sure FMASK is enabled for TC-compat CMASK
5167 - radv: only configure the CMASK tiling for TC-compat on GFX8
5168 - radv: initialize TC-compat CMASK images with the DCC clear code
5169 - radv: enable TC-compat CMASK on GFX10+
5170 - radv: add notccompatcmask debug option
5171 - radv: extend the dirty bits to 64-bit
5172 - ac/surface: init CMASK slice size on GFX9+
5173 - radv: fix clearing CMASK layers on GFX9+
5174 - radv: initialize CMASK with correct clear codes
5175 - radv: restore previous MRT CB_SHADER_MASK logic
5176 - radv: gather if the FS uses perspective or linear interpolations
5177 - radv: determine if a pipeline is candidate for flat shading
5178 - radv: enable VRS 2x2 coarse shading for flat shading on GFX10.3+
5179 - radv: add RADV_DEBUG=novrsflatshading option
5180 - ci: update list of expected CTS failures for RADV
5181 - vulkan: add common entrypoints for VK_KHR_create_renderpass2
5182 - radv: use common entrypoints for VK_KHR_create_renderpass2
5183 - turnip: use common entrypoints for VK_KHR_create_renderpass2
5184 - lavapipe: use common entrypoints for VK_KHR_create_renderpass2
5185 - anv: use common entrypoints for VK_KHR_create_renderpass2
5186 - radv: report that degenerated triangles are not culled
5187 - radv: require DRM 3.35+
5188 - ac/surface: do not allocate FMASK or CMASK for stencil-only surfaces on GFX9+
5189 - radv: do not fixup DCC after compute color resolves if DCC stores enabled
5190 - radv: only set WRITE_COMPRESS_ENABLE for storage image descriptors
5191 - radv: use a sampled image descriptor for reads for the MSAA color decompress
5192 - radv: compress FMASK for all layouts except GENERAL
5193 - radv: cleanup FMASK expand transitions
5194 - radv: do not force enable FMASK during MSAA blits
5195 - radv: use COLOR_ATTACHMENT_OPTIMAL for fast clear/hw resolve operations
5196 - ac: add ac_get_family_name() helper
5197 - radv: change RADV_FORCE_FAMILY to use family name instead of LLVM processor name
5198 - radv: try to keep HTILE compressed with DEPTH_STENCIL_READ_ONLY_OPTIMAL
5199 - radv: clean up fence syncobj code
5200 - ac: add ac_gpu_info::has_image_load_dcc_bug
5201 - aco: fix get_sampler_desc() for image loads
5202 - aco: implement a workaround for the image load DCC hw bug on GFX10.3
5203 - radv: allow DCC for storage images on GFX10.3 with RADV_PERFTEST=dccstores
5204 - radv: handle implicit subpass dependencies per attachment
5205 - radv: init CMASK/FMASK/DCC in parallel
5206 - radv: perform MSAA color decompression for storage images with DCC
5207 - radv: enable DCC stores with MSAA 4x/8x on GFX10+
5208 - radv: simplify a check when enabling DCC for concurrent images
5209 - radv: enable DCC for concurrent images on GFX10
5210 - radv: make sure FMASK decompress and FCE are performed on gfx queue
5211 - radv: add MSAA support to ClearColorImage() on compute queue
5212 - radv: do not clamp framebuffer dimensions to the minimum dimension
5213 - radv: add MSAA support to CopyImage() on compute queue
5214 - radv: use explicit VRS mode when configuring PA_CL_VRS_CNTL
5215 - radv: allow to force VRS rates on GFX10.3 with RADV_FORCE_VRS
5216 - radv: fix needed dynamic state for VRS
5217 - amd/addrlib: expose HTILE address equations to drivers on GFX10+
5218 - ac/surface: rename ac_surface_dcc_address_test.c
5219 - ac/surface: add a test of HtileAddrFromCoord prototype outside of addrlib
5220 - ac/surface: rename gfx9_dcc_equation to gfx9_meta_equation
5221 - ac/surface: increase gfx9_meta_equation::gfx10_bits by 4 elements
5222 - ac/surface: copy the HTILE equations to the surface
5223 - ac/surface: implement HtileAddrFromCoord in NIR
5224 - ac/surface: store the HTILE pitch to the surface
5225 - radv: expose R8_UINT as the only supported format for VRS attachments
5226 - radv: do not allow MSAA with fragment shading rate attachments
5227 - radv: do not enable DCC for fragment shading rate attachments
5228 - radv: determine if attachment VRS is enabled
5229 - radv: configure the VRS HTILE encoding size
5230 - radv: do not use the whole HTILE buffer for depth when VRS is used
5231 - radv: update the HTILE clear word when VRS is used
5232 - radv: allow HTILE for very small images if VRS attachment is used
5233 - radv: create an image for VRS if no depth/stencil attachment is bound
5234 - radv: handle the VRS attachment subpass
5235 - radv: bind our internal depth buffer when not provided by the app
5236 - radv: add support for copying VRS rates into HTILE
5237 - radv: copy VRS rates to HTILE when beginning a subpass
5238 - radv: configure the VRS combiners when an attachment is used
5239 - radv: advertise attachmentFragmentShadingRate on GFX10.3
5240 - ac: add missing BUF_DATA_FORMAT_10_11_11 vertex format on GFX10+
5241 - radv: keep DCC compressed for clears on compute with image stores
5242 - aco: fix opquantize2f16 on GFX6-7
5243 - radv: fix fast clearing depth-only or stencil-only aspects with HTILE
5244 - radv: fix emitting depth bias when beginning a command buffer
5245 - radv: fix emitting default depth bounds state on GFX6
5246 - radv/winsys: fix allocating the number of CS in the sysmem path
5247 - radv/winsys: fix resetting the number of padded IB words
5248 - radv: make sure CP DMA is idle before executing secondary command buffers
5249 - radv: fix various CMASK regressions on GFX9
5250 - radv: fix computation of the number of user SGPRS for NGG GS state
5251 - radv: check if DCC is enabled when resolving different levels
5252 - radv/winsys: fix executing huge secondary command buffers on GFX6
5256 - clover: return CL_INVALID_VALUE when origin or region are NULL
5260 - nouveau/nvc0: fix linear buffer alignment for scan-out/cursors
5261 - nouveau/nv50: fix linear buffer alignment for scan-out/cursors
5262 - frontends/va: extract pipe format to DRM format mapping
5263 - frontends/va: add support for VA_EXPORT_SURFACE_COMPOSED_LAYERS
5264 - frontends/va: add pipe to DRM format mapping for NV12 and P010
5265 - radeonsi/uvd: make format modifiers-aware
5266 - egl: use render node for wl_drm if available
5267 - gbm: fail early when modifier list only contains INVALID
5268 - gbm: remove fprintf calls in gbm_dri_bo_create
5269 - egl/wayland: avoid unnecessary roundtrip when authenticated
5270 - gbm: add gbm_bo_get_fd_for_plane
5271 - egl: fix software flag in _eglAddDevice call on DRM
5272 - egl: only take render nodes into account when listing DRM devices
5273 - Revert "egl: Don't add hardware device if there is no render node v2."
5274 - radv: fix format feature reporting for modifiers
5278 - egl/dri2: enable EGL_WL_bind_wayland_display in EGL device platform
5282 - virgl: Add simple disk cache
5286 - va/picture : Added failure check for stability
5287 - frontends/va: Update conditional checks for code stability.
5288 - frontends/va : Fix memory leaks incase of error returns
5292 - util: add mesa_glthread for Valheim in OpenGL mode.
5296 - mesa: add GL_SR8_EXT, GL_SRG8_EXT for color/srgb format queries
5297 - intel/perf: cleanup, remove duplicate function declaration
5298 - intel/perf: introduce additional ralloc context parameter
5299 - i965: use aligned malloc for context instead of ralloc
5300 - mesa: add check that non base level attachment is mipmap complete
5301 - gitlab-ci: bump piglit commit for windows
5302 - anv: toggle on sample shading if it is set in the shader
5303 - anv/android: fix compilation failure
5304 - anv: fix compilation due to missing vk_format_from_android
5305 - mesa: check cube completeness for cube fbo attachments
5306 - anv/android: fix image creation with external format
5307 - android: add some more stub functions for cross compilation
5308 - intel/common: disable batch decoder on Android platform
5309 - loader: prefer iris on Android
5310 - iris: clamp PointWidth in 3DSTATE_SF like i965 does
5311 - egl: support no error attribute set to false with ES 1.1
5312 - glx: revert "Downgrade sRGB-ful fbconfigs"
5316 - frontends/va/config: Fix check for packed header config
5317 - radeon: Add cropping to encoded H.265 when padding is used
5321 - llvmpipe: increase PIPE_CAP_MAX_TEXTURE_BUFFER_SIZE value
5325 - util/disk_cache: do crc32 check on compressed data for ZSTD
5326 - util/disk_cache: move cache path strdup call back into disk_cache.c
5327 - util/disk_cache: use a new cache dir for the single file cache feature
5328 - util/mesa_sha1: add helper to reconvert sha1 hex strings
5329 - util/fossilize_db: add basic fossilize db util to read/write shader caches
5330 - util/disk_cache: make use of single file cache when env var set
5331 - nir: handle negatives in ffma reassociation optimisation
5332 - util/disk_cache: fix crash in fossilize_db
5333 - util/disk_cache: move cache tests to the util directory
5334 - util/disk_cache: make MESA_DISK_CACHE_READ_ONLY_FOZ_DBS a relative path
5335 - Revert "glsl: default to compat shaders in compat profile"
5336 - glsl: fix declarations of gl_MaxVaryingFloats
5337 - util: create some standalone compression helpers
5338 - util/disk_cache: make use of the new compression helpers
5339 - util/fossilize_db: remove compression from foz db helper
5340 - util/compress: make compression function inputs const
5341 - util/disk_cache: separate file reads from cache item validation
5342 - util/disk_cache: detangle cache item creation from disk writing
5343 - util/disk_cache: add cache item headers to single file cache entries
5344 - glsl: add compilation errors for attribute and varying qualifiers
5345 - glsl: enforce restrictions on builtin functions moved to compat
5346 - mesa: fix incomplete GL_NV_half_float implementation
5347 - util: disable glthread in CSGO
5351 - radv: Only enable sparse features on Polaris and newer.
5352 - tgsi_to_nir: Fix uniform ranges.
5353 - aco: Fix LDS statistics of tess control shaders.
5354 - radv/llvm: Fix reporting LDS stats of tess control shaders.
5355 - aco: Disallow LSHS temp-only I/O when VS output is written indirectly.
5356 - pan/bi: Use correct enum type for NIR intrinsics.
5357 - aco: Use ASSERTED to avoid unused variable warning.
5358 - intel/compiler: Use assume() instead of assert() for array bounds.
5359 - intel/compiler: Make room for maximum dest size in nir_emit_texture.
5360 - anv: Use unreachable() in anv_genX.
5361 - anv: Use ASSERTED for results that are only used in asserts.
5362 - nir: Add new nir_builder helpers for iadd with no_unsigned_wrap.
5363 - nir: Add nir_builder helper for I/O address offset calculations.
5364 - nir: Add a few more algebraic optimizations to help address calculation.
5365 - nir: Fix unsigned upper bound of local_invocation_index for non-CS stages.
5366 - nir: Shrink vectors for load_shared.
5367 - nir: Add unsigned upper bound for TCS load_invocation_id.
5368 - nir: Add default unsigned upper bound configuration.
5369 - nir: Add AMD-specific buffer load/store intrinsics.
5370 - nir: Add nir_opt_offsets to fold const adds into load/store offsets.
5371 - nir: Add tessellation related AMD-specific intrinsics.
5372 - nir: Add AMD-specific Geometry Shader related intrinsics.
5373 - aco: Implement new buffer load/store intrinsics.
5374 - aco: Implement the new tessellation I/O related NIR intrinsics.
5375 - aco: Implement new Geometry Shader intrinsics.
5376 - ac/llvm: Implement AMD-specific buffer load/store intrinsics.
5377 - ac/llvm: Implement the new tessellation intrinsics.
5378 - ac/llvm: Implement new Geometry Shader intrinsics.
5379 - ac/llvm: Make shared loads/stores work correctly for non-CS stages.
5380 - ac/llvm: Make sure to always emit integer comparison for nir_op_ieq.
5381 - ac/llvm: Add constant offset to load/store_shared.
5382 - ac/llvm: Emit more efficient code for load_shared.
5383 - ac: Add NIR passes to lower VS->TCS->TES I/O to memory accesses.
5384 - ac: Add NIR passes to lower ES->GS I/O to memory accesses.
5385 - radv: Lower IO and set driver locations earlier.
5386 - radv: Save I/O usage data to both shader infos for merged stages.
5387 - radv: Calculate tess patches and LDS use outside the backend compilers.
5388 - radv: Determine tcs_in_out_eq in radv_pipeline instead of the compiler.
5389 - radv: Fill some tess shader info earlier.
5390 - radv: Reorder some NIR optimizations in preparation for the I/O changes.
5391 - radv: Use new, NIR-based I/O lowering.
5392 - radv/llvm: Only store TCS outputs where they are really needed.
5393 - radv/llvm: Delete superfluous tess and ESGS I/O code.
5394 - aco: Delete superfluous tess and ESGS I/O code.
5395 - aco: Fix constant address offset calculation for ds_read2 instructions.
5396 - ac/llvm: Fix alignment of shared load intrinsics.
5397 - aco: Optimize workgroup exclusive scan to better avoid bank conflicts.
5398 - aco: Align NGG scratch size to 16 so a single ds_read can always read it.
5399 - aco: Remove useless s_setprio near gs_alloc_req.
5400 - aco: Use s_setprio 3 at the beginning of every VS and TES.
5401 - aco: Extract ngg_nogs_export_prim_id to a separate function.
5402 - aco: Set block_kind_export_end in create_vs/fs_exports.
5403 - aco: Emit fewer branches for NGG VS/TES with late primitive export.
5404 - aco: Add a simple heuristic to decide early or late primitive export.
5405 - aco: Mark VCC clobbered for iadd8 and iadd16 reductions on GFX6-7.
5409 - ci: Fix selection of linker in Android builds
5410 - ci: Move container files into their own dir
5411 - ci: Move out expect files from .gitlab-ci
5412 - ci: Disable two radeonsi jobs
5413 - Revert "ci/panfrost: disable the rest of these jobs temporarily"
5414 - Revert "ci/panfrost: Disable t860/radeonsi testing while the runners are struggling."
5415 - Revert "CI: Disable Panfrost T760"
5416 - ci: Fix visibility property of LAVA jobs
5417 - ci/fdo: Use trimmed traces for Valve games
5418 - gallium/dri2: Pass the resource that corresponds to the plane
5419 - ci: Use a single template for LAVA jobs
5420 - ci: Set more reasonable timeouts for LAVA jobs
5421 - ci: Don't retry failed test runs
5422 - ci: Disable t720 LAVA jobs
5423 - Revert "ci: Disable t720 LAVA jobs"
5424 - Revert "ci: Disable panfrost g52"
5425 - Revert "ci: Disable panfrost t760"
5429 - aco/ra: Update register use bounds before recursing in get_regs_for_copies
5430 - aco/ra: Introduce PhysRegInterval helper class
5431 - aco/ra: Conservatively refactor existing code to use PhysRegInterval
5432 - aco/ra: Remove always-false conditions
5433 - aco/ra: Add iterator interface for PhysRegInterval
5434 - aco/ra: Use std::find_if(_not) to clean up get_reg_simple
5435 - aco/ra: Use std::all_of to simplify a loop
5436 - aco/ra: Conservatively refactor get_reg_specified to use PhysRegInterval
5437 - aco/ra: Move commonly repeated code to a helper function
5438 - aco/ra: Add helpers to test for intersection/containment of reg intervals
5439 - aco/ra: Use std::all_of to simplify a loop
5440 - aco/ra: Remove unused function parameter
5441 - aco/ra: Use PhysReg for member functions of PhysRegInterval
5442 - aco/ra: Use PhysReg when indexing into RegisterFile's containers
5443 - aco/ra: Use PhysRegInterval for collect_vars parameters
5444 - aco/ra: Use PhysRegInterval for count_zero
5445 - aco/ra: Fix print_regs using the wrong constant to check for blocked slots
5446 - aco/ra: Fix build with print_regs enabled
5447 - aco/ra: Remove preprocessor guards for print_regs
5448 - aco/ra: Add helper to get a PhysRegInterval for the register demand
5449 - aco: Fix vector::reserve() being called with the wrong size
5450 - radv: Fix improper max_index_count argument for indexed draws
5451 - ac: Add has_zero_index_buffer_bug to ac_gpu_info
5452 - radv: Skip 0-sized index buffers only when necessary
5453 - aco/ra: Avoid unnecessary copying of std::vectors
5454 - aco/isel: Don't emit unsupported i16<->f16 conversion opcodes on GFX6/7
5455 - aco/isel: Fix i64/u64->float32 conversion for large inputs
5456 - aco/isel: Don't request sign extension when truncating signed integers
5457 - aco/isel: Add documentation and asserts for convert_int
5458 - aco/isel: Fix large inputs being truncated in int32->f16 conversions
5459 - aco/isel: Add documentation for (u)int64->f16 conversion
5460 - ci: skip pipeline_barrier tests that currently crash on RADV
5461 - gitlab: rename RADV bug report template
5462 - aco/spill: Fix improper handling of exec phis
5466 - lima: add precompile debug flag
5467 - lima/ppir: don't use list_length() in loop in regalloc and liveness analysis
5468 - lima: update dEQP fails and skips lists
5469 - lima: relax checks of imported BO
5470 - lima: rename \*_shader_state to \*_compiled_shader
5471 - lima: rename lima_{fs,vs}_bind_state to lima_{fs,vs}_uncompiled_shader
5472 - lima: implement shader disk cache
5473 - lima: compute nir_sha1 for shader key even if disk cache is disabled
5474 - lima: use passed surface to get mipmap level for reload, not cbuf
5475 - lima: limit number of draws per job
5479 - panfrost: Fix typos.
5480 - nouveau: Fix typos.
5481 - nv50/ir: Initialize DataArray members in constructor.
5482 - r600/sfn: Remove StoreMerger unused member b.
5483 - nv50/ir: Add InsertConstraintsPass constructor.
5484 - nv50/ir: Initialize CodeEmitter members in constructor.
5485 - nv50/ir: Initialize RegAlloc member func in constructor.
5486 - clover: Add constructor for global_argument.
5487 - lima: Fix typos.
5488 - v3dv: Fix assert.
5489 - nvc0/ir: Initialize NVC0LoweringPass member gpEmitAddress.
5490 - nvc0/ir: Initialize SchedDataCalculator members in constructor.
5491 - nv50/ir: Initialize BindArgumentsPass member sub in constructor.
5492 - virgl: Convert errno to string.
5493 - r600/sfn: Initialize FragmentShaderFromNir member m_pos_input.
5494 - etnaviv: Fix memory leak in etna_vertex_elements_state_create.
5495 - nv50/ir: Initialize ValueDef member origin in constructors.
5496 - nv50/ir: Initialize Instruction members.
5497 - aco: Initialize ds_state.front.writeMask.
5498 - r600: Fix typos.
5499 - llvmpipe: Fix typos.
5500 - nir/lower_tex: Change coord type to int.
5501 - gv100/ir: Initialize CodeEmitterGV100 members in constructor.
5502 - zink: Remove leftover dead code.
5503 - nv50/ir: Add constructor for NV50LegalizePostRA.
5504 - iris: Fix typos.
5505 - clover: Add constructor for sampler_argument.
5506 - ac: Fix emit_split_buffer_store modulus operation.
5507 - freedreno: Fix file descriptor leak.
5508 - glsl: Initialize parcel_out_uniform_storage members.
5509 - Remove leftover dead code.
5513 - frontends/va/context: don't set max_references with num_render_targets
5517 - lavapipe: Defer lavapipe warning to CreateDevice
5518 - util: Use explicit relaxed reads for u_queue
5519 - radv: memset the alignment hole in cache_entry to 0
5523 - virgl: use atomic operations when increase sub_ctx_id
5527 - turnip: fix alpha to coverage in no color and unused attachment cases
5531 - st/mesa: fix PBO download for TEXTURE_1D_ARRAY textures
5532 - intel/compiler: remove unused member 'input_vue_map'
5536 - iris: only set point sprite overrides if actually using points
5537 - nir/from_ssa: consider defs in sibling blocks
5538 - nir/from_ssa: don't check for interference within the same set
5542 - venus: properly enable WSI for different platforms
5543 - venus: bring up Android support
5544 - venus: implement vn_debug_init_once with os_get_option
5548 - ac/rgp: add ac_msgpack.h/c
5549 - ac/rgp: add rgp co, col, pso data structures
5550 - ac/rgp: add helper function to write rgp elf oject
5551 - ac/rgp: expose data structure to populate co, col, pso database
5552 - ac/rgp,radeonsi,radv: pass struct thread_trace_data to ac_sqtt_dump_data()
5553 - ac/rgp: dump co, col, pso database to rgp profile file
5554 - ac/rgp: set gfxip in elf_hdr.e_flags
5558 - mesa: update oudated members for debug and check
5562 - frontend/dri: fix doesn't support RGBA ordering still expose RGBA in config
5563 - glsl: redeclare built-in variable with separate shader
5564 - virgl: add astc 2d compressed formats