1Mesa 21.0.0 Release Notes / 2021-03-11 2====================================== 3 4Mesa 21.0.0 is a new development release. People who are concerned 5with stability and reliability should stick with a previous release or 6wait for Mesa 21.0.1. 7 8Mesa 21.0.0 implements the OpenGL 4.6 API, but the version reported by 9glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) / 10glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used. 11Some drivers don't support all the features required in OpenGL 4.6. OpenGL 124.6 is **only** available if requested at context creation. 13Compatibility contexts may report a lower version depending on each driver. 14 15Mesa 21.0.0 implements the Vulkan 1.2 API, but the version reported by 16the apiVersion property of the VkPhysicalDeviceProperties struct 17depends on the particular driver being used. 18 19SHA256 checksum 20--------------- 21 22:: 23 24 e6204e98e6a8d77cf9dc5d34f99dd8e3ef7144f3601c808ca0dd26ba522e0d84 mesa-21.0.0.tar.xz 25 26 27New features 28------------ 29 30- GL_EXT_demote_to_helper_invocation on radeonsi 31 32- GL_NV_compute_shader_derivatives on radeonsi 33 34- EGL_MESA_platform_xcb 35 36- Removed GL_NV_point_sprite for classic swrast. 37 38- driconf: remove glx_disable_oml_sync_control, glx_disable_sgi_video_sync, and glx_disable_ext_buffer_age 39 40- Removed support for loading DRI drivers older than Mesa 8.0, including all DRI1 support 41 42- Add support for VK_VALVE_mutable_descriptor_type on RADV 43 44- Removed classic OSMesa in favor of the newly improved gallium OSMesa 45 46- VK_KHR_fragment_shading_rate on RADV (RDNA2 only) 47 48- Freedreno a6xx exposes GL 3.3 49 50- Classic swrast dri driver removed in favor of gallium swrast (llvmpipe or softpipe) 51 52- Panfrost g31/g52/g72 exposes ES 3.0 53 54- Panfrost t760+ exposes GL 3.1 (including on Bifrost) 55 56- Sparse memory support on RADV 57 58- Rapid packed math (16bit-vectorization) on RADV 59 60- None 61 62 63Bug fixes 64--------- 65 66- R8 texture upload / corruption bug on Radeon RX 5700 XT 67- Ambient Occlusion in Two Point Hospital shows black spot artifacts 68- DXVK is broken in latest master 69- mesa/st: Uniforms are not updated after lowering alpha test 70- Regression: Segfault in cso_destroy_context() regression in 20.2 71- \[RADV\] Nioh 2 - The Complete Edition: "Bloom" on lights 72- \[RADV][BISECTED\] The Surge 2 (644830) - In-game assets do not render correctly since 20.3.4. 73- \[iris][icl,tgl][bisected][regression\] failure on piglit.spec.arb_separate_shader_objects.programuniform coverage 74- "radeonsi: Check pitch and offset for validity." is a bad commit 75- RADV: robustBufferAccessUpdateAfterBind is not exposed 76- \[RADV/DXVK\] Shadow artifacts with different games 77- glxgears segfaults with classic i915 78- ANV: Weird jitter in Witcher 1 79- ANV: Weird jitter in Witcher 1 80- ANV: Weird jitter in Witcher 1 81- meson: meson-built libraries have inconsistent compatability / current versions compared to older autotools-built libraries 82- RADV: Extreme overhead in vkQueueSubmit 83- timespec_get used unconditionally / build fails when targeting macOS 10.14 or earlier 84- Graphical glitch of popupping missing texture on Mesa version \>18.0.5 (Padoka Stable + Unstable/Oibaf/ubuntu-x-swat PPAs) 85- occasional corruption issue with RADV in multiple games, disappears after using amdvlk 86- device select layer breaks other layers 87- OpenGL on GMA4500MHD 88- Rage 2: Visual corruption on in-game menu with ACO. 89- GLonD3D12: Crashes and suboptimal fallback 90- GLonD3D12: Crashes and suboptimal fallback 91- GLonD3D12: Crashes and suboptimal fallback 92- \[RADV][REGRESSION][BISECTED\] radv_GetMemoryFdPropertiesKHR returns no valid memory types for vaapi drmbuf 93- anv: vkQueueSubmit with waitSemaphore value of 0 hangs CPU 94- ttn: invalid base/range triggering nir_validate assertion 95- \[RADV][ACO\] Overwatch game crash: amd/compiler/aco_insert_exec_mask.cpp: Failed Assertion 96- Use out encoding for float immediates 97- \[RADV\] Severe performance drop when exceeding VRAM compared to AMDVLK 98- LIBGL_ALWAYS_SOFTWARE=1 picks zink over actual software rasterizers 99- RADV: Occlusion query hangs Big Navi GPU 100- "mesa: don't allocate matrices with malloc" cause eglCreateContext problem on android 7. 101- Metal Gear Solid V: The Phantom Pain: texture issues and vertex stretches 102- miscompiled compute shader loop on llvmpipe (and Iris) 103- Graphics glitches after upgrade to mesa 20.3 on Khadas VIM3 Pro (Mali G52 GPU) 104- glthread crash in \_mesa_glthread_upload 105- Iris driver causing graphics glitch in QEMU spice egl DMA-BUF 106- \[RADV/ACO\] Death Stranding cause a GPU hung (\*ERROR\* Waiting for fences timed out!) 107- \[TGL\] Elder Scrolls Online misrenders 108- \[ANV\] System hang with GRVK demos 109- Rendering artifacts in Barn Finders specifically on Radeon Vega 110- regression in !8152 111- \[bdw][icl][iris\] fails new test \`clearbuffer-depth-cs-probe\` 112- ci: new traces runner needs dashboard links in the job log and junit 113- zink: car model corruption with game TORCS 114- Windows: 32-bit build is broken hard 115- ANV: Not handling separate stencil layouts properly 116- \[Regression][Intel][OpenGL][Bisected\] Copying whole 2D array texture failed on latest driver 117- i915 regressions bisected to "vbo/dlist: use a shared index buffer" 118- radv: dEQP-VK.sparse_resources.\* failures on GFX9 119- radv: dEQP-VK.sparse_resources.\* failures on GFX9 120- Mesa 20.3.x crashes pidgin on AMD RX480 121- libunwind not located / used on macOS 122- Some games using FNA framework show blank screen 123- Intel Vulkan regression of angle_end2end_tests 124- Defer lavapipe warning to queue / command / swapchain buffer creation 125- aco_tests failure with clang build 126- BUG: After issues playing World of Warcraft with RADV 127- Texture views on blits ignore formats 128- mesa-git hangs weston 129- radv: Some MSAA tests fail when DCC is forced. 130- \[RADV/ACO/SIENNA_CICHLID\] Into the game Shadow of the Tomb Raider the flickering artifacts are present on brushes. 131- Memory leak - alloc_prim_store in vbo_save_NewList 132- radv/aco: "Failed to allocate registers" in AC:Valhalla 133- Enable "radeonsi_clamp_div_by_zero" to fix graphical bug in CSGO, "mesa_glthread" for performance 134- master fails to build with "ac_sqtt.h:139:15: error: expected parameter declarator" 135- Conditional rendering implementation conflicts with aux-state tracking 136- regression since !7720 137- regression after !8196 138- Use up to 4 images for IMMEDIATE flip 139- piglit gl-1.0-rendermode-feedback TGSI_FILE_NULL assert on Iris 140- Use LDC and constant buffer state for UBO loads. 141- DOOM crashes on startup with OpenGL on RX 6800 142- Regression with Minecraft/Optifine performance with all VRAM mapped 143- Space Engineers rendering regression after 5f79e4e6 which triggers incorrect optimizations from 053be9f0 144- star conflict crashes on iris, but loads fine on i965, on HD 5500 145- radv: blit/copy tests with A2B10G10R10 SNORM fail when DCC is forced on GFX9 146- freedreno: regression of gl-3.2-layered-rendering-gl-layer-render after e49748521ec9182e8d2eec823182cc463709123f 147- \`gl_FragColor' undeclared (AMDGPU) - tested stable Mesa 20.1 and latest git for 20.3 (Game/Wine/Proton) 148- Mafia III Demo: Artifacts around barrels 149- android: webview crashes after a2fb87eea6d4 150- anv: dEQP-VK.subgroups.ballot_broadcast.compute.subgroupbroadcast_i8vec3_requiredsubgroupsize32 fail 151- Mesa considers the framebuffer with mixed 3D and 2D array attachments to be incomplete. 152- Multiple buffer definitions bound to single OpDecorate::Binding break SPIR-V module. 153- Intel driver segfaults on SPIR-V with OpArrayLength 154- \[g33][bisected][regression\] multiple piglit failures 155- \[v3d][bisected][regression\] Piglit failures on gl-1.0-rendermode-feedback and select 156- Update Mesa CI CTS to latest version 157- Rendering artifacts in Enter The Gungeon on Both RX 590 and Radeon 7 158- No way to turn off "Device" and "Swapchain format" in Vulkan overlay 159- Frames count doesn't turn off in vulkan overlay with frame=0 160- \[bdw][iris][bisected][regression\] failing test on multiple test suites 161- osmesa classic: build failure with Meson and MinGW-W64 162- Crash and slowness in FreeCAD 163- ci: Missing needs: in radeonsi-stoney-\*? 164- Triangles appear from the center of the field on PES2021 with Mesa 20.2.x 165- \[gen9][iris][regression][bisected\] flaky piglit tests 166- \[Intel][OpenGL\] Fail to get correct value when sampling from a texture in depth formats. 167- MESA_VK_DEVICE_SELECT only parses 16-bit vendorID, but in Vulkan is uint32_t 168- lp_test_format test fail on 32-bit mingw builds 169- RADV: Strange clear behavior with multisample arrays 170- Mesa 20.3.0 and older ATi/Radeon cards fails 171- Android building error after commit f08d8c849e 172- OSMesa SEGV in OSMesaGetDepthBuffer 173- osmesa gallium state tracker: Leak of screens and buffers on exit/shared library unload 174- Gallium OSMesa driver is far from being thread-safe 175- OSMesa UAF in OSMesaDestroyContext 176- OSMesaGetDepthBuffer flipped vertically 177- radv,aco: CTS image robustness tests fail to compile 178- 32-bit mesa failing to build inside a chroot due to f88347cd 179- Storing pointer to temporary value inside the Iris driver. 180- \[radeonsi\] DESPERADOS III poor performance when there's lots of animations going on 181- ci: arm64_test build broken (likely by ci-templates bump) 182- New build option to specify default value for shader disk cache size 183- commit f86668f487b32c185388a39e2200c17c298b877a fatal error: util/macros.h: No such file or directory 184- zink: ubo loading problems 185- !7138 broke the D3D12 driver 186- \[icl,tgl][iris][i965][regression][bisected\] piglit failures 187- 15% perf drop in GfxBench Manhattan 3.1 performance 188- \[Intel][OpenGL\] Fail to get correct stencil data from the stencil attachment with glReadPixels() 189- shader-db valgrind error 190- \[AMDGPU NAVI 5700xt\] Large parts of the Blender viewport does not render correctly if an object with hair is moved. 191- \[aco\] problem compiling compute pipeline 192- build failures after simple_mtx helgrind annotations 193- teach helgrind about simple_mtx 194- zink: regression after !7606 195- Chromium browser with VA-API video acceleration got corruption 196- glcpp test 084-unbalanced-parentheses fails with bison 3.6.y 197- \[Intel][OpenGL\] glDepthFunc(GL_EQUAL) doesn't work correctly on Intel Linux Mesa OpenGL drivers 198- d3d12: GPU based validation issue on fbo-clear-formats piglit 199- \[tgl,icl,gen9][bisected\] crucible/vulkancts failures on multiple platforms 200- zink+radv: corruption on pre-game menu in quake3 201- Memory leak in minecraft (many dri/renderD128 regions in /proc/[id]/maps) 202- freedreno: Use nir_opt_large_constants 203- android: amd/common: building error after 0833dd7d1 204- panfrost massive glitches apitrace opengl 2.1 205- freedreno/nir: nir_validate failure after nir_lower_tex 206- \[i965,iris][bisected\] piglit and glcts failures on multiple platforms 207- \[i965,iris][bisected\] piglit and glcts failures on multiple platforms 208- db410c ethernet no longer working 209- Add KHR_display extension to v3dv 210- \[radeonsi\] After 549ae5f84375dfadb86cfd465f0103acfae3249f commit Firefox Nightly Asan begins crashes 211 212 213Changes 214------- 215 216Adam Jackson (36): 217 218- docs: Update Mesa GL enum allocations for EGL_MESA_platform_xcb 219- glx, egl: Add LIBGL_DRI2_DISABLE environment variable 220- glx: Eliminate some stub functions for !GLX_DIRECT_RENDERING 221- glx: Remove unused \__GLXDRIscreen::createContext 222- glx: Check share ctx compatibility in ::create_context_attribs 223- glx: Handle create_context in terms of create_context_attribs 224- glx: Remove DRI1 225- glx: Simplify error handling in glXImportContextEXT 226- glx: Fix the generated error when indirect contexts are not supported 227- glx/indirect: Validate the context version in CreateContextAttribs 228- glx: Claim to support more GL versions in \__glX_send_client_info 229- meson: Make the glvnd vendor name configurable 230- zink: factor out GET_PROC_ADDR and friends to zink_screen.h 231- mesa: Remove silly "dummy_false" extension support 232- zink: Fix indentation in zink_create_instance 233- zink: Factor out winsys awareness from zink_internal_create_screen 234- zink: Factor out zink_get_loader_version() 235- zink: Factor out zink_create_logical_device 236- zink: Simplify MoltenVK support a bit 237- glx/xlib: Build fix 238- swrast: Remove the classic swrast DRI driver 239- treewide: Disambiguate various variables named "debug_options" 240- mesa: Cosmetic cleanups to GL_EXT_texture_sRGB_R8 241- mesa: Implement GL_EXT_texture_sRGB_RG8 for softpipe and llvmpipe 242- zink: Enable GL_EXT_texture_sRGB_R8 243- zink: Enable GL_EXT_texture_sRGB_RG8 244- virgl: Enable GL_EXT_texture_sRGB_RG8 245- drisw: Use debug_screen_wrap like everybody else 246- tests: Fix memory leaks in DispatchSanity 247- mesa: Fix array-format-to-format table on big-endian 248- mesa: Don't make building tests conditional on building DRI drivers 249- nouveau: pacify gcc on ILP32 250- zink: Fix VK_FORMAT_A8B8G8R8_SRGB_PACK32 mapping on big-endian 251- ci: Add a few more drivers to the cross builds 252- osmesa: Pacify MSVC in the test code 253- zink: Fix a thinko in instance setup 254 255Alejandro Piñeiro (12): 256 257- nir/lower_tex: clarify nir_lower_tex_options indexing 258- v3dv: cleanup/remove support for pre-generated variants 259- broadcom/compiler: separate texture/sampler info from v3d_key 260- v3dv: remove combined_idx support 261- v3dv/pipeline: take into account precision for the output_type 262- v3dv: use the common base object type and struct 263- v3dv: implement VK_EXT_private_data 264- turnip: minor tu_queue fixes related to vk_base_object 265- v3dv/cmd_buffer: missing (uint8_t \*) casting when calling memcmp 266- docs/features: update list of v3dv supported features 267- v3dv: remove non-conformant warning 268- v3dv/pipeline: avoid unused warning on release build 269 270Alexander Kanavin (1): 271 272- anv: fix a build race between generating a header and using it 273 274Alexander von Gluck IV (2): 275 276- meson: Add \_GNU_SOURCE for Haiku to activate non-posix functions 277- glsl/builtin_functions: Rename int64 function to int64_avail 278 279Alistair Popple (2): 280 281- gv100/ir: Make emitATOM consistent with emitRED 282- gv100/ir: Use system wide atomics 283 284Alyssa Rosenzweig (170): 285 286- pan/bi: Model writemasks correctly 287- panfrost: Implement linear Z/S for SFBD 288- panfrost: Remove panfrost_can_linear 289- panfrost: Fix out-of-bounds read on SFBD 290- panfrost: Add PAN_GPU_ID debug option 291- panfrost: Enable indirect uniform indexing 292- pan/mdg: Fix shader-db counter 293- pan/bi: Implement sampler1D 294- pan/bi: Fix varying writemask handling 295- pan/bi: Fix off-by-one in RA 296- pan/bi: Ensure TEXC src0 is not marked SSA 297- pan/bi: Implement shader-db stats 298- panfrost: Account for sample count in tib offsets 299- panfrost: Fix RAW8/16/32 component replication 300- docs: Add a stub page for Panfrost 301- docs/panfrost: Fix comment about Lima 302- docs: Update Panfrost in the source tree 303- docs/systems: Update Panfrost link 304- docs/panfrost: Document building Panfrost 305- docs/panfrost: Mention the IRC channel 306- pan/bi: Allow toggling disassembly verbosity 307- pan/bi: Space out disassembly 308- pan/bi: Remove all-0's termination condition 309- pan/bi: Minor styling cleanup in disasm 310- panfrost: Fix LOD mode field on Bifrost 311- pan/bi: Drop on-board packing tests 312- pan/bi: Label shader-db shaders 313- pan/bi: Remove bi_is_live_after 314- pan/bi: Add unused instruction mechanism 315- pan/bi: Add pseudo-instruction mechanism 316- pan/bi: Mark some instructions as unused 317- pan/bi: Defer newline printing in disassembler 318- pan/bi: Use consistent negX/absX naming 319- pan/bi: Use consistent wls naming 320- pan/bi: Use consistent naming of lane/lane0 321- pan/bi: Don't treat extend as per-source 322- pan/bi: Use canonical names for clamps 323- pan/bi: Use canonical names for rounding modes 324- pan/bi: Use canonical varying names 325- pan/bi: Use canonical sample names 326- pan/bi: Use canonical update modes 327- pan/bi: Use canonical min/max semantics 328- pan/bi: Use canonical name for segments 329- pan/bi: Use canonical lane ops 330- pan/bi: Use canonical subgroup size 331- pan/bi: Use canonical inactive result 332- pan/bi: Use consistent neg naming 333- pan/bi: Mark message types in ISA.xml 334- pan/bi: Fix rounding name for HADD in XML 335- pan/bi: Add staging register counts to ISA.xml 336- pan/bi: Add pseudo register formats to XML 337- pan/bi: Rename isa_parse to bifrost_isa 338- pan/bi: Add explicit meson dependency on the ISA helpers 339- pan/bi: Move copyright notice to common code 340- pan/bi: Add helpers for manipulating the ISA 341- pan/bi: Remove reference to 64-bit RA 342- pan/bi: Move modifier prints out of common code 343- pan/bi: Generate bi_opcodes.h 344- pan/bi: Use autogenerated modifiers 345- pan/bi: Generate bi_opcodes.c 346- pan/bi: Merge BIR_INDEX_FAU and BIR_INDEX_BLEND 347- pan/bi: Remove BIR_INDEX_UNIFORM 348- pan/bi: Make BIR_INDEX_ZERO less special 349- pan/bi: Add bi_swizzle enum 350- pan/bi: Add bi_index data structure 351- pan/bi: Add bi_index constructors 352- pan/bi: Add nullity/equality helpers for bi_index 353- pan/bi: Add helper to extract a word from an index 354- pan/bi: Add bi_temp{_reg} for new-style bi_index 355- pan/bi: Add helpers to generate bi_index from NIR 356- pan/bi: Add a helper to convert to old-style nodes 357- pan/bi: Add node_to_index helper 358- pan/bi: Add bi_half and bi_byte selectors 359- pan/bi: Add imm_f32 helper 360- pan/bi: Add bi_imm_u{8, 16} helpers 361- pan/bi: Add bi_{abs, neg} helpers 362- pan/bi: Add new bi_instr data structure 363- pan/bi: Add cursor data structures 364- pan/bi: Add builder data structure 365- ci/panfrost: Skip test with 4096 byte shader 366- pan/bi: Ensure fneg of a constant isn't reached 367- pan/bi: Rename bi_pack_{fma, add} to free up symbols 368- pan/bi: Rename bi_load 369- pan/bi: Add bi_not alias of bi_neg 370- pan/bi: Generate instruction printer 371- pan/bi: Generate builder routines 372- pan/bi: Generate instruction packer for new IR 373- pan/bi: Add bi_count_staging_registers helper 374- pan/bi: Add new style read/writemask helpers 375- pan/bi: Add builder initialization helper 376- pan/bi: Add bi_is_intr_immediate helper 377- pan/bi: Add bi_make_vec_to helper 378- pan/bi: Implement bi_emit_ld_tile via the builder 379- pan/bi: Implement bi_load_sysval via the builder 380- pan/bi: Implement bi_emit_load_const via the builder 381- pan/bi: Implement load_blend_input via the builder 382- pan/bi: Implement bi_reg_fmt_for_nir helper 383- pan/bi: Implement load_vary via the builder 384- pan/bi: Implement BLEND by builder 385- pan/bi: Implement fragment_out by builder 386- pan/bi: Implement store_vary with the builder 387- pan/bi: Implement load_ubo with the builder 388- pan/bi: Implement frag coord with the builder 389- pan/bi: Implement load attribute with the builder 390- pan/bi: Add intrinsic emits for builder 391- pan/bi: Add bi_alu_src_index helper 392- pan/bi: Add bi_nir_round helper 393- pan/bi: Add bi_cmpf_nir helper 394- pan/bi: Implement ALU with the builder 395- pan/bi: Implement jumps with the builder 396- pan/bi: Add TEXS emit with builder 397- pan/bi: Add builder-using helpers for TEXC structs 398- pan/bi: Emit TEXC with builder 399- pan/bi: Fix TEXS/TEXC check prototype 400- pan/bi: Add emit tex for builder 401- pan/bi: Add instruction emit for builder 402- pan/bi: Add bi_message_type_for_instr helper 403- pan/bi: Schedule new instructions singletons 404- pan/bi: Add bi_branch, bi_jump helpers 405- pan/bi: Stub FAU lowering pass 406- pan/bi: Switch to new IR 407- pan/bi: Remove combine lowering 408- pan/bi: Remove old IR packs 409- pan/bi: Remove packing helpers 410- pan/bi: Remove old IR prints 411- pan/bi: Remove old IR spill code 412- pan/bi: Remove old IR scheduling 413- pan/bi: Remove NIR->old IR 414- pan/bi: Remove old IR helpers 415- pan/bi: Remove old IR opcode table 416- pan/bi: Remove old IR instruction emit 417- pan/bi: Use new instruction types 418- pan/bi: Remove old IR 419- pan/mdg: Fix bound setting in RA for sources 420- panfrost: Import render condition check from fd 421- panfrost: Respect the render condition 422- docs: Document extensions exposing GL3.0 423- pan/bi: Fix TEXS register counts 424- pan/bi: Workaround BLEND precolour with explicit moves 425- pan/bi: Pull out bi_dontcare helper 426- pan/bi: Fix ATEST with pure integers 427- pan/bi: Don't suppress Inf/NaN 428- pan/bi: Allow passing thorugh 8-bit scalars 429- pan/bi: Implement scalar i2i8/u2u8 430- pan/bi: Use TEXC for indices \>= 8 431- pan/bi: Parametrize intrinsic immediate limits 432- pan/bi: Assert immediate indices fit 433- panfrost: Disable AFBC of 3D, 2D arrays 434- panfrost: Advertise ES3.0 on Bifrost 435- docs: Add release note for Bifrost GL3.1 436- docs/panfrost: Update GL/ES versions for v5+ 437- docs/features: Mark GL3.1 as done on Panfrost 438- docs/features: Fix missing close paranthesis 439- pan/bi: Implement TEXS for cube maps 440- panfrost: Handle explicit primitive restart 441- panfrost: Add alpha reference to XML 442- panfrost: Implement alpha testing natively 443- pan/bi: Fix assertion 444- pan/bi: Fix 64-bit SSBO addresses 445- pan/bi: Fix RA of node 0 446- pan/bi: Fix printing of node 0 447- pan/bi: Fix M1/M2 decoding in disassembler 448- pan/bi: Fix FLOG_TABLE modifier handling 449- pan/bi: Fix empty shader handling 450- panfrost: Add panfrost_sample_pattern helper 451- panfrost: Set tiler descriptor sampler pattern 452- pan/bi: Use explicit move even for RT#0 of MRT 453- panfrost: Raise TEXTURE_BUFFER_OFFSET_ALIGNMENT 454- panfrost: Don't advertise OES_copy_image 455- panfrost/lcra: Fix constraint counting 456 457Andres Gomez (23): 458 459- ci: update some radv trace checksums 460- ci: update some radv trace checksums 461- .mailmap: add and update aliases for Danylo Piliaiev 462- ci: Bump deqp to current vulkan-cts-1.2.5.0 also in the Lava jobs 463- ci: specify source and build directories with CMake 464- ci: use ephemeral packages when building the build-base image 465- ci: install ci-fairy in the testing images 466- ci: spread the usage of the FDO_UPSTREAM_REPO variable 467- ci: update piglit's version so it features replayer 468- ci: build piglit in the Vulkan testing image 469- ci: specify MinIO's host URL in a global variable 470- ci: add piglit replay jobs and remove tracie ones 471- ci: only modify LD_LIBRARY_PATH when running the piglit cmd 472- ci: add Vulkan piglit traces jobs and remove tracie ones 473- ci: move general build commands to their own section 474- ci: move API specification to driver instead of test suite 475- ci: build piglit inside baremetal and LAVA's rootfs 476- ci: add piglit jobs to LAVA and remove tracie ones 477- ci: refactor arm64 jobs in preparation for piglit addition 478- ci: add piglit job to baremetal and remove tracie ones 479- ci: remove all tracie remains 480- ci: recover tracie dashboard URLs for failing traces 481- ci: correct the trace image URLs in the piglit summary 482 483Andrii Simiklit (6): 484 485- glsl: avoid an out-of-bound access while setting up a location for variable 486- iris: update depth value for stages after fast clear depth 487- glx: lets compare drawing command sizes using MIN3 488- glx: fix spelling issues 489- st/mesa: don't affect original st_CompressedTexSubImage parameters 490- st/mesa: fix pbo upload/download for arrays of textures with only 1 layer 491 492Anuj Phogat (2): 493 494- intel/anv: Fix condition to set MipModeFilter for YUV surface 495- intel/anv: Fix condition for planar yuv surface 496 497Bas Nieuwenhuizen (57): 498 499- radv: Do the sample check for tiling earlier. 500- amd/addrlib: Use signed char for INT_8. 501- radeonsi: Add displayable DCC flushing without explicit flushes. 502- drm-uapi: Add AMD modifiers. 503- amd/common: Add support for modifiers. 504- amd/common: Add modifier tests. 505- radeonsi: Check pitch and offset for validity. 506- radeonsi: Add modifier support. 507- radeonsi: Do not disable DCC when we have it as a modifier. 508- radeonsi: Do not try to disable displayable DCC with modifiers. 509- radeonsi: Add auxiliary plane support. 510- drm/uapi: Fix modifier field mask for AMD modifiers. 511- radv: Use internal drm_fourcc.h 512- gallium/vl: Set modifier field for winsys handle. 513- radv: Dump BO VA ranges on hang. 514- radv: Fix RB+ blending for VK_FORMAT_E5B9G9R9_UFLOAT_PACK32. 515- radv: Fix a hang on CB change by adding flushes. 516- radv: Deal with unused attachments in mip flush 517- radv: Don't invalidate the SCACHE for image barriers. 518- radv: Don't skip layout transitions that only differ in render loop. 519- radv: Never allow fast clears on DCC images that are not compressed. 520- radv: Add option to disable DCC in renderpasses without layout. 521- radv: Disable DCC explicitly for incompatible copies. 522- radv: Enable DCC in the GENERAL layout on GFX10+. 523- radv: Use VRAM for upload buffers if entire VRAM is CPU-visible. 524- radv: Put commandbuffers in VRAM if all VRAM is CPU visible. 525- radv: Use VRAM for the initial gfx cmdbuffer. 526- ac/surf: Prepare for 64-bit flags. 527- ac/surf: Implement PRT layout. 528- ac/surf: Add sparse texture info to radeon_surf. 529- ac/surf: Use correct tilemodes on GFX8 for PRT. 530- radv/winsys: Fix inequality for sparse buffer remapping. 531- radv/winsys: Fix offset in range merging. 532- radv: Create sparse images. 533- radv: Add image sparse memory update implementation. 534- radv: Add sparse image queries. 535- radv: Enable sparse buffer and image support. 536- radv: Add Android module info to linker script. 537- radeonsi: Only set modifier creation function for GFX9+ & with kernel support. 538- radv: Remove redundant WB_L2 flush. 539- radv: Invalidate CB on SHADER_WRITE for meta operations. 540- radv: Do dst invalidations for write accesses. 541- radv: Use access helpers for flushing with meta operations. 542- radv: Use L2 for CP DMA on GFX9+. 543- radv: Use L2 coherency on GFX9+. 544- ac/surface: Fix GFX9 sparse mip info. 545- radv: Do not use a pipe offset for aliased sparse images. 546- radv: Use stricter HW resolve swizzle compat check. 547- radv: Do not hash vk_object_base in descriptor set layout. 548- radv: Improve spilling on discrete GPUs. 549- radv: Fix vram override with fully visible VRAM. 550- radv: Ignore WC flags for VRAM. 551- radv: Do pipe misalignment check per plane. 552- vulkan/device_select: Stop using device properties 2. 553- radv: Don't use dedicated memory info to indicate sharing. 554- radv: Expose robustBufferAccessUpdateAfterBind correctly. 555- frontends/va: Use correct size for secondary planes. 556 557BillKristiansen (1): 558 559- microsoft: add resource state manager utility code 560 561Boris Brezillon (119): 562 563- panfrost: Fix Bifrost blend descriptor emission 564- panfrost: Fix ->reads_frag_coord assignment 565- pan/bi: Extract shadowmap comparator 566- pan/bi: Force BLEND src0 to r0 567- panfrost: Fix panfrost_format_to_bifrost_blend() 568- panfrost: Get rid of the Pixel Format descriptor 569- pan/bi: Store the architecture in the compiler context 570- pan/bi: Expose FAU slots 571- pan/bi: Rename CLPER into CLPER_V7 and add CLPER_V6 572- pan/bi: Add support for the CLPER instructions 573- pan/bi: Add support for derivative instructions 574- pan/bi: Allow vec16 in bi_print_swizzle() 575- pan/bi: Allow lane selections on component 4 and above 576- pan/bi: Add support for tex offsets 577- pan/bi: Don't use TEXS for tex operations with a src that's not lod or coord 578- pan/bi: Support txs operations 579- pan/bi: Support automatic register format 580- pan/bi: Let the GPU pick the right format based on the varying descriptor 581- pan/bi: Set roundmode to RTZ for f2u operations 582- pan/bi: Move LD_VAR packing out of bi_pack_add() 583- pan/bi: Pass LD_VAR update mode explicitly 584- pan/bi: Stop passing special varying names through src0 585- pan/bi: Fix LD_VAR with non-constant index 586- pan/bi: Add a varying_index field to bi_texture 587- pan/bi: Stop extracting the immediate attribute index from src0 588- panfrost: Don't expose fp16 support on Bifrost unless explicitly requested 589- nir: Fix nextafter() for hardware that don't support denorms 590- compiler/spirv: Handle the LocalSizeHint execution modes 591- nir: Make nir_build_deref_offset() support ptr_as_array 592- pan/bi: Emit a combine even if we only pass one staging reg to TEXC 593- nir: Fix LOD source type for txf_ms instructions 594- panfrost: Stop forcing depth to nr_samples 595- panfrost: Get rid of the Sample Count enum 596- panfrost: Fix decoding of texture payloads 597- panfrost: Set depth for 3D textures on Bifrost 598- panfrost: Set sample_count when packing bifrost texture descriptors 599- pan/bi: Only update LOD mode on TEX operations 600- pan/bi: Always emit a LOD/CUBE word for FETCH instructions 601- pan/bi: LOD is a 8.8 fixed point 602- panfrost: Increase blit shader BO size on Bifrost 603- panfrost: Add a minus(1) modifier to the Levels field 604- panfrost: Clarify bit 2:28 meaning in the Midgard texture descriptor 605- panfrost: Add two helpers to calculate the surface pointer and strides 606- panfrost: Set the layer stride 607- panfrost: Unconditionally align strides on 64 bytes for linear resources 608- panfrost: Enable MSAA on bifrost when deqp debug option is set 609- panfrost: Expose panfrost_block_dim() 610- panfrost: Fix panfrost_needs_explicit_stride() for block-based formats 611- panfrost: Calculate the row stride at resource creation time 612- panfrost: Fix stride calculation for Z32_S8X24/X32_S8X24 formats 613- panfrost: Update the resource layout when doing a tile -\> linear conversion 614- panfrost: Update the resource layout before calling util_copy_rect() 615- panfrost: Fix texture payload decoding 616- panfrost: Fix draw descriptor definition 617- panfrost: Only set varyings and varying_buffers when varying_count \\> 0 618- panfrost: Make sure we always add a reader -\> write dependency when needed 619- panfrost: Fix fencing 620- pan/mdg: Add support for multi sample iteration writeout 621- panfrost: Take the number of samples into account in blend shaders 622- panfrost: Preload SampleID when reloading multisample FBs 623- panfrost: Fix provoking vertex selection for lines 624- pan/mdg: Fix texture handling for 2DMS arrays 625- panfrost: Allow 2DMS arrays 626- panfost: Fix depth/stencil writeback on Bifrost v7 627- panfrost: Force ->s_writeback_base to ->zs_writeback_base for Z24S8 buffers 628- panfrost: Reload depth/stencil when they are read 629- gallium/util: Fix depth/stencil blit shaders 630- panfrost: Fix several depth/stencil format mappings 631- pan/bi: Fix ATEST emission 632- panfrost: Move checksum_bo to panfrost_resource 633- panfrost: Group CRC fields in a struct 634- panfrost: Pass a device object to panfrost_new_texture() 635- panfrost: Merge emit_texture_payload() and emit_texture_payload_v7() 636- panfrost: Pass a dev object to panfrost_needs_explicit_stride() 637- panfrost: Define AFBC surface flags 638- panfrost: Adjust the compression tag creation for Bifrost 639- panfrost: Merge panfrost_new_texture() and panfrost_new_texture_bifrost() 640- panfrost: s/panfrost_slice.size0/panfrost_slice.surface_stride/ 641- panfrost: Use PAN_V6_SWIZZLE() in pan_blit.c 642- panfrost: Stop mixing depth and number of samples 643- panfrost: Add a pan_image_layout object 644- panfrost: Move AFBC header_size to a sub-struct 645- panfrost: Fix AFBC header_size and slice size calculation 646- panfrost: Add AFBC slice.body_size and slice.{row,surface}_stride fields 647- panfrost: Adjust surface stride calculation to take AFBC into account 648- panfrost: Add R5G6B5_UNORM entries to the format tables 649- panfrost: Pass a pipe-like swizzle to panfrost_new_texture() 650- panfrost: Adjust the format for AFBC textures on Bifrost v7 651- panfrost: Fix ZS block format v7 definition 652- panfrost: Use proper format for Z16_UNORM 653- panfrost: Fix AFBC support on Bifrost 654- panfrost: Enable AFBC support on Bifrost 655- panfrost: Use panfrost_get_layer_stride() instead of open-coding it 656- panfrost: Initialize AFBC headers to zero 657- panfrost: Fix panfrost_should_linear_convert() 658- panfrost: Allow AFBC on 2D arrays 659- panfrost: Fix calculation of body/header pointers for 3D AFBC 660- panfrost: Allow 3D AFBC on Bifrost v7 661- panfrost: Fix AFBC on Bifrost v6 662- panfrost: Fix UBO count calculation on Bifrost 663- pan/bi: Fix constant slot selection 664- panfrost: Set the RT index when emitting a Bifrost blend descriptor 665- pan/bi: Pass bundle pointers to bi_pack_tuple() 666- pan/bi: Port bi_collect_blend_ret_addr() to the new compiler infra 667- pan/bi: Restrict registers to r0-r15 when compiling blend shaders 668- pan/bi: Use the interference mechanism to describe blend shader reg use 669- pan/bi: Allow non-terminal BLEND operations 670- pan/bi: Lower 8bit fragment outputs to 16bit 671- panfrost: Promote 8b to 16b for blend descriptors 672- panfrost: Test GLES3 on Bifrost 673- panfrost: Get layer stride of level 0 on staging resources 674- panfrost: Pass the resource dimension to panfrost_compression_tag() 675- panfrost: Fix estimate_texture_payload_size() on Bifrost 676- panfrost: Re-enable AFBC on 3D, 2D arrays 677- panfrost: Skip an XFB test that's passing/failing randomly 678- panfrost: Fix panfrost_afbc_format_needs_fixup() 679- pan/bi: Fix the !immediate case in bi_emit_store_vary() 680- panfrost: Fix tiler job injection (again) 681- panfrost: Fix a polygon list corruption in the multi-context case 682 683Boyuan Zhang (2): 684 685- radeon: fix license in header 686- radeon/vcn: use cdw to calculate slice header index 687 688Brendan Dougherty (1): 689 690- mesa: Fix vertex_format_to_pipe_format index. 691 692Caio Marcelo de Oliveira Filho (13): 693 694- intel/fs: Add assert on the brw_STAGE_prog_data downcasts 695- intel/disasm: Don't rely on FALLTHROUGHTs to print unsupported SFID 696- anv: Avoid a couple of warnings related to vk_error macros 697- spirv: Implement OpArrayLength for OpenGL 698- nir: Fix outdated name in comment 699- nir: Remove unused parameter in remove_dead_var_writes 700- nir: Consider pointer initializers in nir_remove_dead_variables 701- spirv: Remove more dead variables 702- spirv2nir: Add --opengl (-g) argument for OpenGL SPIR-V 703- spirv: Don't remove variables used by resource indexing intrinsics 704- nir: Add a data pointer to the callback in nir_remove_dead_variables 705- compiler: Use util/bitset.h for system_values_read 706- spirv: Allow variable pointers pointing to an array of blocks 707 708Chad Versace (24): 709 710- anv/image: Check DISJOINT in vkGetPhysicalDeviceImageFormatProperties2 (v2) 711- anv/image: Fix isl_surf_usage_flags for stencil images 712- isl: Define isl_drm_modifier_get_score() \[v3\] 713- anv/image: Use isl_drm_modifier_get_score() 714- isl: Add isl_format_layout::uniform_channel_type 715- anv/image: Teach anv_get_image_format_features() about modifiers (v3) 716- anv/image: Fill drmFormatModifierTilingFeatures (v2) 717- isl: Make public the list of modifiers 718- anv/image: Refactor iteration over modifiers 719- anv/image: Delete the list of modifier-compatible formats 720- anv/image: Fix VkExternalMemoryProperties for images (v5) 721- anv/image: Rename get_wsi_format_modifier_properties_list() 722- anv/image: Minor refactor of VkImageFormatProperties::sampleCounts 723- anv/image: Fail earlier in anv_get_image_format_properties 724- anv/image: Respect VkImageFormatListCreateInfo for VkImageFormatProperties (v2) 725- anv/image: Drop redundant rejection of YCbCr formats with modifiers 726- anv/image: Emit error message for non-2D DRM images 727- anv/image: Move some DRM code in anv_get_image_format_properties() 728- anv/image: Add more asserts to choose_isl_tiling_flags 729- anv/image: Define add_all_surfaces() 730- anv/image: Further split add_*_surface funcs (v2) 731- anv/image: Rewrite check_surfaces() \[v2\] 732- anv/image: Check surface offsets after adding each surface 733- anv/image: Define anv_image_get_aux_addr (v3) 734 735Chia-I Wu (1): 736 737- virgl: fix modifier truncation 738 739Christian Gmeiner (37): 740 741- ci: sort packages installed via apt-get 742- etnaviv: nir: do not run opt loop after nir_lower_bool_xxx(..) 743- etnaviv: drop nir_print_shader(..) call 744- etnaviv/drm: fix evil-twin etna_drm_table_lock 745- etnaviv/drm: convert to simple_mtx 746- etnaviv/drm: add some locking asserts 747- etnaviv: update fallthrough comments 748- nir: change return type to void 749- etnaviv: rename from immedaite to uniform in some places 750- etnaviv: remove imm\_ prefix from etna_shader_uniform_info members 751- ci: build ARM mesa with X11 OpenGL support 752- ci: build mesa with gbm 753- ci/bare-metal: build full piglit for baremetal ARM targets. 754- ci/fastboot: exclude either deqp or piglit 755- ci/bare-metal: pass thorugh PIGLIT env vars 756- mesa/prog_to_nir: use intrinsic builders 757- tgsi_to_nir: use intrinsic builders 758- nir: use intrinsic builders 759- v3d: use intrinsic builders 760- v3dv: use intrinsic builders 761- ir3: use intrinsic builders 762- st: use intrinsic builders 763- zink: use intrinsic builders 764- tu: use intrinsic builders 765- d3d12: use intrinsic builders 766- iris: use intrinsic builders 767- vc4: use intrinsic builders 768- intel/blorp: use intrinsic builders 769- intel/compiler: use intrinsic builders 770- anv: use intrinsic builders 771- microsoft/compiler: use intrinsic builders 772- pan: use intrinsic builders 773- etnaviv: add set_stream_output_targets(..) stub 774- v3d: drop not use function parameter 775- v3d: update fallthrough comments 776- v3d: mark some variables static const 777- etnaviv: handle NULL views in set_sampler_views 778 779Connor Abbott (17): 780 781- freedreno/ci: Strip location from asserts 782- freedreno/a6xx: Document private memory registers 783- ir3: Expand cat6 a6xx opcode field 784- ir3: Add more a6xx-specific cat6 opcodes 785- ir3: Support assembling & disassembling getspid/getwid 786- ir3: Fix STP/LDP assembly 787- ir3/parser: Fix st{l,lw,g,p} and ld{l,lw,g,p} assembly 788- ir3: Initial support for private memory 789- ir3: Properly validate cat6 half-ness 790- freedreno: Add per-device parameters for private memory 791- tu: Support private memory 792- freedreno/a6xx: Implement private memory 793- ir3: Enable nir_lower_vars_to_scratch on a6xx 794- ir3/ra: Fix array reg liveness in scalar pass 795- ir3: Rename high registers to shared registers 796- ir3: Better rules for shared src copy propagation 797- ir3: Support MOVMSK 798 799Daniel Schürmann (53): 800 801- nir: add strength reduction pattern for imod/irem with pow2 divisor. 802- nir: allow for cheap intrinsics in nir_opt_peephole_select() 803- nir: add nir_phi_get_src_from_block() helper 804- nir/opt_peephole_select: collapse nested IFs if applicable 805- nir/opt_peephole_select: respect selection_control when collapsing ifs 806- nir: don't sink instructions into loops 807- nir/opt_sink: return early when trying to sink unused instructions 808- aco/ra: use get_reg_specified() for p_extract_vector 809- aco: don't create dead exec mask phis on merge blocks 810- aco: fix DCE of rematerializable phi operands 811- aco/spill: only prevent rematerializable vars from being DCE'd if they haven't been renamed 812- aco/ra: fix phi operand renaming 813- nir/opt_if: split ALU from Phi more aggressively 814- aco: don't emit parallelcopy when switching to WQM. 815- aco: make pred_by_exec_mask() accessible in other files 816- aco: allow to schedule SALU/SMEM through exec changes 817- aco: fix def-use distance calculation when scheduling. 818- aco: schedule position exports in the same pass as memory operations 819- aco: create VMEM clauses slightly more aggressive 820- nir/opt_vectorize: use a single instruction per hash entry instead of a vector 821- nir/opt_vectorize: don't hash instructions which are already vectorized 822- nir/opt_vectorize: don't hash filtered instructions 823- nir/opt_vectorize: rehash users of vectorized instructions 824- nir/opt_vectorize: hash whether a swizzle accesses elements beyond the maximum vectorization factor 825- nir/opt_vectorize: fix call to filter function 826- nir,vc4: Lower fneg to fmul(x, -1.0) 827- nir: replace .lower_sub with .has_fsub and .has_isub 828- nir/divergence_analysis: mark load_push_constant as uniform 829- radv: optimize idiv_const for small bitsizes 830- radv: call nir_opt_algebraic_late() after lowering idiv for small bitsizes 831- radv: don't lower_pack() after load-store-vectorization 832- radv: enable .lower_ineg 833- aco: simplify and fix operand/definition sizes 834- aco/ra: fix infinite recursion in get_reg_simple() with subdword registers 835- aco: fix VOP3P assembly, VN and validation 836- aco/RA: fix subdword operands on VOP3P instructions 837- aco: allow constants/literals on every src position for VOP3P 838- aco: allow SGPRs on every src position for VOP3P 839- aco: change usesModifiers() considering opsel_hi on packed instructions 840- aco: create helpers to emit vop3p instructions 841- aco: emit packed 16bit instructions 842- radv: vectorize 16bit instructions 843- aco: simplify multiply-add combining 844- aco: optimize packed mul+add to v_pk_fma_f16 845- aco: optimize packed clamp 846- aco: optimize packed fneg 847- aco: optimize v_pk_fma_f16 -\> v_pk_fmac_f16 on GFX10 848- aco: propagate swizzles when optimizing packed clamp & fma 849- aco: remove divergent branches which only jump over very few instructions 850- aco/optimizer: don't propagate subdword temps of different size 851- aco/optimizer: don't copy-prop logical phis 852- aco: fix nir_intrinsic_ballot with wave32 853- aco: fix shared VGPR allocation on RDNA2 854 855Daniel Stone (17): 856 857- microsoft/clc: Allow building with Clang git 858- microsoft/clc: Disable broken f32 -\> i64/u64 test 859- CI: Add Windows libclc and SPIRV-LLVM-Translator builds 860- CI: Windows: Use 32 vCPUs for Mesa build 861- CI: Remove ludicrous Windows container build timeout 862- CI: Update Windows build for current Meson options 863- CI: Build d3d12 Gallium driver and CLC framework on MSVC 864- CI: Re-enable MSVC build 865- freedreno: Add missing dependency to build 866- CI: Collapse SCons & meson-misc stages into one 867- CI: Collapse llvmpipe & softpipe stages into one 868- CI: Collapse radv & radeonsi stages into one 869- CI: Collapse virgl & d3d12 stages into one 870- CI: Collapse lima & panfrost stages into one 871- CI: Reorder non-hardware stages last 872- CI: Add llvmpipe- prefix to Piglit jobs 873- CI: Add Windows source dependency map 874 875Danylo Piliaiev (22): 876 877- freedreno/a6xx: add support for dual-source blending 878- freedreno/a6xx: Fix typo in height alignment calculation in a6xx layout 879- freedreno/a6xx: add support for ARB_shader_stencil_export 880- tu: Ignore pTessellationState if there is no tesselation shaders 881- tu: pCounterBuffers can be NULL in vkCmd*TransformFeedbackEXT() 882- freedreno/a6xx: Fix assert which checks the count of shader outputs 883- ir3: Allow tesselation to use all 32 varying slots 884- freedreno/a6xx: Fix SP_HS_UNKNOWN_A831 value and document it 885- freedreno/a6xx: bump varyings limit 886- freedreno: Fix FD_MESA_DEBUG=flush debug option 887- freedreno/ir3: remap FRAG_RESULT_COLOR to \_DATA\* for dual-src blending 888- nir/lower_fragcolor: handle dual source blending 889- freedreno/a6xx: fix array pitch for layer-first layouts 890- freedreno/a6xx: add support for gl_Layer in vertex shader 891- freedreno/a6xx: support layered framebuffers in blitter_clear 892- nir: account for point-coord origin when lowering it 893- nir: fix missing nir_lower_pntc_ytransform.c in the makefile 894- freedreno/a6xx: fix transform feedback resuming 895- freedreno/a5xx: implement transform feedback resuming 896- freedreno: Enable GLSL 3.30, updating us to GL 3.3 contexts 897- turnip: remove unused IR3_DP_LOCAL_GROUP_SIZE_* from cs params 898- turnip: implement indirect dispatch 899 900Dave Airlie (69): 901 902- util: add a env getter for versions 903- clover/device: store version in device at constructor. 904- clover: add CL 3.0 CL_DEVICE_NUMERIC_VERSION support 905- clover/platform: move versioning to core object. 906- clover: add CL_PLATFORM_NUMERIC_VERSION support 907- clover: report device CLC versions for 3.0 908- clover: add support for versioned device extensions 909- clover: add platform supported extensions with version 910- clover: add support for opencl C features 911- gallium: handle empty cbuf slots in framebuffer samples helper 912- u_blitter: port radv 3D blit coords logic. 913- lavapipe: enable alpha to one. 914- lavapipe: disable SNORM blending for now 915- llvmpipe: just use draw_regions in draw/line setup. 916- draw: fix tess eval pipeline statistics. 917- gallivm: add float to 8/16 int 918- gallivm/nir: add fsum support 919- gallivm/nir: lower dot products. 920- gallivm: lower vector compares 921- gallivm: fix float atomic exchange. 922- clover: handle memory object properties properly. 923- clover: add support command queue properties 924- clover: add all CL 3.0 API with invalid functions 925- clover: add cl 3.0 SVM invalid support 926- clover: add device/platform info for CL 3.0 927- clover: add 3.0 program properties 928- clover: add CL 3.0 event/queue queries 929- clover/image: handle MEM_KERNEL_READ_AND_WRITE flag. 930- spirv/cl: add enqueued workgroup size. 931- lavapipe: fixup device allocate + enable private data 932- lavapipe: fix wsi acquire fences 933- llvmpipe/setup: move point stats collection earlier. 934- llvmpipe: fix multisample point rendering. 935- llvmpipe: fix multisample lines. 936- lavapipe: fixup mipmap precsion bits 937- lavapipe: enable pipeline stats queries 938- gallium: fix missing bit field in p_state.h 939- zink: allow the backend to optimise shaders. 940- lavapipe: enable VK_EXT_shader_stencil_export 941- lavapipe: enable post depth coverage 942- lavapipe: add support for VK_KHR_indirect_draw_count 943- radeonsi: fix regression on gpus using the radeon winsys. 944- lavapipe: use ralloc for pipeline copies. 945- lavapipe: split out pipeline struct duplication to a macro. 946- lavapipe: don't copy pNext 947- CI: add lavapipe vulkan testing 948- lavapipe: refactor descriptor set binding to support push later. 949- lavapipe: add support for VK_KHR_push_descriptor 950- lavapipe: add support for VK_KHR_descriptor_update_template 951- zink: add some 64-bit conversion ALUs 952- gallium: add an api to retrieve pipe offsets 953- llvmpipe: add support for vulkan streamout offset hook 954- llvmpipe: handle SO statistics multi value query copy. (v2) 955- lavapipe: add transform feedback support 956- gallium: add grid base to dispatch info 957- llvmpipe: add support for grid base 958- llvmpipe: enable lower device id to zero 959- lavapipe: add basic vulkan device group support. 960- util: add printf specifier shared helper code. 961- clover/module: add a printf support to module (v5) 962- clover/nir: hookup printf (v3) 963- intel/isl: move get_tile dims/masks to common isl header 964- device-select-layer: update for vulkan 1.2 965- lavapipe: fix missing piece of VK_KHR_get_physical_device_properties2 966- radv: move queue object to a common base object 967- zink: don't pick a cpu device ever. 968- glsl: fix leak in gl_nir_link_uniform_blocks 969- glx: proposed fix for setSwapInterval 970- lavapipe: fix pipeline vp/scissor mixup. 971 972David McFarland (1): 973 974- radv: fix divide by zero with no tesselation params 975 976David Stevens (6): 977 978- egl/android: don't pass loaderPriv in get_front_bo 979- dri: add image cleanup callback to loader extensions 980- frontend/dri: plumb loader image cleanup callback 981- i965: plumb loader image cleanup callback 982- egl/android: implement image cleanup callback 983- egl/dri2: fix image loaderPrivate type mixup 984 985Duncan Hopkins (4): 986 987- zink: setup version dependent VkPhysicalDeviceVulkan*Features and VkPhysicalDeviceVulkan*Properties. 988- mesa: Undefine ALIGN macro before it is used as a function name. Issues on MacOS. 989- zink: moved vkEnumerateInstanceVersion to create_instance 990- zink. Fixing vkGetPhysicalDeviceProperties2 and vkGetPhysicalDeviceFeatures2 for Vk 1.1 and VK_KHR_get_physical_device_properties2. 991 992Dylan Baker (70): 993 994- Bump version for 21.0 devel 995- Reset new features for 21.0 development cycle 996- meson: Don't add extra values to shader-cache 997- meson: use a feature option for microsoft-clc 998- docs: add release notes for 20.2.3 999- docs: Add relnotes for 20.2.3 1000- docs: update calendar and link releases notes for 20.2.3 1001- release-calender: Update 20.3 1002- docs: add release notes for 20.3.0 1003- docs: Add sha256 sums for 20.3.0 1004- docs: update calendar and link releases notes for 20.3.0 1005- docs: add release schedule for 20.3 1006- docs: add release notes for 20.2.4 1007- relnotes: Add sha256sums for 20.2.4 1008- docs: update calendar and link releases notes for 20.2.4 1009- docs: add release notes for 20.2.5 1010- docs: add sha256 sums for 20.2.5 1011- docs: update calendar and link releases notes for 20.2.5 1012- docs: add release notes for 20.3.1 1013- docs: Add sha256 sums for 20.3.1 1014- docs: update calendar and link releases notes for 20.3.1 1015- docs: add release notes for 20.2.6 1016- docs: Add sha256 sums for 20.2.6 1017- docs: update calendar and link releases notes for 20.2.6 1018- docs: add release notes for 20.3.2 1019- docs: Add sha256 sum for 20.3.2 1020- docs: update calendar and link releases notes for 20.3.2 1021- pick-ui: don't handle the mouse 1022- bin/remove get-pick-list.sh files 1023- docs: store the release-calendar information in csv (and fix tests) 1024- bin: Add script for manipulating the release calendar 1025- bin/gen_calendar_entries: Add support for extending a release 1026- bin/gen_calendar_entries: Add support for making a release 1027- docs: Add calendar entries for 21.0 release candidates. 1028- docs/release-calendar.rsv: Remove spaces 1029- VERSION: bump for 21.0.0-rc1 1030- .pick_status.json: Update to dfe429eb414511170f3dfc960d247c4aa295f924 1031- .pick_status.json: Update to 184bbef33d1fff3520958c130f2b8e4fce17379c 1032- .pick_status.json: Update to c27347b2e1883a30e023347a36bdcf86cdec4a7c 1033- .pick_status.json: Update to 3e13c1f8dfef4a4c0fd5e79bbc364f9e5f998856 1034- VERSION: bump for 21.0.0-rc2 1035- .pick_status.json: Update to af9977a3d5f3378c297965e21389e36491f47e1b 1036- .pick_status.json: Update to c3dbc4df194a15aa1cf09493a3100b59e37e48fe 1037- .pick_status.json: Update to 64f55b82c7f1652e4fae478c0af325fc38b9b53b 1038- .pick_status.json: Update to 3ef89b245e3e1ac4e67fea9c1b13ebeda75769d0 1039- .pick_status.json: Update to d37124b065c2b6c99c042fb402c6a23ce16b034e 1040- .pick_status.json: Mark 8c7d9716669a74159d2eec86490c756c274f663c as backported 1041- .pick_status.json: Mark 45bebc7a9c73f3add08c2290fa1eac237edf5a34 as backported 1042- .pick_status.json: Update to 9052819ebbff07d82c3eb9adf414144df4868644 1043- .pick_status.json: Update to f01ea0aef8a50d2732eb0c64153903e52ed2a757 1044- VERSION: bump for 21.0.0-rc3 1045- .pick_status.json: Update to 86ff78e8fe55b424c6b853ead6979bcd46820d81 1046- .pick_status.json: Update to 9003735b9141fb156d3b2e1133b94cdf14f63424 1047- .pick_status.json: Update to e8707961134daa9b91599840ad5698366a6229b7 1048- .pick_status.json: Update to b609d4677d3f910c546c1d94d8ddfe4511e2f065 1049- bump version for 21.0-rc4 1050- .pick_status.json: Update to 8ed874d73fafcfbcb54730dc5c20e58f24d55f5e 1051- .pick_status.json: Update to 03d3294e35befc2be6ed0ed66ed92fab991c166d 1052- Revert "vulkan: Make vk_debug_report_callback derive from vk_object_base" 1053- VERSION: bump for 21.0.0-rc5 1054- .pick_status.json: Update to 4ded99f99ddbd1103ffddfd9935638fc12e0ecfd 1055- .pick_status.json: Mark 38ce8d4d00c2b0e567b6dd36876cf171acb1dbc7 as backported 1056- .pick_status.json: Update to 9f8a0b797ed9b8ad9bf49af8269a337b1152a744 1057- .pick_status.json: Update to 6ceb6b509e64c54812a5f6a208e7d93cc61119f4 1058- .pick_status.json: Update to ea27f2bf092f462171fe14a44619565d14f43fb8 1059- .pick_status.json: Update to c22267262ee1b6817df368a51168fa82bd17293c 1060- .pick_status.json: Mark 04df0cb4ae7055b0a4a6dc9875aa5926131fe5f4 as backported 1061- .pick_status.json: Mark 942ba4e34124d1058492f544dc8fd42f4012fd12 as backported 1062- .pick_status.json: Mark ea27f2bf092f462171fe14a44619565d14f43fb8 as backported 1063- .pick_status.json: Mark 5f1b3544729178715a1ed0714bd1029737089824 as backported 1064 1065Ella-0 (1): 1066 1067- v3dv: Wayland WSI support 1068 1069Eric Anholt (156): 1070 1071- util/hash_table: Handle NULL ht in \_mesa_hash_table_clear(). 1072- util/hash_table: Clean up the \_mesa_hash_table_clear() implementation. 1073- util/set: Fix the \_mesa_set_clear function to not leave tombstones. 1074- nir/validate: Size the set of blocks to avoid rehashing. 1075- nir_builder: Return a new builder from nir_builder_init_simple_shader(). 1076- nir/builder_tests: Drop unused lin_ctx. 1077- nir/tests: Simplify the mem_ctx setup in our unit tests. 1078- intel: Drop the last uses of a mem_ctx in nir_builder_init_simple_shader(). 1079- nir/builder: Drop the mem_ctx arg from nir_builder_init_simple_shader(). 1080- nir/builder: Add a name format arg to nir_builder_init_simple_shader(). 1081- ci: Move the rust cleanup in lava_build out of the middle of kernel build. 1082- ci: Only install kernel modules for LAVA devices. 1083- ci/freedreno: Group the short a630 dEQP runs into one test job. 1084- ci/deqp: Allow specifying the caselist fraction separate from CI_NODE_INDEX. 1085- ci: Bump deqp to current vulkan-cts-1.2.4 1086- ci: Re-enable the clip_three test on non-freedreno ARMs. 1087- ci/db410c: Fix networking so we get artifacts from our jobs. 1088- gallium/draw: Fix rasterizer_discard for wide points/lines. 1089- freedreno: Fix leak of shader binary on disk cache hits. 1090- nir: Add a size_align helper function for aligning elements to 16 bytes. 1091- freedreno/ir3: Include at least 4 NOPs so that cffdump doesn't disasm junk. 1092- freedreno/ir3: Switch emit_const_ptrs() to take BOs instead of prscs. 1093- freedreno/ir3: Fix incorrect optimization of usage of 16-bit constbuf vals. 1094- freedreno+turnip: Upload large shader constants as a UBO. 1095- freedreno: Disable PIPE_CAP_PREFER_IMM_ARRAYS_AS_CONSTBUF. 1096- turnip: Assert about the storage buffer offset alignment. 1097- ci: Enable -Werror in more clover builds. 1098- freedreno: Fix release build warnings for asserted temp vars. 1099- freedreno/a6xx: Fix use of uninitialized img->level in the SSBO/image path. 1100- freedreno: Fix warning about uninit size for the size==0 special case. 1101- freedreno: Fix uninitialized var warning in afuc using unreachable(). 1102- freedreno: Suppress uninit var warnings from shader stage switch. 1103- ci: Bring freedreno into the "warnings clean release build" fold. 1104- freedreno/afuc: Fix up some sprintf format security warnings. 1105- gallium: Fix leak of the merged driconf options. 1106- freedreno: Fix leak of u_transfer_helper. 1107- egl: Skip closing drivers when building with AddressSanitizer. 1108- meson: Remove old todo comment about pthread stubs. 1109- gallium: Fix leak of bound SSBOs at CSO context destruction. 1110- gallivm: Fix max const buffer count. 1111- gallium: Fix leak of currently bound UBOs at CSO context destruction. 1112- freedreno: Break out of "should we free the entry" loop once we've freed. 1113- xmlconfig: Add unit tests for recent bugs in the driconf rewrite. 1114- xmlconfig: Warn if parsing the engine/app versions fails. 1115- gallium/osmesa: Fix flushing and Y-flipping of the depth buffer. 1116- gallium/osmesa: Remove the broken buffer-reuse scheme. 1117- gallium/osmesa: Fix data race on setting up the ST API. 1118- gallium/osmesa: Fix leak of the ST manager/api on library unload. 1119- gallium/osmesa: Return cleanly for OSMesaGetDepthBuffer() with no depth. 1120- ci/freedreno: Detect the cheza power management bus error and restart. 1121- ci/vc4: Skip VS dynamic loops tests that cause GPU hangs. 1122- softpipe: Fix swizzled texture gather of int textures. 1123- osmesa/test: Clear the stencil bits in the depth test. 1124- docs: Fix the documentation of the OSMesa path. 1125- mesa: Retire classic OSMesa. 1126- ci: Make sure that osmesa stays warnings-clean in release builds. 1127- st/mesa: Replace mesa_to_tgsi() with prog_to_nir() and nir_to_tgsi(). 1128- gallium/ntt: Don't manually reindex instrs. 1129- gallium/ntt: Drop reindexing of SSA defs and regs. 1130- nir: Redefine start/end_ip of blocks to fix NIR-to-TGSI liveness bugs. 1131- etnaviv, v3d: Fix valgrind include paths. 1132- util: Fix memory leak in a hash table unit test. 1133- util/vma: Fix leak of the heap in the unit test. 1134- glx/tests: Remove unused teardown function. 1135- glx/tests: Fix leaks in the unit tests. 1136- freedreno/ir3: Free the compiler at the end of the unit tests. 1137- disk_cache: Fix memory leaks in the unit test. 1138- glsl/general_ir_test: Fix leaks. 1139- glsl/uniform_initializer_tests: Fix memory leak 1140- mapi: Fix symbols check with ASan enabled. 1141- glsl/standalone: Fix memory leaks 1142- driconf: Fix memory leak in the unit test. 1143- amd: Fix leak in ac_surface_modifier_test. 1144- ci: Add an ASan build on x86. 1145- ci/freedreno: Treat all freedreno deqp runs as saving results. 1146- ci/freedreno: Stop specifying the number of deqp threads 1147- mesa/st: Finalize the texture before BlitFramebuffer from it. 1148- freedreno/a6xx: Flush depth at the end of bypass rendering, too. 1149- ci/deqp: Make sure that we pull in all board-specific xfail/skip/flake files. 1150- lvp: Fix vtn warnings about unsupported image read/write without format. 1151- softpipe: count CS invocations for pipeline stats queries. 1152- mesa/st: Fix use-after-free of the draw VS. 1153- ci: Disable the now flaky Portals.trace on a630. 1154- ci/deqp: Move .shader_cache artifacts exclusion to the yml. 1155- ci/deqp: Upgrade the runner, enable junit output. 1156- ci/deqp: Move the load reporting to a quiet block. 1157- mesa/st: Update FP state when textures change with an ATI_fs bound. 1158- mesa/prog_to_nir: Factor out the texture-target-to-sampler-dim helper. 1159- mesa/ati_fs: Clean up writemask handling. 1160- st/mesa: Generate NIR for ATI_fragment_shader instead of TGSI. 1161- gallivm: Use the proper enum for the texture target bitfield. 1162- softpipe: Enable GLSL 400 for compat contexts too. 1163- ci/piglit: Include the updated piglit results list in the job results. 1164- ci/softpipe: Include a piglit run. 1165- gallium/ntt: Fix check for "is there anything in the else block?" 1166- ci/deqp: Fix inverted meaning of DEQP_NO_SAVE_RESULTS. 1167- freedreno: Enable GLSL 1.50, updating us to GL 3.2 contexts. 1168- ci/panfrost: Disable the flaky gimark trace. 1169- gallium/draw: Fix intermittent failure to bind new geometry shaders. 1170- ci/softpipe: Re-enable GS tests that had been banned for being flaky. 1171- gallium/tgsi_exec: Fix shared memory atomic ops. 1172- gallium/tgsi_exec: Reuse the atomic helper for SSBO atomics. 1173- gallium/tgsi_exec: Use the new SSBO lookup interface for SSBO loads. 1174- gallium/tgsi_exec: Move the SSBO store path to tgsi_exec, too. 1175- gallium/tgsi_exec: Replace the SSBO RESQ-specific interface with lookup. 1176- softpipe: Sanity check that the SSBO view offset is within the BO. 1177- ci/softpipe: Skip flaky triangle-rasterization-overdraw. 1178- ci/softpipe: Ban glx-multithread-texture, too. 1179- ci/softpipe: Update the comment about the rasterpos flake. 1180- ci/bare-metal: Drop extra DEQP_PARALLEL settings. 1181- ci/bare-metal: Pass through FDO_CI_CONCURRENT on bare-metal runners. 1182- ci: Add a530 and a630 piglit runs. 1183- gallium/tgsi_exec: Simplify GS output vertex count tracking. 1184- gallium/tgsi_exec: Stop doing the weird allocation of the Addrs array. 1185- gallium/tgsi_exec: Drop the unused scratch temp regs. 1186- gallium/tgsi_exec: Clean up storage of the pixel kill mask. 1187- gallium/tgsi_exec: Remove unused MaxGeometryShaderOutputs. 1188- freedreno/ir3: Deduplicate link_stream_out. 1189- freedreno/a5xx: Drop redundant stream output linking check. 1190- freedreno/a5xx: Move link_stream_out after VPC_VAR_DISABLE like on a6xx. 1191- gallium/tgsi_exec: Fix assertion failure about missing constbufs. 1192- gallium/tgsi_exec: Refactor to fix CS local memory overflow checks. 1193- gallium/tgsi_exec: Add support for PIPE_CAP_LOAD_CONSTBUF. 1194- gallium/ntt: Fix emitting UBO declarations. 1195- gallium/ntt: Fix dynamic indirect indexing of per_vertex_input. 1196- gallium/ntt: Fix load_ubo_vec4 buffer index setup. 1197- gallium/ntt: Add support for PIPE_CAP_LOAD_CONSTBUF. 1198- turnip: Move the limited_z24s8 flag to the shared device info. 1199- freedreno/a6xx: Move the IBO pipe2tex down to where it's used. 1200- freedreno/a6xx: Fix z24s8 non-ubwc blits on a630. 1201- freedreno: Disable UBWC on z24s8 on a630. 1202- freedreno: Mark a615/a618 as also lacking Z24_UINT_S8_UINT support. 1203- freedreno: Add missing dep on u_tracepoints. 1204- ci: Disable the freedreno farm, which went down last night. 1205- gallium/ntt: Drop XXX comment about supporting carry opcodes. 1206- gallium/ntt: Emit SSBO buffer declarations. 1207- gallium/ntt: Emit sample index when necessary for image load/store. 1208- gallium/ntt: Add support for emitting TXF_LZ. 1209- gallium/ntt: Drop comment about needing loop label setup. 1210- gallium/ntt: Drop comment about needing array_id for svga tess. 1211- gallium/ntt: Work around virglrenderer UIF handling bug. 1212- nir/lower_locals_to_regs: Use the imul_imm helper instead of forcing it. 1213- gallium/ntt: Fix leak of the per-instr liveness information. 1214- mesa/st: Free the NIR builtins TGSI tokens after passing to the driver. 1215- mesa/st: Free the ARB_vp/fp nir-to-tgsi temporary tokens. 1216- gallium/ntt: Take ownership of the NIR shader we're passed. 1217- Revert "ci: Disable the freedreno farm, which went down last night." 1218- util/format: Fix pack/unpack of A1R5G5B5_UINT. 1219- swr: Don't report support for shader images. 1220- panfrost: Stub out set_shader_images(). 1221- gallium: Fix leak of shader images on context destruction. 1222- mesa/st: Allocate the gl_context with 16-byte alignment. 1223- vc4: Remove vestiges of alpha test lowering. 1224- v3d: Clean up vestiges of alpha test lowering. 1225- freedreno: Add missing dep on freedreno tracepoints. 1226- r300,i915g: Report no shader buffers or images on non-TCL HW. 1227 1228Eric Engestrom (3): 1229 1230- gitlab-ci: drop deprecated platforms that snuck in when nobody was watching 1231- meson: drop deprecated EGL platform build options 1232- docs: use a single cell for the branch number 1233 1234Erico Nunes (6): 1235 1236- lima: define set_clip_state implementation 1237- mesa: allow half float textures based on ARB_half_float_pixel 1238- lima: add support for half float textures 1239- lima: adjust pp and gp max const buffer size 1240- nir/lower_vec_to_movs: don't vectorize unsupports ops 1241- lima: fix max sampler views 1242 1243Erik Faye-Lund (133): 1244 1245- softpipe: correct signature of get_compiler_options 1246- util/slab: allow usage from c++ code 1247- compiler: add SYSTEM_BIT_FRONT_FACE 1248- microsoft/compiler: add dxil-util code 1249- microsoft/compiler: translate nir to dxil 1250- d3d12: introduce d3d12 gallium driver 1251- d3d12: ensure all compoents of clip-distances are written 1252- d3d12: avoid searching twice for bos 1253- util/u_process: implement util_get_process_name for Windows 1254- d3d12: fix code after simple-shader helper changes 1255- microsoft/compiler: remove unused struct 1256- microsoft/compiler: move c++ higher up 1257- microsoft/compiler: inline some struct-declarations 1258- microsoft/compiler: correct typo 1259- meson: verify that d3d12.h exists when building the d3d12 driver 1260- util: fix unknown pragma warning on msvc 1261- mesa/main: add missing include in glformats.h 1262- docs/features: document d3d12 features 1263- zink: mark general layout as transfer-read/write 1264- zink: always insert barriers for general-layout 1265- zink: more accurately track supported blits 1266- mesa/st: Introduce WINSYS_HANDLE_TYPE_D3D12_RES 1267- d3d12: Support WINSYS_HANDLE_TYPE_D3D12_RES 1268- d3d12: also reject GDI-supporting pixel-formats 1269- llvmpipe: fix arith-test build on msvc 1270- d3d12: transition the right planes 1271- docs: add basic docs for d3d12 driver 1272- zink: fix layered resolves 1273- zink: fall back to util_blitter for scaled resolves 1274- Revert "zink: update shader modules in gfx program when flagged dirty" 1275- Revert "zink: put those shader keys to work fixing up fragment shaders" 1276- Revert "zink: fill in params for fs shader keys and flag shader for rebuild" 1277- Revert "zink: move shader key structs into their own header" 1278- Revert "zink: refcount the shader cache" 1279- Revert "zink: initial implementation of shader keys" 1280- Revert "tgsi: Fix helgrind complaint about one-time init" 1281- Revert "gallium/trace: Fix helgrind complaint about one-time init" 1282- Revert "mesa: Fix helgrind complaint about one-time init" 1283- Revert "util: Fix helgrind complaint about one-time init" 1284- Revert "mesa/st: Use do_once for one-time init" 1285- Revert "gallium/hud: Use do_once for one-time init" 1286- Revert "freedreno/ir3: Use get_once() for one-time init" 1287- Revert "nir: Use get_once() helper for one-time init's" 1288- Revert "util: Add helpers for various one-time-init patters" 1289- docs: document new zink-flag 1290- d3d12: lower bitfield_extract to shifts 1291- d3d12: do not inspect NULL samplers 1292- util/slab: do not dereference NULL-pointer 1293- zink: revert to old load_ubo implementation 1294- docs: break project history out of front-page 1295- docs: move major versions history out of front-page 1296- docs: use external link-references 1297- docs: do not explicitly call out es-versions 1298- docs: mention egl in api-list 1299- docs: inline contents.rst into index.rst 1300- gitlab-ci: store build-artifacts from building mesa 1301- gitlab-ci: build zlib statically on windows 1302- gitlab-ci: build piglit in mesa_deps.ps1 1303- gitlab-ci: run piglit on windows 1304- gitlab-ci: ignore nv_copy_depth_to_color 1305- gitlab-ci: do not clone git-repo for test-job 1306- microsoft/clc: use files-function for source-list 1307- microsoft/clc: add missing dependency 1308- microsoft/clc: increase test-timeout 1309- zink: do not require VK_KHR_external_memory 1310- lavapipe: set some basic usage-flags 1311- gallium/targets/libgl-gdi: prefer d3d12 driver 1312- lavapipe: fix logic-op support 1313- gallium: do not reset buffers for unsupported stages 1314- zink: fix channel ordering in format-mapping 1315- lavapipe: interpret inputRate as an enum-value 1316- lavapipe: implement VK_EXT_vertex_attribute_divisor (v2) 1317- zink: fail if set failed to create 1318- zink: use \_mesa_pointer_set_create for simplicity 1319- gitlab-ci: copy piglit expected results to artifacts 1320- .gitlab-ci: verify that Get-Content worked 1321- mesa: do not allow es2-extension enums for es1 1322- mesa: check for extension instead of desktop GL 1323- gallium/util: make bitcast-helpers explicitly sized 1324- gallium/util: add bitcast helpers for double and uint 1325- zink: force display-targets to be linear 1326- Revert "st/dri: make sure software color-buffers are linear" 1327- zink: use shader-read-only-optimal for samplers 1328- zink: use emit_bitcast helper 1329- zink: ralloc spirv_shader 1330- zink: fix 8 bit index handling code 1331- zink: convert x8-formats in zink_get_format 1332- zink: make zink_format all about raw format-translation 1333- zink: fix format-mapping 1334- zink: add format test 1335- zink: map some more formats 1336- lavapipe: implement VK_EXT_index_type_uint8 1337- zink: nir_op_b2f64 implementation 1338- zink: more conversion ALUs 1339- docs/features: update list of zink features 1340- zink: document some more features for higher GL versions 1341- zink: only emit each cap once 1342- zink: do not open-code CALLOC_STRUCT 1343- zink: factor out zink_batch_release-helper 1344- zink: destroy blitter before destroying batches 1345- zink: release batch memory 1346- zink: do not leak vertex element state 1347- zink: dot leak dummy_buffer 1348- zink: free sets and hash-tables in context 1349- zink: destroy transfer-helper 1350- zink: destroy device and instance 1351- zink: do not use reservations for stream-out 1352- zink: do not reserve or pack fragment outputs 1353- zink: use ConstOffset for nir_tex_src_offset 1354- zink: use lower_scmp instead of open-coding 1355- zink: also lower scmp for soft-fp 1356- zink: remove support for fcsel 1357- gallium/util: do not perform n^2 stencil blits 1358- gallium/ntt: lower uniforms to ubo 1359- zink: disable render_condition_enable during blit 1360- microsoft/compiler: correct dxil fma opcode 1361- microsoft/compiler: do not lower away 64-bit ffma 1362- zink: rename zink vs pipe variables 1363- zink: setup compiler options during init 1364- zink: add missing opcodes 1365- zink: add missing 64-bit integer ops 1366- zink: use hardware int64 when supported 1367- mesa/st: fix regression for basic drivers 1368- zink: handle NULL views in zink_set_sampler_views 1369- zink: fix vertex-stride wrangling 1370- zink: respect feature-cap for independent blending 1371- zink: respect feature-cap for sample-shading 1372- zink: respect feature-cap for multi-draw indirect 1373- zink: make all xfb caps depend on extension 1374- zink: require vulkan memory model for tesselation 1375- zink: respect fragment-shader depth-layout 1376- zink: clone shader before lowering clip_halfz 1377- mesa/main: remove leftover bumpmap code 1378 1379Francisco Jerez (1): 1380 1381- intel/gen12: Fix memory corruption issues in fused Gen12 parts. 1382 1383Georg Lehmann (3): 1384 1385- vulkan/device-select: fix vkGetInstanceProcAddr self-resolving 1386- vulkan/overlay: fix vkGetInstanceProcAddr self-resolving 1387- vulkan/device_select: Only call vkGetPhysicalDeviceProperties2 if the device supports it. 1388 1389Gert Wollny (36): 1390 1391- util/format_zs: Add C++ include handling 1392- nir/print: print GS extra info 1393- r600/sfn: lower bool to int32 only after common optimizations 1394- r600/sfn: use a per stream index register in GS 1395- r600/sfn: Correctly lower all int64 1396- r600/sfn: fix component loading from fixed buffer ID 1397- r600/sfn: Add lowering pass to convert load_interpolated to load for POS 1398- r600/sfn: Add simplified constructors for FS shader inputs. 1399- r600/sfn: lower IO for FS inputs and handle interpolation accordingly 1400- r600/sfn: remove unused FS input deref code 1401- r600/sfn: Fix vertex stage export to accomodate IO lowering 1402- r600/sfn: lower VS output IO 1403- r600/sfn: Lower tess-eval IO 1404- r600/sfn: drop store_deref handling for VS and TES 1405- r600/sfn: lower GS IO 1406- r600/sfn: simplify IO lowering and fix TESS IO lowering 1407- r600/sfn: lower all IO in one pass 1408- r600/sfn: correct error signalling in switch default case 1409- r600/sfn: fix definition of priority queue 1410- r600/sfn: Fix a few warnings in release builds 1411- r600/sfn: remove unused file 1412- r600/sfn: remove leftover debug message 1413- r600/sfn: Fix dest-swizzle for GS vertex loads 1414- r600/sfn: Add support for shader_clock 1415- mesa/st: lower 64 bit ops to scalar before lowering to soft-float 1416- r600/sfn: merge SpecialValue and InlineConstValue 1417- doc: virgl supports ARB_texture_filter_anisotropic already 1418- r600: Support TGSI_OPCODE_I64NEG 1419- r600/sfn: C++ lower-instruct implementation 1420- r600/sfn: Add number for source components for split_y 1421- r600/sfn: add lowering passes to get 64 bit ops lowered to 32 bit vec2 1422- r600/sfn: tie in 64 lowering code 1423- r600: enable support for 64 bit DIVMOD when NIR is used 1424- r600: enable fp64 lowering to softemu with NIR 1425- r600/nir: use "unreachable" instead of "assert" 1426- r600/sfn: fix use of b32all/and 1427 1428Giovanni Mascellani (2): 1429 1430- disk_cache: Fail creation when cannot inizialize queue. 1431- anv: Allow null handle in DestroyDescriptorUpdateTemplate. 1432 1433Hans-Kristian Arntzen (2): 1434 1435- vulkan: Update to 1.2.164. 1436- radv: Implement VK_VALVE_mutable_descriptor_type. 1437 1438Hoe Hao Cheng (11): 1439 1440- zink: define and use \<%guard\> helper in zink_device_info 1441- zink: decouple features and enabling conditions in zink_device_info.py 1442- zink: move blend_operation_advanced conditions to zink_device_info.py 1443- zink: remove useless import in zink_device_info.py 1444- zink: allow Extension/Version to be shared across files 1445- zink: generate instance creation code with a python script 1446- zink: hook zink_instance to build 1447- zink: replace old code with generated zink_instance 1448- zink: fix property detection 1449- zink: add support for VK_EXT_4444_formats 1450- zink: VK_KHR_draw_indirect_count is a device extension 1451 1452Hyunjun Ko (6): 1453 1454- vulkan: Enable VK_KHR_performance_query on android 1455- turnip: Implement VK_KHR_performance_query 1456- turnip: support multipass for performance query. 1457- turnip: enable VK_KHR_performance_query with new debug flag 1458- turnip/kgsl: support VK_KHR_performance_query 1459- turnip: use ir3_compiler_destroy instead of ralloc_free 1460 1461Iago Toral Quiroga (33): 1462 1463- zink: only add MESA WSI structs for specific devices 1464- v3dv: fix typo 1465- v3dv: move authenticated display fd acquisition to swapchain creation time 1466- v3dv: fix width for buffer view texture state 1467- v3dv: add a buffer to image copy path using a texel buffer 1468- v3dv: initialize pipeline layouts for meta operations at driver initialization 1469- v3dv: blit shader clean-ups 1470- v3dv: rename playout and dslayout fields to use underscores. 1471- v3dv: use VkSurface to retrieve an authenticated display fd 1472- v3dv: remove box check from texel buffer copy fragment shader 1473- v3dv: remove redundant free of default pipeline attributes BO 1474- v3dv: only write new uniforms when needed 1475- v3dv: remove obsolete comment 1476- v3dv: fix allocation size for BO handles 1477- v3dv: fix leak in the buffer to image copy via texel buffer 1478- v3dv: batch buffer to image copies with the texel buffer path if possible 1479- v3dv: extend the list of formats supported by the TFU unit 1480- v3dv: remove obsolete disabled code 1481- v3dv: support compressed formats with TFU unit 1482- v3dv: add a format parameter to emit_tfu_job 1483- v3dv: add a TFU path for image copies 1484- v3dv: fix base layer for 3D blits in the TFU path 1485- v3dv: expand format coverage in TFU path for buffer to image copies 1486- v3dv: check return value of drmGetMagic 1487- v3dv: expand the formats that can be handled in the TFU blit path 1488- v3dv: handle Z mirroring in the TFU blit path 1489- v3dv: add a helper to choose a compatible TFU format 1490- v3dv: ignore filter in TFU blit path 1491- v3dv: move error string definition to debug path 1492- v3dv: don't log out of pool memory errors for internal driver pools 1493- v3dv: fix early return from failed drmGetMagic 1494- v3dv: fix incorrect slice selection for TFU jobs 1495- v3dv: fix BO list for TFU jobs 1496 1497Ian Romanick (23): 1498 1499- intel/compiler: Rotate instructions ROR and ROL cannot have source modifiers 1500- intel/compiler: Delete redundant MAC declaration 1501- intel/fs: Silence unused parameter warning in filter_simd 1502- intel/fs: Add support for printing half-float immediate values 1503- util: Add cnd_monotonic to Makefile.sources 1504- nir: Make some notes about fsign versus NaN 1505- nir/algebraic: Make some notes about comparison rearrangements versus infinity 1506- Revert "nir: Replace an odd comparison involving fmin of -b2f" 1507- nir/algebraic: Don't add reordered version of patterns for commutative instructions 1508- nir: Correctly constant fold fsign(NaN) and fsign(-0) 1509- nir/algebraic: Mark some logic-joined comparison reductions as exact 1510- nir/algebraic: Add some compare-with-zero optimizations that are exact 1511- spir-v: Mark floating point comparisons exact 1512- nir/algebraic: Fix broken NaN and -0.0 behavior 1513- nir/algebraic: Mark comparisons generated from lowered fsign precise 1514- nir/algebraic: Move the flrp -\> bcsel rule earlier 1515- i965: Don't parse driconf again 1516- nir/algebraic: Fix a \>\> \#b \<\< \#b for sizes other than 32-bit 1517- intel/compiler: Properly handle shift count for 8-bit sources 1518- intel/compiler: Enable the ability to emit CMPN instructions 1519- intel/compiler: Make the CMPN builder work like the CMP builder 1520- intel/compiler: Use CMPN for min / max on Gen4 and Gen5 1521- nir/algebraic: Fix some min/max of b2f replacements 1522 1523Icecream95 (54): 1524 1525- rbug: Forward get_compiler_options to pipe driver 1526- rbug: Handle non-TGSI shaders 1527- panfrost: Fix AFBC blits of resources with faked RGTC 1528- panfrost: Fix stack shift calculation 1529- pan/mdg: Try demoting uniforms instead of spilling to TLS 1530- panfrost: Split up batches with many jobs 1531- pan/gen_pack: Fix signed integer packing 1532- panfrost: Fix negative LOD bias support on Bifrost 1533- pan/decode: Fix "Access to unknown memory" message formatting 1534- panfrost: Fix precise occlusion queries on Bifrost 1535- panfrost: Fix CLAMP wrap mode 1536- panfrost: Fix the Maximum anisotropy field in the XML 1537- panfrost: Set the anisotropy level when cso->max_anisotropy is set 1538- panfrost: Add a gpu_revision argument to panfrost_get_quirks 1539- panfrost: Expose ARB_texture_filter_anisotropic on supported GPUs 1540- panfrost: Fix panfrost_small_padded_vertex_count for 17 vertices 1541- panfrost: Fix discard behaviour on Bifrost 1542- nir: Handle load_kernel_input in nir_get_io_offset_src 1543- pan/mdg: Fix promoted uniform moves with 64-bit types 1544- pan/mdg: Add load_kernel_input support 1545- pan/mdg: Implement load_global_invocation_id 1546- pan/mdg: Set compute lowering options 1547- panfrost: Stop lowering cs derived sysvals in glsl 1548- panfrost: Add a NIR pass to lower 64-bit vec3 intrinsic loads 1549- pan/mdg: Use the pan_nir_lower_64bit_intrin NIR pass 1550- pan/mdg: Support nir_intrinsic_load_global_constant 1551- pan/mdg: Support nir_intrinsic_group_memory_barrier 1552- panfrost: Allow NULL for some binding functions 1553- pan/mdg: Replace zext with a type enum 1554- pan/mdg: Return false instead of asserting in mir_args_ssa 1555- pan/mdg: Add i2i64 to mir_match_offset 1556- pan/mdg: Pass the memory type to mir_set_offset directly 1557- pan/mdg: Invert the type conditional for load intrinsics 1558- pan/mdg: Support loads and stores to scratch memory 1559- panfrost: Stub out panfrost_render_condition 1560- panfrost: Set conditional render cap 1561- gallium: Add new cap PIPE_CAP_TEXTURE_BUFFER_SAMPLER 1562- docs: Mention PIPE_CAP_TEXTURE_BUFFER_SAMPLER 1563- st/mesa: Use samplers for buffer textures if requested 1564- panfrost: Make the width argument to panfrost_new_texture 32 bits 1565- panfrost: Support buffer sampler views 1566- panfrost: Fix textureSize for buffer textures 1567- panfrost: Enable ARB_texture_buffer_object 1568- panfrost: Dual-source blending on Bifrost 1569- pan/bi: Add a define for the Bifrost shader prefetch size 1570- pan/bi: Add some zero bytes after shaders on Bifrost 1571- panfrost: Fix size assertion in bi_alu_src_index 1572- pan/mdg: Fix spilling when scratch memory is used 1573- pan/bi: Iterate from zero when setting RA interference 1574- pan/decode: Free mapped memory objects on BO unreference 1575- panfrost: Use normal malloc/free instead of ralloc for surfaces 1576- panfrost: Add the tiler heap to fragment jobs 1577- pan/bi: Use the correct size for UBO loads 1578- st/mesa: Update constants on alpha test change if it's lowered 1579 1580Ilia Mirkin (18): 1581 1582- nv50: only support 4 components in separate xfb mode 1583- nv50: fake enough resume support pre-nva0 to pass gles3 requirements 1584- mesa/teximage: show internal format when printing verbose api log 1585- nv50/ir: allow a mov to emit directly to a shader output 1586- nv50: fix instancing of client-side vertex buffers 1587- nv50,nvc0: serialize between before/after using a zeta surface as color 1588- nv50: use 2d blit when m2mf doesn't support the copy 1589- nouveau: change fence destruction logic on screen destroy 1590- nouveau: add drm-shim support 1591- ci: include nouveau in shader-db runs 1592- nouveau: trigger the current fence's work on destroy explicitly 1593- glsl: only expose int64 atomics when extension is enabled 1594- cso: set index_bounds_valid = true for arrays draws 1595- nvc0: index_bias is now only set for indexed draws 1596- st/mesa: fix broken moves for u2i64 and related ops 1597- nv50/ir: clear dnz flag when converting mul/mad to simpler ops 1598- nvc0/ir: add fixup to deal with interpolateAtSample with non-MSAA 1599- nouveau: reinstate fencing on screen destroy 1600 1601Indrajit Kumar Das (3): 1602 1603- radeonsi/gfx10: fix overflow and primitive queries 1604- radeonsi/gfx10: added support for gfx10 conditional rendering 1605- radeonsi/gfx10: fix issue with multiple overflow queries on the same context 1606 1607James Jones (4): 1608 1609- gallium: Add pipe_screen::is_dmabuf_modifier_supported 1610- gallium: Add format modifier plane count query 1611- gallium/dri: Factor out DRI extension setup code 1612- gallium/dri: Use per-screen DRI extension list 1613 1614James Park (54): 1615 1616- radv: Fix radv_queue_init failure handling 1617- c11/threads: Fix Win32 timed functions 1618- c11/threads: Remove Win32 null checks 1619- c11/threads: Remove Windows XP support 1620- util/os_time: Safe os_time_get_nano for Windows 1621- util,radv: Cross-platform monotonic condition variable 1622- radv: Const aco_compiler_statistic_info usage 1623- amd: Simplify ac_addrlib_create 1624- amd: Cast to int for %d snprintf argument 1625- amd: Remove bitfield sizes from enum values 1626- amd: Stub sections that don't have \_WIN32 support 1627- amd: Replace vasprintf with vfprintf 1628- amd: Work around MSVC limit for string literals 1629- amd: Fix signature mismatch 1630- amd: Fix declaration mismatch 1631- amd/common: Check with_tests before adding test 1632- vulkan: Remove GCC pragmas by fixing warnings 1633- vulkan: Replace pthread mutex with mtx_t 1634- vulkan: Portable wsi_common_get_current_time() 1635- util: Add os_localtime 1636- vulkan/util: Consolidate typed_memcpy 1637- aco: Define NOMINMAX in Meson build file 1638- aco: Fix warnings about unsafe integer/bool mix 1639- aco: Add missing C++ includes 1640- aco: Remove nonstandard parentheses 1641- aco: Declare num_reduce_ops for array size 1642- aco: Const correct aco_compiler_statistics 1643- aco: Replace indexed array initialization 1644- aco: Use u_memstream instead of POSIX memstream 1645- aco: Initialize union within Operand for MSVC 1646- aco: Fix warnings for bools in bitwise logic 1647- aco: Stub sections that don't have \_WIN32 support 1648- aco: Avoid extra bitfield padding 1649- radv: Exclude amdgpu driver files for Windows 1650- radv: Update build defines for Windows 1651- radv: Replace VLAs with alloca 1652- radv: Wrap pragmas with \__GNUC_\_ to fix MSVC 1653- radv: Use os_localtime instead of localtime_r 1654- radv: Don't return value in void function 1655- radv: Ignore radv_printflike on Windows 1656- radv: Update radv_assert for MSVC 1657- radv: Fix callback signatures 1658- radv: Fix leak in radv_amdgpu_winsys_destroy() 1659- radv: Fix function parameter types 1660- radv: Use standard \__VA_ARGS_\_ macro 1661- radv: Create shader cache if ENABLE_SHADER_CACHE 1662- radv: Use unsigned with u_bit_scan for MSVC 1663- radv: Replace pthread mutex with mtx_t 1664- radv: Replace pthread thread with thrd_t 1665- radv: Use portable ffs and util_bitcount macros 1666- util: Disable \[[fallthrough]\] for C17 1667- xmlconfig: Disable WITH_XMLCONFIG on Windows 1668- util: Disable memstream for Apple builds 1669- gallium/tessellator: Fix warning suppression 1670 1671Jan Beich (1): 1672 1673- util: unbreak on BSDs after MSVC changes 1674 1675Jason Ekstrand (63): 1676 1677- intel/fs: Fix use of undefined value in fixup_nomask_control_flow 1678- nir/lower_io: Add data OOB asserts to write_constant 1679- nir: Add a more generic helper for gathering constant initializers 1680- nir,clover: Drop nir_lower_mem_constant_vars 1681- nir: Rewrite lower_undef_to_zero 1682- Revert "anv/image: Define anv_image_get_aux_addr (v3)" 1683- vulkan: Update XML and headers to 1.2.162 1684- spirv: Rename some ray-tracing intrinsics to NV 1685- spirv: Update JSON and headers from Khronos main 1686- spirv: Implement OpTraceRayKHR and OpExecuteCallableKHR 1687- spirv: Call repair SSA for OpTerminateInvocation 1688- spirv: Implement OpTerminateRayKHR and OpIgnoreIntersectionKHR 1689- spirv: Implement SpvOpConvertUToAccelerationStructureKHR 1690- nir: Add a halt instruction type 1691- spirv: Emit nir_jump_halt after TerminateRay or IgnoreIntersection 1692- intel/dev: Add a gen_device_info::has_ray_tracing bit 1693- intel/genxml: Add the BINDLESS_SHADER_RECORD data structure 1694- intel/genxml/pack: Stash the cloned address field 1695- intel/genxml: Support truncated addresses 1696- intel/genxml: Add RT_DISPATCH_GLOBALS and RT_*_SBT_HANDLE structs 1697- intel/genxml: Add BVH data structures 1698- nir: Add a helper to get the live set at a cursor 1699- nir/lower_io: Allow ray_hit_attrib in lower_vars_to_explicit_types 1700- nir/lower_io: Support shader_call_data in vars_to_explicit_types 1701- intel/debug: Add a debug flag for ray-tracing shaders 1702- intel/compiler: Add support for bindless shaders 1703- intel/rt: Add a brw_rt.h header with \#defines for basic RT data structures 1704- intel/fs: Add and implement a load_global_const_block intrinsic 1705- intel/rt: Add builder helpers for accessing RT data structures 1706- intel/rt: Add a pass to lower the new ray-tracing intrinsics 1707- intel/rt: Add lowering functions for each ray-tracing stage 1708- intel/rt: Add support for scratch in ray-tracing shaders 1709- intel/rt: Add return instructions at the end of ray-tracing shaders 1710- intel/rt: Add a pass to lower shader call instructions 1711- intel/rt: Add a helper to create a trivial return shader 1712- intel/rt: Implement support for shader call payloads 1713- intel/fs: Add and implement intel-specific ray-tracing intrinsics 1714- intel/rt: Implement traceRay() 1715- intel/rt: Implement the new ray-tracing system values 1716- intel/rt: Add support for shader buffer record memory 1717- intel/rt: Add lowering for ray-walk intrinsics in any-hit shaders 1718- intel/rt: Add lowering for combined intersection/any-hit shaders 1719- intel/rt: Add a helper to create the raygen trampoline shader 1720- intel/rt: Add support for hit attributes 1721- intel/rt: Implement push constants as global memory reads 1722- nir: Use the right argument order for load_scratch_base_ptr 1723- intel/fs: DISCARD_JUMP does not have side-effects 1724- intel/fs: Rename PLACEHOLDER_HALT to HALT_TARGET 1725- intel/fs: Use BRW_OPCODE_HALT for discards 1726- intel/fs: Remove unnecessary HALT_TARGET in opt_redundant_halt() 1727- intel/fs: Emit HALT_TARGET in emit_nir_code() 1728- intel/fs: Implement nir_jump_halt 1729- nir/lower_non_uniform: Refactor for better code organization 1730- nir/lower_non_uniform: Better handle non-derefs 1731- anv: Bump maxGeometryInputComponents to 128 on Gen8+ 1732- intel/compiler: Return 1 for immediates in regs_read 1733- intel/fs: QUAD_SWIZZLE requires packed data 1734- nir: Drop the lower_mem_constant_vars declaration 1735- vulkan: Make vk_debug_report_callback derive from vk_object_base 1736- nir: Don't optimize bcsel-of-shuffle across blocks 1737- nir: Fix parameter order in the bcsel-of-shuffle optimization 1738- intel/fs: Shuffle can't handle source modifiers 1739- anv/formats: Advertise linear sampling on depth formats 1740 1741Jeremy Huddleston (3): 1742 1743- util: Fix pointer to integer conversion error when using libunwind 1744- Fall back on clock_gettime when timespec_get() is unavailable 1745- Adjust dylib compatibility versions to match what was set by mesa-18.3's autotools-based builds 1746 1747Jesse Natalie (105): 1748 1749- microsoft/compiler: Fix reference to renamed intrinsic getter 1750- panfrost/util: Move nir_undef_to_zero into core nir and add 'lower' 1751- nir: Add nir_alu_type -\> glsl_base_type conversion helper 1752- vtn/opencl: Fix alignment for half vload/vstore 1753- nir_load_libclc: Mark libclc shader as internal 1754- spirv: Allow spirv_to_nir callers to provide a float execution mode 1755- microsoft: Add CLC frontend and kernel/compute support to DXIL converter 1756- d3d12: Add glon12 target which only includes d3d12 driver 1757- d3d12: Pipe adapter LUID from callbacks to D3D12 screen init 1758- wgl: Marshal HDC into screen creation and LUID querying 1759- wgl: Implement get_adapter_luid callback 1760- wgl: Add stw_winsys callback to check which PFD flags should be added 1761- wgl: Add PFD flags based on stw_winsys callback response 1762- wgl: Add winsys framebuffer object 1763- wgl: Use winsys framebuffer interface if present 1764- d3d12: Implement winsys framebuffer 1765- winsys/d3d12: Use MakeWindowAssociation to remove DXGI's alt+enter handling 1766- d3d12: Delete unused local variables 1767- microsoft/compiler: Remove dead code/variables 1768- d3d12: Fix brace-initialization issues 1769- d3d12: Fix signed-unsigned comparison warnings 1770- d3d12: Remove Windows-specific macros 1771- d3d12: Clean up d3d12_compiler.h 1772- d3d12: Fix unhandled switch case warnings 1773- microsoft/compiler: Fix unhandled switch case warnings 1774- d3d12: Misc fixes caught by GCC warnings / code inspection 1775- microsoft/compiler: Misc fixes caught by GCC 1776- d3d12: Fix use of incorrect clear color variable 1777- microsoft/compiler: Add missing 'return' to switch case 1778- d3d12: Fix GCC warnings for missing function prototypes 1779- windows: Always set NOMINMAX to remove min/max macros 1780- util: Add os_get_page_size query 1781- driconf: Avoid empty macro resulting in empty initializer braces 1782- gallium: Include winsock lib as a dependency for Windows 1783- gallium: Remove unnecessary forward declaration of swrast_driver_descriptor 1784- clover: Add opencl-native build flag 1785- clover: Support LLVM coming from CMake instead of config-tool 1786- clover: Add version.lib dependency for Clang on Windows 1787- meson: Adjust Clover's required LLVM modules 1788- clover: Fix property_element::as for MSVC 1789- clover/llvm: Work around MSVC quirks 1790- clover/core: Support MSVC 1791- clover/api: Support MSVC 1792- clover: Use .def files for exports on Windows 1793- clover/core: Fix x86 build 1794- gallium: Add optional pipe_context to flush_frontbuffer 1795- d3d12: Fix incorrect fence timeout calculation 1796- CI: Add repeat-wait to Windows Piglit skip 1797- d3d12: Use DirectX-Headers wrap for d3d12.h 1798- d3d12: Refactor screen to abstract DXGI details 1799- d3d12: Add DXCore screen variation 1800- microsoft/compiler: Pick up new dxcapi.h 1801- winsys_handle: Change D3D12 resource handle type to void\* 1802- d3d12: Include wsl/winadapter.h when not compiling for Windows 1803- d3d12: Include dxguids/dxguids.h in files that need \__uuidof 1804- d3d12: Use IID_PPV_ARGS instead of \__uuidof 1805- d3d12: Scope down wrl includes to just client.h 1806- d3d12: Add forward declaration for LUID 1807- d3d12: Use u_dl instead of Windows DLL APIs 1808- d3d12: Only play DLL path tricks on Windows 1809- d3d12: Only support DXGI and GDI APIs on Windows 1810- d3d12: Support Linux eventfds for fences 1811- d3d12: Don't require DXIL for WSL 1812- gallium/dri: Add D3D12 software driver option 1813- d3d12: Flush and wait in flush_frontbuffer 1814- drisw: Add fallback logic for choosing a driver to use 1815- drisw: Prefer hardware-layered sw-winsys drivers over pure sw 1816- nir: Add intrinsic and string ptrs 1817- nir/vtn: Implement printf opcode in terms of intrinsic (v9) 1818- nir: Add a printf lowering pass (v5) 1819- nir: Add an algebraic optimization for float->double->float 1820- microsoft/clc: Hook up printf 1821- microsoft/compiler: Fix warnings produced by GCC in release mode 1822- microsoft/compiler: Fix incorrect size passed to strncpy 1823- d3d12: Unused variable warning indicated bug in bo_unmap 1824- d3d12: Signed/unsigned comparison warning fixes 1825- d3d12: Fix unused local variable warning in release build 1826- d3d12: Fix implicit fallthrough warnings 1827- microsoft/resoure_state_manager: Silence GCC invalid offsetof warning 1828- d3d12: Fix clang warnings from {0} in C++ code 1829- d3d12: Fix uninitialized variable referenced in error case 1830- d3d12: Remove copy/pasted line of array initialization 1831- microsoft/compile: Fix incorrect enum type in function signature 1832- microsoft/compiler: Fix tautological comparison 1833- microsoft/resource_state_manager: Remove unused private variable 1834- microsoft/compiler: Fix clang fallthrough warnings 1835- microsoft/clc: Fix const violations from ralloc_steal 1836- CI: Install DirectX-Headers package for x86 container 1837- CI: Enable d3d12 driver for Linux CI builds 1838- nir: Update saturated float->int/uint conversion algorithm 1839- d3d12: Add a path for mapping of not-directly-mappable buffers 1840- d3d12: Add a slab bufmgr for readback buffers 1841- d3d12: Use buffer pipe usage to inform allocation 1842- d3d12: Use an appropriate pipe resource usage for map intermediates 1843- d3d12: Don't allocate mappable textures 1844- nir: Work around MSVC x86 internal compiler error 1845- drisw: Disable automatic use of layered drivers with LIBGL_ALWAYS_SOFTWARE 1846- wgl: Refactor screen creation to a function 1847- wgl: Add a loop for screen creation with an ordered list of fallbacks 1848- d3d12: Fail screen creation if a shader validator is needed and can't be created 1849- wgl: Disable automatic use of layered drivers with LIBGL_ALWAYS_SOFTWARE 1850- microsoft/clc: Let lower_vars_to_explicit_types fill kernel input driver_location 1851- microsoft/clc: Fix wrap modes for inline samplers for integer textures 1852- microsoft/clc: Move inline samplers to the end of the variable list 1853- microsoft/clc: Use driver_location for metadata instead of re-computing offsets 1854 1855Jonathan Gray (1): 1856 1857- aco: use UINT64_C on 64 bit constant arguments 1858 1859Jonathan Marek (9): 1860 1861- turnip: implement z-scaling and z-mirroring BlitImage 1862- turnip: no linear_to_srgb for alpha channel for gmem clear value packing 1863- turnip: do not include compute stage in pipeline_builder 1864- turnip: always emit LRZ draw state in DIRTY_DRAW_STATE path 1865- turnip: correctly disable draw states outside of renderpasses 1866- turnip: do not emit draw states in draw_cs outside of renderpass 1867- turnip: move up LRZ invalidate in CmdClearAttachments 1868- turnip: always set LRZ registers to zero for 3d clear/blit 1869- turnip: don't always use 3d ops for blit_image 1870 1871Jordan Justen (10): 1872 1873- intel/dev: Use GEN_GEN if defined for gen_device_info_is_9lp 1874- intel/dev: Add gen_device_info_is_12hp 1875- intel/genxml: Copy gen12.xml to gen125.xml 1876- intel/genxml: Build gen 12.5 1877- intel/isl: Build gen 12.5 1878- intel/anv: Build gen 12.5 1879- intel/iris: Build gen 12.5 1880- intel/compiler: Add GEN125 to enum gen 1881- intel/common: Build mi_builder_test for gen 12.5 1882- iris: Fix android build due to missing link to libmesa_iris_gen125 1883 1884Juan A. Suarez Romero (19): 1885 1886- ci: add testing for VC4 drivers (Raspberry Pi 3) 1887- util: function to check for rgbX format 1888- v3d: force alpha to 1 when rendering RGBX formats 1889- v3d: make set tile buffer size function public 1890- v3d: store number of color buffers in job 1891- v3d: split binning start from draw 1892- v3d: add helper to check if format supports TLB resolve 1893- v3d: implement tile buffer blits 1894- v3d: refactor set tile buffer size function 1895- v3d: implement tile-based blit operation 1896- v3d: remove old tile blit code 1897- v3d: use job's nr_cbufs field 1898- v3d: extend the list of formats supported by the TFU unit 1899- ci: Bump deqp to current vulkan-cts-1.2.5.0 1900- doc/features: add VC4 driver 1901- v3d: reinterpret stencil data as uint texture in stencil blit path 1902- v3d: check blit mask inside blit subpaths 1903- v3d: add fast-path tile-based blit for depth/stencil buffers 1904- v3d: fix dest offset in TFU setup 1905 1906Karol Herbst (3): 1907 1908- clover/queue: Flush automatically if applications do not flush themselves 1909- tegra/context: fix regression in tegra_draw_vbo 1910- tegra/context: unwrap indirect_draw_count as well 1911 1912Keith Packard (1): 1913 1914- glx: Provide glvnd wrapper for glXSwapIntervalEXT 1915 1916Kenneth Graunke (16): 1917 1918- intel/compiler: Fix passthrough TCS regressions from program rename 1919- prog_to_nir: Revert name initialization change 1920- intel/compiler: Do interpolateAtOffset coordinate scaling in NIR 1921- intel/fs: Fix sampler message headers on Gen11+ when using scratch 1922- nir/algebraic: Avoid creating new fp64 ops when using softfp64 1923- asm: Fix x86 assembly for inverse matrix operations 1924- asm: Try to fix sparc assembly for inverse matrix operations 1925- nir/lower_non_uniform: Use nir_read_first_invocation helper. 1926- vbo: Don't set node->min_index = max_index = indices_offset when merging 1927- vbo: Only mark merged line strips as lines when actually converting them 1928- tnl: Try not to botch index buffer munging when start \\> 0. 1929- tnl: Respect \`start\` when converting indices to GLuint 1930- tnl: Reset nr_bos to 0 between map/unmap cycles. 1931- Revert "mesa: allow half float textures based on ARB_half_float_pixel" 1932- iris: Consider resolves after changing a resource's aux state 1933- glsl/float64: Bump \#version to 400 1934 1935Krunal Patel (1): 1936 1937- radeon/vce: Bitrate not updated when changing framerate 1938 1939Leo Liu (17): 1940 1941- vl: add AV1 codec picture support 1942- radeon/vcn: add AV1 codec driver firmware interfaces 1943- radeon/vcn: add AV1 support to the decoder 1944- radeon/vcn: add AV1 dpb buffer size 1945- radeon/vcn: add AV1 default tables for the context 1946- radeon/vcn: add AV1 context buffer 1947- radeon/vcn: fill up the context buffer 1948- radeon/vcn: get AV1 message buffer 1949- radeon/vcn: fill up the probs buffer 1950- radeonsi: cap AV1 codec configuration 1951- radeonsi: cap AV1 support to SIENNA CICHLID 1952- frontends/omx/bellagio: add AV1 initial support to omx dec 1953- frontends/omx/av1: add AV1 OBU header parsers 1954- frontends/omx/av1: add AV1 tasks management 1955- frontends/omx/av1: enable AV1 OMX Bellagio support 1956- mesa/st_vdpau: set surface winsys handle modifier 1957- frontends/omx: fix build warning 1958 1959Lionel Landwerlin (21): 1960 1961- intel/dump_gpu: add support for MMAP_OFFSET ioctl 1962- nir: don't consider txf_ms_mcs a query instruction 1963- st: trigger noop if the default value is not true 1964- mesa: add an environment variable to default enable INTEL_blackhole 1965- anv: fix descriptor pool leak in VMA object 1966- nir: wire shading rate variables 1967- compiler/nir: introduce a new helper to get varying name 1968- spirv: add support for KHR_fragment_shading_rate 1969- isl: Fix android build 1970- vulkan/overlay: don't display frame numbers unless required 1971- vulkan/overlay: add new options to display device/swapchain-format 1972- gallium/dri2: Don't forget protected content flag 1973- anv: add transfer usage for color/depth/stencil attachments 1974- intel/mi_builder: fix self modifying batches 1975- anv: Fix stencil layout in render passes 1976- anv: fix invalid programming of BLEND_STATE 1977- anv: only signal wsi fence BO on last command buffer 1978- anv: discard all timeline wait/signal value=0 1979- anv: reset binary syncobj to be signaled before submission 1980- anv: don't wait for completion of work on vkQueuePresent() 1981- anv: Fix wait_count missing increment 1982 1983Louis-Francis Ratté-Boulianne (11): 1984 1985- gallium/nir: Wrap tgsi_to_nir header in extern C 1986- gallium/util: Wrap suballoc.h into extern C 1987- gallium: Wrap some header files into "extern C" 1988- d3d12: Add D3D12 WGL winsys 1989- wgl: Flush in-between resolving buffer and presenting 1990- wgl: Call flush_resource() before presenting 1991- wgl: Wait for fence when not using winsys framebuffer 1992- wgl: Create third buffer when drawing to front buffer 1993- wgl: Wrap stw_pixelformat.h into extern C 1994- d3d12: Release swapchain buffers before resizing them 1995- wgl: Don't crash in stw_make_current if current framebuffer is NULL 1996 1997Lucas Stach (2): 1998 1999- etnaviv: fix disabling of INT filter for real 2000- etnaviv: tex_state: fix miplevel selection 2001 2002Marcin Ślusarz (16): 2003 2004- nir: handle float atomics in copy propagation pass 2005- intel/tools/aubinator_error_decode: exit with an error on unknown option 2006- intel/tools/aubinator_error_decode: allow "-" as an input file 2007- intel/tools/aubinator_error_decode: allow 0 arguments 2008- iris: store copy of the border color in the border color hash table 2009- intel/tools/aubinator_error_decode: cleanup path/file handling 2010- intel/tools/aubinator_error_decode: fix small memory leaks 2011- svga: remove duplicated code 2012- iris: remove redundant check 2013- util/list: add list_is_linked 2014- nine: use list_is_linked 2015- gallium: use list_is_linked 2016- iris: use list_is_linked 2017- r600: use list_is_linked 2018- omx: use list_is_linked 2019- util/list: use helper function in list_is_singular 2020 2021Marek Olšák (278): 2022 2023- st/mesa: fix use-after-free when updating shader info in st_link_nir 2024- nir: optionally shuffle local invocation IDs for compute quad derivatives 2025- nir: rename needs_helper_invocations to needs_quad_helper_invocations 2026- nir: gather shader_info::needs_all_helper_invocations 2027- nir: optimize nir_lower_discard_to_demote to lower discard/demote both ways 2028- ac/llvm: fix demote inside conditional branches 2029- radeonsi: enable GL_EXT_demote_to_helper_invocation 2030- amd: add register enums for VRS 2031- radeonsi: add an option to enable 2x2 coarse shading for non-GUI elements 2032- mesa: add Driver.DrawTransformFeedback 2033- gallium: move count_from_stream_output into pipe_draw_indirect_info 2034- gallium: make pipe_draw_indirect_info \\* a draw_vbo parameter 2035- gallium/u_threaded: lift DIV_ROUND_UP to eliminate it for constant expressions 2036- gallium/u_threaded: clean up direct vs indirect draws 2037- gallium: add pipe_draw_info::index_bounds_valid 2038- gallium/u_threaded: improve draw merging by clearing pipe_draw_info fields 2039- gallium: add missing bits of the direct multi draw interface 2040- gallium: extend draw_vbo to support multi draws 2041- gallium/u_threaded: store start/count in min/max_index for better packing 2042- gallium/u_threaded: add support for multi draws 2043- mesa: clean up Driver.Draw parameter types 2044- mesa: clean up GLboolean types in draw.c 2045- mesa: remove constant drawID parameter from \_mesa_draw_arrays 2046- mesa: move primitive restart enablement determination from st/mesa to main 2047- mesa: index \_RestartIndex with index_size_shift 2048- mesa: add primitive restart state to Driver.Draw parameters 2049- mesa: don't FLUSH_VERTICES from primitive restart changes 2050- radeonsi: don't load DrawID for indirect draws if it's unused 2051- radeonsi: swap DrawId and StartInstance SGPR locations 2052- radeonsi: handle pipe_draw_info::increment_draw_id 2053- radeonsi: fix min_direct_count value 2054- radeonsi: do VGT_FLUSH when switching NGG -\> legacy on Sienna Cichlid 2055- radeonsi: only do VGT_FLUSH for fast launch if previous draw was normal launch 2056- radeonsi: determine correctly if switching from normal launch to fast launch 2057- radeonsi: don't subtract max_verts_per_prim from hw_max_esverts on gfx10.3 2058- radeonsi: read vs_state_bits in vs_prolog correctly 2059- radeonsi: tweak triangle list culling performance for GS fast launch 2060- radeonsi: remove VS input loads when culling with rasterizer discard 2061- radeonsi: add options.inline_uniforms to the shader cache key 2062- ac: add build_alloca with an initializer 2063- ac: fix detection of Pro graphics 2064- ac: fix min/max_good_num_cu_per_sa on gfx10.3 with disabled SEs 2065- ac: rename num_render_backends -\> max_render_backends 2066- ac: rename num_sh_per_se -\> num_sa_per_se 2067- radeonsi: don't do VGT_FLUSH before fast launch on gfx10.3 2068- radeonsi: don't add num_vbos_in_user_sgprs to the shader cache key for non-VS 2069- radeonsi: fix NGG streamout regression 2070- radeonsi: fix scan_instruction for bindless inc_wrap/dec_wrap atomics 2071- winsys/amdgpu: remove amdgpu_winsys_bo::u::sparse::flags 2072- winsys/amdgpu: remove amdgpu_winsys_bo::sparse 2073- winsys/amdgpu: replace amdgpu_winsys_bo::flags with pb_buffer::usage 2074- winsys/amdgpu: replace amdgpu_winsys_bo::initial_domain with pb_buffer::placement 2075- winsys/amdgpu: move amdgpu_winsys_bo::lock for better packing 2076- mesa: add glInternalSetError for glthread 2077- mesa: make error handling for glGetActiveUniform glthread-safe 2078- glthread: make glGetActiveUniform return without syncing 2079- mesa: lock Shared->BufferObjects only once for a glthread batch 2080- mesa: lock Shared->TexMutex only once for a glthread batch 2081- nir: fix gathering TCS cross invocation access with lowered IO 2082- nir: fix gathering patch IO usage with lowered IO 2083- ac/nir: fix a typo in ac_are_tessfactors_def_in_all_invocs 2084- radeonsi: adjust tess SGPRs to allow fully occupied 3 HS waves of triangles 2085- radeonsi: don't leave more than 8 unoccupied lanes in HS 2086- radeonsi: don't allocate LDS for TCS outputs if they are not read 2087- radeonsi: limit HS LDS usage per workgroup to 16K to allow at least 2 WGs/CU 2088- radeonsi: don't generate a dead conditional in si_write_tess_factors on gfx9+ 2089- radeonsi: merge TCS and TCS epilog conditional blocks 2090- radeonsi: always return void from si_build_wrapper_function 2091- radeonsi: if VS and TCS have the same number of threads, merge the conditonals 2092- radeonsi: remove unnecessary NULL checking in NIR tess functions 2093- ac/llvm: prepare for passing VS->TCS IO via VGPRs 2094- radeonsi: pass VS->TCS IO via VGPRs if VS and TCS have the same thread count 2095- radeonsi: don't insert barrier between VS/TCS if all TCS inputs come from VGPRs 2096- radeonsi: don't allocate LDS for TCS inputs if it's not used 2097- radeonsi: implement GS fast launch for indexed triangle strips 2098- mesa: don't duplicate allocation code in \_mesa_new_parameter_list_sized 2099- mesa: track ParameterValues size separately 2100- mesa: properly disallow param list reallocation 2101- mesa: don't print GL errors in release builds if MESA_DEBUG=silent 2102- mesa: call FLUSH_VERTICES before changing sampler uniforms 2103- mesa: move sampler condition for flushing into mesa_flush_vertices_for_uniforms 2104- mesa: skip redundant uniform updates for glUniform 2105- mesa: skip redundant uniform updates for glUniformMatrix 2106- mesa: skip redundant uniform updates for glUniformHandle 2107- mesa: don't read from destination memory when computing state parameter values 2108- mesa: replace \_mesa_problem with unreachable in fetch_state 2109- util: add a common ALIGN16 macro for m_matrix and u_threaded_context 2110- mesa: don't allocate matrices with malloc 2111- mesa: rework matrix statevar enums to remove excessive branching in fetch_state 2112- mesa: remove redundant \_math_matrix_analyse calls in fetch_state 2113- mesa: fix printing state parameters 2114- mesa: allow multi-slot program parameters 2115- mesa: demystify material_attrib() 2116- mesa: optimize setting gl_Light state parameters 2117- mesa: restructure gl_light vars to match the layout of gl_LightSource uniforms 2118- mesa: put constants before state vars for ffvp 2119- mesa: put constants before state vars for ARB programs 2120- mesa: take advantage of sorted parameters in \_mesa_load_state_parameters 2121- mesa: merge matrix state parameters for faster uploads (disabled) 2122- mesa: merge light state parameters for faster uploads (disabled) 2123- mesa: add helpers for drivers to load state parameters into buffers 2124- gallium: add PIPE_CAP_PREFER_REAL_BUFFER_IN_CONSTBUF0 2125- st/mesa: add a faster path for uploading state parameters into constant buffers 2126- st/mesa: replace st_context::state::constants with a mask 2127- mesa: fix crashes in the no_error case of invalid glUniform calls 2128- mesa: skip glMultMatrix if the matrix is identity 2129- mesa: consider glPushMatrix a no-op change from the driver perspective 2130- mesa: canonicalize matrix in glPushMatrix to make glPopMatrix possibly a no-op 2131- mesa: memset matrices at initialization to enable memcpy on it 2132- mesa: treat glPopMatrix as a no-op state change if it doesn't change the matrix 2133- mesa: rewrite glPushAttrib/glPopAttrib to get rid of malloc 2134- mesa: add a fast path for restoring fixed-func tex state in glPopAttrib 2135- mesa: add a fast path for restoring light attributes in glPopAttrib 2136- mesa: reorganize gl_texture and sampler structures for glPush/PopAttrib 2137- mesa: optimize saving/restoring bound textures for glPush/PopAttrib 2138- mesa: reduce the size of gl_texture_attrib_node::Texture by about 90% 2139- mesa: skip \_mesa_set_enable in glPopAttrib if there are no changes 2140- mesa: optimize out no-op calls in glPopAttrib 2141- mesa: more optimizations in glPopAttrib (colormask, drawbuffers, coord replace) 2142- mesa: remove gl_texture_object references from glPush/PopAttrib stack 2143- mesa: allocate the attribute stack on demand 2144- st/mesa: fix uninitialized/random clip plane state vars in lower_ucp 2145- compiler: decrease STATE_LENGTH from 5 to 4 2146- mesa: replace ParameterValueOffset[i\] with Parameters[i].ValueOffset 2147- radeonsi: print more fields in si_dump_shader_key 2148- radeonsi: always use a staging texture for linear 1D textures in VRAM 2149- radeonsi: correct the MAD/FMA support table 2150- radeonsi: use util_logbase2 instead of division by index_size 2151- radeonsi: fix a memory leak in si_create_dcc_retile_cs 2152- radeonsi: fix line stippling with LINES_ADJACENCY without GS 2153- radeonsi: fix max_lds_size warning in release builds 2154- winsys/radeon: don't use debug_get_option_noop in a hot path 2155- winsys/amdgpu: don't use debug_get_option_noop in a hot path 2156- radeonsi: unduplicate code setting MIN_COMPRESSED_BLOCK_SIZE 2157- radeonsi: enable NGG and NGG culling on gfx10.3 APUs by default 2158- radeonsi: add AMD_DEBUG=nofastlaunch for debugging 2159- radeonsi: eliminate shader code for disabled or masked color outputs 2160- radeonsi: fix a nasty bug in si_pm4.c 2161- radeonsi: only mask 1 CU for GS/VS waves on gfx10.3 2162- ac,radeonsi: fix load_first_vertex 2163- radeonsi: don't update indexed flag in SGPR if it's unused 2164- radeonsi: don't update provoking vertex and outprim states in SGPR if unused 2165- ac: enable late allocation on VanGogh to increase perf 2166- radeonsi: disable WGP mode on gfx10.3 to prevent hangs 2167- radeonsi: don't invalidate emitted NUM_INSTANCES for u_blitter 2168- radeonsi: don't set DrawID and StartInstance if they are unused 2169- radeonsi: don't check for GS fast launch for NOT_EOP in the indexed case 2170- Revert "radeonsi: always return void from si_build_wrapper_function" 2171- vbo: remove gl_context dereferences when we can just subtract the pointer 2172- cso: remove unused code 2173- gallium: inline struct u_suballocator to remove dereferences 2174- cso: inline struct cso_cache to remove dereferences 2175- st/mesa: put pipe_screen \\* into st_context and use it 2176- st/mesa: move cso_context next to the other pointers 2177- r300,r600,radeonsi: inline struct radeon_cmdbuf to remove dereferences 2178- draw: add NIR support to draw_create_vertex_shader 2179- st/mesa: don't generate TGSI for the draw VS because it now supports NIR too 2180- st/mesa: remove less useful debug options in hot paths 2181- gallium: fix the PIPE_SHADER_CAP_SUPPORTED_IRS value for all drivers 2182- glthread: use glthread->used instead of glthread->next_batch->used 2183- glthread: use uint64_t to declare the batch buffer instead of align(8) 2184- glthread: change sizes to unsigned or size_t where needed 2185- glthread: count batch space in units of uint64_t elements 2186- gallium/u_threaded: don't pass index bounds to the driver to decrease overhead 2187- gallium/u_threaded: set has_user_indices = false in the driver thread 2188- gallium/u_threaded: don't copy the indexbuf pointer if we overwrite it 2189- gallium/u_threaded: don't make a local copy of pipe_draw_start_count 2190- gallium/u_threaded: optimize set_constant_buffer 2191- mesa: fix glPopAttrib for GL_COORD_REPLACE for r200 2192- mesa: remove code for old (mostly unsupported) GL_NV_point_sprite 2193- mesa: remove MAX_3D_TEXTURE_LEVELS, MAX_CUBE_TEXTURE_LEVELS 2194- radeonsi: move si_screen_clear_buffer into si_compute_blit.c w/o SDMA option 2195- radeonsi: rename buffer functions so as not to reference rings 2196- radeonsi: remove SDMA support 2197- radeonsi: rename SI_TEST_DMA to SI_TEST_BLIT 2198- radeonsi: fix the blit test for SW_64KB_R_X 2199- radeonsi: initialize ctx and gfx_cs first, then allocators 2200- ac: add radeon_info::all_vram_visible for Smart Access Memory 2201- radeons: only force staging uploads for VRAM when all VRAM is not visible 2202- radeonsi: only use staging for linear textures when all VRAM is not visible 2203- radeonsi: unify uploaders and upload to VRAM if all VRAM is visible 2204- radeonsi: map PIPE_USAGE_STREAM to VRAM if all VRAM is visible 2205- winsys/amdgpu: use VRAM for command buffers if all VRAM is visible 2206- ac,radeonsi: implement GL_NV_compute_shader_derivatives 2207- st/mesa: enable compute shader derivatives in SPIR-V 2208- radeonsi: fix a crash in si_fence_server_sync 2209- ac: correct ac_shader_args types, remove sgpr_count 2210- ac: add shader return values into ac_shader_args 2211- radeonsi: split ac_shader_args initialization from LLVM code 2212- radeonsi: move si_create_function into si_shader_llvm.c 2213- radeonsi: move si_build_main_function into si_shader_llvm.c 2214- radeonsi: move si_llvm_compiler_shader and deps into si_shader_llvm.c 2215- ac: unify shader arguments that are duplicated 2216- ac/llvm: handle no_(un)signed_wrap NIR flags 2217- compiler: fix glsl_types.h compile failures when including as C++ in drivers 2218- gallium/util: allow including a few files in C++ 2219- amd/llvm: fix C++ compile failures 2220- radeonsi: allow including a few files from C++ 2221- radeonsi: fix future C++ compile failures and warnings 2222- radeonsi: resolve a tricky C++ failure with goto jumping over initializations 2223- radeonsi: rename si_state_draw.c to .cpp 2224- radeonsi: use a C++ template to decrease draw_vbo overhead by 13 % 2225- radeonsi: fix small primitive culling with MSAA force-disabled and smoothing 2226- radeonsi: disable NGG fast launch with indexed triangle strips to fix a hang 2227- radeonsi: improve a comment about an MSAA bug workaround 2228- nir_to_tgsi: fix NIR options instead of asserting 2229- draw: fix incorrect NIR support code 2230- mesa: fix assertion paramList->LastUniformIndex \\< paramList->FirstStateVarIndex 2231- mesa: remove unused LastUniformIndex 2232- mesa: overallocate program parameter values 2233- mesa: don't restore texture state into unbound textures in glPopAttrib 2234- mesa: call Driver.TexParameter in glPopAttrib to fix r100, r200, old nouveau 2235- gallium: pass pipe_stencil_ref by value (it has only 2 bytes) 2236- gallium: inline pipe_alpha_state to enable better DSA bitfield packing 2237- gallium: inline pipe_depth_state to decrease DSA state size by 4 bytes 2238- cso: don't pass blend_color through cso_context 2239- st/mesa: don't make a local copy of blend color 2240- cso: remove context and delete_state pointers from all CSOs 2241- cso: inline cso_construct_key 2242- gallium/util: fix util_can_blit_via_copy_region for conditional rendering 2243- st/mesa: don't do glCopyPixels via blit if depth bounds test is enabled 2244- st/mesa: relax requirements for doing glCopyPixels via blit 2245- st/mesa: skip glDrawPixels if it's totally clipped for all codepaths 2246- mesa: fix an overflow check for MultiDrawElements 2247- vbo: only set count and end when closing \_mesa_prim 2248- vbo: change the parameters of vbo_get_minmax_index to get rid of \_mesa_prim 2249- mesa: add Driver.DrawGallium\* functions to be used by main/draw.c 2250- gallium: add pipe_draw_info::index::gl_bo 2251- mesa: add a fallback for drivers not implementing Driver.DrawGallium\* 2252- vbo: add vbo_get_minmax_indices_gallium 2253- mesa: switch (Multi)DrawArrays to DrawGallium 2254- mesa: switch Draw(Range)Elements(BaseVertex) calls to DrawGallium 2255- mesa: switch MultiDrawElements(BaseVertex) to DrawGallium\* 2256- vbo: remove \_mesa_prim parameter from vbo_try_prim_conversion 2257- vbo: remove \_mesa_prim parameter from vbo_merge_draws 2258- vbo: remove \_mesa_prim parameter from vbo_copy_vertices 2259- vbo: switch immediate Begin/End to DrawGallium 2260- gallium/u_threaded: clear vertices_per_patch if prim type != PATCHES 2261- gallium: remove and emulate PIPE_CAP_MULTI_DRAW 2262- gallium: fix draw info setup in draw and utilities 2263- freedreno: fixes handling draw info 2264- iris: don't use index_bias if not indexed 2265- nouveau: fix handling draw info 2266- panfrost: don't use index_bias if not indexed 2267- r600: fix handling draw info 2268- swr: fix handling draw info 2269- svga: fix handling draw info 2270- vc4: don't use index_bias if indexed 2271- v3d: don't use index_bias if not indexed 2272- virgl: fix handling draw info 2273- st/mesa: implement Driver.DrawGallium callbacks 2274- gallium: remove PIPE_CAP_INFO_START_WITH_USER_INDICES and fix all drivers 2275- util: add AMD CPU family enums and enable L3 cache pinning on Zen3 2276- ac,radeonsi: limit Smart Access Memory to Zen 3 and GFX10.3 due to perf issues 2277- radeonsi: add driconf options to enable/disable Smart Access Memory 2278- radeonsi: take color interpolation into account for shader variants 2279- util: replace UTIL_MAX_CPUS by util_cpu_caps.num_cpu_mask_bits 2280- st/mesa: simplify checking whether to pin threads to L3 2281- st/mesa: fix a defect when st_validate_state was invoked for unused states 2282- mesa: add STATIC_ASSERTs to the STATE_LIGHT_ATTRIBS case 2283- mesa: fix a bug in merging light state parameters with unpacked uniforms 2284- mesa: fix a second bug in merging light state parameters with unpacked uniforms 2285- radeonsi: fix hang caused by for loop with exec=0 in LS and ES 2286- radeonsi: remove si_gs_prolog_bits::gfx9_prev_is_vs 2287- gallium: skip draws with count == 0 or instance_count == 0 in drivers 2288- mesa: skip draws w/ count == 0 and instance_count == 0 in draw_gallium_fallback 2289- vbo: fix a index buffer map failure with size = 0 in get_minmax_indices_gallium 2290- gallium/u_threaded: skip draws if user index buffer size has size == 0 2291- mesa: always set valid index bounds for non-indexed draws for classic drivers 2292- mesa: fix alpha channel of ETC2_SRGB8 decompression for !bgra 2293- radeonsi: fix centroid with VRS coarse shading 2294- glthread: fix interpreting vertex size == GL_BGRA for vertex attribs 2295- mesa: flush glBegin/End before changing GL_DEPTH_STENCIL_TEXTURE_MODE 2296- i915: use align_calloc for the context to fix m32 crashes 2297- radeon,r200: use align_calloc for the context to fix m32 crashes 2298- nouveau_vieux: use align_calloc for the context to fix m32 crashes 2299- Revert "gallium/u_upload_mgr: allow use of FLUSH_EXPLICIT with persistent mappings" 2300- radeonsi: don't crash on NULL images in si_check_needs_implicit_sync 2301 2302Marek Vasut (1): 2303 2304- etnaviv: Fix rework ZSA into a derived state 2305 2306Marijn Suijten (3): 2307 2308- util: Do not insert uninitialized data if Android property is not set 2309- android: util: Add libcutils to Android.mk shared libs 2310- mesa/math: Fix address of array always returning true 2311 2312Mark Janes (1): 2313 2314- meson: add idep_mesautil to components using simple_mtx.h 2315 2316Martin Peres (1): 2317 2318- driconf: remove the redundant glx-extension-disabling options 2319 2320Matt Turner (2): 2321 2322- glcpp: Handle bison-3.6 error message changes 2323- turnip: Remove unused TU_DEBUG_IR3 flag 2324 2325Mauro Rossi (19): 2326 2327- android: gallium/aux: update old generated sources rules 2328- android: gallium/aux: Add GPU tracepoint mechanism 2329- android: freedreno: Add GPU tracepoints 2330- android: freedreno: Remove fd_log() 2331- android: freedreno/ir3: use python3 in gen rules 2332- android: radv: add libcutils shared dependency 2333- android: spirv: fix '::' typo in gen rules 2334- android: pan/bi: Add explicit dependency on the ISA helpers 2335- android: pan/bi: Generate bi_opcodes.{c,h} 2336- android: pan/bi: Generate instruction printer 2337- android: pan/bi: Generate builder routines 2338- android: pan/bi: Generate instruction packer for new IR 2339- android: pan/bi: Remove combine lowering 2340- android: pan/bi: Remove old IR packs 2341- android: pan/bi: Remove NIR->old IR 2342- android: pan/bi: Remove old IR opcode table 2343- android: ac/radv: fix typo in ac_rgp.h listed in Makefile.sources 2344- android: r600/sfn: add sfn_nir_lower_64bit.cpp to Makefile.sources 2345- android: pan/bi: reorder static dependencies in gallium/dri 2346 2347Michael Forney (1): 2348 2349- meson: add missing dependency on generated git_sha1.h 2350 2351Michael Tang (3): 2352 2353- microsoft/compiler: Add dedicated spirv_to_dxil libraries 2354- util: Implement os_read_file for Windows 2355- microsoft/compiler: Add spirv2dxil executable 2356 2357Michel Dänzer (33): 2358 2359- ac: Don't negate strstr return values in ac_query_gpu_info 2360- ci: Drop ci-templates-sha anchor 2361- ci: Update to current ci-templates 2362- ci: Use ci-fairy docker image instead of local git_archive one 2363- ci: Move sanity stage to the beginning of the pipeline 2364- ci: Squash "check mr/commits" jobs into a single sanity job 2365- ci: Make test-docs job depend on sanity job 2366- ci: Go back to previous ci-templates commit for debian.yml 2367- ci: Run git gc before creating Git cache tarball 2368- ci: Define global variable MESA_TEMPLATES_COMMIT for ci-templates commit 2369- ci: Append $MESA_TEMPLATES_COMMIT to image tags 2370- ci: Drop x86_build_old image 2371- ci: sanity job doesn't need the Git tree 2372- ci: Manual test jobs don't need the Git tree 2373- ci: Run sanity job automatically for forked branches as well 2374- ci: Move BASE_TAG expansion to FDO_BASE_IMAGE assignment 2375- ci: Add .use-base-image template 2376- ci: Adapt armhf_test job to MESA_TEMPLATES_COMMIT related changes 2377- docs: Adapt to FDO_DISTRIBUTION_TAG → MESA_IMAGE_TAG rename 2378- ci: .lava-test:amd64 template needs arm_build 2379- ci: Run sanity job only in pre-merge pipelines 2380- ci: Move deploy stage to the end of the pipeline 2381- wsi/x11: Set recognizable name for WSI swapchain queue thread 2382- wsi/x11: Always link against xcb-xrandr 2383- wsi/x11: Detect Xwayland 2384- wsi/x11: Use PresentOptionAsync for MAILBOX present mode with Xwayland 2385- wsi/x11: Treat IMMEDIATE present mode the same as MAILBOX for Xwayland 2386- ci: Rule out scheduled pipelines in .windows-build-rules 2387- ci: Add \*ignore_scheduled_pipelines to mesa/gallium rules templates 2388- wsi/x11: Use wsi_x11_get_connection in x11_present_to_x11_dri3 2389- wsi/x11: Always free randr_reply in wsi_x11_connection_create 2390- wsi/x11: Make sure wsi_x11_connection::is_xwayland is always initialized 2391- wsi/x11: Use get_screen_resources_current in wsi_x11_detect_xwayland 2392 2393Michel Zou (16): 2394 2395- zink: fix build on windows 2396- util: fix -Wshift-count-overflow warning 2397- zink: fix unused variable warning 2398- libgl-gdi: add zink support 2399- spirv: workaround setjmp/longjmp crash on MinGW 2400- glsl: Drop mingw -O1 workaround for GCC>=7.3 2401- util: fix mingw format-extra-args warning 2402- glapi: fix unused-function warning 2403- glsl: fix redefinition warning on win32 2404- wgl: fix maybe-uninitialized warning 2405- softpipe: fix maybe-uninitialized warning 2406- gallium/tests: fix unused-but-set-variable warning 2407- llvmpipe: work around mingw compiler optimization bug 2408- meson: fix multiline string warning 2409- llvmpipe: fix unused variables warnings 2410- drisw: fix unused variables warnings 2411 2412Mike Blumenkrantz (113): 2413 2414- util/threaded_context: use driver's ubo alignment for constant buffer uploads 2415- zink: initial implementation of shader keys 2416- zink: refcount the shader cache 2417- zink: move shader key structs into their own header 2418- zink: fill in params for fs shader keys and flag shader for rebuild 2419- zink: put those shader keys to work fixing up fragment shaders 2420- zink: update shader modules in gfx program when flagged dirty 2421- zink: handle arbitrary border colors using VK_EXT_custom_border_color 2422- zink: track custom border color samplers and verify against device limits 2423- zink: add alternate ubo loader in ntv 2424- zink: assert all index values in ntv OpAccessChain constructor 2425- zink: initial shader key implementation 2426- zink: change a memcmp==0 to !memcmp 2427- zink: use shader keys for samplemask 2428- mesa/st: set reserved storage for params+values to 16 2429- zink: fix direct image mapping offset 2430- zink: really fix direct image mapping offset (I mean it this time) 2431- st/pbo: fix pbo uploads without PIPE_CAP_TGSI_VS_LAYER_VIEWPORT 2432- st/mesa: set drawpixels swizzle before creating sampler view 2433- glsl/float64: make this compatible with glsl 330 2434- zink: support frem shader op 2435- zink: add nir pass for splitting 64bit vertex attribs which cross slot boundaries 2436- zink: be more paranoid about array strides in ntv 2437- zink: add get_storage_class() ntv util 2438- zink: handle struct derefs in ntv 2439- zink: ntv formatting 2440- zink: add struct type support for ntv 2441- zink: add handling for 64bit values in spirv_builder 2442- zink: support nir_op_f2f32 2443- zink: add handlers for some bitfield ops in ntv 2444- zink: set 64bit shader caps in ntv 2445- zink: change function params and asserts to permit 64bit types in ntv 2446- zink: add 64bit glsl basetype handling in ntv 2447- zink: handle 64bit constant loading in ntv 2448- zink: split ubo loading for 64bit types into 2x32bit loads 2449- zink: set nir options for 64bit handling based on feature presence 2450- zink: enable 64bit pipe caps 2451- mesa/st: run nir_lower_point_size_mov on geometry shaders based on cap 2452- mesa/st: do not run lower_psiz_mov on vertex shader if geometry shader is present 2453- mesa/st: tabs -\> spaces in st_program 2454- mesa/st: handle running nir lower passes for ucp and psiz in tess stage 2455- mesa/st: flag ST_NEW_CONSTANTS upon running nir_lower_point_size_mov 2456- mesa/st: set lower_point_size for tes/gs during program update 2457- zink: force stencil format for stencil-only samplers and swizzle the right component 2458- zink: add nir_op_bit_count to ntv 2459- zink: handle nir_op_ibitfield_extract: in ntv 2460- zink: handle nir_op_find_lsb and nir_op_ifind_msb in ntv 2461- zink: move rp hash functions further up in file 2462- zink: fix rp hash table 2463- zink: fix gl_SampleMaskIn handling 2464- zink: don't always run nir_lower_io_arrays_to_elements_no_indirects 2465- zink: add ntv handling for tess shader i/o variables 2466- zink: add handling for tess shader intrinsics 2467- zink: set up ntv init for tess shaders 2468- zink: set scoped barrier flag in nir options 2469- zink: pull xfb info from tess shader when applicable 2470- zink: set tess info in pipeline creation 2471- zink: support PIPE_PRIM_PATCHES 2472- zink: add handling for tcs and tes shader states 2473- zink: only run nir_lower_clip_halfz for last vertex processing stage 2474- zink: add push constant handling to get_storage_class() 2475- zink: add stubs for tess outer/inner level handling 2476- zink: implement passthrough tcs shader injection 2477- zink: handle partial writes to shader outputs 2478- zink: export tess shader pipe caps 2479- doc/features: mark off tessellation for zink 2480- zink: zero VkMemoryRequirements on init 2481- zink: fix debug utils init 2482- zink: handle null ubos 2483- zink: handle 0 as valid pipeline hash value 2484- zink: fix more instance detection stuff 2485- st/pbo: fix pbo uploads without PIPE_CAP_TGSI_VS_LAYER_VIEWPORT and skip gs 2486- zink: avoid replacing valid tcs with injected one 2487- zink: require KHR_maintenance2 for tessellation and set bottom-left origin 2488- zink: fix tess shader i/o variables 2489- zink: add KHR_draw_indirect_count detection 2490- zink: hook up IndirectCount draw commands 2491- zink: enable PIPE_CAP_MULTI_DRAW_INDIRECT(_PARAMS) caps 2492- features: mark off multidraw for zink 2493- radv: avoid oob read during clear 2494- zink: handle dynamic sampler array indexing for arb_gpu_shader5 2495- zink: run nir_lower_tex for offsets if shaderImageGatherExtended is missing 2496- zink: use Offset param for txf ops 2497- zink: implement ARB_texture_gather 2498- zink: handle textureGather with Shadow-type samplers 2499- zink: enable PIPE_CAP_MAX_TEXTURE_GATHER_COMPONENTS 2500- features: mark off textureGather for zink 2501- zink: handle fs interpolation functions in ntv 2502- zink: set PIPE_CAP_MAX_VIEWPORTS 2503- zink: handle gl_SampleMaskIn loading in ntv 2504- zink: always load (gl_InstanceID - gl_BaseInstance) when loading gl_InstanceID 2505- zink: enable PIPE_CAP_START_INSTANCE 2506- zink: handle vertex streams 2507- zink: run nir_lower_dynamic_bo_access 2508- zink: handle arrays of ubos 2509- zink: GLSL 4.00 2510- features: mark off GL 4.0 for zink 2511- zink: GLSL 410 2512- features: mark off GL 4.1 for zink 2513- zink: handle non-const offsets for txf/tg4 ops 2514- nir: preserve explicit_binding in lower_atomics_to_ssbo 2515- zink: clamp shader input/output max values 2516- glcpp: disable 'windows' tests 2517- zink: flag gfx pipeline dirty using newer mechanism 2518- radv: null bo list pointer for null descriptors on update 2519- radv: zero the bo descriptor array when allocating a new set 2520- zink: fix streamout for tess stage 2521- zink: fix slot mapping for legacy gl io with tess stages 2522- zink: handle 1bit undef values in ntv 2523- gallium/trace: add a pipe_screen::get_compiler_options method 2524- mesa/st: clamp scissored clear regions to fb size 2525- zink: unset generated TCS if its parent TESS is unset 2526- zink: fix streamout emission for super-enhanced layouts 2527 2528Nanley Chery (32): 2529 2530- mesa: Add and use \_mesa_has_depth_float_channel 2531- mesa: Clamp some depth values in glClearBufferfv 2532- mesa: Clamp some depth values in glClearBufferfi 2533- iris: Add and use convert_depth_value 2534- iris: Use converted depth in clear_depth_stencil 2535- iris: Disable color fast-clears in iris_copy_region 2536- i965: Disable color fast-clears for miptree copy 2537- intel/blorp: Delete clear color conversions during copies 2538- iris: Stop quantizing the depth clear value 2539- iris: Fix resource ptr in resolve_sampler_views 2540- iris: Drop res variable in resolve_sampler_views 2541- iris: Stop using blorp_hiz_stencil_op 2542- intel/blorp: Drop support for STC_CCS resolves 2543- iris: Move STC case in get_copy_region_aux_settings 2544- iris: Support clears in more GPU-based copies 2545- iris: Don't prepare depth for stencil-aspect blits 2546- iris: Move depth-format assertion out of iris_blit 2547- iris: Use texture preparation helper in iris_blit 2548- iris: Increase use of pipe_resources in iris_blit 2549- iris: Loop through an aspect mask in iris_blit 2550- iris: Blit non-stencil according to aspect_mask 2551- iris: Use single-aspect formats more in iris_blit 2552- iris: Blit stencil according to aspect_mask 2553- iris: Explain how conditional aux accesses work 2554- iris: Make can_fast_clear_depth return constants 2555- iris: Disable conditional fast clears 2556- iris: Delete iris_resolve_conditional_render 2557- iris: Drop fast_clear_color's blorp_flags param 2558- dri: Restrict glthread for CS:GO to radeonsi 2559- gallium: Map \_DRI_IMAGE_FORMAT_NONE to NULL 2560- gallium: Flush GL API resources in eglCreateImage 2561- iris: Disable aux as needed in iris_flush_resource 2562 2563Neha Bhende (3): 2564 2565- meson: Don't build svgadrm on windows 2566- meson.build: Use SSE math for MinGW X86 build as per sse2 option 2567- meson.build: Disable zlib as per -Dzlib option 2568 2569Neil Armstrong (1): 2570 2571- kmsro: sync Android.mk GALLIUM_TARGET_DRIVERS 2572 2573Pavel Asyutchenko (1): 2574 2575- vulkan/overay: fix violation of VUID-VkDeviceCreateInfo-pNext-00373 2576 2577Pierre Moreau (17): 2578 2579- clover: rename platform/device apis using strings 2580- clover/llvm: don't use strings for version handling. 2581- clover/spirv: avoid strings for version handling 2582- clover/api: Add extended versioning query for built-in kernels 2583- clover/api: Add extended versioning query for OpenCL C 2584- clover/spirv: Add version conversion utilities 2585- clover/spirv: Add function checking whether a binary contains SPIR-V 2586- clover/spirv: Change API to use std::string binaries 2587- clover/spirv: Add function checking the SPIR-V version 2588- clover/spirv: Use cl_version for SPIR-V versions (v2) 2589- clover: List supported ILs versions 2590- clover: Implement clCreateProgramWithILKHR 2591- clover: Handle CL_PROGRAM_IL in clGetProgramInfo 2592- clover/api: Implement CL_DEVICE_IL_VERSION 2593- clover: Advertise cl_khr_il_program 2594- clover: Implement clCreateProgramWithIL from OpenCL 2.1 2595- clover: Expose cl_khr_extended_versioning 2596 2597Pierre-Eric Pelloux-Prayer (74): 2598 2599- radeonsi: remove unused NO_RB_PLUS flag 2600- radeonsi: remove AMD_DEBUG=zerovram flag 2601- mesa/gallium: add MESA_MAP_ONCE / PIPE_MAP_ONCE 2602- winsys/amdgpu: make RADEON_ALL_BOS a debug only feature 2603- amdgpu_bo: make cache_entry a extensible array 2604- radeonsi/gfx10: flush gfx cs on ngg -\> legacy transition 2605- ac: use bigger storage for ac_arg::arg_index / ac_shader_args::arg_count 2606- util: add a FALLTROUGH macro 2607- nir: update fallthrough comments 2608- gallium: update fallthrough comments 2609- xxhash: update fallthrough comments 2610- src/mesa: update fallthrough comments 2611- compiler/spirv: update fallthrough comments 2612- radeonsi: update fallthrough comments 2613- gallium/winsys: update fallthrough comments 2614- vbo: update fallthrough comments 2615- gallium/util: update fallthrough comments 2616- softpipe: update fallthrough comments 2617- gallium: update fallthrough comments 2618- radeon: update fallthrough comments 2619- llvmpipe: update fallthrough comments 2620- gallivm: update fallthrough comments 2621- nir/ntt: update fallthrough comments 2622- amd/ac: update fallthrough comments 2623- egl: update fallthrough comments 2624- tgsi: update fallthrough comments 2625- glx: update fallthrough comments 2626- Revert "Revert "radeonsi: use staging buffer uploads for most VRAM buffers"" 2627- gallium/u_threaded: fix staging and non-staging conflicts 2628- gallium/u_threaded: disable forced staging upload at runtime 2629- dlist: do not call \_mesa_lookup_list twice 2630- vbo/dlist: create an index buffer in compile_vertex_list 2631- vbo/dlist: convert LINE_STRIPS to LINES 2632- vbo/dlist: implement primitive merging 2633- util/hash_table: add \_mesa_hash_data_with_seed function 2634- mesa: optimize \_mesa_program_resource_location 2635- vbo/dlist: refactor prim_store/vertex_store allocations 2636- vbo/dlist: avoid splitting draw commands in multiple draws 2637- vbo/dlist: only use merged primitives when it's ok to do so 2638- driconf: add allow_incorrect_primitive_id option 2639- radeonsi: fix si_get_draw_start_count count value 2640- gallium/u_threaded: set has_user_indices = false for merged draws 2641- gallium/u_threaded: fix pipe_resource leak for staging transfer 2642- st/mesa: disable line stippling if pattern is all 1's 2643- driconf: add workaround for Enter The Gungeon 2644- egl: fix EGL_EXT_protected_content/surface mixup 2645- vbo/dlist: use a shared index buffer 2646- vdpau: fix -Wabsolute-value warning 2647- vdpau: fix invalid enum usage 2648- amd/addrlib: use cpp.has_argument() to filter compiler arguments 2649- tesselator: remove unused variable 2650- gallium/vl: merge identical h264/h265 enums 2651- radeonsi: fix redundant initializations 2652- mesa/st: fix redundant initialization 2653- radeonsi: pass radeon_cmdbuf to emit_cache_flush 2654- radeonsi: pass radeon_cmdbuf to si_cp_dma_wait_for_idle 2655- ac/sqtt: add ac_thread_trace_data 2656- ac/radv: move sqtt structs and helpers to amd/common 2657- ac/radv: move radv_rgp.c to ac 2658- ac/sqtt: move rgp/sqtt def to ac 2659- ac/sqtt: move ac_is_thread_trace_complete to ac 2660- ac/sqtt: move radv_get_expected_buffer_size to ac 2661- radeonsi: add radeon_set_uconfig_reg_seq_perfctr 2662- radeonsi: implement SQTT support 2663- ac/rgp: add missing include 2664- dri: enable glthread + radeonsi workaround for CS:GO 2665- st/mesa: consider texture view format for fbo blits 2666- mesa/fbo: don't check_end_texture_render on fb read change 2667- st/mesa: use the correct src format in ReadPixels 2668- radeonsi: invalidate compute sgprs in si_rebind_buffer 2669- radeonsi: inhibit clockgating when using SQTT 2670- radeonsi: properly set SPI_SHADER_PGM_HI_ES 2671- radeonsi: fix read from compute / write from draw sync 2672- radeonsi: fix si_check_render_feedback 2673 2674Rhys Perry (148): 2675 2676- radv/winsys: set has_dedicated_vram in the null winsys 2677- aco: don't combine precise max(min()) to med3 2678- aco: fix combine_constant_comparison_ordering() NaN check with 16/64-bit 2679- aco: disallow various v_add_u32 opts if modifiers are used 2680- aco/tests: initialize debug function 2681- aco/tests: expand optimize.const_comparison_ordering tests 2682- aco/tests: add some more clamp combining tests 2683- nir: add nir_var_mem_ubo to nir_var_read_only_modes 2684- nir: allow reordering of loads from read-only modes 2685- aco: disable omod if the sign of zeros should be preserved 2686- aco: fix fp16 \*0.5 omod 2687- aco/tests: add output modifier tests 2688- aco: don't use SMEM for SSBO stores 2689- aco: create v_mad_u32_u24 2690- nir: add nir_var_vec_indexable_modes 2691- nir/copy_prop_vars,nir/dead_write_vars: ignore read-only loads 2692- nir/loop_analyze: initialize loop variables on demand 2693- nir/search: check instr type before adding to worklist 2694- nir/search: check for changes before adding uses to worklist 2695- nir/deref: add helpers to lazily create paths 2696- nir/copy_prop_vars: use nir_deref_and_path 2697- nir/copy_prop_vars: avoid a duplicate lookup if src == vec_src 2698- aco: don't create v_mov_b32 in v_mul_imm() 2699- aco: count v_mul_lo_u32 as 16 cycles 2700- aco: create vgpr constant copies using v_bfrev_b32 2701- aco: copy constant to sgpr in Builder::v_mul_imm() 2702- aco: try harder to not create v_mul_lo_u32 2703- aco: use v_mul_imm() for some nir_op_imul 2704- aco/tests: add Builder::v_mul_imm() tests 2705- aco: fix v_mul_hi_u32_u24 format 2706- nir/unsigned_upper_bound: fix buffer overflow in search_phi_bcsel 2707- nir/unsigned_upper_bound: decrement num_sources_left before recursing 2708- radv/llvm,aco/ngg: fix large shift exponent in ngg_gs_vertex_lds_addr 2709- aco: fix GS with no outputs 2710- aco/ngg: fix division-by-zero in assertion 2711- nir/lower_non_uniform: improve code with the same texture, sampler indices 2712- nir: fix sampler_lod_parameters_pan indices 2713- nir: use a single canonical list of intrinsic indices 2714- nir: add bit_size_src for when the destination bit size matches a source 2715- nir: add destination bit-size information to more intrinsics 2716- nir: remove useless nir_builder_opcodes.h include 2717- nir: move nir_load_system_value() to nir_builder.h 2718- nir: add generated intrinsic builders 2719- spirv: use intrinsic builders 2720- glsl_to_nir: use intrinsic builders 2721- nir: use intrinsic builders 2722- radv: use intrinsic builders 2723- nir: make intrinsic order in nir_print consistent 2724- nir: fix intrinsic builders on MSVC C++ 2725- nir: fix nir_builder.h on MSVC C++ and GCC7. 2726- d3d12: remove hand-written intrinsic builders 2727- nir: add helpers for chasing resource bindings 2728- nir/opt_load_store_vectorize: use resource binding chasing helpers 2729- ac/nir: use binding chasing helpers 2730- aco: use binding chasing helpers 2731- radv: use FALLTHROUGH macro 2732- aco: use FALLTHROUGH macro 2733- nir/opt_sink: use common instruction removal/insertion helpers 2734- aco: don't assume src=lower when splitting self-intersecting copies 2735- aco: test self-intersecting copies when src=higher 2736- aco: remove sign-extension in constantValue64() 2737- aco: allow 64-bit literals if they can be sign/zero-extended from 32-bit 2738- aco: add get_const/is_constant_representable helpers 2739- aco: use v_lshrrev_b64 for 64-bit VGPR copies on GFX10+ 2740- aco: coalesce constant copies 2741- aco: clear operands in update_renames() 2742- aco: don't fill killed operands in update_renames() 2743- aco: remove rollback code in get_reg_create_vector() 2744- aco: repeat get_reg_create_vector() with increased register demand if fail 2745- aco: use clear() helper instead of writing reg file directly 2746- aco: simplify get_reg_impl() 2747- aco: remove rollback code around parallelcopy creation 2748- aco: remove rollback code for blocked fixed definitions 2749- aco: move update_renames() out of get_reg() 2750- aco: remove rollback code when making an instruction vop3 2751- nir/lower_non_uniform: remove non_uniform flags after lowering 2752- nir: improve divergence analysis for loads with non-uniform resources 2753- nir/opt_access: don't ignore image arrays in process_variable() 2754- nir/opt_access: ignore barriers and coherent qualifier 2755- nir/opt_access: check restrict before marking a variable as readonly 2756- nir/opt_access: don't check restrict in can_reorder() 2757- nir/opt_access: rename can_reorder() and set ACCESS_NON_WRITEABLE in it 2758- nir/opt_access: add basic Vulkan support 2759- nir/opt_access: handle variable pointers 2760- nir/opt_access: consider global stores 2761- nir/opt_access: infer writeonly 2762- compiler: update gl_access_qualifier comments 2763- aco: fix various s_subb_u32 operands to SCC 2764- aco: rename s_subb_u32 operands to borrow 2765- nir/opt_access: don't ignore infer_non_readable 2766- aco: fix mbcnt_amd with wave32 2767- aco: allow divergent mbcnt_amd masks 2768- aco: add block to worklist in mark_block_wqm() 2769- ac/llvm: insert phis before demote kill 2770- aco: fix incorrect address calculation for load_barycentric_at_sample 2771- ac/nir: use llvm.readcyclecounter for LLVM9+ 2772- nir/tests: fix callback for load/store vectorizer tests 2773- nir: allow 5 component vectors 2774- nir,spirv: add sparse texture fetches 2775- nir,spirv: add sparse image loads 2776- nir,spirv: implement SpvOpImageSparseTexelsResident 2777- nir: add sparse_residency_code_and 2778- nir/lower_tex: fix lower_tg4_offsets with sparse fetches 2779- vtn: support SpvCapabilitySparseResidency 2780- radv: implement CREATE_REQUIRE_FULL_SUBGROUPS_BIT with cswave32 2781- nir: gather whether a compute shader uses non-quad subgroup intrinsics 2782- radv: workaround games which assume full subgroups if cswave32 is enabled 2783- nir/load_store_vectorize: don't ignore subgroup memory barriers 2784- nir: add nir_load_store_vectorize_options 2785- nir/load_store_vectorize: add data as callback args 2786- radv: vectorize shader I/O 2787- nir,radv: add and use nir_vectorize_tess_levels() 2788- aco: fix unreachable() for uniform 8/16-bit nir_op_mov from VGPR 2789- aco: fix MIMG_instruction::lwe comment 2790- aco: move MIMG VDATA to its own operand 2791- aco: implement nir_op_vec5 2792- aco: implement sparse texture fetches 2793- aco: implement sparse image loads 2794- aco: form sparse load clauses 2795- ac/nir: implement nir_op_vec5 2796- ac/nir: implement sparse image/texture loads 2797- radv: implement is_sparse_texels_resident and sparse_residency_code_and 2798- radv: support SpvCapabilitySparseResidency 2799- radv/winsys: set has_packed_math_16bit in null winsys 2800- nir/opt_vectorize: fix typo in instr_can_rewrite() 2801- nir/opt_vectorize: fix srcs_equal() with two different non-const 2802- aco: try to better align 8+ dword SGPR vectors 2803- aco: remove can_reorder semantic in get_sync_info_with_hack 2804- radv: add RADV_DEBUG=invariantgeom 2805- radv: set invariantgeom for Shadow of the Tomb Raider 2806- aco: improve nir_op_vec with constant operands 2807- aco/tests: don't rely on argument evaluation order 2808- nir/loop_unroll: unroll more aggressively if it can improve load scheduling 2809- aco: fix convert_to_SDWA() check in add_subdword_definition() 2810- radv,aco: don't use MUBUF for multi-channel loads on GFX8 with robustness2 2811- aco: don't consider a phi trivial if same's register doesn't match the def 2812- radv: round-up num_records division in radv_flush_vertex_descriptors 2813- radv: correctly enable WGP_MODE for NGG and GS 2814- radv: correctly enable WGP_MODE for tessellation control 2815- aco: always set exec_live=false 2816- aco: do not flag all blocks WQM to ensure we enter all nested loops in WQM 2817- aco: add fallback algorithm in get_reg() 2818- aco/lower_phis: fix all_preds_uniform with continue_or_break 2819- aco: add missing usable_read2 check 2820- nir/opt_shrink_vectors: add option to skip shrinking image stores 2821- radv: don't shrink image stores for The Surge 2 2822- radv: don't set sx_blend_opt_epsilon for V_028C70_COLOR_10_11_11 2823- aco: calculate all p_as_uniform and v_readfirstlane_b32 sources in WQM 2824 2825Rob Clark (93): 2826 2827- freedreno: Drop fd_context_lock() and friends 2828- freedreno/drm: Convert to simple_mtx 2829- freedreno: debug cleanup 2830- freedreno: Convert to mesa_log*() 2831- freedreno: Fix spurious flush 2832- freedreno: batch-cache locking 2833- freedreno/a6xx: Texture cache locking 2834- freedreno: Use ctx seqno in batch cache key 2835- freedreno/drm: Make ring refcnt atomic again 2836- freedreno/batch: Move fd_batch_get_prologue() 2837- freedreno: Make fd_context_batch() return a reference 2838- freedreno: Add submit lock 2839- freedreno/drm: Drop growable submit_bos table 2840- freedreno/batch: Cleanup submit immediately after flush 2841- freedreno/drm: Rework APPEND() macro 2842- freedreno: Protect gmem_cache ralloc allocations 2843- mesa/fbo: Fix valgrind complaints 2844- mesa/bufferobj: Fix valgrind complaints 2845- nir: Fix nir_validate fail after nir_lower_tex 2846- freedreno/drm: Add some locking asserts 2847- freedreno/ir3: Add pass to deal with load_uniform base offsets 2848- freedreno/ir3: Fix crash in shader compile fail path 2849- freedreno: emit_marker() cleanup 2850- freedreno: Convert one last mtx_t -\> simple_mtx_t 2851- freedreno/a6xx: Clear control mem at context create 2852- freedreno/drm: Quiet timedout error msg 2853- freedreno/ir3: Fix valgrind complaint about streamout state 2854- util: Add helgrind support for simple_mtx 2855- util: Add helpers for various one-time-init patters 2856- nir: Use get_once() helper for one-time init's 2857- freedreno/ir3: Use get_once() for one-time init 2858- gallium/hud: Use do_once for one-time init 2859- mesa/st: Use do_once for one-time init 2860- util: Fix helgrind complaint about one-time init 2861- mesa: Fix helgrind complaint about one-time init 2862- gallium/trace: Fix helgrind complaint about one-time init 2863- tgsi: Fix helgrind complaint about one-time init 2864- mesa: Synchronize get_gl_override() 2865- util: Add property_get() fallback for android 2866- mesa: Use os_get_option() for MESA_*_OVERRIDE 2867- egl/surfaceless: glthread support 2868- egl/dri2: Drop some pointless ifdeffery 2869- util: Add helper to get FILE\* options 2870- gallium/aux: Add GPU tracepoint mechanism 2871- freedreno: Small log-parser.py cleanup 2872- freedreno: Remove unused fxn 2873- freedreno: Don't emit log/trace points in gmem for nondraw 2874- freedreno: Add GPU tracepoints 2875- freedreno: Add trace-parser.py 2876- freedreno: Remove fd_log() 2877- gallium/aux: Avoid creating queue when traces not enabled 2878- gallium/aux: Split u_tracepoints.[ch\] generation 2879- gallium/aux: Update scons build for u_tracepoints.[ch\] 2880- util: Promote \__builtin_types_compatible_p compat 2881- util: Allow STATIC_ASSERT() everywhere 2882- util+treewide: container_of() cleanup 2883- freedreno/ir3: Fix half-immed decoding issues 2884- freedreno/ir3: Fix mova1 disasm 2885- freedreno/ir3: Add some more disasm test vectors 2886- freedreno/ir3: Move assembler error handling 2887- freedreno/ir3/parser: Reset lexer when input changes 2888- freedreno/ir3: Various cat0 updates 2889- freedreno/ir3/parser: Add new cat0 instructions 2890- freedreno/ir3/parser: cat1 instructions can write relative GPR 2891- freedreno/ir3/parser: cat1 updates (mova1, movmsk) 2892- freedreno/ir3/parser: Handle half-immed 2893- freedreno/ir3: Clean up instruction creation 2894- freedreno/ir3: Cleanup cat6 load instructions 2895- freedreno/ir3/parser: Fix cat6 store encoding 2896- freedreno/ir3/parser: Fix dsxpp/dsypp encoding 2897- freedreno/ir3/parser: Fixup cat5 s2en instructions 2898- freedreno/ir3: Don't set bit for dest conversion for p0.c 2899- freedreno/ir3/parser: Add missing (sat) modifier 2900- freedreno/ir3/parser: Relative gpr/const can have modifiers too 2901- freedreno/ir3/parser: Add initial cat6 IBO instructions 2902- freedreno/ir3: Tweak ldib/resinfo encoding 2903- freedreno/ir3: Add parsing and assembler testing 2904- freedreno/ir3: Don't leak disk_cache 2905- freedreno/ir3: Disambiguate a6xx+ "bindless" instructions 2906- freedreno/ir3: Add cat5/cat6 nonuniform flag 2907- freedreno/ir3/parser: Add ldc support 2908- freedreno/ir3/parser: Fix atomic support 2909- freedreno/ir3/parser: Fix pre-a6xx resinfo 2910- freedreno/ir3/parser: Add ldgb support 2911- freedreno/ir3/parser: Add stgb support 2912- freedreno/ir3/parser: Fixup stg parsing and add more tests 2913- freedreno/ir3: Fix ldg decoding/parsing 2914- freedreno/ir3: Explicitly flag disasm test vectors that don't parse 2915- freedreno/ir3: Fix pre-a6xx ldgb/stib parsing 2916- freedreno/ir3/parser: a6xx ldib/stib parsing 2917- freedreno/ir3/parser: Fix pre-a6xx stib parsing 2918- mesa: Remove \_mesa_destroy_context() 2919- util/u_queue: Ensure num_cpu_mask_bits is valid 2920 2921Robin Ole Heinemann (1): 2922 2923- anv: Add DRM_RDWR flag in anv_gem_handle_to_fd 2924 2925Ruijing Dong (4): 2926 2927- radeon/vcn: hevc main10 profile decoding pitch fix 2928- radeon/vcn: add 0x02 to enc emulation prevention 2929- radeon/vcn: support hevc SAO enc for VCN2+ 2930- radeon/vcn: fix hevc 10bit profile error 2931 2932Ryan Neph (2): 2933 2934- virgl: fix BGRA emulation artifacts during window resize 2935- Revert "virgl: fix BGRA emulation artifacts during window resize" 2936 2937Sagar Ghuge (2): 2938 2939- anv: Invalidate the correct AUX-TT entry 2940- anv: Skip CCS ambiguate which preceed fast-clears 2941 2942Samuel Iglesias Gonsálvez (3): 2943 2944- turnip: implement VK_KHR_depth_stencil_resolve support 2945- turnip: pCounterBufferOffsets can be NULL on vkCmd*TransformFeedbackEXT() 2946- turnip: fix cube map array image size calculation 2947 2948Samuel Pitoiset (155): 2949 2950- aco: fix combining add/sub to b2i if a new dest needs to be allocated 2951- nir/algebraic: optimize bitfield_select(a, iand(a, b), c) 2952- aco/tests: add some tests for combining s_add+s_lshl to s_lshl<n>_add 2953- aco: combine more s_add+s_lshl to s_lshl<n>_add by ignoring uses 2954- aco: introduce a generic label for labelling instructions 2955- aco: add a new Operand flag to indicate that is 16-bit 2956- aco: optimize v_mad_u32_u16 with acc=0 to v_mul_u32_u24 2957- aco: select v_mad_u32_u16 for 16-bit multiplications on GFX9+ 2958- aco: select v_mul_lo_u16 for 16-bit multiplications that can't overflow 2959- aco: optimize v_add_u32(v_mul_lo_u16) -\> v_mad_u32_u16 2960- aco: optimize v_add(v_bcnt(a, 0), b) to v_bcnt(a, b) 2961- ci: update the list of skipped tests for RAVEN 2962- ci: update the list of expected failures for RADV 2963- aco: remove v_{add,sub,subrev}_u32 on GFX8 2964- radv: do VGT_FLUSH when switching NGG -\> legacy on Sienna Cichlid 2965- radv: fix applying the NGG minimum vertex count requirement 2966- radv: don't count unusable vertices to the NGG LDS size 2967- radv: don't subtract max_verts_per_prim from hw_max_esverts on gfx10.3 2968- aco: fix combining max(-min(a, b), c) if a or b uses the neg modifier 2969- radv/winsys: fill real PCIID for Sienna Cichlid and Navy Flounder 2970- radv/winsys: add missing Van Gogh and Dimgrey Cavefish in the null winsys 2971- ci: add list of expected failures for Sienna Cichlid 2972- radv: ignore other blend targets if dual-source blending is enabled 2973- radv: print more debug messages when generating a hang report 2974- radv: append a time string to the hang report dump directory 2975- radv: dump application info in the GPU hang report 2976- radv: add RADV_DEBUG=noumr to disable UMR logs during GPU hang detection 2977- radv: dump BO ranges into bo_ranges.log instead of stderr 2978- ci: fix name of the Sienna Cichlid expected failures file 2979- nir: fix gathering cross invocation info 2980- radv: add new vk_format_is_*() helpers 2981- ac,radv: use better export formats for 8-bit when RB+ isn't allowed 2982- aco/tests: extend the optimize.add_lshl tests to GFX8 2983- aco: add a new Operand flag to indicate that is 24-bit 2984- aco: allow to use the range analysis UB in emit_{sop2,vop2}_instruction() 2985- aco: optimize v_add+s_lshl to v_mad_u32_u24 on GFX6-8 2986- aco: optimize v_add+v_lshlrev to v_mad_u32_u24 on GFX6-8 2987- ac: add gpu_info::has_32bit_predication 2988- radv: use 32-bit predication for conditional rendering on GFX10.3+ 2989- radv: always use 32-bit predication on compute queues 2990- radv: fix missing initialization of the predication value 2991- radv/winsys: fix the sysmem submission path for GFX6 2992- radv: disable SQTT support for unsupported GPUs 2993- radv: fix using bitfields for debug/perftest options 2994- radv: save and dump vertex descriptors during GPU hang detection 2995- radv: enable NGG on GFX10.3 APUs by default 2996- radv: only disable CU2 & CU3 when NGG is enabled 2997- radv: only mask 1 CU for GS/VS waves on GFX10.3 2998- radv: disable WGP_MODE for NGG on GFX10.3 2999- radv/llvm,aco: always split typed vertex buffer loads on GFX6 and GFX10+ 3000- ci: disable check-commits 3001- Revert "radv/llvm,aco: always split typed vertex buffer loads on GFX6 and GFX10+" 3002- vulkan: add missing src_inc to the device select layer 3003- ci: build the Vulkan device select layer 3004- nir: gather if a fragment shader uses sample shading 3005- radv: reduce maxTransformFeedbackBufferDataSize to 512 3006- radv: mark GFX10.3 as a non-conformant Vulkan implementation 3007- radv: fix exporting multiviews with NGG 3008- radv: set the predication boolean as 32-bit if necessary 3009- radv: use 32-bit predication for skipping FCE on GFX10.3+ 3010- radv: fix using FS sample shading if the linker optimized inputs away 3011- ci: update the list of expected failures for RADV/FIJI 3012- radv: enable using MSAA2x and MSAA4x sample locations on GFX10+ 3013- radv: advertise VK_EXT_sample_locations on GFX10+ 3014- ac/surface: initialize the FMASK slice size for GFX9+ 3015- radv: fix clearing FMASK for layered MSAA images on GFX9+ 3016- radv: disable alphaToOne feature 3017- amd/registers: add missing VRS registers 3018- radv: add VK_KHR_fragment_shading_rate but leave it disabled 3019- radv: implement VK_KHR_fragment_shading_rate 3020- radv/llvm: implement fragment shading rate 3021- aco: implement fragment shading rate 3022- radv: track if VRS is enabled to apply a workaround on GFX10.3 3023- radv/llvm: implement a workaround for gl_FragCoord.z with VRS on GFX10.3 3024- aco: implement a workaround for gl_FragCoord.z with VRS on GFX10.3 3025- radv: advertise VK_KHR_fragment_shading_rate on GFX10.3+ 3026- radv: add support for resolving layered depth/stencil images 3027- radv: add missing DB flush after depth/stencil resolve operations 3028- radv: enable TC-compat HTILE for D32_SFLOAT+MSAA on GFX10+ 3029- radv: adjust the maximum number of coverage samples for VRS 3030- radv: fix maxFragmentShadingRateRasterizationSamples 3031- radv: remove useless push constants data when resolving ds attachments 3032- radv: ignore the mutable bit for TC-compatible HTILE 3033- radv: enable VK_EXT_line_rasterization on GFX9 3034- radv: sort the extension table like Khronos 3035- radv: add code that checks if the extension table is sorted correctly 3036- radv: make sure FMASK compression is enabled for MSAA copies 3037- Revert "radv: use 32-bit predication for skipping FCE on GFX10.3+" 3038- radv: dump VA ranges history when a GPU hang is detected 3039- radv: add a Python script to check if a VA was ever valid 3040- radv: disable stippledBresenhamLines on GFX9 3041- nir: fix determining if an addition might overflow for phi sources 3042- radv: disable A2 SNORM/SSCALED/SINT for texel buffers & images on all gens 3043- radv: fix clearing images with vkCmdClear{Color,DepthStencil}Image() 3044- radv: remove unused radv_image::aspects 3045- radv: always clear the SR0/SR1 bits of the HTILE buffer 3046- radv: fix potential HTILE issues for TC-compat images on GFX8 3047- radv: add radv_htile_get_initial_value() and document the HTILE dword 3048- radv: fix TC-compat HTILE images with DST_OPTIMAL on the compute queue 3049- radv: clean up radv_layout_is_htile_compressed() 3050- radv: only load the DS fast clear values for compressed rendering 3051- radv: enable TC-compat HTILE in GENERAL on GFX10+ 3052- aco: fix creating the dest vector when 16-bit vertex fetches are splitted 3053- radv/llvm,aco: always split typed vertex buffer loads on GFX6 and GFX10+ 3054- radv: configure the texture descriptor for TC-compat CMASK on GFX10+ 3055- radv: fix enabling TC-compat HTILE in GENERAL for writes on GFX10+ 3056- radv: fix performance regression by restoring TC-compat HTILE in GENERAL 3057- radv: determine at creation if an image view can be fast cleared 3058- radv: do not predicate FMASK decompression when DCC+MSAA is used 3059- ci: re-mark some depth/stencil resolve CTS as expected failures 3060- radv: fix crashes when fast-clearing in a secondary command buffer 3061- radv: disable TC-compat HTILE in GENERAL for Detroit: Become Human 3062- radv: re-initialize HTILE properly after depth/stencil compute resolves 3063- radv: only re-initialize HTILE after ds compute resolves if compressed 3064- ac/surface: initialize dcc_slice_size on GFX9+ 3065- radv: add support for fast-clearing DCC layers on GFX9+ 3066- radv: clean up radv_decompress_dcc_compute() 3067- radv: do not use predication when the range doesn't cover the whole image 3068- radv: enable DCC for layered color images on GFX10+ 3069- radv: mark VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT as unsupported on GFX6-7 3070- aco: fix inserting expcnt for MIMG on GFX6 3071- ci: mark some sparse tests as expected failures on Pitcairn (GFX6) 3072- radv: mark some sparse texture CTS as expected failures on GFX9 3073- radv: set depth to 1 for subpass resolves using the compute path 3074- radv: decompress DCC for partial resolves using the compute path 3075- radv: fixup DCC after color resolves using the compute path 3076- radv: fix color resolves if the dest image has DCC 3077- radv: fix clearing DCC on GFX9 3078- radv: only use predication if the FCE value is allocated 3079- radv: allocate and initialize the FCE predicate value for CMASK too 3080- radv: update the FCE predicate for fast clears using CMASK 3081- radv: skip fast-clear eliminate for CMASK based on a predicate 3082- ac/surface: store DCC mip info into the surface 3083- radv: prevent fast-clearing uncompressed DCC levels 3084- radv: add support for fast-clearing DCC levels on GFX10+ 3085- radv: do not enable DCC for 3D images with mipmaps on GFX10+ 3086- radv: enable DCC for mipmaps on GFX10+ 3087- radv: disable VK_EXT_sample_locations again on GFX10+ 3088- radv: enable DCC for MSAA on GFX10+ 3089- radv: do not invalidate the L2 metadata cache on compute queues 3090- radv: flush L2 metadata as part of CB/DB flush instead of CS_DONE on GFX9 3091- radv: restore invalidating the vector cache for internal meta operations 3092- radv: flush L2 for images affected by the pipe misaligned issue on GFX10+ 3093- ci: exclude one CTS test that timeout most of the time for RADV CI 3094- radv: fix a sync issue with geometry shader primitives query on GFX10+ 3095- radv: fix overflow when computing the SQTT buffer size 3096- radv: inhibit clock gating when tracing with SQTT 3097- radv: fix separate depth/stencil layout in render pass 3098- radv,aco: fix shifting input VGPRs for the LS VGPR init bug on GFX9 3099- nir/algebraic: mark more optimization with fsat(NaN) as inexact 3100- radv: fix centroid with VRS coarse shading 3101- radv: fix waiting on the last enabled RB for occlusion queries 3102- radv: only apply the MRT output NaN fixup to non-meta shaders 3103- radv: set correct value for OFFCHIP_BUFFERING on GFX10+ 3104- radv: do not scale the depth bias for D16_UNORM depth surfaces 3105 3106Serge Martin (1): 3107 3108- clover: add core clover printf support (v12) 3109 3110Simon Ser (11): 3111 3112- amd/common: introduce ac_surface_print_info 3113- radeonsi: use ac_surface_print_info in si_print_texture_info 3114- radv: add img debug flag 3115- egl: fix typo in wl_drm error message 3116- egl/wayland: remove libwayland \\< 1.18 workaround 3117- ci: skip failing test on lavapipe 3118- radv: fix access to uninitialized radeon_bo_metadata 3119- egl/wayland: add a NULL guard for the authenticate callback 3120- radv: only set BO metadata for the first plane 3121- nouveau/nvc0: fix linear buffer alignment for scan-out/cursors 3122- nouveau/nv50: fix linear buffer alignment for scan-out/cursors 3123 3124Steven Houston (1): 3125 3126- v3dv: VK_KHR_display extension support 3127 3128Tapani Pälli (7): 3129 3130- egl/dri2: fix race between image create and egl_image_target_texture 3131- iris: initialize shared screen->vtbl only once 3132- mesa/st: choose S/D format depending on gl_format passed for readpixels 3133- anv: fix calculation of buffer size in case dynamic size is used 3134- mesa: fix layered framebuffer attachment target check 3135- vbo/dlist: free prim_store->prims when vbo_save is destroyed 3136- i965: use aligned malloc for context instead of ralloc 3137 3138Theogen Ratkin (1): 3139 3140- docs: grammar fixes 3141 3142Thong Thai (4): 3143 3144- frontends/va/postproc: Use the actual image height when blitting 3145- frontends/va/postproc: Convert destination when deinterlacing 3146- gallium: Fix VAAPI postproc blit 3147- frontends/va: Return an error if non-interlaced buffer is not supported 3148 3149Timothy Arceri (1): 3150 3151- glsl: default to compat shaders in compat profile 3152 3153Timur Kristóf (16): 3154 3155- nir: Use src_is_invocation_id in get_deref_info. 3156- aco/optimizer: Only set scc_needed when it is actually needed. 3157- aco/optimizer: Propagate scc_needed label through p_wqm. 3158- aco: Fix NGG GS assert failure from the WG scan. 3159- aco: Skip TCS s_barrier when VS outputs are not stored in the LDS. 3160- aco: Use program->num_waves as maximum in scheduler. 3161- aco: Keep live-though variables and constants spilled. 3162- aco: Spill more optimally before loops. 3163- aco: Note if rasterization can start early. 3164- aco: Wait for stores when NGG or legacy VS can finish early. 3165- ci: Add an expected failures list for Oland (GFX6) 3166- radv: Only enable sparse features on Polaris and newer. 3167- tgsi_to_nir: Fix uniform ranges. 3168- radv/llvm: Fix reporting LDS stats of tess control shaders. 3169- aco: Disallow LSHS temp-only I/O when VS output is written indirectly. 3170- aco: Fix LDS statistics of tess control shaders. 3171 3172Tomeu Vizoso (3): 3173 3174- ci: Temporarily disable jobs on the Collabora lab 3175- Revert "ci: Temporarily disable jobs on the Collabora lab" 3176- ci: Only run the sanity job if there's a MR 3177 3178Tony Wasserka (22): 3179 3180- glsl: Fix -Wshadow warning 3181- util: Fix/silence variable shadowing warnings 3182- meson: Treat LLVM headers as a system dependency 3183- aco: Fix -Wshadow warnings 3184- aco/tests: Fix -Wshadow warnings 3185- aco/tests: Fix -Wunused warnings in release mode 3186- radv: Fix -Wshadow warnings 3187- radv,aco: Compile with -Wshadow when available 3188- radv/query: Avoid hardcoding array size constants 3189- radv/winsys: Fix use of nonexisting struct type in sizeof 3190- aco: Annotate switch fallthroughs 3191- radv,aco: Compile with -Wimplicit-fallthrough when available 3192- gitlab: add RADV bug report template 3193- aco/ra: Add policy parameter to select implementation details for testing 3194- aco/tests: Fix GFX10_3 being printed as gfx11 3195- aco/tests: Allow specifiying the test subvariant in setup_cs 3196- aco/tests: Fix deadlock for too large test lists 3197- aco: Add tests for subdword register allocation 3198- aco/ra: Add some documentation 3199- aco/ra: Fix register allocation for subdword operands 3200- aco/ra: Avoid redundant RegisterFile copies in get_reg_impl 3201- aco: Fix vector::reserve() being called with the wrong size 3202 3203Trevor Woerner (1): 3204 3205- docs/egl.rst: switch true→enabled 3206 3207Vinson Lee (55): 3208 3209- swr: Initialize FetchJit member mpFetchInfo in constructor. 3210- turnip: Remove pipeline NULL check. 3211- draw: Clean up single-use goto statements. 3212- glsl: Initialize ir_variable member field data.is_xfb. 3213- glsl: Fix typos in comments. 3214- microsoft/compiler: Add dxil_nir_lower_16bit_conv prototype. 3215- turnip: Fix file descriptor return. 3216- nvir/gm107: Initialize SchedDataCalculatorGM107 member score. 3217- vdpau: Add missing printf format specifier. 3218- v3dv: Remove unsigned comparison to zero. 3219- frontends/va: Fix \*num_entrypoints check. 3220- clover/spirv: Add missing break for SpvOpExecutionMode case. 3221- turnip: Close sync_fd only if it is a valid file descriptor. 3222- nv50/ir: Initialize GCRA members in constructor. 3223- microsoft/compiler: Add struct dxil_features forward declaration. 3224- microsoft/compiler: Add struct glsl_type forward declaration. 3225- microsoft/compiler: Add scope for declaration in case statement. 3226- r600/sfn: Fix typos. 3227- r600/sfn: Initialize ShaderFromNir members in constructor. 3228- r600/sb: Initialize sb_context members in constructor. 3229- clover: Initialize command_queue member \_props. 3230- nv50/ir: Initialize Program members in constructor. 3231- clover: Fix typo in comment. 3232- scons: Fix build with llvm-12. 3233- amd/addrlib: Initialize Lib members in constructors. 3234- util: Add os_get_page_size support for macOS. 3235- meson: Fix Clang microsoft-enum-value detection. 3236- meson: Fix build with llvm-12. 3237- r600/sfn: Initialize ShaderInputVarying members in constructors. 3238- mesa: Remove extra texObj. 3239- intel/genxml: Avoid generating identical 12.5 and 12 branches. 3240- mesa: Remove cmd_size \\< 0 check. 3241- zink: Fix typos. 3242- glsl: Fix typos in comments. 3243- glsl: Initialize glsl_type member name. 3244- vc4: Fix typos. 3245- d3d12: Fix memory leak if create_gfx_pipeline_state failed. 3246- d3d12: Fix memory leak if create_root_signature failed. 3247- v3d: Fix typos. 3248- nir/tests: Initialize nir_serialize_test member dup. 3249- d3d12: Fix memory leak if state is NULL. 3250- d3d12: Initialize TransitionableResourceState m_SupportsSimultaneousAccess. 3251- turnip: Remove unsigned nonnegative check. 3252- svga: Fix typos in comments. 3253- d3d12: Initialize local_resource member mapped in constructor. 3254- swr: Fix typos. 3255- virgl: Fix typos. 3256- softpipe: Fix typos. 3257- radeonsi: Fix typos. 3258- freedreno/afuc: Replace readfile with os_read_file. 3259- r300: Fix typos. 3260- clover: Add constructor for clover::module. 3261- nv50/ir: Initialize CodeEmitterGM107 members in constructor. 3262- etnaviv: Fix memory leak in etna_vertex_elements_state_create. 3263- aco: Initialize ds_state.front.writeMask. 3264 3265Víctor Manuel Jáquez Leal (1): 3266 3267- frontends/va/context: don't set max_references with num_render_targets 3268 3269Witold Baryluk (3): 3270 3271- zink: Cap PIPE_SHADER_CAP_MAX_CONST_BUFFERS to 32 3272- vulkan/device_select: Store Vulkan vendorID and deviceID as uint32_t 3273- lavapipe: Defer lavapipe warning to CreateDevice 3274 3275X512 (13): 3276 3277- util: implement GET_PROGRAM_NAME for Haiku 3278- util/meson: Add libnetwork dependency for Haiku 3279- targets/haiku-softpipe/meson: add libswpipe.so to install directory 3280- hgl/meson: add version to libGL.so 3281- meson: fix Haiku EGL build; no dri requirement 3282- include: fix export in Haiku OpenGL kit headers 3283- hgl: use local headers instead of system header 3284- frontends/hgl: set state_manager 3285- frontends/hgl: set framebuffer id 3286- aux/driver_ddebug: Normalize pid type from Haiku 3287- targets/haiku-softpipe: Restore GalliumContext 3288- hgl: Major refactor and cleanup 3289- util/u_thread: Disable pthread_barrier_t on Haiku 3290 3291Yevhenii Kharchenko (2): 3292 3293- meson: Add build option to specify default shader disk cache max-size 3294- st/mesa: fix PBO download for TEXTURE_1D_ARRAY textures 3295 3296Yevhenii Kolesnikov (3): 3297 3298- intel/fs: don't spill a register, set by undef 3299- iris: only set point sprite overrides if actually using points 3300- nir/from_ssa: consider defs in sibling blocks 3301 3302Yogesh mohan marimuthu (1): 3303 3304- radeonsi: enable vrs2x2 coarse shading if flat shading (v9) 3305 3306Yuxuan Shui (1): 3307 3308- Add EGL xcb platform 3309 3310Zack Rusin (1): 3311 3312- meson.build: Order the flex/bison by odds of them working 3313 3314cheyang (5): 3315 3316- android: fix build failure with libbacktrace 3317- symbol_table:fix mesa symbol table return scope error 3318- glsl: remove unused state variable 3319- virgl: next_handle variable modify to atomic inc in virgl_object_assign_handle 3320- mesa: glProgramBinary add resource_hash 3321 3322jzielins (5): 3323 3324- swr: Pass draw start information to state update mechanism 3325- swr: fix crashes caused by incorrectly reporting SSBO support 3326- gallium/swr: Fix Windows build 3327- swr: Fix building with LLVM12 3328- swr: Fix crashes on Windows 3329 3330nia (1): 3331 3332- util: Avoid pthread_setaffinity_np on NetBSD 3333 3334yshi18 (1): 3335 3336- iris: fix memleak for query_buffer_uploader 3337