Lines Matching +full:freedreno +full:- +full:rules

1 Mesa 20.0.0 Release Notes / 2020-02-19
21 ---------------
25      bb6db3e54b608d2536d4000b3de7dd3ae115fc114e8acbb5afff4b3bbed04b34  mesa-20.0.0.tar.xz
28 ------------
30 -  OpenGL 4.6 on radeonsi.
31 -  GL_ARB_gl_spirv on radeonsi.
32 -  GL_ARB_spirv_extensions on radeonsi.
33 -  GL_EXT_direct_state_access for compatibility profile.
34 -  VK_AMD_device_coherent_memory on RADV.
35 -  VK_AMD_mixed_attachment_samples on RADV.
36 -  VK_AMD_shader_explicit_vertex_parameter on RADV.
37 -  VK_AMD_shader_image_load_store_lod on RADV.
38 -  VK_AMD_shader_fragment_mask on RADV.
39 -  VK_EXT_subgroup_size_control on RADV/LLVM.
40 -  VK_KHR_separate_depth_stencil_layouts on Intel, RADV.
41 -  VK_KHR_shader_subgroup_extended_types on RADV.
42 -  VK_KHR_swapchain_mutable_format on RADV.
43 -  VK_KHR_shader_float_controls on RADV/ACO.
44 -  GFX6 (Southern Islands) and GFX7 (Sea Islands) support on RADV/ACO.
45 -  Wave32 support for GFX10 (Navi) on RADV/ACO.
46 -  Compilation of Geometry Shaders on RADV/ACO.
47 -  Vulkan 1.2 on Intel, RADV.
48 -  GL_INTEL_shader_integer_functions2 and
52 ---------
54 -  drisw crashes on calling NULL putImage on EGL surfaceless platform
56 -  [radeonsi][vaapi][bisected] invalid VASurfaceID when playing
58 -  [RADV] GPU hangs while the cutscene plays in the game Assassin's
60 -  ACO: The Elder Scrolls Online crashes on startup (Navi)
61 -  Broken rendering of glxgears on S/390 architecture (64bit, BigEndian)
62 -  aco: sun flickering with Assassins Creeds Origins
63 -  !1896 broke ext_image_dma_buf_import piglit tests with radeonsi
64 -  aco: wrong geometry with Assassins Creed Origins on GFX6
65 -  valgrind errors since commit a8ec4082a41
66 -  OSMesa osmesa_choose_format returns a format not supported by
68 -  Build error with VS on WIN
69 -  Using EGL_KHR_surfaceless_context causes spurious "libEGL warning:
71 -  !3460 broke texsubimage test with piglit on zink+anv
72 -  The screen is black when using ACO
73 -  [Regression] JavaFX unbounded VRAM+RAM usage
74 -  radv: implement VK_AMD_shader_explicit_vertex_parameter
75 -  Civilization VI crashes when loading game (AMD Vega Mobile)
76 -  [radeonsi] X-Server crashes when trying to start Guild Wars 2 with
78 -  aco: implement GFX6 support
79 -  Add support for VK_KHR_swapchain_mutable_format
80 -  radv: The Surge 2 crashes in ac_get_elem_bits()
81 -  [Regression] JavaFX unbounded VRAM+RAM usage
82 -  Use the OpenCL dispatch defnitions from OpenCL_Headers
83 -  [regression][ilk,g965,g45] various dEQP-GLES2.functional.shaders.\*
85 -  aco: Dead Rising 4 crashes in lower_to_hw_instr() on GFX6-GFX7
86 -  libvulkan_radeon.so crash with \`free(): double free detected in
88 -  Commit be08e6a causes crash in com.android.launcher3 (Launcher)
89 -  anv: Regression causing issues for radv when there are no Intel
91 -  Mesa no longer compiles with GCC 10
92 -  [Navi/aco] Guild Wars 2 - ring gfx timeout with commit 3bca0af2
93 -  [radv/aco] Regression is causing a soft crash in The Witcher 3
94 -  [bisected] [radeonsi] GPU hangs/resets while playing interlaced
96 -  [radeonsi] MSAA image not copied properly after image store through
98 -  T-Rex and Manhattan onscreen performance issue on Android
99 -  VkSamplerCreateInfo compareEnable not respected
100 -  VkSamplerCreateInfo compareEnable not respected
101 -  Freedreno drm softpin driver implementation leaks memory
102 -  [POLARIS10] VRAM leak involving glTexImage2D with non-NULL data
104 -  [regression][bisected][ivb/byt] crucible test
105    func.push-constants.basic.q0 causes gpu hang
106 -  MR 3096 broke lots of piglit ext_framebuffer_object tests on Raven
107 -  Rise of the Tomb Raider benchmark crash on Dell XPS 7390 2-in-1 w/
109 -  Raven Ridge (2400G): Resident Evil 2 crashes my machine
110 -  Common practice of glGetActiveUniform leads to O(N²) behavior in Mesa
111 -  Rocket League ingame artifacts
112 -  [radv] SteamVR direct mode no longer works
113 -  [ANV] unused create parameters not properly ignored
114 -  [Bisected] Mesa fails to start alacritty with the wayland backend
116 -  [iris] piglit test clip-distance-vs-gs-out fails due to VUE map
117    mismatch between VS <-> GS stages
118 -  [radv] SteamVR direct mode no longer works
119 -  Blocky corruption in The Surge 2
120 -  radeonsi: Floating point exception on R9 270 gpu for a set of traces
121 -  [RADV] [Navi] LOD artifacting in Halo - The Master Chief Collection
123 -  [CTS]
124    dEQP-VK.api.image_clearing.core.clear_color_image.2d.linear.single_layer.r32g32b32\_\*
125    fail on GFX6-GFX8
126 -  Vulkan: Please consider adding another sample count to
128 -  Navi10: Bitrate based encoding with VAAPI/RadeonSI unusable
129 -  [RADV] create parameters not properly ignored
130 -  [regression][bdw,gen9,hsw,icl][iris] gltcs failures on
132 -  Bugs in RadeonSI VAAPI implementation
133 -  [GFX10] Glitch rendering Custom Avatars in Beat Saber
134 -  intel/fs: Check for 16-bit immediates in
136 -  i965/iris: assert when destroy GL context with active query
137 -  Visuals without alpha bits are not sRGB-capable
138 -  swapchain throttling: wait for fence has 1ns timeout
139 -  radeonsi: OpenGL app always produces page fault in gfxhub on Navi 10
140 -  [regression]
141    KHR-GLES31.core.geometry_shader.api.program_pipeline_vs_gs_capture
143 -  [CTS]
144    dEQP-VK.spirv_assembly.instruction.spirv1p4.entrypoint.tess_con_pc_entry_point
146 -  [RADV] SPIR-V warning when compiling shader using storage
148 -  [RADV] The Dead Rising 4 is causing a GPU hang with LLVM backend
149 -  macOS u_thread.h:156:4: error: implicit declaration of function
151 -  [Wine / Vulkan] Doom 2016 Hangs on Main Menu
152 -  NULL resource when playing VP9 video through VDPAU on RX 570
153 -  radeonsi: mpv --vo=vaapi incorrect rendering on gfx9+
154 -  [BSW/BDW] skia lcdblendmode & lcdoverlap test failure
155 -  Create a way to prefer iris vs i965 via driconf
156 -  [Bisected] i965: CS:GO crashes in emit_deref_copy_load_store with
158 -  radv/aco Jedi Fallen Order hair rendering buggy
159 -  Inaccurate information on https://www.mesa3d.org/repository.html
161 -  [RADV] VK_KHR_timeline_semaphore balloons in runtime
162 -  Shadow of Mordor has randomly dancing black shadows on Talion's face
163 -  gen7 crucible failures func.push-constants.basic.q0 and
164    func.shader-subgroup-vote.basic.q0
165 -  GL_EXT_disjoint_timer_query failing with GL_INVALID_ENUM
166 -  Unreal 4 Elemental and MatineeFightScene demos misrender
167 -  gputest gimark has unwanted black liquorice flakes
168 -  triangle strip clipping with GL_FIRST_VERTEX_CONVENTION causes wrong
170 -  [bisected][regression][g45,g965,ilk] piglit arb_fragment_program kil
172 -  glcts crashes since the enablement of ARB_shading_language_include
173 -  Android build broken
174 -  ld.lld: error: duplicate symbol (mesa-19.3.0-rc1)
175 -  Divinity: Original Sin Enhanced Edition(Native) crash on start
176 -  HSW. Tropico 6 and SuperTuxKart have shadows flickering
177 -  GL_EXT_disjoint_timer_query failing with GL_INVALID_ENUM
178 -  glxgears segfaults on POWER / Xvnc
179 -  [regression][bdw,gen9,icl][iris] piglit failures on mesa
181 -  Redundant builds of libmesa_classic and libmesa_gallium
182 -  [IVB,BYT] [Regression] [Bisected] Core dump at launching
183    arb_compute_shader/linker/bug-93840.shader_test
184 -  Vulkan drivers need access to format utils of gallium
185 -  Disabling lower_fragdata_array causes shader-db to crash for some
187 -  GL_EXT_disjoint_timer_query failing with GL_INVALID_ENUM
188 -  Android build broken by commit 9020f51 "util/u_endian: Add error
190 -  radv secure compile feature breaks compilation of RADV on armhf EABI
191    (19.3-rc1)
192 -  radv_debug.c warnings when compiling on 32 bits : cast to pointer
194 -  Meson: Mesa3D build failure with standalone Mingw-w64 multilib
195 -  [regression][bisected] KHR46 VertexArrayAttribFormat has unexpectedly
197 -  textureSize(samplerExternalOES, int) missing in desktop mesa 19.1.7
199 -  zink: implicly casting integers to pointers, warnings on 32-bit
201 -  Objects leaving trails in Firefox with antialias and
205 -------
209 -  clover/llvm: fix build after llvm 10 commit 1dfede3122ee
213 -  drisw: Cache the depth of the X drawable
217 -  pan/midgard: Optimize comparisions with similar operations
218 -  pan/midgard: Move midgard_is_branch_unit to helpers
219 -  pan/midgard: Optimize branches with inverted arguments
220 -  pan/midgard: Fix midgard_compile.h includes
224 -  intel/perf: adapt to platforms like Solaris without d_type in struct
229 -  v3d: adds an extra MOV for any sig.ld\*
230 -  mesa/main/util: moving gallium u_mm to util, remove main/mm
231 -  nir/opt_peephole_select: remove unused variables
232 -  turnip: remove unused descriptor state dirty
236 -  egl: Fix \_eglPointerIsDereferencable w/o mincore()
240 -  haiku/hgl: Fix build via header reordering
244 -  pipe-loader: Build kmsro loader for with all kmsro targets
245 -  pan/midgard: Remove OP_IS_STORE_VARY
246 -  pan/midgard: Add a dummy source for loads
247 -  pan/midgard: Refactor swizzles
248 -  pan/midgard: Eliminate blank_alu_src
249 -  pan/midgard: Use fp32 blend shaders
250 -  pan/midgard: Validate tags when branching
251 -  pan/midgard: Fix quadword_count handling
252 -  pan/midgard: Compute bundle interference
253 -  pan/midgard: Add bizarre corner case
254 -  pan/midgard: offset_swizzle doesn't need dstsize
255 -  pan/midgard: Extend offset_swizzle to non-32-bit
256 -  pan/midgard: Extend swizzle packing for vec4/16-bit
257 -  pan/midgard: Extend default_phys_reg to !32-bit
258 -  panfrost/ci: Update T760 expectations
259 -  pan/midgard: Fix printing of half-registers in texture ops
260 -  pan/midgard: Disassemble half-steps correctly
261 -  pan/midgard: Pass shader stage to disassembler
262 -  pan/midgard: Switch base for vertex texturing on T720
263 -  nir: Add load_output_u8_as_fp16_pan intrinsic
264 -  pan/midgard: Identify ld_color_buffer_u8_as_fp16\*
265 -  pan/midgard: Implement nir_intrinsic_load_output_u8_as_fp16_pan
266 -  pan/midgard: Pack load/store masks
267 -  panfrost: Select format-specific blending intrinsics
268 -  pan/midgard: Add blend shader selection bits for MRT
269 -  pan/midgard: Implement linearly-constrained register allocation
270 -  pan/midgard: Integrate LCRA
271 -  pan/midgard: Remove util/ra support
272 -  pan/midgard: Compute spill costs
273 -  pan/lcra: Use Chaitin's spilling heuristic
274 -  pan/midgard: Copypropagate vector creation
275 -  pan/midgard: Fix copypropagation for textures
276 -  pan/midgard: Generalize texture registers across GPUs
277 -  pan/midgard: Fix vertex texturing on early Midgard
278 -  pan/midgard: Use texture, not textureLod, on early Midgard
279 -  pan/midgard: Disassemble with old pipeline always on T720
280 -  pan/midgard: Prioritize texture registers
281 -  pan/midgard: Expand 64-bit writemasks
282 -  pan/midgard: Implement i2i64 and u2u64
283 -  pan/midgard: Fix mir_round_bytemask_down for !32b
284 -  pan/midgard: Pack 64-bit swizzles
285 -  pan/midgard: Use generic constant packing for 8/64-bit
286 -  pan/midgard: Implement non-aligned UBOs
287 -  pan/midgard: Expose more typesize helpers
288 -  pan/midgard: Fix masks/alignment for 64-bit loads
289 -  pan/midgard: Represent ld/st offset unpacked
290 -  pan/midgard: Use shader stage in mir_op_computes_derivative
291 -  panfrost: Stub out clover callbacks
292 -  panfrost: Pass kernel inputs as uniforms
293 -  panfrost: Disable tiling for GLOBAL resources
294 -  panfrost: Set PIPE_COMPUTE_CAP_ADDRESS_BITS to 64
295 -  pan/midgard: Introduce quirks checks
296 -  panfrost: Add the lod_bias field
297 -  nir: Add load_sampler_lod_paramaters_pan intrinsic
298 -  pan/midgard: Implement load_sampler_lod_paramaters_pan
299 -  pan/midgard: Add LOD bias/clamp lowering
300 -  pan/midgard: Describe quirk MIDGARD_BROKEN_LOD
301 -  pan/midgard: Enable LOD lowering only on buggy chips
302 -  panfrost: Add lcra.c to Android.mk
303 -  pan/midgard: Use lower_tex_without_implicit_lod
304 -  panfrost: Add information about T720 tiling
305 -  panfrost: Implement pan_tiler for non-hierarchy GPUs
306 -  panfrost: Simplify draw_flags
307 -  pan/midgard: Splatter on fragment out
308 -  gitlab-ci: Remove non-default skips from Panfrost
309 -  panfrost: Remove blend shader hack
310 -  panfrost: Update SET_VALUE with information from igt
311 -  panfrost: Rename SET_VALUE to WRITE_VALUE
312 -  gallium/util: Support POLYGON in u_stream_outputs_for_vertices
313 -  pan/midgard: Move spilling code out of scheduler
314 -  pan/midgard: Split spill node selection/spilling
315 -  pan/midgard: Simplify spillability test
316 -  pan/midgard: Remove spill cost heuristic
317 -  pan/midgard: Move bounds checking into LCRA
318 -  pan/midgard: Remove consecutive_skip code
319 -  pan/midgard: Remove code marked "TODO: remove me"
320 -  pan/midgard: Dynamically allocate r26/27 for spills
321 -  pan/midgard: Use no_spill bitmask
322 -  pan/midgard: Don't use no_spill for memory spill src
323 -  pan/midgard: Force alignment for csel_v
324 -  pan/midgard: Don't try to free NULL in LCRA
325 -  pan/midgard: Simplify and fix vector copyprop
326 -  pan/midgard: Fix shift for TLS access
327 -  panfrost: Describe thread local storage sizing rules
328 -  panfrost: Rename unknown_address_0 -> scratchpad
329 -  panfrost: Split stack_shift nibble from unk0
330 -  panfrost: Add routines to calculate stack size/shift
331 -  panfrost: Factor out panfrost_query_raw
332 -  panfrost: Query core count and thread tls alloc
333 -  panfrost: Route stack_size from compiler
334 -  panfrost: Emit SFBD/MFBD after a batch, instead of before
335 -  panfrost: Handle minor cppcheck issues
336 -  pan/midgard: Remove unused ld/st packing hepers
337 -  pan/midgard: Handle misc. cppcheck warnings
338 -  panfrost: Calculate maximum stack_size per batch
339 -  panfrost: Pass size to panfrost_batch_get_scratchpad
340 -  pandecode: Add cast
341 -  panfrost: Move nir_undef_to_zero to Midgard compiler
342 -  panfrost: Move property queries to \_encoder
343 -  panfrost: Add panfrost_model_name helper
344 -  panfrost: Report GPU name in es2_info
345 -  ci: Remove T760/T860 from CI temporarily
346 -  panfrost: Pass blend RT number through
347 -  pan/midgard: Add schedule barrier after fragment writeout
348 -  pan/midgard: Writeout per render target
349 -  pan/midgard: Fix liveness analysis with multiple epilogues
350 -  pan/midgard: Set r1.w magic
351 -  panfrost: Fix FBD issue
352 -  ci: Reinstate Panfrost CI
353 -  panfrost: Remove fbd_type enum
354 -  panfrost: Pack invocation_shifts manually instead of a bit field
355 -  panfrost: Remove asserts in panfrost_pack_work_groups_compute
356 -  panfrost: Simplify sampler upload condition
357 -  panfrost: Don't double-create scratchpad
358 -  panfrost: Add PAN_MESA_DEBUG=precompile for shader-db
359 -  panfrost: Let precompile imply shaderdb
360 -  panfrost: Handle empty shaders
361 -  pan/midgard: Use a reg temporary for mutiple writes
362 -  pan/midgard: Hoist temporary coordinate for cubemaps
363 -  pan/midgard: Set .shadow for shadow samplers
364 -  pan/midgard: Set Z to shadow comparator for 2D
365 -  pan/midgard: Add uniform/work heuristic
366 -  pan/midgard: Implement textureOffset for 2D textures
367 -  pan/midgard: Fix crash with txs
368 -  pan/midgard: Lower txd with lower_tex
369 -  panfrost: Decode shader types in pantrace shader-db
370 -  pan/decode: Skip COMPUTE in blobber-db
371 -  pan/decode: Prefix blobberdb with MESA_SHADER\_\*
372 -  pan/decode: Append 0:0 spills:fills to blobber-db
373 -  pan/midgard: Fix disassembler cycle/quadword counting
374 -  pan/midgard: Bounds check lcra_restrict_range
375 -  pan/midgard: Extend IS_VEC4_ONLY to arguments
376 -  pan/midgard: Clamp LOD register swizzle
377 -  pan/midgard: Expand swizzle for texelFetch
378 -  pan/midgard: Fix fallthrough from offset to comparator
379 -  pan/midgard: Do witchcraft on texture offsets
380 -  pan/midgard: Generalize temp coordinate to non-2D
381 -  pan/midgard: Implement shadow cubemaps
382 -  pan/midgard: Enable lower_(un)pack\_\* lowering
383 -  pan/midgard: Support loads from R11G11B10 in a blend shader
384 -  pan/midgard: Add mir_upper_override helper
385 -  pan/midgard: Compute destination override
386 -  panfrost: Rename pan_instancing.c -> pan_attributes.c
387 -  panfrost: Factor batch/resource out of instancing routines
388 -  panfrost: Move instancing routines to encoder/
389 -  panfrost: Factor out panfrost_compute_magic_divisor
390 -  panfrost: Fix off-by-one in pan_invocation.c
391 -  pan/decode: Fix reference computation for invocations
392 -  panfrost: Slight cleanup of Gallium's pan_attribute.c
393 -  panfrost: Remove pan_shift_odd
394 -  pan/decode: Handle gl_VertexID/gl_InstanceID
395 -  panfrost: Unset vertex_id_zero_based
396 -  pan/midgard: Factor out emit_attr_read
397 -  pan/midgard: Lower gl_VertexID/gl_InstanceID to attributes
398 -  panfrost: Extend attribute_count for vertex builtins
399 -  panfrost: Route gl_VertexID through cmdstream
400 -  pan/midgard: Fix minor typo
401 -  panfrost: Remove MALI_SPECIAL_ATTRIBUTE_BASE defines
402 -  panfrost: Update information on fixed attributes/varyings
403 -  panfrost: Remove MALI_ATTR_INTERNAL
404 -  panfrost: Inline away MALI_NEGATIVE
405 -  panfrost: Implement remaining texture wrap modes
406 -  panfrost: Add pan_attributes.c to Android.mk
407 -  panfrost: Add missing #include in common header
408 -  panfrost: Remove mali_alt_func
409 -  panfrost; Update comment about work/uniform_count
410 -  panfrost: Remove 32-bit next_job path
411 -  glsl: Set .flat for gl_FrontFacing
412 -  pan/midgard: Promote tilebuffer reads to 32-bit
413 -  pan/midgard: Use type-appropriate st_vary
414 -  pan/midgard: Implement flat shading
415 -  panfrost: Identify glProvokingVertex flag
416 -  panfrost: Disable some CAPs we want lowered
417 -  panfrost: Implement integer varyings
418 -  panfrost: Remove MRT indirection in blend shaders
419 -  panfrost: Respect glPointSize()
420 -  pan/midgard: Convert fragment writeout to proper branches
421 -  pan/midgard: Remove prepacked_branch
422 -  panfrost: Handle RGB16F colour clear
423 -  panfrost: Pack MRT blend shaders into a single BO
424 -  pan/midgard: Fix memory corruption in constant combining
425 -  pan/midgard: Use better heuristic for shader termination
426 -  pan/midgard: Generalize IS_ALU and quadword_size
427 -  pan/midgard: Generate MRT writeout loops
428 -  pan/midgard: Remove old comment
429 -  pan/midgard: Identity ld_color_buffer as 32-bit
430 -  pan/midgard: Use upper ALU tags for MFBD writeout
431 -  panfrost: Texture from Z32F_S8 as R32F
432 -  panfrost: Support rendering to non-zero Z/S layers
433 -  panfrost: Implement sRGB blend shaders
434 -  panfrost: Cleanup tiling selection logic
435 -  panfrost: Report MSAA 4x supported for dEQP
436 -  panfrost: Handle PIPE_FORMAT_R10G10B10A2_USCALED
437 -  panfrost: Respect constant buffer_offset
438 -  panfrost: Adjust for mismatch between hardware/Gallium in arrays/cube
439 -  pan/midgard: Account for z/w flip in texelFetch
440 -  panfrost: Don't double-flip Z/W for 2D arrays
441 -  pan/midgard: Support indirect UBO offsets
442 -  panfrost: Fix linear depth textures
443 -  pan/midgard: Bytemasks should round up, not round down
444 -  panfrost: Identify un/pack colour opcodes
445 -  pan/midgard: Fix recursive csel scheduling
446 -  panfrost: Expose some functionality with dEQP flag
447 -  panfrost: Compile tiling routines with -O3
448 -  panfrost,lima: De-Galliumize tiling routines
449 -  panfrost: Rework linear<--->tiled conversions
450 -  panfrost: Add pandecode entries for ASTC/ETC formats
451 -  panfrost: Fix crash in compute variant allocation
452 -  panfrost: Drop mysterious zero=0xFFFF field
453 -  panfrost: Don't use implicit mali_exception_status enum
454 -  pan/decode: Remove last_size
455 -  pan/midgard: Remove pack_color define
456 -  pan/decode: Remove SHORT_SLIDE indirection
457 -  panfrost: Fix 32-bit warning for \`indices\`
458 -  pan/decode: Drop MFBD compute shader stuff
459 -  pan/midgard: Record TEXTURE_OP_BARRIER
460 -  pan/midgard: Disassemble barrier instructions
461 -  pan/midgard: Validate barriers use a barrier tag
462 -  pan/midgard: Handle tag 0x4 as texture
463 -  pan/midgard: Remove float_bitcast
464 -  pan/midgard: Fix missing prefixes
465 -  pan/midgard: Don't crash with constants on unknown ops
466 -  pan/midgard: Use fprintf instead of printf for constants
470 -  lima: Beautify stream dumps
471 -  lima: Parse VS and PLBU command stream while making a dump
472 -  lima/streamparser: Fix typo in vs semaphore parser
473 -  lima/streamparser: Add findings introduced with gl_PointSize
474 -  lima/parser: Some fixes and cleanups
475 -  lima/parser: Add RSW parsing
476 -  lima/parser: Add texture descriptor parser
477 -  lima: Rotate dump files after each finished pp frame
478 -  lima: Fix dump file creation
479 -  lima/parser: Fix rsw parser
480 -  lima/parser: Fix VS cmd stream parser
481 -  lima/parser: Make rsw alpha blend parsing more readable
482 -  lima: Add stencil support
483 -  lima: Fix alpha blending
487 -  vulkan/wsi: disable the hardware cursor
491 -  main: fix several 'may be used uninitialized' warnings
492 -  glsl: fix an incorrect max_array_access after optimization of
494 -  glsl: fix a binding points assignment for ssbo/ubo arrays
495 -  glsl/nir: do not change an element index to have correct block name
496 -  mesa/st: fix a memory leak in get_version
500 -  util: import xxhash
501 -  util: move fnv1a hash implementation into its own header
502 -  util/hash_table: replace \_mesa_hash_data's fnv1a hash function with
504 -  util/hash_table: added hash functions for integer types
505 -  util/hash_table: update users to use new optimal integer hash
510 -  intel: Add device info for 1x4x6 Jasper Lake
511 -  intel: Add pci-ids for Jasper Lake
515 -  lima: fix stride in texture descriptor
516 -  lima: add layer_stride field to lima_resource struct
517 -  lima: introduce ppir_op_load_coords_reg to differentiate between
520 -  lima: add cubemap support
521 -  lima/ppir: add lod-bias support
525 -  radv: Fix timeout handling in syncobj wait.
526 -  radv: Remove \_mesa_locale_init/fini calls.
527 -  turnip: Remove \_mesa_locale_init/fini calls.
528 -  anv: Remove \_mesa_locale_init/fini calls.
529 -  radv: Fix disk_cache_get size argument.
530 -  radv: Close all unnecessary fds in secure compile.
531 -  radv: Do not change scratch settings while shaders are active.
532 -  radv: Allocate cmdbuffer space for buffer marker write.
533 -  radv: Enable VK_KHR_buffer_device_address.
534 -  amd/llvm: Refactor ac_build_scan.
535 -  radv: Unify max_descriptor_set_size.
536 -  radv: Fix timeline semaphore refcounting.
537 -  radv: Fix RGBX Android<->Vulkan format correspondence.
538 -  amd/common: Fix tcCompatible degradation on Stoney.
539 -  amd/common: Always use addrlib for HTILE tc-compat.
540 -  radv: Limit workgroup size to 1024.
541 -  radv: Expose all sample counts for integer formats as well.
542 -  amd/common: Handle alignment of 96-bit formats.
543 -  nir: Add clone/hash/serialize support for non-uniform tex
545 -  nir: print non-uniform tex fields.
546 -  amd/common: Always initialize gfx9 mipmap offset/pitch.
547 -  turnip: Use VK_NULL_HANDLE instead of NULL.
548 -  meson: Enable -Werror=int-conversion.
549 -  Revert "amd/common: Always initialize gfx9 mipmap offset/pitch."
550 -  radv: Only use the gfx mipmap level offset/pitch for linear textures.
551 -  spirv: Fix glsl type assert in spir2nir.
552 -  radv: Emit a BATCH_BREAK when changing pixel shaders or
554 -  radv: Use new scanout gfx9 metadata flag.
555 -  radv: Disable VK_EXT_sample_locations on GFX10.
556 -  radv: Remove syncobj_handle variable in header.
557 -  radv: Expose VK_KHR_swapchain_mutable_format.
558 -  radv: Allow DCC & TC-compat HTILE with
560 -  radv: Do not set SX DISABLE bits for RB+ with unused surfaces.
564 -  llvmpipe: use ppc64le/ppc64 Large code model for JIT-compiled shaders
568 -  util/os_socket: Include unistd.h to fix build error
572 -  panfrost: MALI_DEPTH_TEST is actually MALI_DEPTH_WRITEMASK
573 -  panfrost: Destroy the upload manager allocated in
575 -  panfrost: Release the ctx->pipe_framebuffer ref
576 -  panfrost: Move BO cache related fields to a sub-struct
577 -  panfrost: Try to evict unused BOs from the cache
578 -  gallium: Fix the ->set_damage_region() implementation
579 -  panfrost: Make sure we reset the damage region of RTs at flush time
580 -  panfrost: Remove unneeded phi nodes
581 -  panfrost/midgard: Fix swizzle for store instructions
582 -  panfrost/midgard: Print the actual source register for store
584 -  panfrost/midgard: Use a union to manipulate embedded constants
585 -  panfrost/midgard: Rework mir_adjust_constants() to make it type/size
587 -  panfrost/midgard: Make sure promote_fmov() only promotes 32-bit imovs
588 -  panfrost/midgard: Factorize f2f and u2u handling
589 -  panfrost/midgard: Add f2f64 support
590 -  panfrost/midgard: Fix mir_print_instruction() for branch instructions
591 -  panfrost/midgard: Add 64 bits float <-> int converters
592 -  panfrost/midgard: Add missing lowering passes for type/size
594 -  panfrost/midgard: Add a condense_writemask() helper
595 -  panfrost/midgard: Prettify embedded constant prints
596 -  panfrost: Fix the damage box clamping logic
600 -  turnip: Update tu_query_pool with turnip-specific fields
601 -  turnip: Implement vkCreateQueryPool for occlusion queries
602 -  turnip: Implement vkCmdBeginQuery for occlusion queries
603 -  turnip: Implement vkCmdEndQuery for occlusion queries
604 -  turnip: Update query availability on render pass end
605 -  turnip: Implement vkGetQueryPoolResults for occlusion queries
606 -  turnip: Implement vkCmdResetQueryPool
607 -  turnip: Implement vkCmdCopyQueryPoolResults for occlusion queries
608 -  anv: Properly fetch partial results in vkGetQueryPoolResults
609 -  anv: Handle unavailable queries in vkCmdCopyQueryPoolResults
610 -  turnip: Enable occlusionQueryPrecise
611 -  turnip: Free event->bo on vkDestroyEvent
612 -  turnip: Fix vkGetQueryPoolResults with available flag
613 -  turnip: Fix vkCmdCopyQueryPoolResults with available flag
617 -  s/APIENTRY/GLAPIENTRY/ in teximage.c
618 -  nir: fix a couple signed/unsigned comparison warnings in
620 -  Call shmget() with permission 0600 instead of 0777
621 -  nir: no-op C99 \_Pragma() with MSVC
625 -  util/vector: Fix u_vector_foreach when head rolls over
629 -  spirv: Don't leak GS initialization to other stages
630 -  glsl: Check earlier for MaxShaderStorageBlocks and MaxUniformBlocks
631 -  glsl: Check earlier for MaxTextureImageUnits and MaxImageUniforms
632 -  anv: Initialize depth_bounds_test_enable when not explicitly set
633 -  spirv: Consider the sampled_image case in wa_glslang_179 workaround
634 -  intel/fs: Lower 64-bit MOVs after lower_load_payload()
635 -  intel/fs: Fix lowering of dword multiplication by 16-bit constant
636 -  intel/vec4: Fix lowering of multiplication by 16-bit constant
637 -  anv/gen12: Temporarily disable VK_KHR_buffer_device_address (and EXT)
638 -  spirv: Implement SPV_KHR_non_semantic_info
639 -  panfrost: Fix Makefile.sources
640 -  anv: Drop unused function parameter
641 -  anv: Ignore some CreateInfo structs when rasterization is disabled
642 -  intel/fs: Only use SLM fence in compute shaders
643 -  spirv: Drop EXT for PhysicalStorageBuffer symbols
644 -  spirv: Handle PhysicalStorageBuffer in memory barriers
645 -  nir: Add missing nir_var_mem_global to various passes
646 -  intel/fs: Add FS_OPCODE_SCHEDULING_FENCE
647 -  intel/fs: Add workgroup_size() helper
648 -  intel/fs: Don't emit fence for shared memory if only one thread is
650 -  intel/fs: Don't emit control barrier if only one thread is used
651 -  anv: Always initialize target_stencil_layout
652 -  intel/compiler: Add names for SHADER_OPCODE_[IU]SUB_SAT
653 -  nir: Make nir_deref_path_init skip trivial casts
657 -  egl: Mention if swrast is being forced
661 -  drm-shim: fix EOF case
662 -  etnaviv: rs: upsampling is not supported
663 -  etnaviv: add drm-shim
664 -  etnaviv: drop not used config_out function param
665 -  etnaviv: use a more self-explanatory param name
666 -  etnaviv: handle 8 byte block in tiling
667 -  etnaviv: add support for extended pe formats
668 -  etnaviv: fix integer vertex formats
669 -  etnaviv: use NORMALIZE_SIGN_EXTEND
670 -  etnaviv: fix R10G10B10A2 vertex format entries
671 -  etnaviv: handle integer case for GENERIC_ATTRIB_SCALE
672 -  etnaviv: remove dead code
673 -  etnaviv: remove not used etna_bits_ones(..)
674 -  etnaviv: drop compiled_rs_state forward declaration
675 -  etnaviv: update resource status after flushing
676 -  gallium: add PIPE_CAP_MAX_VERTEX_BUFFERS
677 -  etnaviv: check if MSAA is supported
678 -  etnaviv: gc400 does not support any vertex sampler
679 -  etnaviv: use a better name for FE_VERTEX_STREAM_UNK14680
680 -  etnaviv: move state based texture structs
681 -  etnaviv: move descriptor based texture structs
682 -  etnaviv: add deqp debug option
683 -  etnaviv: drop default state for PE_STENCIL_CONFIG_EXT2
684 -  etnaviv: drm-shim: add GC400
688 -  nir: Fix non-determinism in lower_global_vars_to_local
689 -  radv: Rename ac_arg_regfile
690 -  ac: Add a shared interface between radv, radeonsi, LLVM and ACO
691 -  ac/nir, radv, radeonsi: Switch to using ac_shader_args
692 -  radv: Move argument declaration out of nir_to_llvm
693 -  aco: Constify radv_nir_compiler_options in isel
694 -  aco: Use radv_shader_args in aco_compile_shader()
695 -  aco: Split vector arguments at the beginning
696 -  aco: Make num_workgroups and local_invocation_ids one argument each
697 -  radv: Replace supports_spill with explict_scratch_args
698 -  aco: Use common argument handling
699 -  aco: Make unused workgroup id's 0
700 -  nir: Maintain the algebraic automaton's state as we work.
701 -  a6xx: Add more CP packets
702 -  freedreno: Use new macros for CP_WAIT_REG_MEM and CP_WAIT_MEM_GTE
703 -  freedreno: Fix CP_MEM_TO_REG flag definitions
704 -  freedreno: Document CP_COND_REG_EXEC more
705 -  freedreno: Document CP_UNK_A6XX_55
706 -  freedreno: Document CP_INDIRECT_BUFFER_CHAIN
710 -  panfrost: Fix panfrost_bo_access memory leak
711 -  panfrost: Fix headers and gpu_headers memory leak
715 -  aco: fix immediate offset for spills if scratch is used
716 -  aco: only use single-dword loads/stores for spilling
717 -  aco: fix accidential reordering of instructions when scheduling
718 -  aco: workaround Tonga/Iceland hardware bug
719 -  aco: fix invalid access on Pseudo_instructions
720 -  aco: preserve kill flag on moved operands during RA
721 -  aco: rematerialize s_movk instructions
722 -  aco: check if SALU instructions are predeceeded by exec when
724 -  aco: value number instructions using the execution mask
725 -  aco: use s_and_b64 exec to reduce uniform booleans to one bit
726 -  amd/llvm: Add Subgroup Scan functions for SI
727 -  radv: Enable Subgroup Arithmetic and Clustered for SI
728 -  aco: don't value-number instructions from within a loop with ones
730 -  aco: don't split live-ranges of linear VGPRs
731 -  aco: fix a couple of value numbering issues
732 -  aco: refactor visit_store_fs_output() to use the Builder
733 -  aco: Initial GFX7 Support
734 -  aco: SI/CI - fix sampler aniso
735 -  aco: fix SMEM offsets for SI/CI
736 -  aco: implement nir_op_fquantize2f16 for SI/CI
737 -  aco: only use scalar loads for readonly buffers on SI/CI
738 -  aco: implement nir_op_isign on SI/CI
739 -  aco: move buffer_store data to VGPR if needed
740 -  aco: implement quad swizzles for SI/CI
741 -  aco: recognize SI/CI SMRD hazards
742 -  aco: fix disassembly of writelane instructions.
743 -  aco: split read/writelane opcode into VOP2/VOP3 version for SI/CI
744 -  aco: implement 64bit VGPR shifts for SI/CI
745 -  aco: make 1/2*PI a literal constant on SI/CI
746 -  aco: implement 64bit i2b for SI /CI
747 -  aco: implement 64bit ine/ieq for SI/CI
748 -  aco: disable disassembly for SI/CI due to lack of support by LLVM
749 -  radv: only flush scalar cache for SSBO writes with ACO on GFX8+
750 -  aco: flush denorms after fmin/fmax on pre-GFX9
751 -  aco: don't use a scalar temporary for reductions on GFX10
752 -  aco: implement (clustered) reductions for SI/CI
753 -  aco: implement inclusive_scan for SI/CI
754 -  aco: implement exclusive scan for SI/CI
755 -  radv: disable Youngblood app profile if ACO is used
756 -  aco: return to loop_active mask at continue_or_break blocks
757 -  radv: Enable ACO on GFX7 (Sea Islands)
758 -  aco: use soffset for MUBUF instructions on SI/CI
759 -  aco: improve readfirstlane after uniform ssbo loads on GFX7
760 -  aco: propagate temporaries into expanded vectors
761 -  nir: fix printing of var_decl with more than 4 components.
762 -  aco: compact various Instruction classes
763 -  aco: compact aco::span<T> to use uint16_t offset and size instead of
765 -  aco: fix unconditional demote_to_helper
766 -  aco: rework lower_to_cssa()
767 -  aco: handle phi affinities transitively through parallelcopies
768 -  aco: ignore parallelcopies to the same register on jump threading
769 -  aco: fix combine_salu_not_bitwise() when SCC is used
770 -  aco: reorder VMEM operands in ACO IR
771 -  aco: fix register allocation with multiple live-range splits
772 -  aco: simplify adjust_sample_index_using_fmask() & get_image_coords()
773 -  aco: simplify gathering of MIMG address components
774 -  docs: add new features for RADV/ACO.
775 -  aco: fix image_atomic_cmp_swap
779 -  Revert "st/dri: do FLUSH_VERTICES before calling flush_resource"
780 -  Revert "gallium: add st_context_iface::flush_resource to call
785 -  intel/blorp: Fix usage of uninitialized memory in key hashing
786 -  i965/program_cache: Lift restriction on shader key size
787 -  intel/blorp: Fix usage of uninitialized memory in key hashing
788 -  intel/fs: Do not lower large local arrays to scratch on gen7
789 -  i965: Unify CC_STATE and BLEND_STATE atoms on Haswell as a workaround
790 -  glsl: Add varyings to "zero-init of uninitialized vars" workaround
791 -  drirc: Add glsl_zero_init workaround for GpuTest
792 -  iris/query: Implement PIPE_QUERY_GPU_FINISHED
793 -  iris: Fix value of out-of-bounds accesses for vertex attributes
794 -  i965: Do not set front_buffer_dirty if there is no front buffer
795 -  st/mesa: Handle the rest renderbuffer formats from OSMesa
796 -  st/nir: Unify inputs_read/outputs_written before serializing NIR
800 -  nir/serialize: pack function has name and entry point into flags.
801 -  nir/serialize: fix serializing functions with no implementations.
802 -  spirv: don't store 0 to cs.ptr_size for non kernel stages.
803 -  spirv: get the correct type for function returns.
804 -  spirv/nir/opencl: handle some multiply instructions.
805 -  nir: add 64-bit ufind_msb lowering support. (v2)
806 -  nouveau: request ufind_msb64 lowering in the frontend.
807 -  vtn/opencl: add clz support
808 -  nir: fix deref offset builder
809 -  llvmpipe: initial query buffer object support. (v2)
810 -  docs: add llvmpipe to ARB_query_buffer_object.
811 -  gallivm: split out the flow control ir to a common file.
812 -  gallivm: nir->tgsi info convertor (v2)
813 -  gallivm: add popcount intrinsic wrapper
814 -  gallivm: add cttz wrapper
815 -  gallivm: add selection for non-32 bit types
816 -  gallivm: add nir->llvm translation (v2)
817 -  draw: add nir info gathering and building support
818 -  gallium: add nir lowering passes for the draw pipe stages. (v2)
819 -  gallivm: add swizzle support where one channel isn't defined.
820 -  llvmpipe: add initial nir support
821 -  nir/samplers: don't zero samplers_used/txf.
822 -  llvmpipe/images: handle undefined atomic without crashing
823 -  gallivm/llvmpipe: add support for front facing in sysval.
824 -  llvmpipe: enable texcoord semantics
825 -  gallium/scons: fix graw-xlib build on OSX.
826 -  llvmpipe: add queries disabled flag
827 -  llvmpipe: disable occlusion queries when requested by state tracker
828 -  draw: add support for collecting primitives generated outside
830 -  llvmpipe: enable support for primitives generated outside streamout
831 -  aco: handle gfx7 int8/10 clamping on exports
832 -  gallivm: add bitfield reverse and ufind_msb
833 -  llvmpipe/nir: handle texcoord requirements
834 -  gallivm: fix transpose for when first channel isn't created
835 -  gallivm: fix perspective enable if usage_mask doesn't have 0 bit set
836 -  gallivm/nir: cleanup code and call cmp wrapper
837 -  gallivm/nir: copy compare ordering code from tgsi
838 -  gallivm: add base instance sysval support
839 -  gallivm/draw: add support for draw_id system value.
840 -  gallivm: fixup base_vertex support
841 -  llvmpipe: enable ARB_shader_draw_parameters.
842 -  vtn: convert vload/store to single value loops
843 -  vtn/opencl: add shuffle/shuffle support
844 -  gallivm/nir: wrap idiv to avoid divide by 0 (v2)
845 -  llvmpipe: switch to NIR by default
846 -  nir: sanitize work group intrinsics to always be 32-bit.
847 -  gallivm: add 64-bit const int creator.
848 -  llvmpipe/gallivm: add kernel inputs
849 -  gallivm: add support for 8-bit/16-bit integer builders
850 -  gallivm: pick integer builders for alu instructions.
851 -  gallivm/nir: allow 8/16-bit conversion and comparison.
852 -  tgsi/mesa: handle KERNEL case
853 -  gallivm/llvmpipe: add support for work dimension intrinsic.
854 -  gallivm/llvmpipe: add support for block size intrinsic
855 -  gallivm/llvmpipe: add support for global operations.
856 -  llvmpipe: handle serialized nir as a shader type.
857 -  llvmpipe: add support for compute shader params
858 -  llvmpipe/nir: use nir_max_vec_components in more places
859 -  gallivm: handle non-32 bit undefined
860 -  llvmpipe: lower hadd/add_sat
861 -  gallivm/nir: lower packing
862 -  gallivm/nir: add vec8/16 support
863 -  llvmpipe: add debug option to enable OpenCL support.
864 -  gallivm: fixup const int64 builder.
865 -  llvmpipe: enable ARB_shader_group_vote.
866 -  gallium/util: add multi_draw_indirect to util_draw_indirect.
867 -  llvmpipe: enable driver side multi draw indirect
868 -  llvmpipe: add support for ARB_indirect_parameters.
869 -  llvmpipe: add ARB_derivative_control support
870 -  gallivm: fix gather component handling.
871 -  llvmpipe: fix some integer instruction lowering.
872 -  galllivm: fix gather offset casting
873 -  gallivm: fix find lsb
874 -  gallivm/nir: add missing break for isub.
878 -  .mailmap: use correct email address
882 -  virgl: support emulating planar image sampling
886 -  gallium/swr: Enable support bptc format.
887 -  docs/features: mark GL_ARB_texture_compression_bptc as done for
892 -  gallium: enable INTEL_PERFORMANCE_QUERY
893 -  iris: INTEL performance query implementation
894 -  gallium: check all planes' pipe formats in case of multi-samplers
898 -  radeonsi: Clear uninitialized variable
902 -  st_get_external_sampler_key: improve error message
906 -  zink: make sure src image is transfer-src-optimal
910 -  Bump VERSION to 20.0.0-devel
911 -  docs/new_features: Empty the feature list for the 20.0 cycle
912 -  nir: correct use of identity check in python
913 -  r200: use preprocessor for big vs little endian checks
914 -  r100: Use preprocessor to select big vs little endian paths
915 -  dri/osmesa: use preprocessor for selecting endian code paths
916 -  util/u_endian: Use \_WIN32 instead of \_MSC_VER
917 -  util/u_endian: set PIPE_ARCH_*_ENDIAN to 1
918 -  mesa/main: replace uses of \_mesa_little_endian with preprocessor
919 -  mesa/swrast: replace instances of \_mesa_little_endian with
921 -  mesa/main: delete now unused \_mesa_little_endian
922 -  gallium/osmesa: Use PIPE_ARCH_*_ENDIAN instead of little_endian
924 -  util: rename PIPE_ARCH_*_ENDIAN to UTIL_ARCH_*_ENDIAN
925 -  util/u_endian: Add error checks
926 -  meson: Add dep_glvnd to egl deps when building with glvnd
927 -  docs: add release notes for 19.2.3
928 -  docs: add sha256 sum to 19.2.3 release notes
929 -  docs: update calendar, add news item and link release notes for
931 -  meson: gtest needs pthreads
932 -  gallium/osmesa: Convert osmesa test to gtest
933 -  osmesa/tests: Extend render test to cover other working cases
934 -  util: Use ZSTD for shader cache if possible
935 -  docs: Add release notes for 19.2.4
936 -  docs: Add SHA256 sum for for 19.2.4
937 -  docs: update calendar, add news item and link release notes for
939 -  docs: Add relnotes for 19.2.5
940 -  docs/relnotes/19.2.5: Add SHA256 sum
941 -  docs: update calendar, add news item and link release notes for
943 -  docs/release-calendar: Update for extended 19.3 rc period
944 -  docs: Add release notes for 19.2.6
945 -  docs: Add SHA256 sum for 19.2.6
946 -  docs: update calendar, add news item and link release notes for
948 -  gallium/auxiliary: Fix uses of gnu struct = {} extension
949 -  meson: Add -Werror=gnu-empty-initializer to MSVC compat args
950 -  docs: Add release notes for 19.2.7
951 -  docs: Add SHA256 sums for 19.2.7
952 -  docs: update calendar, add news item and link release notes for
954 -  docs: Update mesa 19.3 release calendar
955 -  meson/broadcom: libbroadcom_cle needs expat headers
956 -  meson/broadcom: libbroadcom_cle also needs zlib
957 -  docs: add release notes for 19.3.0
958 -  docs/19.3.0: Add SHA256 sums
959 -  docs: Update release notes, index, and calendar for 19.3.0
960 -  dcos: add releanse notes for 19.3.1
961 -  docs: Add release notes, update calendar, and add news for 19.3.1
962 -  docs: add relnotes for 19.2.8
963 -  docs/relnotes/19.2.8: Add SHA256 sum
964 -  docs: Add release notes, news, and update calendar for 19.2.8
965 -  docs: Add release notes for 19.3.2
966 -  docs: add SHA256 sums for 19.3.2
967 -  docs: Add release notes for 19.3.2, update calendar and home page
968 -  docs: Update release calendar for 20.0
969 -  docs: Add relnotes for 19.3.3 release
970 -  docs: Add SHA 256 sums for 19.3.3
971 -  docs: update news, calendar, and link release notes for 19.3.3
972 -  VERSION: bump to 20.0.0-rc1
973 -  bin/pick-ui: Add a new maintainer script for picking patches
974 -  .pick_status.json: Update to 0d14f41625fa00187f690f283c1eb6a22e354a71
975 -  .pick_status.json: Update to b550b7ef3b8d12f533b67b1a03159a127a3ff34a
976 -  .pick_status.json: Update to 9afdcd64f2c96f3fcc1a28912987f2e8066aa995
977 -  .pick_status.json: Update to 7eaf21cb6f67adbe0e79b80b4feb8c816a98a720
978 -  VERSION: bump to 20.0-rc2
979 -  .pick_status.json: Update to d8bae10bfe0f487dcaec721743cd51441bcc12f5
980 -  .pick_status.json: Update to 689817c9dfde9a0852f2b2489cb0fa93ffbcb215
981 -  .pick_status.json: Update to 23037627359e739c42b194dec54875aefbb9d00b
982 -  VERSION: bump for 20.0.0-rc3
983 -  .pick_status.json: Update to 2a98cf3b2ecea43cea148df7f77d2abadfd1c9db
984 -  .pick_status.json: Update to 946eacbafb47c8b94d47e7c9d2a8b02fff5a22fa
985 -  .pick_status.json: Update to bee5c9b0dc13dbae0ccf124124eaccebf7f2a435
989 -  turnip: Remove failed command buffer from pool
990 -  turnip: Fix issues in tu_compute_pipeline_create() that may lead to
995 -  Docs: remove duplicate meson docs for windows
996 -  docs: fix ascii html representation
997 -  nir/algebraic: i2f(f2i()) -> trunc()
998 -  nir/algebraic: sqrt(x)*sqrt(x) -> fabs(x)
1002 -  intel/compiler: Return early if read() failed
1006 -  ci: Make lava inherit the ccache setup of the .build script.
1007 -  ci: Switch over to an autoscaling GKE cluster for builds.
1008 -  Revert "ci: Switch over to an autoscaling GKE cluster for builds."
1009 -  mesa/st: Add mapping of MESA_FORMAT_RGB_SNORM16 to gallium.
1010 -  gallium: Add defines for FXT1 texture compression.
1011 -  gallium: Add some more channel orderings of packed formats.
1012 -  gallium: Add an equivalent of MESA_FORMAT_BGR_UNORM8.
1013 -  gallium: Add equivalents of packed MESA_FORMAT_*UINT formats.
1014 -  mesa: Stop defining a full separate format for RGBA_UINT8.
1015 -  mesa/st: Test round-tripping of all compressed formats.
1016 -  mesa: Prepare for the MESA_FORMAT\_\* enum to be sparse.
1017 -  mesa: Redefine MESA_FORMAT\_\* in terms of PIPE_FORMAT_*.
1018 -  mesa/st: Gut most of st_mesa_format_to_pipe_format().
1019 -  mesa/st: Make st_pipe_format_to_mesa_format an effective no-op.
1020 -  u_format: Fix swizzle of A1R5G5B5.
1021 -  ci: Use several debian buster packages instead of hand-building.
1022 -  ci: Make the skip list regexes match the full test name.
1023 -  ci: Use cts_runner for our dEQP runs.
1024 -  ci: Enable all of GLES3/3.1 testing for softpipe.
1025 -  ci: Remove old commented copy of freedreno artifacts.
1026 -  ci: Disable flappy blit tests on a630.
1027 -  ci: Expand the freedreno blit skip regex to cover more cases.
1028 -  util: Move gallium's PIPE_FORMAT utils to /util/format/
1029 -  mesa: Move compile of common Mesa core files to a static lib.
1030 -  mesa/st: Simplify st_choose_matching_format().
1031 -  mesa: Don't put sRGB formats in the array format table.
1032 -  mesa/st: Reuse st_choose_matching_format from st_choose_format().
1033 -  util: Add a mapping from VkFormat to PIPE_FORMAT.
1034 -  turnip: Drop the copy of the formats table.
1035 -  ci: Move freedreno's parallelism to the runner instead of gitlab-ci
1037 -  ci: Use a tag from the parallel-deqp-runner repo.
1038 -  nir: Add a scheduler pass to reduce maximum register pressure.
1039 -  nir: Refactor algebraic's block walk
1040 -  nir: Make algebraic backtrack and reprocess after a replacement.
1041 -  freedreno: Introduce a fd_resource_layer_stride() helper.
1042 -  freedreno: Introduce a fd_resource_tile_mode() helper.
1043 -  freedreno: Introduce a resource layout header.
1044 -  freedreno: Convert the slice struct to the new resource header.
1045 -  freedreno/a6xx: Log the tiling mode in resource layout debug.
1046 -  turnip: Disable timestamp queries for now.
1047 -  turnip: Fix unused variable warnings.
1048 -  turnip: Drop redefinition of VALIDREG now that it's in ir3.h.
1049 -  turnip: Reuse tu6_stage2opcode() more.
1050 -  turnip: Add basic SSBO support.
1051 -  turnip: Refactor the graphics pipeline create implementation.
1052 -  turnip: Add a helper function for getting tu_buffer iovas.
1053 -  turnip: Sanity check that we're adding valid BOs to the list.
1054 -  turnip: Move pipeline BO list adding to BindPipeline.
1055 -  turnip: Add support for compute shaders.
1056 -  ci: Disable egl_ext_device_drm tests in piglit.
1057 -  freedreno: Enable texture upload memory throttling.
1058 -  freedreno: Stop forcing ALLOW_MAPPED_BUFFERS_DURING_EXEC off.
1059 -  freedreno: Track the set of UBOs to be uploaded in UBO analysis.
1060 -  freedreno: Drop the extra offset field for mipmap slices.
1061 -  freedreno: Refactor the UBWC flags registers emission.
1062 -  freedreno: Move UBWC layout into a slices array like the non-UBWC
1064 -  tu: Move our image layout into a freedreno_layout struct.
1065 -  freedreno: Move a6xx's setup_slices() to a shareable helper function.
1066 -  freedreno: Switch the 16-bit workaround to match what turnip does.
1067 -  tu: Move UBWC layout into fdl6_layout() and use that function.
1068 -  turnip: Lower usub_borrow.
1069 -  turnip: Drop unused variable.
1070 -  turnip: Add support for descriptor arrays.
1071 -  turnip: Fix support for immutable samplers.
1072 -  ci: Fix caselist results archiving after parallel-deqp-runner rename.
1073 -  mesa: Fix detection of invalidating both depth and stencil.
1074 -  mesa/st: Deduplicate the NIR uniform lowering code.
1075 -  mesa/st: Move the vec4 type size function into core GLSL types.
1076 -  mesa/prog: Reuse count_vec4_slots() from ir_to_mesa.
1077 -  mesa/st: Move the dword slot counting function to glsl_types as well.
1078 -  i965: Reuse the new core glsl_count_dword_slots().
1079 -  nir: Fix printing of ~0 .locations.
1080 -  turnip: Refactor linkage state setup.
1081 -  mesa: Make atomic lowering put atomics above SSBOs.
1082 -  gallium: Pack the atomic counters just above the SSBOs.
1083 -  nir: Drop the ssbo_offset to atomic lowering.
1084 -  compiler: Add a note about how num_ssbos works in the program info.
1085 -  freedreno: Stop scattered remapping of SSBOs/images to IBOs.
1086 -  radeonsi: Remove a bunch of default handling of pipe caps.
1087 -  r600: Remove a bunch of default handling of pipe caps.
1088 -  r300: Remove a bunch of default handling of pipe caps.
1089 -  radeonsi: Drop PIPE_CAP_TGSI_ANY_REG_AS_ADDRESS.
1090 -  turnip: Fix some whitespace around binary operators.
1091 -  turnip: Refactor the intrinsic lowering.
1092 -  turnip: Add limited support for storage images.
1093 -  turnip: Disable UBWC on images used as storage images.
1094 -  turnip: Add support for non-zero (still constant) UBO buffer indices.
1095 -  turnip: Add support for uniform texel buffers.
1096 -  freedreno/ir3: Plumb the ir3_shader_variant into legalize.
1097 -  turnip: Add support for fine derivatives.
1098 -  turnip: Fix execution of secondary cmd bufs with nothing in primary.
1099 -  freedreno: Add some missing a6xx address declarations.
1100 -  freedreno: Fix OUT_REG() on address regs without a .bo supplied.
1101 -  turnip: Port krh's packing macros from freedreno to tu.
1102 -  turnip: Convert renderpass setup to the new register packing macros.
1103 -  turnip: Convert the rest of tu_cmd_buffer.c over to the new pack
1105 -  vulkan/wsi: Fix compiler warning when no WSI platforms are enabled.
1106 -  iris: Silence warning about AUX_USAGE_MC.
1107 -  mesa/st: Fix compiler warnings from INTEL_shader_integer_functions.
1108 -  ci: Enable -Werror on the meson-i386 build.
1109 -  tu: Fix binning address setup after pack macros change.
1110 -  Revert "gallium: Fix big-endian addressing of non-bitmask array
1115 -  meson: split out idep_xmlconfig_headers from idep_xmlconfig
1116 -  anv: add missing xmlconfig headers dependency
1117 -  radv: drop unnecessary xmlpool_options_h
1118 -  pipe-loader: drop unnecessary xmlpool_options_h
1119 -  loader: replace xmlpool_options_h with idep_xmlconfig_headers
1120 -  targets/omx: replace xmlpool_options_h with idep_xmlconfig_headers
1121 -  targets/va: replace xmlpool_options_h with idep_xmlconfig_headers
1122 -  targets/vdpau: replace xmlpool_options_h with idep_xmlconfig_headers
1123 -  targets/xa: replace xmlpool_options_h with idep_xmlconfig_headers
1124 -  targets/xvmc: replace xmlpool_options_h with idep_xmlconfig_headers
1125 -  dri: replace xmlpool_options_h with idep_xmlconfig_headers
1126 -  i915: replace xmlpool_options_h with idep_xmlconfig_headers
1127 -  nouveau: replace xmlpool_options_h with idep_xmlconfig_headers
1128 -  r200: replace xmlpool_options_h with idep_xmlconfig_headers
1129 -  radeon: replace xmlpool_options_h with idep_xmlconfig_headers
1130 -  meson: move idep_xmlconfig_headers to xmlpool/
1131 -  gitlab-ci: build a recent enough version of GLVND (ie. 1.2.0)
1132 -  meson: require glvnd 1.2.0
1133 -  meson: revert glvnd workaround
1134 -  meson: add variable to control the symbols checks
1135 -  meson: move the generic symbols check arguments to a common variable
1136 -  meson: add windows support to symbols checks
1137 -  meson: require \`nm\` again on Unix systems
1138 -  mesa/imports: let the build system detect strtok_r()
1139 -  egl: fix \_EGL_NATIVE_PLATFORM fallback
1140 -  egl: move #include of local headers out of Khronos headers
1141 -  gitlab-ci: build libdrm using meson instead of autotools
1142 -  gitlab-ci: auto-cancel CI runs when a newer commit is pushed to the
1144 -  CL: sync C headers with Khronos
1145 -  CL: sync C++ headers with Khronos
1146 -  vulkan: delete typo'd header
1147 -  egl: use EGL_CAST() macro in eglmesaext.h
1148 -  anv: add missing "fall-through" annotation
1149 -  vk_util: drop duplicate formats in vk_format_map[]
1150 -  meson: drop duplicate \`lib\` prefix on libiris_gen\*
1151 -  meson: drop \`intel_\` prefix on imgui_core
1152 -  docs: reword a bit and list HTTPS before FTP
1153 -  intel: add mi_builder_test for gen12
1154 -  intel/compiler: add ASSERTED annotation to avoid "unused variable"
1156 -  intel/compiler: replace \`0\` pointer with \`NULL\`
1157 -  util/simple_mtx: don't set the canary when it can't be checked
1158 -  anv: drop unused #include
1159 -  travis: autodetect python version instead of hard-coding it
1160 -  util/format: remove left-over util_format_description_table
1162 -  util/format: add PIPE_FORMAT_ASTC_*x*x*_SRGB to
1164 -  util/format: add trivial srgb<->linear conversion test
1165 -  u_format: move format tests to util/tests/
1166 -  amd: fix empty-body issues
1167 -  nine: fix empty-body-issues
1168 -  meson: simplify install_megadrivers.py invocation
1169 -  mesa: avoid returning a value in a void function
1170 -  meson: use github URL for wraps instead of completely unreliable
1172 -  egl: drop confusing mincore() error message
1173 -  llvmpipe: drop LLVM < 3.4 support
1174 -  util/atomic: fix return type of p_atomic_add_return() fallback
1175 -  util/os_socket: fix header unavailable on windows
1176 -  freedreno/perfcntrs: fix fd leak
1177 -  util/disk_cache: check for write() failure in the zstd path
1181 -  lima: fix nir shader memory leak
1182 -  lima: fix bo submit memory leak
1183 -  lima/ppir: enable lower_fdph
1184 -  gallium/util: add alignment parameter to util_upload_index_buffer
1185 -  lima: allocate separate bo to store varyings
1186 -  lima: refactor indexed draw indices upload
1187 -  vc4: move the draw splitting routine to shared code
1188 -  lima: split draw calls on 64k vertices
1189 -  lima/ppir: fix lod bias src
1190 -  lima/ppir: remove assert on ppir_emit_tex unsupported feature
1191 -  lima: set shader caps to optimize control flow
1192 -  lima/ppir: remove orphan load node after cloning
1193 -  lima/ppir: implement full liveness analysis for regalloc
1194 -  lima/ppir: handle write to dead registers in ppir
1195 -  lima/ppir: fix ssa undef emit
1196 -  lima/ppir: split ppir_op_undef into undef and dummy again
1197 -  lima/ppir: fix src read mask swizzling
1199 Erik Faye-Lund (82):
1201 -  zink: heap-allocate samplers objects
1202 -  zink: emit line-width when using polygon line-mode
1203 -  anv: remove incorrect polygonMode=point early-out
1204 -  zink: use actual format for render-pass
1205 -  zink: always allow mutating the format
1206 -  zink: do not advertize coherent mapping
1207 -  zink: disable fragment-shader texture-lod
1208 -  zink: transition resources before resolving
1209 -  zink: always allow sampling of images
1210 -  zink: use u_blitter when format-reinterpreting
1211 -  zink/spirv: drop temp-array for component-count
1212 -  zink/spirv: support loading bool constants
1213 -  zink/spirv: implement bany_fnequal[2-4]
1214 -  zink/spirv: implement bany_inequal[2-4]
1215 -  zink/spirv: implement ball_iequal[2-4]
1216 -  zink/spirv: implement ball_fequal[2-4]
1217 -  zink: do advertize integer support in shaders
1218 -  zink/spirv: add support for nir_op_flrp
1219 -  zink: correct depth-stencil format
1220 -  nir: patch up deref-vars when lowering clip-planes
1221 -  zink: always allow transfer to/from buffers
1222 -  zink: implement buffer-to-buffer copies
1223 -  zink: remove no-longer-needed hack
1224 -  zink: move format-checking to separate source
1225 -  zink: move filter-helper to separate helper-header
1226 -  zink: move blitting to separate source
1227 -  zink: move drawing separate source
1228 -  st/mesa: unmap pbo after updating cache
1229 -  zink: use true/false instead of TRUE/FALSE
1230 -  zink: reject invalid sample-counts
1231 -  zink: fix crash when restoring sampler-states
1232 -  zink: delete query rather than allocating a new one
1233 -  zink: do not try to destroy NULL-fence
1234 -  zink: handle calloc-failure
1235 -  zink: avoid NULL-deref
1236 -  zink: avoid NULL-deref
1237 -  zink: avoid NULL-deref
1238 -  zink: error-check right variable
1239 -  zink: silence coverity error
1240 -  zink: enable PIPE_CAP_MIXED_COLORBUFFER_FORMATS
1241 -  zink: implement nir_texop_txd
1242 -  zink: implement txf
1243 -  zink: implement some more trivial opcodes
1244 -  zink: simplify front-face type
1245 -  zink: factor out builtin-var creation
1246 -  zink: implement load_vertex_id
1247 -  zink: use nir_fmul_imm
1248 -  zink: remove unused code-path in lower_pos_write
1249 -  nir/zink: move clip_halfz-lowering to common code
1250 -  etnaviv: use nir_lower_clip_halfz instead of open-coding
1251 -  st/mesa: use uint-samplers for sampling stencil buffers
1252 -  zink: fixup initialization of operand_mask / num_extra_operands
1253 -  util: initialize float-array with float-literals
1254 -  st/wgl: eliminate implicit cast warning
1255 -  gallium: fix a warning
1256 -  mesa/st: use float literals
1257 -  docs: fix typo in html tag name
1258 -  docs: fix paragraphs
1259 -  docs: open paragraph before closing it
1260 -  docs: use code-tag instead of pre-tag
1261 -  docs: use code-tags instead of pre-tags
1262 -  docs: use code-tags instead of pre-tags
1263 -  docs: move paragraph closing tag
1264 -  docs: remove double-closed definition-list
1265 -  docs: do not double-close link tag
1266 -  docs: do not use definition-list for sub-topics
1267 -  docs: use figure/figcaption instead of tables
1268 -  docs: remove trailing header
1269 -  docs: remove leading spaces
1270 -  docs: remove trailing newlines
1271 -  docs: use [1] instead of asterisk for footnote
1272 -  docs: remove pointless, stray newline
1273 -  docs: fixup indentation
1274 -  zink: implement nir_texop_txs
1275 -  zink: support offset-variants of texturing
1276 -  zink: avoid incorrect vector-construction
1277 -  zink: store image-type per texture
1278 -  zink: support sampling non-float textures
1279 -  zink: support arrays of samplers
1280 -  zink: set compareEnable when setting compareOp
1281 -  st/mesa: use uint-result for sampling stencil buffers
1282 -  Revert "nir: Add a couple trivial abs optimizations"
1286 -  radv/winsys: set IB flags prior to submit in the sysmem path
1290 -  glsl: Fix software 64-bit integer to 32-bit float conversions.
1291 -  intel/fs/gen11+: Handle ROR/ROL in lower_simd_width().
1292 -  intel/fs/gen8+: Fix r127 dst/src overlap RA workaround for EOT
1294 -  intel/fs: Fix nir_intrinsic_load_barycentric_at_sample for SIMD32.
1295 -  intel/fs/cse: Fix non-deterministic behavior due to inaccurate
1297 -  intel/fs: Make implied_mrf_writes() an fs_inst method.
1298 -  intel/fs: Try to vectorize header setup in lower_load_payload().
1299 -  intel/fs: Generalize fs_reg::is_contiguous() to register files other
1301 -  intel/fs: Rework fs_inst::is_copy_payload() into multiple
1303 -  intel/fs: Extend copy propagation dataflow analysis to copies with
1305 -  intel/fs: Add partial support for copy-propagating FIXED_GRFs.
1306 -  intel/fs: Add support for copy-propagating a block of multiple
1308 -  intel/fs: Allow limited copy propagation of a LOAD_PAYLOAD into
1310 -  intel/fs/gen4-6: Allocate registers from aligned_pairs_class based on
1312 -  intel/fs/gen6: Constrain barycentric source of LINTERP during bank
1314 -  intel/fs/gen6: Generalize aligned_pairs_class to SIMD16 aligned
1316 -  intel/fs/gen6: Use SEL instead of bashing thread payload for unlit
1318 -  intel/fs: Split fetch_payload_reg() into separate helper for
1320 -  intel/fs: Introduce barycentric layout lowering pass.
1321 -  intel/fs: Switch to standard vector layout for barycentrics at
1323 -  intel/fs/cse: Make HALT instruction act as CSE barrier.
1324 -  intel/fs/gen7: Fix fs_inst::flags_written() for
1326 -  intel/fs: Add virtual instruction to load mask of live channels into
1328 -  intel/fs/gen12: Workaround unwanted SEND execution due to broken
1330 -  intel/fs/gen12: Fixup/simplify SWSB annotations of SIMD32 scratch
1332 -  intel/fs/gen12: Workaround data coherency issues due to broken NoMask
1337 -  freedreno: reorder format check
1341 -  Correctly wait in the fragment stage until all semaphores are
1343 -  Vulkan Overlay: Don't try to change the image layout to present twice
1344 -  Vulkan overlay: use the corresponding image index for each swapchain
1348 -  r600: Disable eight bit three channel formats
1349 -  virgl: Increase the shader transfer buffer by doubling the size
1350 -  gallium/tgsi_from_mesa: Add 'extern "C"' to be able to include from
1352 -  nir: make nir_get_texture_size/lod available outside nir_lower_tex
1353 -  gallium: tgsi_from_mesa - handle VARYING_SLOT_FACE
1354 -  r600: Add functions to dump the shader info
1355 -  r600: Make it possible to include r600_asm.h in a C++ file
1356 -  r600/sb: Correct SB disassambler for better debugging
1357 -  r600: Fix maximum line width
1358 -  r600: Make SID and unsigned value
1359 -  r600: Delete vertex buffer only if there is actually a shader state
1360 -  mesa/st: glsl_to_nir: don't lower atomics to SSBOs if driver supports
1365 -  etnaviv: drm: Don't miscalculate timeout
1366 -  freedreno/drm: Don't miscalculate timeout
1370 -  drirc: set allow_higher_compat_version for Faster Than Light
1371 -  virgl/drm: update UAPI
1372 -  teximage: split out helper from EGLImageTargetTexture2DOES
1373 -  glapi / teximage: implement EGLImageTargetTexStorageEXT
1374 -  dri_util: add driImageFormatToSizedInternalGLFormat function
1375 -  i965: track if image is created by a dmabuf
1376 -  i965: refactor intel_image_target_texture_2d
1377 -  i965: support EXT_EGL_image_storage
1378 -  st/dri: track if image is created by a dmabuf
1379 -  st/mesa: refactor egl image binding a bit
1380 -  st/mesa: implement EGLImageTargetTexStorage
1384 -  freedreno/ir3: cleanup by removing repeated code
1385 -  freedreno: support 16b for the sampler opcode
1386 -  freedreno/ir3: fix printing output registers of FS.
1387 -  freedreno/ir3: fixup when changing to mad.f16
1388 -  freedreno/ir3: enable half precision for pre-fs texture fetch
1389 -  turnip: fix invalid VK_ERROR_OUT_OF_POOL_MEMORY
1390 -  freedreno/ir3: put the conversion back for half const to the right
1395 -  v3d: rename vertex shader key (num)_fs_inputs fields
1396 -  mesa/st: make sure we remove dead IO variables before handing NIR to
1398 -  glsl: add missing initialization of the location path field
1399 -  v3d: fix indirect BO allocation for uniforms
1400 -  v3d: actually root the first BO in a command list in the job
1401 -  v3d: add missing plumbing for VPM load instructions
1402 -  v3d: add debug assert
1403 -  v3d: enable debug options for geometry shader dumps
1404 -  v3d: remove unused variable
1405 -  v3d: add initial compiler plumbing for geometry shaders
1406 -  v3d: fix packet descriptions for geometry and tessellation shaders
1407 -  v3d: emit geometry shader state commands
1408 -  v3d: implement geometry shader instancing
1409 -  v3d: add 1-way SIMD packing definition
1410 -  v3d: compute appropriate VPM memory configuration for geometry shader
1412 -  v3d: we always have at least one output segment
1413 -  v3d: add support for adjacency primitives
1414 -  v3d: don't try to render if shaders failed to compile
1415 -  v3d: predicate geometry shader outputs inside non-uniform control
1417 -  v3d: save geometry shader state for blitting
1418 -  v3d: support transform feedback with geometry shaders
1419 -  v3d: remove obsolete assertion
1420 -  v3d: do not limit new CL space allocations with branch to 4096 bytes
1421 -  v3d: support rendering to multi-layered framebuffers
1422 -  v3d: move layer rendering to a separate helper
1423 -  v3d: handle writes to gl_Layer from geometry shaders
1424 -  v3d: fix primitive queries for geometry shaders
1425 -  v3d: disable lowering of indirect inputs
1426 -  v3d: support precompiling geometry shaders
1427 -  v3d: expose OES_geometry_shader
1428 -  u_vbuf: don't try to delete NULL driver CSO
1429 -  v3d: fix bug when checking result of syncobj fence import
1433 -  intel/compiler: Report the number of non-spill/fill SEND messages on
1435 -  nir/algebraic: Add the ability to mark a replacement as exact
1436 -  nir/algebraic: Mark other comparison exact when removing a == a
1437 -  intel/fs: Disable conditional discard optimization on Gen4 and Gen5
1438 -  nir/range-analysis: Add pragmas to help loop unrolling
1439 -  nir/range_analysis: Make sure the table validation only occurs once
1440 -  nir/opt_peephole_select: Don't count some unary operations
1441 -  intel/compiler: Increase nir_opt_peephole_select threshold
1442 -  nir/algebraic: Simplify some Inf and NaN avoidance code
1443 -  nir/algebraic: Rearrange bcsel sequences generated by
1445 -  intel/compiler: Fix 'comparison is always true' warning
1446 -  mesa: Silence 'left shift of negative value' warning in BPTC
1448 -  mesa: Silence unused parameter warning
1449 -  anv: Fix error message format string
1450 -  mesa: Extension boilerplate for INTEL_shader_integer_functions2
1451 -  glsl: Add new expressions for INTEL_shader_integer_functions2
1452 -  glsl_types: Add function to get an unsigned base type from a signed
1454 -  glsl: Add built-in functions for INTEL_shader_integer_functions2
1455 -  nir: Add new instructions for INTEL_shader_integer_functions2
1456 -  nir/algebraic: Add lowering for uabs_usub and uabs_isub
1457 -  nir/algebraic: Add lowering for 64-bit hadd and rhadd
1458 -  nir/algebraic: Add lowering for 64-bit usub_sat
1459 -  nir/algebraic: Add lowering for 64-bit uadd_sat
1460 -  nir/algebraic: Add lowering for 64-bit iadd_sat and isub_sat
1461 -  compiler: Translate GLSL IR to NIR for new
1463 -  intel/fs: Don't lower integer multiplies that don't need lowering
1464 -  intel/fs: Add SHADER_OPCODE_[IU]SUB_SAT pseudo-ops
1465 -  intel/fs: Implement support for NIR opcodes for
1467 -  nir/spirv: Translate SPIR-V to NIR for new
1469 -  spirv: Silence a bunch of unused parameter warnings
1470 -  spirv: Add support for IntegerFunctions2INTEL capability
1471 -  i965: Enable INTEL_shader_integer_functions2 on Gen8+
1472 -  gallium: Add a cap bit for OpenCL-style extended integer functions
1473 -  gallium: Add a cap bit for integer multiplication between 32-bit and
1474    16-bit
1475 -  iris: Enable INTEL_shader_integer_functions2
1476 -  anv: Enable SPV_INTEL_shader_integer_functions2 and
1478 -  nir/algebraic: Optimize some 64-bit integer comparisons involving
1480 -  relnotes: Add GL_INTEL_shader_integer_functions2 and
1482 -  intel/fs: Don't count integer instructions as being possibly coissue
1486 -  gallium/auxiliary: Reduce conversions in
1488 -  gallium/auxiliary: Handle count == 0 in
1490 -  panfrost: Add negative lod bias support
1491 -  panfrost: Compact the bo_access readers array
1492 -  panfrost: Dynamically allocate shader variants
1493 -  panfrost: Add ETC1/ETC2 texture formats
1494 -  panfrost: Add ASTC texture formats
1495 -  pan/midgard: Fix bundle dynarray leak
1496 -  pan/midgard: Fix a memory leak in the disassembler
1497 -  pan/midgard: Support disassembling to a file
1498 -  pan/bifrost: Support disassembling to a file
1499 -  pan/decode: Support dumping to a file
1500 -  pan/decode: Dump to a file
1501 -  pan/decode: Rotate trace files
1502 -  panfrost: Don't copy uniforms when the size is zero
1503 -  pan/midgard: Fix a liveness info leak
1507 -  lima: support indexed draw with bias
1508 -  lima: fix lima_set_vertex_buffers()
1512 -  gm107/ir: fix loading z offset for layered 3d image bindings
1513 -  nv50/ir: mark STORE destination inputs as used
1514 -  nv50,nvc0: fix destination coordinates of blit
1515 -  nvc0: add dummy reset status support
1516 -  gm107/ir: avoid combining geometry shader stores at 0x60
1517 -  nvc0: treat all draws without color0 broadcast as MRT
1518 -  nvc0: disable xfb's which don't have a stride
1522 -  intel/compiler: remove old comment
1526 -  intel/compiler: Don't change hstride if not needed
1527 -  anv: Export filter_minmax support only when it's really supported
1528 -  anv: Export VK_KHR_buffer_device_address only when really supported
1529 -  anv: Enable Vulkan 1.2 support
1533 -  iris: try to set the specified tiling when importing a dmabuf
1534 -  gallium: dmabuf support for yuv formats that are not natively
1536 -  gallium: let the pipe drivers decide the supported modifiers
1540 -  clover: Initialize Asm Parsers
1541 -  clover: Use explicit conversion from llvm::StringRef to std::string
1545 -  gallium/swr: Fix depth values for blit scenario
1546 -  swr/rasterizer: Add tessellator implementation to the rasterizer
1547 -  gallium/swr: Fix Windows build
1548 -  gallium/gallivm/tgsi: enable tessellation shaders
1549 -  gallium/gallivm: enable linking lp_bld_printf function with C++ code
1550 -  gallium/swr: implementation of tessellation shaders compilation
1551 -  gallium/swr: fix tessellation state save/restore
1552 -  docs: Update SWR tessellation support
1556 -  util: Add a util_sparse_array data structure
1557 -  anv: Move refcount to anv_bo
1558 -  anv: Use a util_sparse_array for the GEM handle -> BO map
1559 -  anv: Fix a relocation race condition
1560 -  anv: Stop storing the GEM handle in anv_reloc_list_add
1561 -  anv: Declare the bo in the anv_block_pool_foreach_bo loop
1562 -  anv: Inline anv_block_pool_get_bo
1563 -  anv: Replace ANV_BO_EXTERNAL with anv_bo::is_external
1564 -  anv: Handle state pool relocations using "wrapper" BOs
1565 -  anv: Fix a potential BO handle leak
1566 -  anv: Rework anv_block_pool_expand_range
1567 -  anv: Use anv_block_pool_foreach_bo in get_bo_from_pool
1568 -  anv: Rework the internal BO allocation API
1569 -  anv: Choose BO flags internally in anv_block_pool
1570 -  anv/tests: Zero-initialize instances
1571 -  anv/tests: Initialize the BO cache and device mutex
1572 -  anv: Allocate block pool BOs from the cache
1573 -  anv: Use the query_slot helper in vkResetQueryPoolEXT
1574 -  anv: Allocate query pool BOs from the cache
1575 -  anv: Set more flags on descriptor pool buffers
1576 -  anv: Allocate descriptor buffers from the BO cache
1577 -  util: Add a free list structure for use with util_sparse_array
1578 -  anv: Allocate batch and fence buffers from the cache
1579 -  anv: Allocate scratch BOs from the cache
1580 -  anv: Allocate misc BOs from the cache
1581 -  anv: Drop anv_bo_init and anv_bo_init_new
1582 -  anv: Add a device parameter to anv_execbuf_add_bo
1583 -  anv: Set the batch allocator for compute pipelines
1584 -  anv: Use a bitset for tracking residency
1585 -  anv: Zero released anv_bo structs
1586 -  anv: Use the new BO alloc API for Android
1587 -  anv: Don't delete fragment shaders that write sample mask
1588 -  anv: Don't claim the null RT as a valid color target
1589 -  anv: Stop compacting render targets in the binding table
1590 -  anv: Move the RT BTI flush workaround to begin_subpass
1591 -  spirv: Remove the type from sampled_image
1592 -  spirv: Add a vtn_decorate_pointer helper
1593 -  spirv: Sort out the mess that is sampled image
1594 -  nir/builder: Add a nir_extract_bits helper
1595 -  nir: Add tests for nir_extract_bits
1596 -  intel/nir: Use nir_extract_bits in lower_mem_access_bit_sizes
1597 -  intel/fs: Add DWord scattered read/write opcodes
1598 -  intel/fs: refactor surface header setup
1599 -  intel/nir: Plumb devinfo through lower_mem_access_bit_sizes
1600 -  intel/fs: Implement the new load/store_scratch intrinsics
1601 -  intel/fs: Lower large local arrays to scratch
1602 -  anv: Lock around fetching sync file FDs from semaphores
1603 -  anv: Plumb timeline semaphore signal/wait values through from the API
1604 -  spirv: Fix the MSVC build
1605 -  anv/pipeline: Assume layout != NULL
1606 -  genxml: Mark everything in genX_pack.h always_inline
1607 -  anv: Input attachments are always single-plane
1608 -  anv: Flatten descriptor bindings in anv_nir_apply_pipeline_layout
1609 -  anv: Delete dead shader constant pushing code
1610 -  anv: Stop bounds-checking pushed UBOs
1611 -  anv: Pre-compute push ranges for graphics pipelines
1612 -  intel/compiler: Add a flag to avoid compacting push constants
1613 -  anv: Re-arrange push constant data a bit
1614 -  anv: Rework push constant handling
1615 -  anv: Use a switch statement for binding table setup
1616 -  anv: More carefully dirty state in BindDescriptorSets
1617 -  anv: More carefully dirty state in BindPipeline
1618 -  anv: Use an anv_state for the next binding table
1619 -  anv: Emit a NULL vertex for zero base_vertex/instance
1620 -  nir: Validate that variables are in the right lists
1621 -  iris: Re-enable param compaction
1622 -  Revert "i965/fs: Merge CMP and SEL into CSEL on Gen8+"
1623 -  vulkan/enum_to_str: Handle out-of-order aliases
1624 -  anv/entrypoints: Better handle promoted extensions
1625 -  vulkan: Update the XML and headers to 1.1.129
1626 -  anv: Push constants are relative to dynamic state on IVB
1627 -  anv: Set up SBE_SWIZ properly for gl_Viewport
1628 -  anv: Respect the always_flush_cache driconf option
1629 -  iris: Stop setting up fake params
1630 -  anv: Drop bo_flags from anv_bo_pool
1631 -  anv: Add a has_softpin boolean
1632 -  blorp: Pass the VB size to the VF cache workaround
1633 -  anv: Always invalidate the VF cache in BeginCommandBuffer
1634 -  anv: Apply cache flushes after setting index/draw VBs
1635 -  anv: Use PIPE_CONTROL flushes to implement the gen8 VF cache WA
1636 -  anv: Don't leak when set_tiling fails
1637 -  util/atomic: Add a \_return variant of p_atomic_add
1638 -  anv: Disallow allocating above heap sizes
1639 -  anv: Stop tracking VMA allocations
1640 -  anv: Set up VMA heaps independently from memory heaps
1641 -  anv: Stop advertising two heaps just for the VF cache WA
1642 -  anv: Add an explicit_address parameter to anv_device_alloc_bo
1643 -  util/vma: Factor out the hole splitting part of util_vma_heap_alloc
1644 -  util/vma: Add a function to allocate a particular address range
1645 -  anv: Add allocator support for client-visible addresses
1646 -  anv: Use a pNext loop in AllocateMemory
1647 -  anv: Implement VK_KHR_buffer_device_address
1648 -  util/atomic: Add p_atomic_add_return for the unlocked path
1649 -  vulkan/wsi: Provide the implicitly synchronized BO to vkQueueSubmit
1650 -  vulkan/wsi: Add a hooks for signaling semaphores and fences
1651 -  anv: Always add in EXEC_OBJECT_WRITE when specified in extra_flags
1652 -  anv: Use submit-time implicit sync instead of allocate-time
1653 -  anv: Add a fence_reset_reset_temporary helper
1654 -  anv: Use BO fences/semaphores for AcquireNextImage
1655 -  anv: Return VK_ERROR_OUT_OF_DEVICE_MEMORY for too-large buffers
1656 -  anv: Re-capture all batch and state buffers
1657 -  anv: Re-emit all compute state on pipeline switch
1658 -  ANV: Stop advertising smoothLines support on gen10+
1659 -  anv: Flush the queue on DeviceWaitIdle
1660 -  anv: Unconditionally advertise Vulkan 1.1
1661 -  anv: Bump the advertised patch version to 129
1662 -  i965: Enable GL_EXT_gpu_shader4 on Gen6+
1663 -  anv: Properly advertise sampledImageIntegerSampleCounts
1664 -  anv: Drop unneeded struct keywords
1665 -  blorp: Stop whacking Z24 depth to BGRA8
1666 -  blorp: Allow reading with HiZ
1667 -  i965/blorp: Don't resolve HiZ unless we're reinterpreting
1668 -  intel/blorp: Use the source format when using blorp_copy with HiZ
1669 -  anv: Allow HiZ in TRANSFER_SRC_OPTIMAL on Gen8-9
1670 -  i965: Allow HiZ for glCopyImageSubData sources
1671 -  intel/nir: Add a memory barrier before barrier()
1672 -  intel/disasm: Fix decoding of src0 of SENDS
1673 -  genxml: Remove a non-existant HW bit
1674 -  anv: Don't add dynamic state base address to push constants on Gen7
1675 -  anv: Flag descriptors dirty when gl_NumWorkgroups is used
1676 -  anv: Re-use flush_descriptor_sets in flush_compute_state
1677 -  intel/vec4: Support scoped_memory_barrier
1678 -  nir: Handle more barriers in dead_write and copy_prop
1679 -  nir: Handle barriers with more granularity in combine_stores
1680 -  llmvpipe: No-op implement more barriers
1681 -  nir: Add a new memory_barrier_tcs_patch intrinsic
1682 -  spirv: Add a workaround for OpControlBarrier on old GLSLang
1683 -  spirv: Add output memory semantics to OpControlBarrier in TCS
1684 -  nir/glsl: Emit memory barriers as part of barrier()
1685 -  intel/nir: Stop adding redundant barriers
1686 -  nir: Rename nir_intrinsic_barrier to control_barrier
1687 -  nir/lower_atomics_to_ssbo: Also lower barriers
1688 -  anv: Drop an unused variable
1689 -  intel/blorp: Fill out all the dwords of MI_ATOMIC
1690 -  anv: Don't over-advertise descriptor indexing features
1691 -  anv: Memset array properties
1692 -  vulkan/wsi: Add a driconf option to force WSI to advertise
1694 -  vulkan: Update the XML and headers to 1.2.131
1695 -  turnip: Pretend to support Vulkan 1.2
1696 -  anv: Bump the patch version to 131
1697 -  anv,nir: Lower quad_broadcast with dynamic index in NIR
1698 -  anv: Implement the new core version feature queries
1699 -  anv: Implement the new core version property queries
1700 -  relnotes: Add Vulkan 1.2
1701 -  anv: Drop some VK_IMAGE_TILING_OPTIMAL checks
1702 -  anv: Support modifiers in GetImageFormatProperties2
1703 -  vulkan/wsi: Move the ImageCreateInfo higher up
1704 -  vulkan/wsi: Use the interface from the real modifiers extension
1705 -  vulkan/wsi: Filter modifiers with ImageFormatProperties
1706 -  vulkan/wsi: Implement VK_KHR_swapchain_mutable_format
1707 -  anv/blorp: Rename buffer image stride parameters
1708 -  anv: Canonicalize buffer formats for image/buffer copies
1709 -  anv: Add an anv_physical_device field to anv_device
1710 -  anv: Take an anv_device in vk_errorf
1711 -  anv: Take a device in anv_perf_warn
1712 -  anv: Stop allocating WSI event fences off the instance
1713 -  anv: Drop the instance pointer from anv_device
1714 -  anv: Move the physical device dispatch table to anv_instance
1715 -  anv: Drop separate chipset_id fields
1716 -  anv: Re-arrange physical_device_init
1717 -  anv: Allow enumerating multiple physical devices
1718 -  anv/apply_pipeline_layout: Initialize the nir_builder before use
1719 -  intel/blorp: resize src and dst surfaces separately
1720 -  anv: Use TRANSFER_SRC_OPTIMAL for depth/stencil MSAA resolves
1721 -  anv: Add a layout_to_aux_state helper
1722 -  anv: Use isl_aux_state for HiZ resolves
1723 -  anv: Add a usage parameter to anv_layout_to_aux_usage
1724 -  anv: Allow HiZ in read-only depth layouts
1725 -  anv: Improve BTI change cache flushing
1726 -  intel/fs: Don't unnecessarily fall back to indirect sends on Gen12
1727 -  intel/disasm: Properly disassemble indirect SENDs
1728 -  intel/isl: Plumb devinfo into isl_genX(buffer_fill_state_s)
1729 -  intel/isl: Add a hack for the Gen12 A0 texture buffer bug
1730 -  anv: Rework the meaning of anv_image::planes[]::aux_usage
1731 -  anv: Replace aux_surface.isl.size_B checks with aux_usage checks
1732 -  intel/aux-map: Add some #defines
1733 -  intel/aux-map: Factor out some useful helpers
1734 -  anv: Delete a redundant calculation
1735 -  isl: Add a helper for calculating subimage memory ranges
1736 -  anv: Add another align_down helper
1737 -  anv: Make AUX table invalidate a PIPE\_\* bit
1738 -  anv: Make anv_vma_alloc/free a lot dumber
1739 -  anv: Rework CCS memory handling on TGL-LP
1740 -  intel/blorp: Add support for CCS_E copies with UNORM formats
1741 -  intel/isl: Allow CCS_E on more formats
1742 -  intel/genxml: Make SO_DECL::"Hole Flag" a Boolean
1743 -  anv: Insert holes for non-existant XFB varyings
1744 -  intel/blorp: Handle bit-casting UNORM and BGRA formats
1745 -  anv: Replace one more aux_surface.isl.size_B check
1746 -  intel/mi_builder: Force write completion on Gen12+
1747 -  anv: Set actual state pool sizes when we have softpin
1748 -  anv: Re-use one old BT block in reset_batch_bo_chain
1749 -  anv/block_pool: Ensure allocations have contiguous maps
1750 -  anv: Rename a variable
1751 -  genxml: Add a new 3DSTATE_SF field on gen12
1752 -  anv,iris: Set 3DSTATE_SF::DerefBlockSize to per-poly on Gen12+
1753 -  intel/genxml: Drop SLMEnable from L3CNTLREG on Gen11
1754 -  iris: Set SLMEnable based on the L3$ config
1755 -  iris: Store the L3$ configs in the screen
1756 -  iris: Use the URB size from the L3$ config
1757 -  i965: Re-emit l3 state before BLORP executes
1758 -  intel: Take a gen_l3_config in gen_get_urb_config
1759 -  intel/blorp: Always emit URB config on Gen7+
1760 -  iris: Consolodate URB emit
1761 -  anv: Emit URB setup earlier
1762 -  intel/common: Return the block size from get_urb_config
1763 -  intel/blorp: Plumb deref block size through to 3DSTATE_SF
1764 -  anv: Plumb deref block size through to 3DSTATE_SF
1765 -  iris: Plumb deref block size through to 3DSTATE_SF
1766 -  anv: Always fill out the AUX table even if CCS is disabled
1767 -  intel/fs: Write the address register with NoMask for MOV_INDIRECT
1768 -  anv/blorp: Use the correct size for vkCmdCopyBufferToImage
1772 -  winsys/amdgpu: avoid double simple_mtx_unlock()
1773 -  i965: update Makefile.sources for perf changes
1774 -  util/futex: use futex syscall on OpenBSD
1775 -  util/u_thread: don't restrict u_thread_get_time_nano() to \__linux_\_
1779 -  freedreno: add Adreno 640 ID
1780 -  freedreno/ir3: disable texture prefetch for 1d array textures
1781 -  freedreno/registers: fix a6xx_2d_blit_cntl ROTATE
1782 -  etnaviv: blt: use only for tiling, and add missing formats
1783 -  etnaviv: separate PE and RS formats, use only RS only for tiling
1784 -  etnaviv: blt: set TS dirty after clear
1785 -  turnip: add display wsi
1786 -  turnip: add x11 wsi
1787 -  turnip: implement CmdClearColorImage/CmdClearDepthStencilImage
1788 -  turnip: fix sRGB GMEM clear
1789 -  util: add missing R8G8B8A8_SRGB format to vk_format_map
1790 -  freedreno/regs: update UBWC related bits
1791 -  turnip: implement UBWC
1792 -  etnaviv: avoid using RS for 64bpp formats
1793 -  etnaviv: implement 64bpp clear
1794 -  etnaviv: blt: fix partial ZS clears with TS
1795 -  etnaviv: support 3d/array/integer formats in texture descriptors
1796 -  turnip: fix integer render targets
1797 -  freedreno/registers: add missing MH perfcounter enum for a2xx
1798 -  freedreno/perfcntrs: add a2xx MH counters
1799 -  freedreno/perfcntrs/fdperf: fix u64 print on 32-bit builds
1800 -  freedreno/perfcntrs/fdperf: add missing a20x compatible
1801 -  freedreno/perfcntrs/fdperf: add missing a2xx case in select_counter
1802 -  turnip: fix display wsi fence timing out
1803 -  turnip: don't skip unused attachments when setting up tiling config
1804 -  turnip: implement CmdClearAttachments
1805 -  turnip: don't set unused BLIT_DST_INFO bits for GMEM clear
1806 -  turnip: MSAA resolve directly from GMEM
1807 -  turnip: allow writes to draw_cs outside of render pass
1808 -  turnip: add function to allocate aligned memory in a substream cs
1809 -  turnip: improve emit_textures
1810 -  turnip: implement border color
1811 -  turnip: add hw binning
1812 -  turnip: fix incorrectly failing assert
1813 -  freedreno/ir3: add GLSL_SAMPLER_DIM_SUBPASS to tex_info
1814 -  freedreno/registers: add a6xx texture format for stencil sampler
1815 -  turnip: fix hw binning render area
1816 -  turnip: fix tile layout logic
1817 -  turnip: update tile_align_w/tile_align_h
1818 -  turnip: set load_layer_id to zero
1819 -  turnip: set FRAG_WRITES_SAMPMASK bit
1820 -  turnip: fix VK_IMAGE_ASPECT_STENCIL_BIT image view
1821 -  turnip: no 8x msaa on 128bpp formats
1822 -  turnip: add dirty bit for push constants
1823 -  turnip: subpass rework
1824 -  turnip: CmdClearAttachments fixes
1825 -  turnip: implement subpass input attachments
1826 -  etnaviv: remove sRGB formats from format table
1827 -  etnaviv: sRGB render target support
1828 -  etnaviv: set output mode and saturate bits
1829 -  etnaviv: update INT_FILTER choice for GLES3 formats
1830 -  etnaviv: disable integer vertex formats on pre-HALTI2 hardware
1831 -  etnaviv: remove swizzle from format table
1832 -  etnaviv: add missing formats
1833 -  etnaviv: add missing vs_needs_z_div handling to NIR backend
1834 -  turnip: use single substream cs
1835 -  turnip: use common blit path for buffer copy
1836 -  turnip: don't require src image to be set for clear blits
1837 -  turnip: implement CmdFillBuffer/CmdUpdateBuffer
1838 -  freedreno/ir3: lower mul_2x32_64
1839 -  turnip: fix emit_textures for compute shaders
1840 -  turnip: remove compute emit_border_color
1841 -  turnip: fix emit_ibo
1842 -  turnip: change emit_ibo to be like emit_textures
1843 -  turnip: remove duplicate A6XX_SP_CS_CONFIG_NIBO
1844 -  nir: add option to lower half packing opcodes
1845 -  freedreno/ir3: lower pack/unpack ops
1846 -  turnip: don't set LRZ enable at end of renderpass
1847 -  freedreno/ir3: update prefetch input_offset when packing inlocs
1848 -  turnip: add cache invalidate to fix input attachment cases
1849 -  turnip: don't set SP_FS_CTRL_REG0_VARYING if only fragcoord is used
1850 -  freedreno/ir3: fix vertex shader sysvals with pre_assign_inputs
1851 -  freedreno/registers: document vertex/instance id offset bits
1852 -  freedreno/ir3: support load_base_instance
1853 -  turnip: emit base instance vs driver param
1854 -  turnip: emit_compute_driver_params fixes
1855 -  turnip: compute gmem offsets at renderpass creation time
1856 -  turnip: implement secondary command buffers
1857 -  nir: fix assign_io_var_locations for vertex inputs
1858 -  turnip: minor warning fixes
1859 -  util/format: add missing vulkan formats
1860 -  turnip: disable B8G8R8 vertex formats
1861 -  etnaviv: fix incorrectly failing vertex size assert
1862 -  etnaviv: update headers from rnndb
1863 -  etnaviv: HALTI2+ instanced draw
1864 -  etnaviv: implement gl_VertexID/gl_InstanceID
1865 -  etnaviv: remove unnecessary vertex_elements_state_create error
1867 -  st/mesa: don't lower YUV when driver supports it natively
1868 -  st/mesa: run st_nir_lower_tex_src_plane for lowered xyuv/ayuv
1869 -  freedreno/ir3: allow inputs with the same location
1870 -  turnip: remove tu_sort_variables_by_location
1871 -  turnip: fix array/matrix varyings
1872 -  turnip: hook up GetImageDrmFormatModifierPropertiesEXT
1873 -  turnip: set linear tiling for scanout images
1874 -  vulkan/wsi: remove unused image_get_modifier
1875 -  turnip: simplify tu_physical_device_get_format_properties
1876 -  etnaviv: implement UBOs
1877 -  turnip: hook up cmdbuffer event set/wait
1881 -  iris: Add IRIS_DIRTY_RENDER_BUFFER state flag
1882 -  iris/gen11+: Move flush for render target change
1883 -  iris: Allow max dynamic pool size of 2GB for gen12
1884 -  intel: Remove unused Tigerlake PCI ID
1885 -  iris: Fix some indentation in iris_init_render_context
1886 -  iris: Emit CS Stall before Instruction Cache flush for gen12 WA
1887 -  anv: Emit CS Stall before Instruction Cache flush for gen12 WA
1891 -  v3d: Fix predication with atomic image operations
1895 -  nir/lower_double_ops: relax lower mod()
1896 -  Revert "nir/lower_double_ops: relax lower mod()"
1897 -  nir/spirv: skip unreachable blocks in Phi second pass
1901 -  nir: fix unused variable warning in nir_lower_vars_to_explicit_types
1902 -  nir: fix unused variable warning in
1904 -  nir: fix unused function warning in src/compiler/nir/nir.c
1905 -  intel/gen_decoder: Fix unused-but-set-variable warning
1909 -  nv50/ir: fix crash in isUniform for undefined values
1910 -  nir/validate: validate num_components on registers and intrinsics
1911 -  nir/serialize: fix vec8 and vec16
1912 -  nir/tests: add serializer tests
1913 -  nir/tests: MSVC build fix
1914 -  spirv: handle UniformConstant for OpenCL kernels
1915 -  clover/nir: treat UniformConstant as global memory
1916 -  clover/nir: set spirv environment to OpenCL
1917 -  clover/spirv: allow Int64 Atomics for supported devices
1918 -  nir: handle nir_deref_type_ptr_as_array in
1920 -  nv50/ir: implement global atomics and handle it for nir
1921 -  nir/serialize: cast swizzle before shifting
1922 -  aco: use NIR_MAX_VEC_COMPONENTS instead of 4
1923 -  nv50ir/nir: support vec8 and vec16
1927 -  iris: Fix "Force Zero RTA Index Enable" setting again
1928 -  nir: Handle image arrays when setting variable data
1929 -  Revert "intel/blorp: Fix usage of uninitialized memory in key
1931 -  iris: Properly move edgeflag_out from output list to global list
1932 -  iris: Wrap iris_fix_edge_flags in NIR_PASS
1933 -  mesa: Handle GL_COLOR_INDEX in \_mesa_format_from_format_and_type().
1934 -  iris: Change keybox parenting
1935 -  iris: Stop mutating the resource in get_rt_read_isl_surf().
1936 -  iris: Drop 'old_address' parameter from iris_rebind_buffer
1937 -  iris: Create an "iris_surface_state" wrapper struct
1938 -  iris: Maintain CPU-side SURFACE_STATE copies for views and surfaces.
1939 -  iris: Update SURFACE_STATE addresses when setting sampler views
1940 -  iris: Disable VF cache partial address workaround on Gen11+
1941 -  driconf, glsl: Add a vs_position_always_invariant option
1942 -  drirc: Set vs_position_always_invariant for Shadow of Mordor on Intel
1943 -  st/mesa: Add GL_TDFX_texture_compression_FXT1 support
1944 -  iris: Map FXT1 texture formats
1945 -  meson: Add a "prefer_iris" build option
1946 -  main: Change u_mmAllocMem align2 from bytes (old API) to bits (new
1948 -  meson: Include iris in default gallium-drivers for x86/x86_64
1949 -  util: Detect use-after-destroy in simple_mtx
1950 -  intel/genxml: Add a partial TCCNTLREG definition
1951 -  iris: Enable Gen11 Color/Z write merging optimization
1952 -  anv: Enable Gen11 Color/Z write merging optimization
1953 -  intel/decoder: Make get_state_size take a full 64-bit address and a
1955 -  iris: Create smaller program keys without legacy features
1956 -  iris: Default to X-tiling for scanout buffers without modifiers
1957 -  iris: Alphabetize source files after iris_perf.c was added
1958 -  drirc: Final Fantasy VIII: Remastered needs
1960 -  iris: Make helper functions to turn iris shader keys into brw keys.
1961 -  iris: Fix shader recompile debug printing
1962 -  iris: Avoid replacing backing storage for buffers with no contents
1963 -  intel: Drop Gen11 WaBTPPrefetchDisable workaround
1964 -  st/nir: Optionally unify inputs_read/outputs_written when linking.
1965 -  iris: Set nir_shader_compiler_options::unify_interfaces.
1966 -  st/mesa: Allow ASTC5x5 fallbacks separately from other ASTC LDR
1968 -  iris: Disable ASTC 5x5 support on Gen9 for now.
1969 -  iris: Delete remnants of the unimplemented ASTC 5x5 workaround
1970 -  iris: Allow HiZ for copy_region sources
1971 -  anv: Only enable EWA LOD algorithm when doing anisotropic filtering.
1972 -  Revert "nir: assert that nir_lower_tex runs after lowering derefs"
1973 -  i965: Simplify brw_get_renderer_string()
1974 -  iris: Simplify iris_get_renderer_string()
1975 -  intel: Use similar brand strings to the Windows drivers
1976 -  intel/compiler: Fix illegal mutation in get_nir_image_intrinsic_image
1977 -  iris: Fix export of fences that have already completed.
1978 -  st/mesa: Allocate full miplevels if MaxLevel is explicitly set
1979 -  iris: Drop some workarounds which are no longer necessary
1980 -  anv: Drop some workarounds that are no longer necessary
1981 -  intel: Fix aux map alignments on 32-bit builds.
1982 -  meson: Prefer 'iris' by default over 'i965'.
1983 -  loader: Check if the kernel driver is i915 before loading iris
1984 -  iris: Drop 'engine' from iris_batch.
1985 -  iris: Make iris_emit_default_l3_config pull devinfo from the batch
1986 -  iris: Support multiple chained batches.
1987 -  i965: Use brw_batch_references in tex_busy check
1988 -  loader: Fix leak of kernel driver name
1992 -  freedreno/registers: Fix typo
1993 -  freedreno/registers: Move SP_PRIMITIVE_CNTL and SP_VS_VPC_DST
1994 -  freedreno/registers: Add comments about primitive counters
1995 -  freedreno/a6xx: Fix primitive counters again
1996 -  freedreno/a6xx: Clear sysmem with CP_BLIT
1997 -  freedreno: Add nogmem debug option to force bypass rendering
1998 -  freedreno/a6xx: Fix layered texture type enum
1999 -  freedreno/a6x: Rename z/s formats
2000 -  freedreno/a6xx: Add register offset for STG/LDG
2001 -  freedreno/ir3: Emit link map as byte or dwords offsets as needed
2002 -  freedreno/ir3: Add load and store intrinsics for global io
2003 -  freedreno: Don't count primitives for patches
2004 -  freedreno/ir3: Add ir3 intrinsics for tessellation
2005 -  freedreno/ir3: Use imul24 in offset calculations
2006 -  freedreno/ir3: Add tessellation field to shader key
2007 -  freedreno/ir3: Extend geometry lowering pass to handle tessellation
2008 -  freedreno/ir3: Add new synchronization opcodes
2009 -  freedreno/ir3: End TES with chsh when using GS
2010 -  freedreno/ir3: Implement tess coord intrinsic
2011 -  freedreno/ir3: Implement TCS synchronization intrinsics
2012 -  freedreno/ir3: Setup inputs and outputs for tessellation stages
2013 -  freedreno/ir3: Don't assume binning shader is always VS
2014 -  freedreno/ir3: Pre-color TCS header and primitive ID inputs
2015 -  freedreno/ir3: Allocate const space for tessellation parameters
2016 -  freedreno/a6xx: Build the right draw command for tessellation
2017 -  freedreno/a6xx: Allocate and program tessellation buffer
2018 -  freedreno/a6xx: Emit constant parameters for tessellation stages
2019 -  freedreno/a6xx: Program state for tessellation stages
2020 -  freedreno: Use bypass rendering for tessellation
2021 -  freedreno/a6xx: Only set emit.hs/ds when we're drawing patches
2022 -  freedreno/blitter: Save tessellation state
2023 -  freedreno/a6xx: Only use merged regs and four quads for VS+FS
2024 -  freedreno/a6xx: Turn on tessellation shaders
2025 -  freedreno/ir3: Use regid() helper when setting up precolor regs
2026 -  freedreno/registers: Remove duplicate register definitions
2027 -  freedreno: New struct packing macros
2028 -  freedreno/registers: Add 64 bit address registers
2029 -  freedreno/a6xx: Drop stale include
2030 -  freedreno/a6xx: Include fd6_pack.h in a few files
2031 -  freedreno/a6xx: Convert emit_mrt() to OUT_REG()
2032 -  freedreno/a6xx: Convert emit_zs() to OUT_REG()
2033 -  freedreno/a6xx: Convert VSC pipe setup to OUT_REG()
2034 -  freedreno/a6xx: Convert gmem blits to OUT_REG()
2035 -  freedreno/a6xx: Convert some tile setup to OUT_REG()
2036 -  freedreno/a6xx: Silence warning for unused perf counters
2037 -  freedreno/a6xx: Document the CP_SET_DRAW_STATE enable bits
2038 -  freedreno/a6xx: Make DEBUG_BLIT_FALLBACK only dump fallbacks
2039 -  freedreno: Add debug flag for forcing linear layouts
2040 -  freedreno/a6xx: Program sampler swap based on resource tiling
2041 -  freedreno/a6xx: Pick blitter swap based on resource tiling
2042 -  freedreno/a6xx: Add fd_resource_swap() helper
2043 -  freedreno/a6xx: Use blitter for resolve blits
2044 -  freedreno/a6xx: RB6_R8G8B8 is actually 32 bit RGBX
2045 -  freedreno/a6xx: Use A6XX_SP_2D_SRC_FORMAT_MASK macro
2046 -  freedreno/a6xx: Handle srgb blits on the blitter
2047 -  freedreno/a6xx: Move handle_rgba_blit() up
2048 -  freedreno/a6xx: Rewrite compressed blits in a helper function
2049 -  freedreno/a6xx: Set up multisample sysmem MRTs correctly
2050 -  st/mesa: Lower vars to ssa and constant prop before
2052 -  ir3: Set up full/half register conflicts correctly
2053 -  iris: Advertise PIPE_CAP_NATIVE_FENCE_FD
2054 -  iris: Print warning and return \*out = NULL when fd to syncobj fails
2058 -  gallium/swr: Fix GS invocation issues - Fixed proper setting
2059    gl_InvocationID. - Fixed GS vertices output memory overflow.
2060 -  gallium/swr: Enable some ARB_gpu_shader5 extensions Enable / add to
2061    features.txt: - Enhanced textureGather. - Geometry shader instancing.
2062    - Geometry shader multiple streams.
2063 -  gallium/swr: Fix crash when use GL_TDFX_texture_compression_FXT1
2065 -  gallivm: add TGSI bit arithmetic opcodes support
2066 -  gallium/swr: Fix glVertexPointer race condition.
2067 -  gallium/swr: Disable showing detected arch message.
2068 -  docs/GL4: update gallium/swr features
2069 -  gallium/swr: add option for static link
2070 -  gallium/swr: Fix gcc 4.8.5 compile error
2071 -  gallium/swr: simplify environmental variabled expansion code
2075 -  freedreno/drm: Fix memory leak in softpin implementation
2079 -  egl: avoid local modifications for eglext.h Khronos standard header
2084 -  ac: add missing Arcturus to the info of pc lines
2088 -  gallium: dri2: Use index as plane number.
2089 -  android: mesa: Revert "android: mesa: revert "Enable asm
2094 -  intel/dev: set default num_eu_per_subslice on gen12
2095 -  intel/perf: add TGL support
2096 -  intel/perf: fix Android build
2097 -  mesa: check draw buffer completeness on
2099 -  vulkan: bump headers/registry to 1.1.127
2100 -  anv: Properly handle host query reset of performance queries
2101 -  anv: implement VK_KHR_separate_depth_stencil_layouts
2102 -  mesa: check framebuffer completeness only after state update
2103 -  anv: invalidate file descriptor of semaphore sync fd at vkQueueSubmit
2104 -  anv: remove list items on batch fini
2105 -  anv: detach batch emission allocation from device
2106 -  anv: expose timeout helpers outside of anv_queue.c
2107 -  anv: move queue init/finish to anv_queue.c
2108 -  anv: allow NULL batch parameter to anv_queue_submit_simple_batch
2109 -  anv: prepare driver to report submission error through queues
2110 -  anv: refcount semaphores
2111 -  anv: prepare the driver for delayed submissions
2112 -  anv/wsi: signal the semaphore in the acquireNextImage
2113 -  anv: implement VK_KHR_timeline_semaphore
2114 -  intel/dev: flag the Elkhart Lake platform
2115 -  intel/perf: add EHL performance query support
2116 -  intel/perf: fix invalid hw_id in query results
2117 -  intel/perf: set read buffer len to 0 to identify empty buffer
2118 -  intel/perf: take into account that reports read can be fairly old
2119 -  intel/perf: simplify the processing of OA reports
2120 -  intel/perf: fix improper pointer access
2121 -  anv: fix missing gen12 handling
2122 -  anv: fix incorrect VMA alignment for CCS main surfaces
2123 -  anv: fix fence underlying primitive checks
2124 -  anv: fix assumptions about temporary fence payload
2125 -  intel/perf: drop batchbuffer flushing at query begin
2126 -  i965/iris: perf-queries: don't invalidate/flush 3d pipeline
2127 -  anv: constify pipeline layout in nir passes
2128 -  anv: drop unused parameter from apply layout pass
2129 -  vulkan/wsi: error out when image fence doesn't signal
2130 -  mesa: avoid triggering assert in implementation
2131 -  i965/iris/perf: factor out frequency register capture
2132 -  loader: fix close on uninitialized file descriptor value
2133 -  anv: don't close invalid syncfd semaphore
2134 -  anv: fix intel perf queries availability writes
2135 -  anv: set stencil layout for input attachments
2136 -  iris: Implement Gen12 workaround for non pipelined state
2137 -  anv: Implement Gen12 workaround for non pipelined state
2138 -  anv: only use VkSamplerCreateInfo::compareOp if enabled
2139 -  anv: fix pipeline switch back for non pipelined states
2140 -  genxml: add new Gen11+ PIPE_CONTROL field
2141 -  iris: handle new PIPE_CONTROL field
2142 -  iris: implement another workaround for non pipelined states
2143 -  anv: implement another workaround for non pipelined states
2144 -  intel/perf: expose timestamp begin for mdapi
2145 -  intel/perf: report query split for mdapi
2146 -  anv: enable VK_KHR_swapchain_mutable_format
2147 -  anv: don't report error with other vendor DRM devices
2148 -  anv: ensure prog params are initialized with 0s
2149 -  anv/iris: warn gen12 3DSTATE_HS restriction
2150 -  intel: Implement Gen12 workaround for array textures of size 1
2151 -  isl: drop CCS row pitch requirement for linear surfaces
2152 -  isl: add gen12 comment about CCS for linear tiling
2153 -  anv: implement gen9 post sync pipe control workaround
2154 -  anv: set MOCS on push constants
2158 -  radv: fix radv secure compile feature breaks compilation on armhf
2163 -  etnaviv: Fix assert when try to accumulate an invalid fd
2167 -  glsl: encode/decode types using a union with bitfields for
2169 -  glsl: encode vector_elements and matrix_columns better
2170 -  glsl: encode explicit_stride for basic types better
2171 -  glsl: encode array types better
2172 -  glsl: encode struct/interface types better
2173 -  st/mesa: call nir_opt_access only once
2174 -  st/mesa: call nir_lower_flrp only once per shader
2175 -  compiler: make variable::data::binding unsigned
2176 -  nir: pack nir_variable::data::stream
2177 -  nir: pack nir_variable::data::xfb\_\*
2178 -  radeonsi: use IR SHA1 as the cache key for the in-memory shader cache
2179 -  radeonsi: don't keep compute shader IR after compilation
2180 -  radeonsi: keep serialized NIR instead of nir_shader in
2182 -  nir: pack the rest of nir_variable::data
2183 -  nir/serialize: don't expand 16-bit variable state slots to 32 bits
2184 -  nir/serialize: store 32-bit object IDs instead of 64-bit
2185 -  nir/serialize: pack nir_variable flags
2186 -  mesa: expose SPIR-V extensions in the Compatibility profile too
2187 -  util: add blob_finish_get_buffer
2188 -  radeonsi/nir: call nir_serialize only once per shader
2189 -  radeonsi/nir: fix compute shader crash due to nir_binary == NULL
2190 -  glsl/linker: pass shader_info to analyze_clip_cull_usage directly
2191 -  compiler: pack shader_info from 160 bytes to 96 bytes
2192 -  st/mesa: fix Sanctuary and Tropics by disabling ARB_gpu_shader5 for
2194 -  st/mesa: rename DEBUG_TGSI -> DEBUG_PRINT_IR
2195 -  st/mesa: remove \\n being only printed in debug builds after printed
2197 -  st/mesa: print TCS/TES/GS/CS TGSI in the right place & keep disk
2199 -  st/mesa: add ST_DEBUG=nir to print NIR shaders
2200 -  st/mesa: remove unused TGSI-only debug printing functions
2201 -  gallium/noop: call finalize_nir
2202 -  radeonsi/nir: remove dead function temps
2203 -  radeonsi/nir: call nir_lower_flrp only once per shader
2204 -  radeonsi/nir: don't lower fma, instead, fuse fma
2205 -  mesa: enable glthread for 7 Days To Die
2206 -  st/mesa: rename delete_basic_variant -> delete_common_variant
2207 -  st/mesa: decrease the size of st_fp_variant_key from 48 to 40 bytes
2208 -  st/mesa: start deduplicating some program code
2209 -  st/mesa: initialize affected_states and uniform storage earlier in
2211 -  st/mesa: consolidate and simplify code flagging
2213 -  st/mesa: trivially merge st_vertex_program into st_common_program
2214 -  st/mesa: rename st_common_program to st_program
2215 -  st/mesa: cleanups after unification of st_vertex/common program
2216 -  st/mesa: rename occurences of stcp to stp to correspond to st_program
2217 -  st/mesa: more cleanups after unification of st_vertex/common_program
2218 -  st/mesa: subclass st_vertex_program for VP-specific members
2219 -  st/mesa: call nir_sweep in st_finalize_nir
2220 -  st/mesa: keep serialized NIR instead of nir_shader in st_program
2221 -  st/mesa: call nir_serialize only once per shader
2222 -  nir: move data.image.access to data.access
2223 -  nir/print: only print image.format for image variables
2224 -  glsl_to_nir: rename image_access to mem_access
2225 -  nir: move data.descriptor_set above data.index for better packing
2226 -  nir: don't use GLenum16 in nir.h
2227 -  ac: add radeon_info::num_rings and move ring_type to amd_family.h
2228 -  ac: fill num_rings for remaining IPs
2229 -  winsys/amdgpu: detect noop dependencies on the same ring correctly
2230 -  nir: strip as we serialize to remove the nir_shader_clone call
2231 -  nir/serialize: do ctx = {0} instead of manual initializations
2232 -  util/blob: add 8-bit and 16-bit reads and writes
2233 -  nir/serialize: pack instructions better
2234 -  nir/serialize: pack src better and limit the object count to 1M from
2236 -  nir/serialize: don't serialize var->data for temporaries
2237 -  nir/serialize: deduplicate serialized var types by reusing the last
2239 -  nir/serialize: try to store a diff in var data locations instead of
2241 -  nir/serialize: pack load_const with non-64-bit constants better
2242 -  nir/serialize: pack 1-component constants into 20 bits if possible
2243 -  nir/serialize: pack nir_intrinsic_instr::const_index[] better
2244 -  nir/serialize: try to pack two alu srcs into 1 uint32
2245 -  nir/serialize: don't store deref types if not needed
2246 -  nir/serialize: don't serialize mode for deref non-cast instructions
2247 -  nir/serialize: try to put deref->var index into the unused bits of
2249 -  nir/serialize: cleanup - fold nir_deref_type_var cases into switches
2250 -  nir/serialize: try to pack both deref array src into 32 bits
2251 -  nir/serialize: remove up to 3 consecutive equal ALU instruction
2253 -  nir/serialize: reuse the writemask field for 2 src X swizzles of SSA
2255 -  nir/serialize: serialize swizzles for vec8 and vec16
2256 -  nir/serialize: serialize writemask for vec8 and vec16
2257 -  nir/serialize: don't serialize redundant
2259 -  nir/serialize: use 3 unused bits in intrinsic for
2261 -  nir/serialize: support any num_components for remaining instructions
2262 -  ac: set swizzled bit in cache policy as a hint not to merge
2264 -  radeonsi: initialize the per-context compiler on demand
2265 -  radeonsi/nir: don't run si_nir_opts again if there is no change
2266 -  st/mesa: don't serialize all streamout state if there are no SO
2268 -  st/mesa: don't use redundant stp->state.ir.nir
2269 -  st/mesa: don't call ProgramStringNotify in glsl_to_nir
2270 -  st/mesa: propagate gl_PatchVerticesIn from TCS to TES before linking
2272 -  st/mesa: simplify looping over linked shaders when linking NIR
2273 -  st/mesa: don't use \*\* in the st_nir_link_shaders signature
2274 -  st/mesa: add st_variant base class to simplify code for shader
2276 -  ac/nir: don't rely on data.patch for tess factors
2277 -  radeonsi/nir: implement subgroup system values for SPIR-V
2278 -  radeonsi: simplify the interface of
2280 -  radeonsi: simplify get_tcs_tes_buffer_address_from_generic_indices
2281 -  radeonsi/nir: validate is_patch because SPIR-V doesn't set it for
2283 -  radeonsi/nir: don't rely on data.patch for tess factors
2284 -  radeonsi/nir: fix location_frac handling for TCS outputs
2285 -  radeonsi/nir: support interface output types to fix SPIR-V xfb
2287 -  radeonsi: enable SPIR-V and GL 4.6 for NIR
2288 -  util/driconfig: print ATTENTION if MESA_DEBUG=silent is not set
2289 -  radeonsi/gfx10: simplify some duplicated NGG GS code
2290 -  radeonsi/gfx10: fix the vertex order for triangle strips emitted by a
2292 -  llvmpipe: implement TEX_LZ and TXF_LZ opcodes
2293 -  gallivm: implement LOAD with CONSTBUF but don't enable it for
2295 -  st/mesa: support UBOs for Selection/Feedback/RasterPos
2296 -  st/mesa: save currently bound vertex samplers and sampler views in
2298 -  st/mesa: support samplers for Selection/Feedback/RasterPos
2299 -  st/mesa: support SSBOs for Selection/Feedback/RasterPos
2300 -  st/mesa: support shader images for Selection/Feedback/RasterPos
2301 -  st/mesa: use a separate VS variant for the draw module
2302 -  st/mesa: remove st_vp_variant::num_inputs
2303 -  st/mesa: remove struct st_vp_variant in favor of st_common_variant
2304 -  st/mesa: don't generate VS TGSI if NIR is enabled
2305 -  draw, st/mesa: generate TGSI for ffvp/ARB_vp if draw lacks LLVM
2306 -  st/mesa: release the draw shader properly to fix driver crashes
2308 -  st/dri: assume external consumers of back buffers can write to the
2310 -  radeonsi: enable NIR by default and document GL 4.6 support
2311 -  radeonsi/gfx10: disable vertex grouping
2312 -  radeonsi/gfx10: simplify the tess_turns_off_ngg condition
2313 -  radeonsi: don't rely on CLEAR_STATE to set PA_SC_GENERIC_SCISSOR\_\*
2314 -  ac: fix ac_get_i1_sgpr_mask for Wave32
2315 -  ac: fix the return value in cull_bbox when bbox culling is disabled
2316 -  radeonsi: deduplicate ES and GS thread enablement code
2317 -  radeonsi: disallow compute-based culling if polygon mode is enabled
2318 -  radeonsi: set is_monolithic for VS prologs when the shader is really
2320 -  radeonsi: don't wrap the VS prolog in if (ES thread) .. endif
2321 -  radeonsi/gfx10: don't insert NGG streamout atomics if they are never
2323 -  radeonsi: allow generating VS prologs with 0 inputs
2324 -  radeonsi: fix determining whether the VS prolog is needed
2325 -  radeonsi: reset more fields in si_llvm_context_set_ir to fix reusing
2327 -  radeonsi/gfx10: fix ngg_get_ordered_id
2328 -  amd/addrlib: update to the latest version
2329 -  ac/surface: fix an assertion failure on gfx9 in CMASK computation
2330 -  radeonsi/gfx10: don't declare any LDS for NGG if it's not used
2331 -  radeonsi/gfx10: enable NGG passthrough for eligible shaders
2332 -  radeonsi/gfx10: improve performance for TES using PrimID but not
2334 -  Revert "u_vbuf: Regard non-constant vbufs with non-instance elements
2336 -  winsys/radeon: initialize pte_fragment_size
2337 -  radeonsi: preserve the scanout flag for shared resources on gfx9 and
2339 -  radeonsi: ignore PIPE_BIND_SCANOUT for imported textures
2340 -  radeonsi: remove the "display_dcc_offset == 0" assertion
2341 -  radeonsi: rename SDMA debug flags
2342 -  radeonsi: remove broken and unused SI SDMA image copy code
2343 -  radeonsi: add AMD_DEBUG=nodmaclear for debugging
2344 -  radeonsi: add AMD_DEBUG=nodmacopyimage for debugging
2345 -  radeonsi: rename dma_cs -> sdma_cs
2346 -  radeonsi: move SI and CIK+ SDMA code into 1 common function for
2348 -  radeonsi: disable SDMA on gfx8 to fix corruption on RX 580
2349 -  radeonsi: remove TGSI
2350 -  gallium: put u_vbuf_get_caps return values into u_vbuf_caps
2351 -  gallium/cso_context: move non-vbuf vertex buffer and element code
2353 -  gallium: bypass u_vbuf if it's not needed (no fallbacks and no user
2355 -  ac/gpu_info: always use distributed tessellation on gfx10
2356 -  radeonsi: fix monolithic pixel shaders with two-sided colors and
2358 -  radeonsi: fix context roll tracking in si_emit_shader_vs
2359 -  radeonsi: test polygon mode enablement accurately
2360 -  radeonsi: determine accurately if line stippling is enabled for
2362 -  radeonsi: clean up messy si_emit_rasterizer_prim_state
2363 -  ac: unify build_sendmsg_gs_alloc_req
2364 -  ac: unify primitive export code
2365 -  ac/gpu_info: add pc_lines and use it in radeonsi
2366 -  ac: add 128-bit bitcount
2367 -  ac: add ac_build_s_endpgm
2368 -  radeonsi/gfx9: force the micro tile mode for MSAA resolve correctly
2370 -  radeonsi: rename desc_list_byte_size -> vb_desc_list_alloc_size
2371 -  radeonsi: add si_context::num_vertex_elements
2372 -  radeonsi: don't allow draw calls with uninitialized VS inputs
2373 -  radeonsi: simplify si_set_vertex_buffers
2374 -  ac,radeonsi: increase the maximum number of shader args and return
2376 -  radeonsi: put up to 5 VBO descriptors into user SGPRs
2377 -  radeonsi: don't enable VBOs in user SGPRs if compute-based culling
2379 -  radeonsi: fix assertion and other failures in
2381 -  radeonsi: actually enable VBOs in user SGPRs
2382 -  radeonsi: don't adjust depth and stencil PS output locations
2383 -  radeonsi: rename DBG_NO_TGSI -> DBG_NO_NIR
2384 -  radeonsi: remove TGSI from comments
2385 -  radeonsi: rename si_shader_info -> si_shader_binary_info
2386 -  radeonsi: fork tgsi_shader_info and tgsi_tessctrl_info
2387 -  radeonsi: merge si_tessctrl_info into si_shader_info
2388 -  radeonsi: clean up si_shader_info
2389 -  radeonsi: rename si_compile_tgsi_main -> si_build_main_function
2390 -  radeonsi: rename si_shader_create -> si_create_shader_variant for
2392 -  radeonsi: fold si_create_function into si_llvm_create_func
2393 -  radeonsi: remove always constant ballot_mask_bits from
2395 -  radeonsi: move PS LLVM code into si_shader_llvm_ps.c
2396 -  radeonsi: separate code computing info for small primitive culling
2397 -  ac/cull: don't read Position.Z if it's not needed for culling
2398 -  radeonsi: make si_insert_input\_\* functions non-static
2399 -  radeonsi: move VS_STATE.LS_OUT_PATCH_SIZE a few bits higher to make
2401 -  radeonsi/gfx10: separate code for getting edgeflags from the
2403 -  radeonsi/gfx10: separate code for determining the number of vertices
2405 -  radeonsi: fix si_build_wrapper_function for compute-based primitive
2407 -  radeonsi: work around an LLVM crash when using
2409 -  radeonsi: move si_insert_input\_\* functions
2410 -  radeonsi: move tessellation shader code into si_shader_llvm_tess.c
2411 -  radeonsi: remove llvm_type_is_64bit
2412 -  radeonsi: move geometry shader code into si_shader_llvm_gs.c
2413 -  radeonsi: move code for shader resources into
2415 -  radeonsi: remove useless #includes
2416 -  radeonsi: merge si_compile_llvm and si_llvm_compile functions
2417 -  gallium: add st_context_iface::flush_resource to call FLUSH_VERTICES
2418 -  st/dri: do FLUSH_VERTICES before calling flush_resource
2419 -  Revert "radeonsi: unbind image before compute clear"
2420 -  radeonsi: clean up how internal compute dispatches are handled
2421 -  radeonsi: don't invoke decompression inside internal launch_grid
2422 -  radeonsi: fix doubles and int64
2423 -  radeonsi: turn an assertion into return in si_nir_store_output_tcs
2424 -  ac: add prefix bitcount functions
2425 -  ac: add ac_build_readlane without optimization barrier
2426 -  radeonsi/gfx10: update comments and remove invalid TODOs
2427 -  radeonsi/gfx10: correct VS PrimitiveID implementation for NGG
2428 -  radeonsi/gfx10: move s_sendmsg gs_alloc_req to the beginning of
2430 -  radeonsi/gfx10: export primitives at the beginning of VS/TES
2431 -  radeonsi/gfx10: merge main and pos/param export IF blocks into one if
2433 -  radeonsi/gfx10: don't initialize VGPRs not used by NGG passthrough
2434 -  radeonsi/gfx10: move GE_PC_ALLOC setting to shader states
2435 -  radeonsi/gfx10: implement NGG culling for 4x wave32 subgroups
2436 -  ac: add helper ac_build_triangle_strip_indices_to_triangle
2437 -  radeonsi/gfx10: rewrite late alloc computation
2438 -  radeonsi/gfx10: enable GS fast launch for triangles and strips with
2440 -  radeonsi: use ctx->ac. for types and integer constants
2441 -  radeonsi: move non-LLVM code out of si_shader_llvm.c
2442 -  radeonsi: move VS shader code into si_shader_llvm_vs.c
2443 -  radeonsi: move si_shader_llvm_build.c content into si_shader_llvm.c
2444 -  radeonsi: minor cleanup in si_shader_internal.h
2445 -  radeonsi: move si_nir_build_llvm into si_shader_llvm.c
2446 -  radeonsi: fold si_shader_context_set_ir into si_build_main_function
2447 -  radeonsi: move more LLVM functions into si_shader_llvm.c
2448 -  radeonsi: make si_compile_llvm return bool
2449 -  radeonsi: make si_compile_shader return bool
2450 -  radeonsi: change prototypes of si_is_multi_part_shader &
2452 -  radeonsi: separate LLVM compilation from non-LLVM code
2453 -  util/simple_mtx: add a missing include to get ASSERTED
2454 -  gallium/util: add a cache of live shaders for shader CSO
2456 -  radeonsi: use the live shader cache
2457 -  radeonsi: restructure si_shader_cache_load_shader
2458 -  radeonsi: print shader cache stats with AMD_DEBUG=cache_stats
2459 -  radeonsi: expose shader cache stats to the HUD
2460 -  radeonsi: make screen available to shader part compilation
2461 -  radeonsi: fix a regression since the addition of si_shader_llvm_vs.c
2462 -  Revert "winsys/amdgpu: Close KMS handles for other DRM file
2464 -  Revert "winsys/amdgpu: Re-use amdgpu_screen_winsys when possible"
2465 -  radeonsi: don't report that multi-plane formats are supported
2466 -  radeonsi: fix the DCC MSAA bug workaround
2467 -  radeonsi: don't wait for shader compilation to finish when destroying
2472 -  etnaviv: Replace bitwise OR with logical OR
2473 -  etnaviv: tgsi: Fix gl_FrontFacing support
2474 -  etnaviv: Report correct number of vertex buffers
2475 -  etnaviv: Do not filter out PIPE_FORMAT_S8_UINT_Z24_UNORM on
2476    pre-HALTI2
2477 -  etnaviv: Destroy rsc->pending_ctx set in etna_resource_destroy()
2481 -  Revert "st/mesa: call nir_serialize only once per shader"
2482 -  Revert "st/mesa: keep serialized NIR instead of nir_shader in
2484 -  iris: separating out common perf code
2488 -  mapi/glapi: Generate sizeof() helpers instead of fixed sizes.
2489 -  mesa/glthread: Implement ARB_multi_bind.
2490 -  drirc: Enable glthread for dolphin/citra/yuzu.
2494 -  etnaviv: update Android build files
2498 -  egl: Implement getImage/putImage on pbuffer swrast.
2502 -  intel/compiler: Use ARRAY_SIZE()
2503 -  intel/compiler: Extract GEN\_\* macros into separate file
2504 -  intel/compiler: Split has_64bit_types into float/int
2505 -  intel/compiler: Don't disassemble align1 3-src operands on Gen < 10
2506 -  intel/compiler: Limit compaction unit tests to specific gens
2507 -  intel/compiler: Add NF some more places
2508 -  intel/compiler: Add a INVALID_{,HW_}REG_TYPE macros
2509 -  intel/compiler: Split hw_type tables
2510 -  intel/compiler: Handle invalid inputs to brw_reg_type_to_*()
2511 -  intel/compiler: Handle invalid compacted immediates
2512 -  intel/compiler: Factor out brw_validate_instruction()
2513 -  intel/compiler: Validate some instruction word encodings
2514 -  intel/compiler: Add unit tests for new EU validation checks
2515 -  intel/compiler: Validate fuzzed instructions
2516 -  intel/compiler: Test compaction on Gen <= 12
2517 -  gitlab-ci: Skip ext_timer_query/time-elapsed
2518 -  intel/compiler: Move Gen4/5 rounding to visitor
2519 -  util: Explain BITSET_FOREACH_SET params
2520 -  util: Remove tmp argument from BITSET_FOREACH_SET macro
2524 -  android: aco: fix Lower to CSSA
2525 -  android: radeonsi: fix build error due to wrong u_format.csv file
2527 -  android: util/format: fix include path list
2528 -  android: radeonsi: fix build after vl refactoring (v2)
2529 -  android: nir: add a load/store vectorization pass
2530 -  android: util: Add a mapping from VkFormat to PIPE_FORMAT.
2531 -  android: radv: fix vk_format_table.c generated source build
2532 -  android: radeonsi,ac: fix building error due to ac changes
2533 -  android: radv: build radv_shader_args.c
2537 -  gitlab-ci: Set arm job CCACHE_DIR properly
2538 -  gitlab-ci: Use separate arm64 build/test docker images
2539 -  gitlab-ci: Don't build libdrm for ARM
2540 -  gitlab-ci: Use ninja -j4 for building dEQP
2541 -  gitlab-ci: Move artifact preparation to separate script
2542 -  gitlab-ci: Share dEQP build process between x86 & ARM test image
2544 -  gitlab-ci: Sort packages in debian-install.sh
2545 -  gitlab-ci: Run piglit tests with llvmpipe
2546 -  gitlab-ci: Use separate docker images for x86 build/test jobs
2547 -  gitlab-ci: Delete install/bin from artifacts as well
2548 -  gitlab-ci: Document that ci-templates refs must be in sync
2549 -  gitlab-ci: Use functional container job names
2550 -  gitlab-ci: Rename container install scripts to match job names
2552 -  gitlab-ci: Organize images using new REPO_SUFFIX templates feature
2553 -  gitlab-ci: Directly use host-mapped directory for ccache
2554 -  gitlab-ci: Stop reporting piglit test results via JUnit
2555 -  gitlab-ci: Stop storing piglit test results as JUnit
2556 -  gitlab-ci: Put HTML summary in artifacts for failed piglit jobs
2557 -  gitlab-ci: Update to current ci-templates master
2558 -  gitlab-ci: Run piglit glslparser & quick_shader tests separately
2559 -  glsl/tests: Use splitlines() instead of strip()
2560 -  gitlab-ci: Use the common run policy for LAVA jobs as well again
2561 -  gitlab-ci: Overhaul job run policy
2562 -  gitlab-ci: Don't exclude any piglit quick_shader tests
2563 -  gitlab-ci: Test against LLVM / clang 9 on x86
2564 -  gitlab-ci: Stop using manual jobs for merge requests
2565 -  gitlab-ci: Set GIT_STRATEGY to none for the dummy job
2566 -  gitlab-ci: Use single if for manual job rules entry
2567 -  winsys/amdgpu: Keep a list of amdgpu_screen_winsyses in amdgpu_winsys
2568 -  winsys/amdgpu: Keep track of retrieved KMS handles using hash tables
2569 -  winsys/amdgpu: Only re-export KMS handles for different DRM FDs
2570 -  util: Add os_same_file_description helper
2571 -  winsys/amdgpu: Re-use amdgpu_screen_winsys when possible
2572 -  winsys/amdgpu: Close KMS handles for other DRM file descriptions
2573 -  winsys/amdgpu: Re-use amdgpu_screen_winsys when possible
2574 -  winsys/amdgpu: Close KMS handles for other DRM file descriptions
2578 -  Meson: Check for dladdr with MinGW
2579 -  disk_cache_get_function_timestamp: check for dladdr
2580 -  Meson: Add llvm>=9 modules
2582 Miguel Casas-Sanchez (1):
2584 -  i965: Ensure that all 2101010 image imports can pass framebuffer
2589 -  gallium/dri2: Fix creation of multi-planar modifier images
2590 -  gallium: Store the image format in winsys_handle
2591 -  iris: Fix import of multi-planar surfaces with modifiers
2595 -  egl/android: Restrict minimum triple buffering for android
2600 -  llvmpipe: Check thread creation errors
2604 -  st/mesa: release tgsi tokens for shader states
2605 -  svga: fix size of format_conversion_table[]
2606 -  svga: Use pipe_shader_state_from_tgsi to set shader state
2610 -  Add support for T820 CI Jobs
2611 -  ci: Remove T820 from CI temporarily
2612 -  gitlab-ci/lava: add pipeline information in the lava job name
2616 -  nir/opcodes: Add a helper function to generate the comparison binops
2617 -  nir/opcodes: Add a helper function to generate reduce opcodes
2618 -  nir: Add a 16-bit bool type
2619 -  nir: Add a 8-bit bool type
2620 -  nir/lower_alu_to_scalar: Support lowering 8- and 16-bit reduce ops
2621 -  freedreno/ir3: Support 16-bit comparison instructions
2622 -  freedreno/ir3: Add implementation of nir_op_b16csel
2623 -  freedreno/ir3: Implement f2b16 and i2b16
2624 -  freedreno/ir3: Enabling lowering 16-bit flrp
2628 -  kmsro: Extend to include ingenic-drm
2629 -  u_vbuf: Mark vbufs incompatible if more were requested than HW
2631 -  u_vbuf: Only create driver CSO if no incompatible elements
2632 -  u_vbuf: Regard non-constant vbufs with non-instance elements as free
2633 -  u_vbuf: Return true in u_vbuf_get_caps if nb of vbufs is below
2638 -  state_tracker: Handle texture view min level in st_generate_mipmap()
2642 -  intel/compiler: remove the operand restriction for src1 on GLK
2643 -  intel/compiler: fix nir_op_{i,u}*32 on ICL
2647 -  radeonsi: make si_fence_server_signal flush pipe without work
2651 -  svga: check return value of define_query_vgpu{9,10}
2655 -  compiler/spirv: Fix uses of gnu struct = {} extension
2656 -  include/CL: Update OpenCL headers to latest
2657 -  clover: Use the dispatch table type from the OpenCL headers
2658 -  clover/meson: Define OpenCL header macros
2660 Pierre-Eric Pelloux-Prayer (54):
2662 -  radeonsi: tell the shader disk cache what IR is used
2663 -  mesa: enable msaa in clear_with_quad if needed
2664 -  mesa: pass vao as a function paramter
2665 -  mesa: add EXT_dsa glVertexArray\* functions declarations
2666 -  mesa: rework \_mesa_lookup_vao_err to allow usage from EXT_dsa
2667 -  mesa: add vao/vbo lookup helper for EXT_dsa
2668 -  mesa: add EXT_dsa glVertexArray\* functions implementation
2669 -  mesa: add gl_vertex_array_object parameter to client state helpers
2670 -  mesa: add EXT_dsa glEnableVertexArrayEXT / glDisableVertexArrayEXT
2671 -  mesa: add EXT_dsa EnableVertexArrayAttribEXT /
2673 -  mesa: extract helper function from \_mesa_GetPointerv
2674 -  mesa: add EXT_dsa glGetVertexArray\* 4 functions
2675 -  mesa: fix call to \_mesa_lookup_vao_err
2676 -  radeonsi: fix shader disk cache key
2677 -  radeonsi: enable mesa_glthread for GfxBench
2678 -  mesa: update features.txt to reflect EXT_dsa status
2679 -  mesa: add ARB_framebuffer_no_attachments named functions
2680 -  mesa: add ARB_vertex_attrib_64bit VertexArrayVertexAttribLOffsetEXT
2681 -  mesa: add ARB_clear_buffer_object named functions
2682 -  mesa: add ARB_gpu_shader_fp64 selector-less functions
2683 -  mesa: add ARB_instanced_arrays EXT_dsa function
2684 -  mesa: add ARB_texture_buffer_range glTextureBufferRangeEXT function
2685 -  mesa: implement ARB_texture_storage_multisample + EXT_dsa functions
2686 -  mesa: extend vertex_array_attrib_format to support EXT_dsa
2687 -  mesa: add ARB_vertex_attrib_binding glVertexArray\* functions
2688 -  mesa: add ARB_sparse_buffer NamedBufferPageCommitmentEXT function
2689 -  mesa: enable EXT_direct_state_access
2690 -  mesa: fix warning in 32 bits build
2691 -  radeonsi: implement sdma for GFX9
2692 -  radeonsi: display cs blit count for AMD_DEBUG=testdma
2693 -  radeonsi: use gfx9.surf_offset to compute texture offset
2694 -  radeonsi: fix multi plane buffers creation
2695 -  radeonsi: dcc dirty flag
2696 -  st/mesa: add a notify_before_flush callback param to flush
2697 -  st/dri: use st->flush callback to flush the backbuffer
2698 -  radeonsi: disable dcc for 2x MSAA surface and bpe < 4
2699 -  gallium: refuse to create buffers larger than UINT32_MAX
2700 -  radeon/vcn2: enable rate control for hevc encoding
2701 -  radeonsi: check ctx->sdma_cs before using it
2702 -  radeonsi: release saved resources in si_retile_dcc
2703 -  radeonsi: release saved resources in si_compute_expand_fmask
2704 -  radeonsi: release saved resources in si_compute_clear_render_target
2705 -  radeonsi: release saved resources in si_compute_copy_image
2706 -  radeonsi: release saved resources in si_compute_clear_12bytes_buffer
2707 -  radeonsi: release saved resources in si_compute_do_clear_or_copy
2708 -  radeonsi: fix fmask expand compute shader
2709 -  radeonsi: make sure fmask expand is done if needed
2710 -  radeonsi: unbind image before compute clear
2711 -  radeonsi: drop the negation from fmask_is_not_identity
2712 -  util: call bind_sampler_states before setting sampler_views
2713 -  radeonsi: move AMD_DEBUG tests to AMD_TEST
2714 -  docs: document AMD_DEBUG variable
2715 -  radeonsi: stop using the VM_ALWAYS_VALID flag
2716 -  radeonsi/ngg: add VGT_FLUSH when enabling fast launch
2718 Prodea Alexandru-Liviu (2):
2720 -  Meson: Remove lib prefix from graw and osmesa when building with
2722 -  Appveyor: Quickly fix meson build. As this required use of Python
2727 -  lima: sync lima_drm.h with kernel
2728 -  lima: create heap buffer with new interface if available
2729 -  lima: add noheap debug option
2733 -  intel/isl: Add MOCS settings to isl_device.
2734 -  anv: Use mocs settings from isl_dev.
2735 -  iris: Use mocs from isl_dev.
2736 -  intel: Add workaround for stencil state.
2737 -  intel/genxml: Add 3DSTATE_CONSTANT_ALL packet.
2738 -  intel/aubinator: Decode 3DSTATE_CONSTANT_ALL.
2739 -  intel/blorp: Use 3DSTATE_CONSTANT_ALL to setup push constants.
2740 -  iris: Rework push constants emitting code.
2741 -  iris: Use 3DSTATE_CONSTANT_ALL when possible.
2742 -  anv: Move gen8+ push constant packet workaround.
2743 -  anv: Add get_push_range_address() helper.
2744 -  anv: Move code for emitting push constants into its own function.
2745 -  anv: Use 3DSTATE_CONSTANT_ALL when possible.
2746 -  iris: Add restriction to 3DSTATE_CONSTANT\_ packets.
2747 -  util/os_socket: Add socket related functions.
2748 -  vulkan/overlay: Add a control socket.
2749 -  vulkan/overlay: Add support for a control socket.
2750 -  vulkan/overlay: Add a command to start capturing data to a file.
2751 -  vulkan/overlay: Add basic overlay control script.
2752 -  vulkan/overlay: Update docs.
2753 -  iris: Implement WA for push constants.
2754 -  utils/os_socket: Define ssize_t on windows.
2755 -  intel: Load the driver even if I915_PARAM_REVISION is not found.
2759 -  radv: adjust loop unrolling heuristics for int64
2760 -  aco: add Instruction::usesModifiers() and add more checks in the
2762 -  radv: fix radv_nir_get_max_workgroup_size when nir=NULL
2763 -  aco: use DPP instead of exec modification when lowering GFX10
2765 -  aco: fix shuffle with uniform operands
2766 -  nir/divergence: improve DA of shuffle
2767 -  aco: fix read_invocation with VGPR lane index
2768 -  aco: don't propagate vgprs into v_readlane/v_writelane
2769 -  aco: combine read_invocation and shuffle implementations
2770 -  radv: enable FP16/FP64 denormals earlier and only for LLVM
2771 -  aco: don't combine literals into v_cndmask_b32/v_subb/v_addc
2772 -  aco: fix 64-bit fsign with 0
2773 -  aco: implement VK_KHR_shader_float_controls
2774 -  aco: refactor reduction lowering helpers
2775 -  aco: implement 64-bit integer reductions
2776 -  radv/aco: enable VK_KHR_shader_subgroup_extended_types
2777 -  nir: make nir_variable::{num_members,num_state_slots} a uint16_t
2778 -  nir: add nir_variable::index and nir_index_vars
2779 -  nir/large_constants: use nir_index_vars and nir_variable::index
2780 -  docs: update features.txt for RADV
2781 -  aco: improve waitcnt insertion around loops
2782 -  aco: fix copy+paste error
2783 -  aco: fix waitcnts for barriers at block ends
2784 -  nir: add nir_num_variable_modes and nir_var_mem_push_const
2785 -  radv: set alignment for load_ssbo/store_ssbo in meta shaders
2786 -  nir: add a load/store vectorization pass
2787 -  nir: add load/store vectorizer tests
2788 -  aco: enable load/store vectorizer
2789 -  aco: allow constant offsets for global/scratch instructions on GFX10
2790 -  aco: set dlc/glc correctly for image loads
2791 -  aco: propagate p_wqm on an image_sample's coordinate p_create_vector
2792 -  aco: fix i2i64
2793 -  aco: fix incorrect cast in parse_wait_instr()
2794 -  aco: add v_nop inbetween exec write and VMEM/DS/FLAT
2795 -  aco: improve WAR hazard workaround with >64bit stores
2796 -  aco: fix GFX10 opcodes for some global/flat atomics
2797 -  aco: fix assembly of FLAT/GLOBAL atomics
2798 -  aco: fix SADDR with FLAT on GFX10
2799 -  aco: don't enable store_global for helper invocations
2800 -  aco: improve FLAT/GLOBAL scheduling
2801 -  aco: implement global atomics
2802 -  ac/llvm: fix pointer type for global atomics
2803 -  ac/llvm: improve sync scope for global atomics
2804 -  radv: set writes_memory for global memory stores/atomics
2805 -  aco: validate the CFG
2806 -  aco: handle loop exit and IF merge phis with break/discard
2807 -  aco: fix block_kind_discard s_andn2 definition to exec
2808 -  nir/lower_io_to_vector: don't create arrays when not needed
2809 -  nir/load_store_vectorize: fix combining stores with aliasing loads
2811 -  aco/wave32: fix comparison optimizations
2812 -  aco: improve jump threading with wave32
2813 -  aco: fix vgpr alloc granule with wave32
2814 -  aco: limit register usage for large work groups
2815 -  aco: set vm for pos0 exports on GFX10
2816 -  aco: fix imageSize()/textureSize() with large buffers on GFX8
2817 -  aco: fix uninitialized data in the binary
2818 -  aco: handle VOP3 modifiers when combining a constant comparison's NaN
2820 -  aco: handle omod successors with the constant in the first operand
2821 -  aco: check usesModifiers() when identifying a neg/abs
2822 -  aco: better handle neg/abs of sgprs
2823 -  aco: set exec_potentially_empty for demotes
2824 -  aco: don't DCE atomics with return values
2825 -  aco: disable add combining for ds_swizzle_b32
2826 -  aco: check if multiplication/clamp is live when applying output
2828 -  nir/divergence: handle load_primitive_id in GS
2829 -  nir/lower_gs_intrinsics: add option for per-stream counts
2830 -  aco: update IR validator
2831 -  aco: apply literals to split mads
2832 -  aco: combine two sgprs into a VALU if they're the same
2833 -  aco: improve can_use_VOP3()
2834 -  aco: rewrite literal combining
2835 -  aco: rewrite apply_sgprs()
2836 -  aco: add check_vop3_operands()
2837 -  aco: be more careful with literals in combine_salu_{n2,lshl_add}
2838 -  aco: follow through temporary when merging tests into constant
2840 -  aco: allow applying two sgprs to an instruction
2841 -  aco: allow an extra SGPR with multiple uses to be applied to VOP3
2842 -  aco: take advantage of GFX10's constant bus limit and VOP3 literals
2843 -  aco: improve creation of v_madmk_f32/v_madak_f32
2844 -  aco: fix clamp optimization
2845 -  aco: improve clamp optimization
2846 -  aco: add min(-max(), ) and max(-min(), ) optimization
2847 -  aco: don't move literal to reg when making an instruction VOP3 on
2849 -  aco: allow input modifiers on v_cndmask_b32
2850 -  aco: replace extract_vector with copies
2851 -  aco: improve readfirstlane after uniform LDS loads
2852 -  aco: add integer min/max to can_swap_operands
2853 -  nir/sink,nir/move: move/sink load_per_vertex_input
2854 -  nir/sink,nir/move: move/sink nir_op_mov
2855 -  nir/algebraic: a & ~(a >> 31) -> imax(a, 0)
2856 -  aco: fix stack buffer overflow in apply_sgprs()
2857 -  aco: fix fall-through test in try_remove_simple_block() with
2858    back-edges
2859 -  aco: fix operand kill flags when a temporary is used more than once
2860 -  aco: fix off-by-one error when initializing sgpr_live_in
2861 -  radv: move gs copy shader creation before other variants
2862 -  aco: improve support for s_sendmsg
2863 -  radv/aco,aco: implement GS on GFX9+
2864 -  aco: implement GS on GFX7-8
2865 -  radv/aco: allow ACO for GS
2866 -  aco: explicitly mark end blocks for exports
2867 -  aco: remove needs_instance_id
2868 -  aco: implement GS copy shaders
2869 -  radv/aco: use ACO for GS copy shaders
2870 -  aco: use nir_move_copies
2871 -  aco: fix WaR check for >64-bit FLAT/GLOBAL instructions
2872 -  aco: fix operand to scc when selecting SGPR ufind_msb/ifind_msb
2873 -  aco: always add sgprs to sgpr_ids when choosing literals
2874 -  aco: fix literal application with v_cndmask_b32/v_addc_co_u32/etc
2875 -  amd/common,radv: move vertex_format_table to ac_shader_util.{h,c}
2876 -  aco: rework vertex fetching a bit
2877 -  aco: skip unused channels at the start when fetching vertices
2878 -  aco: handle unaligned vertex fetch on GFX10
2879 -  aco: value-number MUBUF instructions
2880 -  aco: use MUBUF in some situations instead of splitting vertex fetches
2881 -  aco: fix rebase error from GS copy shader support
2882 -  aco: ensure predecessors' p_logical_end is in WQM when a p_phi is in
2884 -  aco: run p_wqm instructions in WQM
2885 -  nir/algebraic: add patterns for a >> #b << #b
2886 -  nir/algebraic: add some half packing optimizations
2887 -  aco: fix target calculation when vgpr spilling introduces sgpr
2889 -  aco: don't consider loop header blocks branch blocks in
2891 -  aco: don't update demand in add_coupling_code() for loop headers
2892 -  aco: only create parallelcopy to restore exec at loop exit if needed
2893 -  aco: don't always add logical edges from continue_break blocks to
2895 -  aco: error when block has no logical preds but VGPRs are live at the
2897 -  aco: set exec_potentially_empty after continues/breaks in nested IFs
2898 -  aco: improve assertion at the end of spiller
2899 -  aco: fill reg_demand with sensible information in add_coupling_code()
2900 -  aco: parallelcopy exec mask before s_wqm
2901 -  aco: fix exec mask consistency issues
2902 -  aco: fix gfx10_wave64_bpermute
2906 -  anv: Unify GetDeviceQueue and GetDeviceQueue2
2910 -  freedreno/ir3: split pre-coloring to it's own function
2911 -  freedreno/ir3: use SSA flag on dest register too
2912 -  freedreno/ir3: ir3_print tweaks
2913 -  freedreno/ir3/ra: move regs_count==0 check
2914 -  freedreno/ir3/ra: remove ir print after livein/out
2915 -  freedreno/ir3: remove obsolete comment
2916 -  freedreno/a3xx: fix SP_FS_MRT_REG.HALF_PRECISION
2917 -  freedreno/a4xx: fix SP_FS_MRT_REG.HALF_PRECISION
2918 -  freedreno/ir3: sync disasm changes from envytools
2919 -  freedreno/ir3: also track # of nops for shader-db
2920 -  freedreno: fix eglDupNativeFenceFD error
2921 -  freedreno/ir3: fix valgrind complaint with STLW
2922 -  freedreno/ir3: remove half-precision output
2923 -  freedreno/ir3: rename fanin/fanout to collect/split
2924 -  freedreno/ir3: remove impossible condition
2925 -  freedreno/ir3: add input/output iterators
2926 -  freedreno/ir3: show input/output wrmask's in disasm
2927 -  freedreno/ir3: helper to print ir if debug enabled
2928 -  freedreno/ir3: remove first-vertex sysval
2929 -  freedreno/ir3: simplify creating sysval inputs
2930 -  freedreno/ir3: re-work shader inputs/outputs
2931 -  freedreno/ir3: only tex instructions have wrmask
2932 -  freedreno/ir3: fix gpu hang with pre-fs-tex-fetch
2933 -  freedreno/ir3: legalize cleanups
2934 -  freedreno/ir3: remove unused parameter
2935 -  freedreno/perfcntrs: small cleanup
2936 -  freedreno/perfcntrs: remove gallium dependencies
2937 -  freedreno/perfcntrs: move to shared location
2938 -  freedreno/perfcntrs: add accessor to get per-gen tables
2939 -  freedreno/perfctrs/a2xx: move CP to be first group
2940 -  freedreno/perfcntrs/a6xx: remove RBBM counters
2941 -  freedreno/perfcntrs: add fdperf
2942 -  freedreno/perfctrs/fdperf: periodically restore counters
2943 -  gitlab-ci: update deqp build so we can generate xml
2944 -  gitlab-ci/deqp: preserve full list of unexpected results
2945 -  gitlab-ci/deqp: preserve caselists for blocks with fails
2946 -  gitlab-ci/deqp: detect and report flakes
2947 -  gitlab-ci: bump arm test container
2948 -  gitlab-ci/deqp: generate xml results for fails/flakes
2949 -  gitlab-ci/deqp: generate junit results
2950 -  gitlab-ci/freedreno/a6xx: remove most of the flakes
2951 -  freedreno: use rsc->slice accessor everywhere
2952 -  freedreno: switch to layout helper
2953 -  gitlab-ci: disable junit results for deqp
2954 -  freedreno/ir3: remove store_output lowered to store_shared_ir3
2955 -  freedreno/ir3: fix neverball assert in case of unused VS inputs
2956 -  nir/lower_clip: Fix incorrect driver loc for clipdist outputs
2957 -  freedreno/fdperf: use drmOpen()
2958 -  freedreno/a6xx: disable LRZ when blending
2959 -  freedreno/a5xx+a6xx: split LRZ layout to per-gen
2960 -  freedreno/a6xx: fix LRZ layout
2961 -  freedreno/a6xx: fix LRZ logic
2962 -  freedreno/a6xx: enable LRZ by default
2963 -  spirv: add OpLifetime\*
2964 -  freedreno/ir3: add last-baryf shaderdb stat
2965 -  freedreno/ir3: add scheduler traces
2966 -  freedreno/ir3: add iterator macros
2967 -  freedreno/a6xx: fix OUT_REG() vs growable cmdstream
2968 -  nir+vtn: vec8+vec16 support
2969 -  freedreno/ir3: fix flat shading again
2970 -  nir: assert that nir_lower_tex runs after lowering derefs
2971 -  mesa/st: lower samplers before nir_lower_tex
2972 -  freedreno/ir3: rename instructions
2973 -  gitlab-ci: fix missing caselist.css/xsl
2974 -  freedreno/a6xx: limit scratch/debug markers to debug builds
2975 -  freedreno/a6xx: cleanup rasterizer state
2976 -  freedreno/a6xx: separate rast stateobj for prim restart
2977 -  freedreno/a6xx: drop a few more per-draw registers
2978 -  freedreno/a6xx: move dynamic program state to streaming stateobj
2979 -  freedreno/a6xx: add PROG_FB_RAST stateobj
2980 -  freedreno/drm: fix invalid-cmdstream-size with older kernels
2981 -  freedreno: use PIPE_CAP_RGB_OVERRIDE_DST_ALPHA_BLEND
2982 -  mesa/st: random whitespace cleanup
2983 -  freedreno/a6xx: remove special handling based on MRT format
2984 -  freedreno/a6xx: convert blend state to stateobj
2985 -  freedreno: extract vsc pipe bo from GMEM state
2986 -  freedreno: consolidate GMEM state
2987 -  freedreno: constify fd_tile
2988 -  freedreno: constify fd_vsc_pipe
2989 -  freedreno/a6xx: constify gmem state
2990 -  freedreno/a5xx: constify gmem state
2991 -  freedreno/a4xx: constify gmem state
2992 -  freedreno/a3xx: constify gmem state
2993 -  freedreno/a2xx: constify gmem state
2994 -  freedreno: get GMEM state from batch
2995 -  freedreno: add gmem state cache
2996 -  freedreno: add gmem_lock
2997 -  freedreno: remove flush-queue
2998 -  freedreno: allow ctx->batch to be NULL
3002 -  nir: Build nir_lower_point_size.c in libmesa_nir
3003 -  android: Add panfrost support to build scripts
3004 -  android: Fix u_format_table.c being generated twice
3005 -  panfrost: Prefix schedule_program to prevent collision
3006 -  android: Fix whitespace issue
3010 -  gitlab-ci: Use lavacli from packages
3014 -  gallium/scons: fix graw_gdi build
3015 -  util/atomic: Fix p_atomic_add for unlocked and msvc paths
3016 -  winsys/svga: use new ioctl for logging
3020 -  Android: Fix build issue without LLVM
3021 -  panfrost: Fix Android build
3025 -  intel: limit shader geometry on BDW GT1
3029 -  intel/compiler: Clear accumulator register before EOT
3033 -  main: fix coverity error in \_mesa_program_resource_find_name()
3037 -  radv: declare NGG scratch for VS or TES and only on GFX10
3038 -  radv: fix compute pipeline keys when optimizations are disabled
3039 -  docs: document all RADV environment variables
3040 -  radv: add a note about perftest/debug options
3041 -  radv: fix 32-bit compiler warnings
3042 -  nir: fix packing of nir_variable
3043 -  radv/gfx10: enable wave32 for compute based on shader's wavesize
3044 -  radv: hardcode the number of waves for the GFX6 LS-HS bug
3045 -  radv: determine shaders wavesize at pipeline level
3046 -  radv: rely on shader's wavesize when computing NGG info
3047 -  radv: implement VK_EXT_subgroup_size_control
3048 -  radv/gfx10: fix primitive indices orientation for NGG GS
3049 -  ac: handle pointer types to LDS in ac_get_elem_bits()
3050 -  gitlab-ci: build a specific libdrm version for ARM64
3051 -  gitlab-ci: build RADV on ARM64
3052 -  ac: fix build with recent LLVM
3053 -  radv: remove useless RADV_DEBUG=unsafemath debug option
3054 -  radv: make sure to not clear the ds attachment after resolves
3055 -  ac: add radeon_info::has_l2_uncached
3056 -  radv: implement VK_AMD_device_coherent_memory
3057 -  spirv: fix lowering of OpGroupNonUniformAllEqual
3058 -  ac: remove useless cast in ac_build_set_inactive()
3059 -  ac: add 8-bit and 16-bit supports to ac_build_shuffle()
3060 -  ac: add 8-bit and 16-bit supports to ac_build_readlane()
3061 -  ac: add 8-bit and 16-bit supports to ac_build_set_inactive()
3062 -  ac: add 8-bit and 16-bit supports to ac_build_dpp()
3063 -  ac: add 8-bit and 16-bit supports to ac_build_swizzle()
3064 -  ac: add 8-bit and 16-bit supports to get_reduction_identity()
3065 -  ac: add 8-bit and 16-bit supports to ac_build_wwm()
3066 -  ac: add 8-bit and 16-bit supports to ac_build_optimization_barrier()
3067 -  ac: add 16-bit float support to ac_build_alu_op()
3068 -  radv: advertise VK_KHR_shader_subgroup_extended_types on GFX8-GFX9
3069 -  radv: enable VK_KHR_shader_subgroup_extended_types on GFX6-GFX7
3070 -  docs: add missing new features for RADV
3071 -  pipe-loader: check that the pointer to driconf_xml isn't NULL
3072 -  gitlab-ci: move building piglit into a separate script
3073 -  gitlab-ci: fix ldd check for Vulkan drivers
3074 -  gitlab-ci: add a job that only build things needed for testing
3075 -  gitlab-ci: do not build with debugoptimized for meson-main
3076 -  gitlab-ci: build swr in meson-main
3077 -  gitlab-ci: build GLVND in meson-clang
3078 -  gitlab-ci: remove now useless meson-swr-glvnd build job
3079 -  gitlab-ci: reduce the number of scons build
3080 -  radv: disable subgroup shuffle operations on GFX10
3081 -  ac/llvm: fix the local invocation index for wave32
3082 -  meson: only build imgui when needed
3083 -  radv: set the image view aspect mask during subpass transitions
3084 -  radv: set the image view aspect mask before resolves
3085 -  radv: rework creation of decompress/resummarize meta pipelines
3086 -  radv: create decompress pipelines for separate depth/stencil layouts
3087 -  radv: select the depth decompress path based on the aspect mask
3088 -  ac/llvm: fix warning in ac_build_canonicalize()
3089 -  radv: fix reporting subgroup size with
3091 -  radv: fix enabling sample shading with SampleID/SamplePosition
3092 -  radv/gfx10: fix implementation of exclusive scans
3093 -  ac: add 8-bit and 16-bit supports to ac_build_permlane16()
3094 -  radv: enable VK_KHR_shader_subgroup_extended_types on GFX10
3095 -  ac/llvm: convert src operands to pointers if necessary
3096 -  radv: add more constants to avoid using magic numbers
3097 -  radv,ac/nir: lower deref operations for shared memory
3098 -  aco: drop useless lowering of deref operations for shared memory
3099 -  ac/llvm: fix atomic var operations if source isn't a deref
3100 -  radv: remove dead shader input/output variables
3101 -  radv: simplify a check in radv_fixup_vertex_input_fetches()
3102 -  radv/gfx10: fix the vertex order for triangle strips emitted by a GS
3103 -  gitlab-ci: rename build-deqp.sh to build-deqp-gl.sh
3104 -  gitlab-ci: add a gl suffix to the x86 test image and all test jobs
3105 -  gitlab-ci: add a new job that builds a base test image for VK
3106 -  gitlab-ci: build cts_runner in the x86 test image for VK
3107 -  gitlab-ci: build dEQP VK 1.1.6 in the x86 test image for VK
3108 -  gitlab-ci: add a new base test job for VK
3109 -  gitlab-ci: allow to run dEQP Vulkan with DEQP_VER
3110 -  gitlab-ci: configure the Vulkan ICD export with VK_DRIVER
3111 -  gitlab-ci: build RADV in meson-testing
3112 -  gitlab-ci: add a job that runs Vulkan CTS with RADV conditionally
3113 -  radv: do not use VK_TRUE/VK_FALSE
3114 -  radv: move emission of two PA_SC\_\* registers to the pipeline CS
3115 -  radv: fix possibly wrong PA_SC_AA_CONFIG value for conservative rast
3116 -  radv: synchronize after performing a separate depth/stencil fast
3118 -  radv: do not init HTILE as compressed state when dst layout allows it
3119 -  radv: initialize HTILE for separate depth/stencil aspects
3120 -  radv: implement VK_KHR_separate_depth_stencil_layouts
3121 -  gitlab-ci: set RADV_DEBUG=checkir for RADV test jobs
3122 -  ac/nir: fix out-of-bound access when loading constants from global
3123 -  radv: enable SpvCapabilityImageMSArray
3124 -  radv: handle unaligned vertex fetches on GFX6/GFX10
3125 -  radv/gfx10: fix ngg_get_ordered_id
3126 -  radv/gfx10: fix the out-of-bounds check for vertex descriptors
3127 -  ac: declare an enum for the OOB select field on GFX10
3128 -  radv: init a default multisample state for the resolve FS path
3129 -  radv: ignore pMultisampleState if rasterization is disabled
3130 -  radv: ignore pTessellationState if the pipeline doesn't use tess
3131 -  radv: ignore pDepthStencilState if rasterization is disabled
3132 -  radv: tidy up radv_pipeline_init_blend_state()
3133 -  radv: ignore pColorBlendState if rasterization is disabled
3134 -  radv: rely on pipeline layout when creating push descriptors with
3136 -  radv: return the correct pitch for linear mipmaps on GFX10
3137 -  radv: record number of color/depth samples for each subpass
3138 -  radv: implement VK_AMD_mixed_attachment_samples
3139 -  ac/surface: use uint16_t for mipmap level pitches
3140 -  radv: do not fill keys from fragment shader twice
3141 -  spirv: add SpvCapabilityImageReadWriteLodAMD
3142 -  spirv,nir: add new lod parameter to image_{load,store} intrinsics
3143 -  amd/llvm: handle nir_intrinsic_image_deref_{load,store} with lod
3144 -  aco: handle nir_intrinsic_image_deref_{load,store} with lod
3145 -  radv: advertise VK_AMD_shader_image_load_store_lod
3146 -  radv/gfx10: disable vertex grouping
3147 -  radv/gfx10: determine if a pipeline is eligible for NGG passthrough
3148 -  radv/gfx10: do not declare LDS for NGG if useless
3149 -  radv/gfx10: add support for NGG passthrough mode
3150 -  radv/gfx10: improve performance for TES using PrimID but not
3152 -  radv: only use VkSamplerCreateInfo::compareOp if enabled
3153 -  radv/gfx10: enable all CUs if NGG is never used
3154 -  radv/gfx10: simplify some duplicated NGG GS code
3155 -  vulkan/overlay: Fix for Vulkan 1.2
3156 -  radv: update VK_EXT_descriptor_indexing for Vulkan 1.2
3157 -  radv: update VK_EXT_host_query_reset for Vulkan 1.2
3158 -  radv: update VK_EXT_sampler_filter_minmax for Vulkan 1.2
3159 -  radv: update VK_EXT_scalar_block_layout for Vulkan 1.2
3160 -  radv: update VK_KHR_8bit_storage for Vulkan 1.2
3161 -  radv: update VK_KHR_buffer_device_address for Vulkan 1.2
3162 -  radv: update VK_KHR_create_renderpass2 for Vulkan 1.2
3163 -  radv: update VK_KHR_depth_stencil_resolve for Vulkan 1.2
3164 -  radv: update VK_KHR_draw_indirect_count for Vulkan 1.2
3165 -  radv: update VK_KHR_driver_properties for Vulkan 1.2
3166 -  radv: update VK_KHR_image_format_list for Vulkan 1.2
3167 -  radv: update VK_KHR_imageless_framebuffer for Vulkan 1.2
3168 -  radv: update VK_KHR_shader_atomic_int64 for Vulkan 1.2
3169 -  radv: update VK_KHR_shader_float16_int8 for Vulkan 1.2
3170 -  radv: update VK_KHR_shader_float_controls for Vulkan 1.2
3171 -  radv: update VK_KHR_shader_subgroup_extended_types for Vulkan 1.2
3172 -  radv: update VK_KHR_uniform_buffer_standard_layout for Vulkan 1.2
3173 -  radv: update VK_KHR_timeline_semaphore for Vulkan 1.2
3174 -  radv: implement Vulkan 1.1 features and properties
3175 -  radv: implement Vulkan 1.2 features and properties
3176 -  radv: enable Vulkan 1.2
3177 -  aco: fix emitting SMEM instructions with no operands on GFX6-GFX7
3178 -  aco: do not select 96-bit/128-bit variants for ds_read/ds_write on
3180 -  aco: do not combine additions of DS instructions on GFX6
3181 -  aco: implement stream output with vec3 on GFX6
3182 -  aco: fix emitting slc for MUBUF instructions on GFX6-GFX7
3183 -  aco: print assembly with CLRXdisasm for GFX6-GFX7 if found on the
3185 -  aco: fix constant folding of SMRD instructions on GFX6
3186 -  aco: do not use the vec3 variant for stores on GFX6
3187 -  aco: do not use the vec3 variant for loads on GFX6
3188 -  aco: add new addr64 bit to MUBUF instructions on GFX6-GFX7
3189 -  aco: implement nir_intrinsic_load_barycentric_at_sample on GFX6
3190 -  radv: fix double free corruption in radv_alloc_memory()
3191 -  radv: add explicit external subpass dependencies to meta operations
3192 -  radv: handle missing implicit subpass dependencies
3193 -  spirv: add SpvCapabilityFragmentMaskAMD
3194 -  nir: add two new texture ops for multisample fragment color/mask
3196 -  spirv: add support for SpvOpFragment{Mask}FetchAMD operations
3197 -  nir/lower_input_attachments: lower nir_texop_fragment_{mask}_fetch
3198 -  ac/nir: add support for nir_texop_fragment_{mask}_fetch
3199 -  aco: add support for nir_texop_fragment_{mask}_fetch
3200 -  radv: advertise VK_AMD_shader_fragment_mask
3201 -  aco: fix printing assembly with CLRXdisasm on GFX6
3202 -  aco: fix wrong IR in nir_intrinsic_load_barycentric_at_sample
3203 -  aco: implement nir_intrinsic_store_global on GFX6
3204 -  aco: implement nir_intrinsic_load_global on GFX6
3205 -  aco: implement nir_intrinsic_global_atomic\_\* on GFX6
3206 -  aco: implement 64-bit nir_op_ftrunc on GFX6
3207 -  aco: implement 64-bit nir_op_fceil on GFX6
3208 -  aco: implement 64-bit nir_op_fround_even on GFX6
3209 -  aco: implement 64-bit nir_op_ffloor on GFX6
3210 -  aco: implement nir_op_f2i64/nir_op_f2u64 on GFX6
3211 -  ac/llvm: fix missing casts in ac_build_readlane()
3212 -  aco: combine MRTZ (depth, stencil, sample mask) exports
3213 -  aco: fix a hardware bug for MRTZ exports on GFX6
3214 -  aco: fix a hazard with v_interp\_\* and v_{read,readfirst}lane\_\* on
3216 -  aco: copy the literal offset of SMEM instructions to a temporary
3217 -  radv: enable ACO support for GFX6
3218 -  radv: print NIR shaders after lowering FS inputs/outputs
3219 -  radv: do not allow sparse resources with multi-planar formats
3220 -  radv: enable VK_AMD_shader_fragment_mask on GFX6-GFX7
3221 -  compiler: add a new explicit interpolation mode
3222 -  spirv: add support for SpvDecorationExplicitInterpAMD
3223 -  compiler: add PERSP to the existing barycentric system values
3224 -  compiler: add new SYSTEM_VALUE_BARYCENTRIC\_\*
3225 -  spirv: add support for SpvBuiltInBaryCoord\*
3226 -  nir: add nir_intrinsic_load_barycentric_model
3227 -  nir: lower SYSTEM_VALUE_BARYCENTRIC\_\* to nir_load_barycentric()
3228 -  nir: add nir_intrinsic_interp_deref_at_vertex
3229 -  nir: lower interp_deref_at_vertex to load_input_vertex
3230 -  spirv: implement SPV_AMD_shader_explicit_vertex_parameter
3231 -  ac/llvm: implement VK_AMD_shader_explicit_vertex_parameter
3232 -  aco: implement VK_AMD_shader_explicit_vertex_parameter
3233 -  radv: gather which input PS variables use an explicit interpolation
3235 -  radv: implement VK_AMD_shader_explicit_vertex_parameter
3236 -  radv: bump conformance version to 1.2.0.0
3237 -  radv: remove the non conformant VK implementation warning on GFX10
3238 -  aco: fix VS input loads with MUBUF on GFX6
3239 -  radv/gfx10: add a separate flag for creating a GDS OA buffer
3240 -  radv/gfx10: implement NGG GS queries
3241 -  radv/gfx10: re-enable NGG GS
3242 -  radv: refactor physical device properties
3243 -  aco: fix MUBUF VS input loads when expanding vec3 to vec4 on GFX6
3244 -  aco: do not use ds_{read,write}2 on GFX6
3245 -  aco: fix waiting for scalar stores before "writing back" data on
3246    GFX8-GFX9
3247 -  aco: fix creating v_madak if v_mad_f32 has two sgpr literals
3248 -  nir: do not use De Morgan's Law rules for flt and fge
3252 -  loader: #define PATH_MAX when undefined (eg. Hurd)
3253 -  util: Do not fail to build on unknown pthread_setname_np
3254 -  meson: Do not require libdrm for DRI2 on hurd
3258 -  radeon/vcn: Handle crop parameters for encoder
3262 -  radeonsi: use compute shader for clear 12-byte buffer
3266 -  kmsro: Add "mcde" entry point
3270 -  nir: fix couple of compile warnings
3271 -  util/android: fix android build errors
3272 -  Revert "egl: implement new functions from
3274 -  Revert "egl: handle EGL_IMAGE_EXTERNAL_FLUSH_EXT"
3275 -  Revert "st/dri: add support for EGL_EXT_image_flush_external"
3276 -  Revert "st/dri: assume external consumers of back buffers can write
3278 -  Revert "dri_interface: add interface for
3280 -  mesa: allow bit queries for EXT_disjoint_timer_query
3281 -  Revert "mesa: allow bit queries for EXT_disjoint_timer_query"
3282 -  mesa: allow bit queries for EXT_disjoint_timer_query
3283 -  gitlab-ci: update Piglit commit, update skips
3284 -  mapi: add GetInteger64vEXT with EXT_disjoint_timer_query
3285 -  glsl: handle max uniform limits with lower_const_arrays_to_uniforms
3286 -  gitlab-ci: bump piglit checkout commit
3287 -  glsl: additional interface redeclaration check for SSO programs
3288 -  intel/compiler: add newline to limit_dispatch_width message
3289 -  intel/compiler: force simd8 when dual src blending on gen8
3290 -  dri: add \__DRI_IMAGE_FORMAT_SXRGB8
3291 -  i965: expose MESA_FORMAT_B8G8R8X8_SRGB visual
3292 -  mesa/st/i965: add a ProgramResourceHash for quicker resource lookup
3293 -  mesa: create program resource hash in a single place
3294 -  iris: set depth stall enabled when depth flush enabled on gen12
3295 -  anv: set depth stall enabled when depth flush enabled on gen12
3296 -  isl/gen12: add reminder comment about missing WA with 3D surfaces
3297 -  anv: fix assert in GetImageDrmFormatModifierPropertiesEXT
3298 -  anv: add assert for isl_mod_info in choose_isl_tiling_flags
3299 -  anv: initialize clear_color_is_zero_one
3300 -  egl/android: fix buffer_count for applications setting max count
3301 -  anv/android: setup gralloc1 usage from gralloc0 usage manually
3302 -  anv/android: make format_supported_with_usage static
3303 -  intel/vec4: fix valgrind errors with vf_values array
3304 -  glsl: fix a memory leak with resource_set
3305 -  iris: fix aux buf map failure in 32bits app on Android
3309 -  winsys/svga: Enable transhuge pages for buffer objects
3310 -  svga: Avoid discard DMA uploads
3311 -  gallium/util: Increase the debug_flush map depth
3312 -  svga: Fix banded DMA upload
3316 -  st/va: Convert interlaced NV12 to progressive
3317 -  util/format: Add the P010 format used for 10-bit videos
3318 -  gallium: Add PIPE_FORMAT_P010 support
3319 -  st/va: Add support for P010, used for 10-bit videos
3320 -  radeon: Use P010 for decoding of 10-bit videos
3321 -  r600: Remove HEVC related code since HEVC is not supported
3322 -  mesa: Prevent \_MaxLevel from being less than zero
3323 -  Revert "st/va: Convert interlaced NV12 to progressive"
3327 -  glsl: just use NIR to lower outputs when driver can't read outputs
3328 -  glsl: disable lower_fragdata_array() for NIR drivers
3329 -  mesa: add ARB_shading_language_include stubs
3330 -  glsl: add infrastructure for ARB_shading_language_include
3331 -  mesa: add ARB_shading_language_include infrastructure to
3333 -  mesa: add helper to validate tokenise shader include path
3334 -  mesa: add \_mesa_lookup_shader_include() helper
3335 -  mesa: add copy_string() helper
3336 -  mesa: add glNamedStringARB() support
3337 -  mesa: implement glGetNamedStringARB()
3338 -  mesa: make error checking optional in \_mesa_lookup_shader_include()
3339 -  mesa: implement glIsNamedStringARB()
3340 -  mesa: implement glGetNamedStringivARB()
3341 -  mesa: split \_mesa_lookup_shader_include() in two
3342 -  mesa: implement glDeleteNamedStringARB()
3343 -  glsl: add ARB_shading_language_include support to #line
3344 -  glsl: pass gl_context to glcpp_parser_create()
3345 -  glsl: add preprocessor #include support
3346 -  glsl: error if #include used while extension is disabled
3347 -  glsl: add can_skip_compile() helper
3348 -  glsl: delay compilation skip if shader contains an include
3349 -  mesa: add support cursor support for relative path shader includes
3350 -  mesa: add shader include lookup support for relative paths
3351 -  mesa: implement glCompileShaderIncludeARB()
3352 -  mesa: enable ARB_shading_language_include
3353 -  gitlab-ci: bump piglit checkout commit
3354 -  gitlab-ci: update for arb_shading_language_include
3355 -  compiler: move build definition of pp_standalone_scaffolding.c
3356 -  radv: add some infrastructure for fresh forks for each secure compile
3357 -  radv: add a secure_compile_open_fifo_fds() helper
3358 -  radv: create a fresh fork for each pipeline compile
3359 -  docs: update source code repository documentation
3360 -  glsl: move calculate_array_size_and_stride() to link_uniforms.cpp
3361 -  glsl: don't set uniform block as used when its not
3362 -  glsl: make use of active_shader_mask when building resource list
3363 -  glsl/nir: iterate the system values list when adding varyings
3364 -  docs: remove mailing list as way of submitting patches
3365 -  glsl: move nir_remap_dual_slot_attributes() call out of glsl_to_nir()
3366 -  glsl: copy the how_declared field when converting to nir
3367 -  nir: add some fields to nir_variable_data
3368 -  glsl: copy the new data fields when converting to nir
3369 -  glsl: add support for named varyings in
3371 -  glsl: add subroutine support to nir_build_program_resource_list()
3372 -  st/glsl_to_nir: call gl_nir_lower_buffers() a little later
3373 -  st/glsl_to_nir: use nir based program resource list builder
3374 -  st/glsl_to_nir: fix SSO validation regression
3375 -  glsl: rename gl_nir_link() to gl_nir_link_spirv()
3376 -  glsl: add gl_nir_link_check_atomic_counter_resources()
3377 -  glsl: add new gl_nir_link_glsl() helper
3378 -  glsl: reorder link_and_validate_uniforms() calls
3379 -  mesa: add new UseNIRGLSLLinker constant
3380 -  glsl: use nir linker to link atomics
3381 -  glsl: add check_image_resources() for the nir linker
3382 -  glsl: use nir version of check_image_resources() for nir linker
3383 -  glsl: move check_subroutine_resources() into the shared util code
3384 -  glsl: call check_subroutine_resources() from the nir linker
3385 -  glsl: move uniform resource checks into the common linker code
3386 -  glsl: call uniform resource checks from the nir linker
3387 -  glsl: move calculate_subroutine_compat() to shared linker code
3388 -  glsl: call calculate_subroutine_compat() from the nir linker
3389 -  glsl: fix potential bug in nir uniform linker
3390 -  glsl: remove bogus assert in nir uniform linking
3391 -  glsl: fix check for matrices in blocks when using nir uniform linker
3392 -  glsl: count uniform components and storage better in nir linking
3393 -  glsl_to_nir: update interface type properly
3394 -  glsl: fix gl_nir_set_uniform_initializers() for image arrays
3398 -  ac: Handle invalid GFX10 format correctly in ac_get_tbuffer_format.
3399 -  aco: Make sure not to mistakenly propagate 64-bit constants.
3400 -  aco: Treat all booleans as per-lane.
3401 -  aco: Optimize out trivial code from uniform bools.
3402 -  aco: Fix operand of s_bcnt1_i32_b64 in emit_boolean_reduce.
3403 -  aco: Remove superfluous argument from emit_boolean_logic.
3404 -  aco: Remove lower_linear_bool_phi, it is not needed anymore.
3405 -  aco: Optimize load_subgroup_id to one bit field extract instruction.
3406 -  aco/wave32: Change uniform bool optimization to work with wave32.
3407 -  aco/wave32: Replace hardcoded numbers in spiller with wave size.
3408 -  aco/wave32: Introduce emit_mbcnt which takes wave size into account.
3409 -  aco/wave32: Add wave size specific opcodes to aco_builder.
3410 -  aco/wave32: Use lane mask regclass for exec/vcc.
3411 -  aco/wave32: Fix load_local_invocation_index to support wave32.
3412 -  aco/wave32: Use wave_size for barrier intrinsic.
3413 -  aco/wave32: Allow setting the subgroup ballot size to 64-bit.
3414 -  aco/wave32: Fix reductions.
3415 -  aco: Fix uniform i2i64.
3416 -  ac/llvm: Fix ac_build_reduce in wave32 mode.
3417 -  aco/wave32: Set the definitions of v_cmp instructions to the lane
3419 -  aco: Implement 64-bit constant propagation.
3420 -  aco: Allow optimizing vote_all and nir_op_iand.
3421 -  aco: Don't skip combine_instruction when definitions[1] is used.
3422 -  aco: Optimize out s_and with exec, when used on uniform bitwise
3424 -  aco: Flip s_cbranch / s_cselect to optimize out an s_not if possible.
3425 -  nouveau/nvc0: add extern keyword to nvc0_miptree_vtbl.
3426 -  intel/compiler: Fix array bounds warning on GCC 10.
3427 -  radeon: Move si_get_pic_param to radeon_vce.c
3428 -  r600: Move get_pic_param to radeon_vce.c
3429 -  gallium: Fix a couple of multiple definition warnings.
3430 -  radeon: Fix multiple definition error with radeon_debug
3431 -  aco: Fix -Wstringop-overflow warnings in aco_span.
3432 -  aco: Fix maybe-uninitialized warnings.
3433 -  aco: Fix signedness compare warning.
3434 -  aco: Make a better guess at which instructions need the VCC hint.
3435 -  aco: Transform uniform bitwise instructions to 32-bit if possible.
3436 -  aco/gfx10: Fix VcmpxExecWARHazard mitigation.
3437 -  aco: Fix the meaning of is_atomic.
3438 -  aco/optimizer: Don't combine uniform bool s_and to s_andn2.
3442 -  gallium/swr: Fix arb_transform_feedback2
3446 -  gitlab-ci: Disable lima jobs
3447 -  gitlab-ci: Run only LAVA jobs in special-named branches
3448 -  panfrost: Add checksum fields to SFBD descriptor
3449 -  panfrost: Set 0x10 bit on mali_shader_meta.unknown2_4 on T720
3450 -  panfrost: Rework format encoding on SFBD
3451 -  panfrost: Take into account texture layers in SFBD
3452 -  panfrost: Decode blend shaders for SFBD
3453 -  panfrost: Generate polygon list manually for SFBD
3454 -  panfrost: Print the right zero field
3455 -  panfrost: Pipe the GPU ID into compiler and disassembler
3456 -  panfrost: Set depth and stencil for SFBD based on the format
3457 -  panfrost: Multiply offset_units by 2
3458 -  panfrost: Make sure the shader descriptor is in sync with the GL
3460 -  gitlab-ci: Remove limit on kernel logging
3461 -  panfrost: Just print tiler fields as-is for Tx20
3462 -  panfrost: Rework buffers in SFBD
3463 -  gitlab-ci: Fix dir name for VK-GL-CTS sources
3464 -  panfrost: Don't print the midgard_blend_rt structs on SFBD
3465 -  panfrost: Add quirks system to cmdstream
3466 -  panfrost: Simplify shader patching
3467 -  panfrost: White list the Mali T720
3468 -  gitlab-ci: Test Panfrost on T720 GPUs
3469 -  panfrost: Add PAN_MESA_DEBUG=sync
3470 -  panfrost: Hold a reference to sampler views
3471 -  pan/midgard: Remove undefined behavior
3472 -  nir: Don't copy empty array
3473 -  util: Don't access members of NULL pointers
3474 -  panfrost: Don't lose bits!
3475 -  st/mesa: Don't access members of NULL pointers
3476 -  panfrost: Handle Z24_UNORM_S8_UINT as MALI_Z32_UNORM
3477 -  panfrost: Increase PIPE_SHADER_CAP_MAX_OUTPUTS to 16
3478 -  panfrost: Dynamically allocate array of texture pointers
3479 -  panfrost: Map with size of first layer for 3D textures
3480 -  panfrost: Store internal format
3481 -  gitlab-ci: Update kernel for LAVA to 5.5-rc1 plus fixes
3482 -  gitlab-ci: Switch LAVA jobs to use shared dEQP runner
3483 -  gitlab-ci: Upgrade kernel for LAVA jobs to v5.5-rc5
3484 -  gitlab-ci: Consolidate container and build stages for LAVA
3488 -  panfrost: free last_read/write tables in mir_create_dependency_graph
3489 -  panfrost: free allocations in schedule_block
3490 -  panfrost: add lcra_free() to free lcra state
3491 -  panfrost: free spill cost table in mir_spill_register
3495 -  lima: add debug prints for BO cache
3496 -  lima: align size before trying to fetch BO from cache
3497 -  lima: ignore flags while looking for BO in cache
3498 -  lima: set dithering flag when necessary
3499 -  lima: add support for gl_PointSize
3500 -  lima: enable tiling
3501 -  lima: handle DRM_FORMAT_MOD_INVALID in resource_from_handle()
3502 -  lima: expose tiled format modifier in query_dmabuf_modifiers()
3503 -  lima: use single BO for GP outputs
3504 -  lima: drop suballocator
3505 -  lima: fix allocation of GP outputs storage for indexed draw
3506 -  lima: postpone PP stream generation
3507 -  lima: don't reload and redraw tiles that were not updated
3508 -  lima: fix PP stream terminator size
3509 -  lima: use linear layout for shared buffers if modifier is not
3511 -  lima: add debug flag to disable tiling
3512 -  lima: drop support for R8G8B8 format
3513 -  lima: fix PLBU_CMD_PRIMITIVE_SETUP command
3514 -  lima: fix viewport clipping
3515 -  lima: implement polygon offset
3516 -  lima: fix PIPE_CAP\_\* to mark features that aren't supported yet
3517 -  lima: add new findings to texture descriptor
3518 -  lima: fix handling of reverse depth range
3519 -  ci: lava: pass CI_NODE_INDEX and CI_NODE_TOTAL to lava jobs
3520 -  ci: Re-enable CI for lima on mali450
3521 -  lima: implement invalidate_resource()
3522 -  nir: don't emit ishl in \_nir_mul_imm() if backend doesn't support
3524 -  lima: use imul for calculations with intrinsic src
3525 -  lima: ppir: don't delete root ld_tex nodes without successors in
3527 -  lima: ppir: always create move and update ld_tex successors for all
3529 -  lima: disable early-z if fragment shader uses discard
3533 -  swr: Fix build with llvm-10.0.
3534 -  panfrost: Fix gnu-empty-initializer build errors.
3535 -  scons: Bump C standard to gnu11 on macOS 10.15.
3536 -  util/u_thread: Restrict u_thread_get_time_nano on macOS.
3537 -  swr: Fix build with llvm-10.0.
3538 -  swr: Fix build with llvm-10.0.
3539 -  lima: Fix build with GCC 10.
3540 -  swr: Fix GCC 4.9 checks.
3541 -  panfrost: Remove unused anonymous enum variables.
3545 -  u_vbuf: add logic to use a limited number of vbufs
3546 -  u_vbuf: use single vertex buffer if it's not possible to have
3551 -  util/u_thread: Fix build under Haiku
3555 -  glsl: Enable textureSize for samplerExternalOES
3556 -  meson: Fix linkage of libgallium_nine with libgalliumvl
3557 -  meta: Cleanup function for DrawTex
3558 -  main: allow external textures for BindImageTexture
3559 -  meta: Add cleanup function for Bitmap
3563 -  Revert "draw: revert using correct order for prim decomposition."
3567 -  zink: confused compilation macro usage for zink in target helpers.