Lines Matching +full:iris +full:- +full:rules
1 Mesa 24.2.0 Release Notes / 2024-08-14
20 ---------------
24 c02bb72cea290f78b11895a0c95c7c92394f180d7ff66d4a762ec6950a58addf mesa-24.2.0.tar.xz
28 ------------
30 - VK_KHR_dynamic_rendering_local_read on RADV
31 - VK_EXT_legacy_vertex_attributes on lavapipe, ANV, Turnip and RADV
32 - VK_MESA_image_alignment_control on RADV
33 - VK_EXT_shader_replicated_composites on ANV, dozen, hasvk, lavapipe, nvk, RADV, and Turnip
34 - VK_KHR_maintenance5 on v3dv
35 - VK_KHR_maintenance7 on RADV
36 - VK_EXT_depth_clamp_zero_one on v3dv
37 - GL_ARB_depth_clamp on v3d
38 - Defaulting to a new shader cache implementation, reducing filesystem overhead.
42 ---------
44 - u_debug_stack_test.capture_not_overwritten fails on i386
45 - turnip: vulkaninfo crashed
46 - turnip-freereno Build error (/usr/local/include/xf86drm.h:40:10: fatal error: drm.h: No such file…
47 - tu: compilation failing when compiling turnip with only kgsl and X11 enabled
48 - samplerCube constructor in fragment shader no longer converting uvec2 bindless texture handles ( …
49 - anv: gen9.5 flakiness in dEQP-VK.multiview.dynamic_rendering.depth.*
50 - vaapi decoding corruption with green blocks
51 - interpolateAt precision lowering unhandled by glsl_to_nir()
52 - [anv] CS2 crashes on LNL
53 - [anv] Dota2 does not start on LNL
54 - [radeonsi][bisected][regression] glClientWaitSync() quickly times out with INT64_MAX timeout
55 - d3d10umd: Build regression in 24.2.0-devel
56 - zink/tu: glcts flake on a750
57 - nouveau: advertises GL_EXT_memory_object without implementing \`*UID` callbacks
58 - LIBGL_DRIVERS_PATH gone
59 - [Regression][Vulkan][TGL][Bisected]vkCmdCopyQueryPoolResults failed to write buffer with compute …
60 - Worms Revolution: not rendering explosion effects
61 - crash on pushbuf_validate nvc0_blit do_blit_framebuffer
62 - piglit: cl-api-build-program crashes
63 - i915g: glGenerateMipmap() fails with 2048×2048 textures
64 - [radeonsi] Asterix & Obelix XXLRomastered: river misrendered (completely black)
65 - Build fails without Vulkan
66 - No dependency check for PyYAML in meson.build
67 - GPU Hang in Metal Gear Rising Revengeance
68 - VK_ERROR_DEVICE_LOST A770 DXVK Fallout 3
69 - [Bisected] Recent compile issue in libnak
70 - anv: Wrong push constant values for bytes?
71 - anv: dEQP-VK.protected_memory tests GPU hang on MTL
72 - RustiCL (or maybe not...): radeonsi freezes after 2 hours of simulation, zink works just fine
73 - ci_run_n_monitor.py doesn't monitor manual jobs
74 - Crash in util_idalloc_resize due to glBindTexture with a way-too-large ID
75 - mesa-24.1.2 fails to compile: ast_to_hir.cpp:5371:39: error: '%s' directive argument is null
76 - [regression][bisected] VMware Xv video displays as black rectangle
77 - Blender 4.2,4.3 crashes when rendering with motion blur on RDNA3 cards (OpenGL/radeonsi)
78 - nvk: regression with multiple games crashing
79 - Transparent background in Blender 3D view with nouveau
80 - turnip: latest git does not build anymore
81 - ACO Unimplemented intrinsic instr
82 - RADV/ACO: assert on per-sample interpolation
83 - radv: large descriptor layout creation is slow
84 - Gnome shell (wayland) crashes when opening any window
85 - DRI Intel drivers fix a problem in Redhat 7 (Mesa 18), but are not included for Redhat 8 (Mesa ve…
86 - Vulkan: ../src/nouveau/vulkan/nvk_physical_device.c:1109: VK_ERROR_INCOMPATIBLE_DRIVER
87 - RADV: Smooth lines affect triangle rendering
88 - [armhf build error][regression] error: ‘StringMapIterator’ was not declared in this scope; did yo…
89 - Build fails with latest llvm 19: error: no matching function for call to unwrap(LLVMOpaqueModule*…
90 - tu: support KHR_8bit_storage
91 - Incorrect colours on desktop and apps
92 - nir: Incorrect nir_opt_algebraic semantics for signed integer constants causing end-to-end miscom…
93 - src/gallium/frontends/clover/meson.build:93:40: ERROR: Unknown variable "idep_mesaclc".
94 - panfrost: mpv is broken on T604
95 - Nightly CI is broken
96 - [radv] GPU hang in Starfield on RX 5700 XT
97 - anv, isl, iris: Clarify and improve CCS + FCV on gfx12
98 - isl: CPCB horizontal and vertical alignment requirements unknown
99 - Indika: flickering black artifacting on the snow
100 - intel/isl: Split Xe2 changes into new files of Xe2.
101 - rusticl: Generated rusticl_mesa_bindings.c fails to find include
102 - isl: Remove 512B pitch requirement for non-displayable CCS
103 - MESA 24.1 - broken zink OpenGL under Windows
104 - Blue flickering rectangles on AMD RX 7600
105 - GPU hangs on AMD Radeon RX 6400 on a fragment shader
106 - v3dv: vkcube-wayland crashes
107 - intel/brw: scoreboarding regression
108 - regression in !29436 for radv+angle on stoney
109 - [radv][regression] Starfield invisible terrain on a 7900 XTX
110 - free_zombie_shaders() leave context in a bad state (access violation occurs)
111 - r300: X11 fails to start with the modesetting driver (glamor is broken with R300/R400 gpus).
112 - [NINE]Far Cry 1 trees flicker regression[bisected][traces]
113 - Vulkan: Most sync2 implementations are missing new access flags
114 - Incorrect buffer_list advance when writing disjoint image descriptors
115 - ANV: Block shadows in Cyberpunk on Intel A770
116 - ACO ERROR: Temporary never defined or are defined after use
117 - [ANV] Graphics memory allocation in Total War: Warhammer 3
118 - DG2: God of War trace fails to play
119 - Borderlands trace fails to play on dg2
120 - NVK: Vulkan apps simply terminated with segfault under wayland and Xwayland
121 - NVK: VK_ERROR_OUT_OF_DEVICE_MEMORY on swapchain creation
122 - anv/zink regression: piglit.spec.arb_fragment_layer_viewport.layer-no-gs
123 - [anv] failures when upgrading vulkancts 1.3.6 -> 1.3.7 on intel mesa ci
124 - RustiCL: deadlock when calling clGetProfilingInfo() on callbacks
125 - [Intel][Vulkan][Gen12] Vulkan compute shader is 3x slower than the same OpenCL kernel
126 - turnip: Broken AHB support
127 - zink: nir validation failures in Sparse code
128 - nir: nir_opt_varyings uses more stack than musl libc has
129 - dEQP-VK.pipeline.pipeline_library.shader_module_identifier.pipeline_from_id.graphics regression
130 - freedreno + perfetto missing dependency on adreno_common.xml.h
131 - anv: unbounded shader cache
132 - radv: Crash due to nir validation fail in Enshrouded
133 - vulkan/wsi/wayland: valgrid reports invalid read in \`vk_free` call in \`wsi_wl_surface_analytics…
134 - android: sRGB configs no longer exist after !27709
135 - bisected: turnip: deqp regressions
136 - aco: Radeonsi unable to use rusticl
137 - anv: clean up default_pipeline_cache in anv_device
138 - [24.1-rc4] fatal error: intel/dev/intel_wa.h: No such file or directory
139 - Turnip driver is crashing since turnip: ANB/AHB support got merged
140 - vcn: rewinding attached video in Totem cause [mmhub] page fault
141 - When using amd gpu deinterlace, tv bt709 properties mapping to 2 chroma
142 - a530: ir3_context_error assertion (unknown vertex shader output name: VARYING_SLOT_EDGE)
143 - VCN decoding freezes the whole system
144 - [RDNA2 [AV1] [VAAPI] hw decoding glitches in Thorium 123.0.6312.133 after https://gitlab.freedesk…
145 - radv regression bisected: Flickering textures (vega)
146 - [Regression][Bisected] EGL/Wayland: QT applications terminated by SIGSEGV (Address boundary error…
147 - WSI: Support VK_IMAGE_ASPECT_MEMORY_PLANE_i_BIT_EXT for DRM Modifiers in Vulkan
148 - nvk: Tracker issue for gamescope support
149 - nvk: Implement VK_EXT_image_drm_format_modifier
150 - nvk: NVK_DEBUG=zero_memory is hitting an assert
151 - nvk: Implement VK_EXT_conservative_rasterization
152 - zink sparse: Improper semaphore handling
153 - zink sparse: Reference to mip tails should be refcounted
154 - radv: Enshrouded GPU hang on RX 6800
155 - NVK Zink: Wrong color in Unigine Valley benchmark
156 - intel vulkan incremental build takes forever
157 - intel vulkan incremental build takes forever
158 - 24.0.6: build fails
159 - shader with multidimensional array in shader storage buffer takes too long to compile
160 - panforst: T604 issue with using u32 for flat varyings
161 - lp_screen.c:601:4: error: 'snprintf' will always be truncated; specified size is 16, but format s…
162 - [anv] FINISHME: support YUV colorspace with DRM format modifiers
163 - gen9/11 test became flaky: piglit.spec.!opengl 1_4.blendminmax
164 - mesa 24 intel A770 KOTOR black shadow smoke scenes
165 - nvk: Implement VK_EXT_pipeline_robustness
166 - [bisected][regression] kitty fails to start due to \`glfwWindowHint(GLFW_SRGB_CAPABLE,true)`
167 - r600: bisected 5eb0136a3c561 breaks a number of piglits
168 - [bdw][bisected][regression] assertion failure in nir_validate.c
169 - Graphical glitches in RPCS3 after updating Vulkan Intel drivers
170 - [R600] OpenGL and VDPAU regression in Mesa 23.3.0 - some bitmaps get distorted.
171 - VAAPI radeonsi: VBAQ broken with HEVC
172 - tu: weird fail in packing tests
173 - radv/video: 10-bit support
174 - radv: vkCmdWaitEvents2 is broken
175 - anv: add support for EDS3::extendedDynamicState3AlphaToCoverageEnable
176 - ci: switch from CI_JOB_JWT to id_tokens
177 - Zink: enabled extensions and features may not match
178 - anv: share embedded samplers
182 -------
186 - mesa: Enable EXT_shadow_samplers for GLES2
187 - gallium: Rename ${target}/target.c to ${target}/{$target}_target.c
188 - treewide: Include mesa_interface.h not dri_interface.h
189 - mesa_interface: Set ourselves free
190 - mesa_interface: Move out of GL/internal/
191 - gallium/dril: Compatibility stub for the legacy DRI loader interface
192 - dri: Let dril handle the DRI driver link farm
193 - gallium/meson: Deconflate swrast/softpipe/llvmpipe
197 - Revert "egl/wayland: Remove EGL_WL_create_wayland_buffer_from_image"
201 - v3dv/cmd_buffer: always bind pipeline static state
202 - v3dv/ci: dEQP-VK.dynamic_state.*.double_static_bind are fixed now
203 - v3dv: port dynamic state tracking to use Mesa Vulkan
204 - v3dv: provide implementation for vkCmdBindVertexBuffers2
205 - v3dv: provide implementation for CmdSetViewportWithCount
206 - v3dv: CullMode and FrontFace are dynamic now
207 - v3dv: DepthBoundsTestEnable is dynamic now
208 - v3dv: move depth CFG bits setting to cmd buffer emission
209 - v3dv: ez_state/incompatible_ez_test could be recomputed at cmd_buffer
210 - v3dv: PrimitiveTopology is now dynamic
211 - v3dv: StencilOp and StencilTestEnable are now dynamic
212 - v3dv/ci: update expected list due VK_EXT_extended_dynamic_state
213 - v3dv: enable VK_EXT_extended_dynamic_state
214 - v3dv/cmd_buffer: missing updates due PrimitiveTopology being dynamic now
215 - v3dv: fixes StencilTestEnable handling
216 - v3dv: PrimitiveRestartEnable is now dynamic.
217 - v3dv: DepthBiasEnable is dynamic now
218 - v3dv: SetRasterizerDiscardEnable is dynamic now
219 - v3dv: enable VK_EXT_extended_dynamic_state2
220 - v3dv: add debug option to disable custom pipeline caches for meta operations
221 - v3dv/meta_clear: take into account multiview for the custom clear pipeline caches
222 - v3dv/meta_clear: use v3dv_renderpass used as parameter
223 - v3dv/device: compute maxDescriptorSet*Limits multiplying per-stage by 4
224 - v3dv/device: set DescriptorUpdateAfterBind limits
225 - v3d/devinfo: unify comment style
226 - broadcom: move HW-dependant constants to v3d_device_info
227 - v3d,v3dv: document cl_emit_with_prepacked
228 - v3dv/pipeline: ensure vk_graphics_pipeline_all_state alive when still needed
229 - drm-shim: stub synobj_timeline_wait and query ioctl
233 - lavapipe: fixes for macOS support
234 - lavapipe: build "Windows" check should use the host machine, not the \`platforms` option.
235 - util: fix memory related OS calls on macOS
236 - wsi: fix compilation on macOS
237 - util: macOS support for cnd_monotonic
241 - egl/gbm: Walk device list to initialize DRM platform
245 - ac/surface: fix version check for gfx12 DCC
249 - pan/mdg: quirk to disable auto32
250 - panfrost: implement SFBD raw format support on v4
254 - egl/wayland: Pass dri2_wl_formats to create_dri_image
255 - egl/wayland: Fail EGL surface creation if opaque format is unsupported
259 - lavapipe: Set ICD api_version to 1.3
263 - vulkan: optimize vk_dynamic_graphics_state_any_dirty
264 - vulkan: add helper to fill out spirv caps automatically
265 - nir/lower_subgroups: add filter
266 - nir/lower_subgroups: add generic scan/reduce lower
267 - nir/lower_subgroups: relax ballot_type_to_uint
268 - nir/lower_robust_access: also handle image derefs
269 - docs: add header-stub for vk_enum_to_str
270 - vulkan: add vk_debug_ignored_stype helper
271 - nvk: use common stype debug
272 - broadcom: use common stype debug
273 - pvr: use common stype debug
274 - anv,hasvk: use common stype debug
275 - dzn: use common stype debug
276 - nir: add is_first_fan_agx sysval
277 - nir: add texops for AGX border colour emulation
278 - nir: add quad_ballot_agx intrinsic
279 - nir,agx: add depth=never workaround
280 - nir,agx: fix load_active_subgroup_index
281 - compiler: add ACCESS_IN_BOUNDS_AGX
282 - agx: optimize and/or with booleans
283 - agx: enable more lowering
284 - agx: fix query LOD of array
285 - agx: fix some ms texture packing
286 - agx: handle cross-workgroup memory barriers
287 - agx: allow 8-bit bcsel
288 - agx: fix phi translation corruption
289 - agx: fix load_helper_invocation with sample shading
290 - agx: fix frag sidefx with sample shading
291 - agx: handle subgroup barriers
292 - agx: fix spilling inside sample loop
293 - agx: switch to demote internally
294 - agx: lower nir_intrinsic_load_num_subgroups
295 - agx: delete unreachable blocks
296 - agx: model more subgroup ops
297 - agx: lower shuffle
298 - agx: handle non-immediate shuffles in divergent CF
299 - agx: handle quad_broadcast
300 - agx: handle quad swaps
301 - agx: add missing b2b16 implementation
302 - agx: forbid uniforms on ballots
303 - agx: lower 8-bit subgroups
304 - agx: flesh out subgroup lowering
305 - agx: report uses_txf
306 - agx: expose agx_link_libagx
307 - agx: document another sample_mask restriction
308 - agx: reserve scratch registers for mem<-->mem swaps
309 - agx: optimize txf with lod 0
310 - agx: fix bogus unit test
311 - agx: stash early_fragment_test info
312 - agx: handle quad reduce
313 - agx: implement quad_ballot
314 - agx: lower more quad ops
315 - agx: optimize elect()
316 - agx: fix UB in cursor comparison
317 - ail: constify everything
318 - asahi: mark eMRT loads as in-bounds
319 - asahi: calculate validity when unpacking
320 - asahi: agx_translate_sample_count
321 - asahi: assert bo size > 0
322 - asahi: unwrap pointless null check
323 - asahi: implement PIPE_CAP_QUERY_MEMORY_INFO
324 - asahi: rm unused #include
325 - asahi: resize key
326 - asahi: cleanup fs epilog link info
327 - asahi: move agx_link_varyings_vs_fs
328 - asahi: fix prolog emit
329 - asahi: pack UVS key properly
330 - asahi: plumb shader stage into info
331 - asahi: get debug in common
332 - asahi: rm deadcode
333 - asahi: drop rgb10a2_sint rendering
334 - asahi: add missing rgba4 format
335 - asahi: fix 1D array atomics
336 - asahi: fix txf/image_load robustness with arrays
337 - asahi: rework VBO lower for divisor=0
338 - asahi: mv AGX_MAX_OCCLUSION_QUERIES define
339 - asahi: handle agx_ppp_fragment_face_2 with no info
340 - asahi: clarify format code in image lowering
341 - asahi: fix rgb565 blending
342 - asahi: fix store_output component/offset
343 - asahi: fix sample ID with multiblock epilogs
344 - asahi: lower texture instructions with epilogs
345 - asahi: fix cull unknown bits
346 - asahi: simplify image atomic lowering
347 - asahi: move primitive MSAA field
348 - asahi: free libagx if we don't use a ralloc memctx
349 - asahi: eliminate troublesome empty uniforms
350 - asahi: rearrange VS uniforms
351 - asahi: set src_type for store_output
352 - asahi: rm dead code
353 - asahi: add agx_index_size_to_B helper
354 - asahi: move some GS lowering into lower_gs
355 - asahi: don't use load_num_vertices in geometry shaders
356 - asahi: mv vertex_id_for_topology_class into GS lowering
357 - asahi: rm another num_vertices use
358 - asahi: rm dated comment
359 - asahi: rm unused lower
360 - asahi: rm num_vertices uses
361 - asahi: rm redundant input_vertices
362 - asahi: mv initialization of grid z for indirect GS
363 - asahi: rm more dead lowering
364 - asahi: rm always true param
365 - asahi: update comment
366 - asahi: update comment for maint5
367 - asahi: eliminate num_workgroups for VS->GS + VS->TCS
368 - asahi: drop bogus assertion
369 - asahi: pack tilebuffer usc word ahead-of-time
370 - asahi: add agx_ppp_push_merged helper
371 - asahi: use ppp_merge
372 - asahi: don't allocate varyings ourselves
373 - asahi: don't allocate for ppp updates
374 - asahi: extend varying linking for tri fan weirdness
375 - asahi: plumb tri fan flatshading through common
376 - asahi: don't ralloc in agx_fast_link
377 - asahi: extend epilog key for force early frag handling
378 - asahi: don't reserve extra UVS space for layer
379 - libagx: use sub_group_scan_inclusive_add
380 - libagx: add query copy kernel
381 - libagx: don't use get_group_id()
382 - asahi/decode: QoL improvements
383 - asahi: track imports for decode
384 - asahi: clean up bg/eot counts
385 - asahi: rename meta -> bg/eot
386 - asahi: don't allocate for USC words
387 - asahi: split frag shader words
388 - asahi: split CDM Launch words
389 - asahi: unify naming for COUNTS structs
390 - nir/tex_instr_result_size: handle subpass_ms
391 - nir/lower_point_size: support lowered i/o
392 - asahi/decode: drop Apple-specific decode check
393 - libagx: rm unused field
394 - libagx: fix static assert
395 - libagx: fix triangle fan + prim restart + GS/XFB
396 - libagx: drop unused !indexed path
397 - libagx: add libagx_copy_xfb_counters helper
398 - asahi: be robust against out of sync shader info
399 - agx: fix draw param gather for sw vs
400 - asahi: split param structs for GS internal kernel
401 - agx: rework libagx I/O lowering
402 - asahi: add missing lowerings
403 - asahi: force bindless for eMRT
404 - asahi: bounds check eMRT stores
405 - asahi: support bigger buffer textures
406 - asahi: add AGX_TEXTURE_FLAG_CLAMP_TO_0 flag
407 - agx: handle discard with force early tests
408 - asahi: pack blend key
409 - agx: switch to combined clip/cull
410 - asahi: add flag controlling sample mask without MSAA
411 - asahi: use scalar outputs for rast shaders
412 - asahi: move null descriptor routines to common
413 - asahi: implement rba2 for uniform texel buffers
414 - asahi: implement rba2 for storage texel buffers
415 - agx: prepare for lower_wpos_center
416 - asahi: extract agx_calculate_vbo_clamp
417 - agx: fix indirect CF accounting
418 - mesa: fix duplicate initializer
419 - nir/lower_wpos_center: clean up
420 - nir/builtin_builder: factor out nir_build_texture_query
421 - asahi: use nir_build_texture_query
422 - gallium: remove ability to probe asahi
423 - asahi: add broken bits of unstable Linux UAPI
424 - agx: fix 64-bit bcsel ingestion
425 - agx: fix fmin/fmax with (-0, 0) pair
426 - libagx: fix uint8_t definition
427 - libagx: make index buffer fetch robust
428 - libagx: generalize query copies
429 - asahi: implement rba2 semantics for vbo
430 - asahi: fix vbo clamp with stride=0
431 - asahi: implement robustness2 for msaa image stores
432 - asahi: be more clever about GS side effects
433 - nir/lower_robust_access: handle MSAA images
434 - nir: add nir_metadata_control_flow
435 - treewide: use nir_metadata_control_flow
436 - nir: document restriction on load_smem_amd constantness
437 - vulkan: reference count vk_descriptor_update_template
438 - vulkan: handle push DUT with emulated secondaries
439 - vulkan: fix potential UAF with vk_cmd_enqueue_CmdPushDescriptorSetKHR
440 - vulkan: handle enqueueing CmdPushConstants2KHR
441 - vulkan: handle enqueueing CmdPushDescriptorSet2KHR
442 - lvp: use common push constant enqueue
443 - lvp: use common push descriptor set enqueue
444 - lvp: fix silly casting for sampler desc updates
445 - lvp: use common descriptor update templates
446 - nir/format_convert: remove unorm bit size assert
447 - nir: add nir_def_replace helper
448 - treewide: use nir_def_replace sometimes
449 - agx: fix insidious ballot optimizer bug
450 - agx: add unit test for ballot bug
451 - agx: set discard_is_demote
452 - nir: add nir_break_if helper
453 - treewide: use nir_break_if
454 - nir: fix miscompiles with rules with INT32_MIN
455 - nir/algebraic: explicitly suffix constants
456 - nir/opt_constant_folding: fix array size define
457 - zink: move print_pipeline_stats
458 - zink: print pipeline stats for compute shader-db
459 - util: add dui/uid helpers
460 - nir: add nir_alu_instr float controls queries
461 - nir/search: use ALU float control helpers
462 - nir: use MIN2/MAX2 opcodes for imin/umax folding
463 - nir: strengthen fmin/fmax definitions with signed zero
464 - glsl/float64: handle signed zero with min/max
465 - nir/lower_double_ops: handle signed zero with min/max
466 - nir/lower_alu: remove dead #define
467 - nir: add lower_fminmax_signed_zero
468 - agx: set lower_fminmax_signed_zero
469 - agx: do not flush denorms for fp16 fmin/fmax
470 - asahi: drop old comment
471 - asahi: drop stale comment
472 - asahi: make agx_pack opencl compatible
473 - asahi: tuck in null query check
474 - libagx: specify heap size explicitly
475 - asahi,libagx: tessellate on device
476 - libagx: add kernel for incrementing CS counter
477 - asahi: handle CS pipeline stat with indirect dispatch
478 - libagx: handle VS/IA pipeline stats on GPU
479 - asahi: eliminate load_num_workgroups from TCS unrolled ID
480 - nouveau/drm-shim: set ram_user
481 - nvk: add instruction count exec property
482 - nir/opt_constant_folding: fix array size define, pt 2
483 - zink: remove extraneous \\n with shaderdb
484 - zink: match shader-db report.py format
488 - tu: Disable depth and stencil tests when attachment state requires it
492 - android: Fix zink build failure
496 - nine: avoid using post-compacted indices with state expecting pre-compacted ones
500 - drisw: fix build without dri3
504 - meson: fix deprecation warning in create-android-cross-file.sh
505 - android: simplify building libgallium_dri on Android
509 - nvk: update 3d classes for conservative raster
510 - nvk: implement VK_EXT_conservative_rasterization
511 - nvk: import SPH headers files from open-gpu-doc
512 - nvk: generate Rust bindings from SPH header files
516 - asahi: Make asahi_clc build work on x86_64->x86 builds
520 - radv: Use zerovram for Enshrouded.
521 - util/disk_cache: Fix cache marker refresh.
522 - util/disk_cache: Delete the old multifile cache if using the default.
523 - util/cache_test: Add tests for old cache deletion.
524 - relnotes: Add an entry about the new cache default.
528 - vk/graphics_state: Add last bits for extraPrimitiveOverestimationSize
532 - panvk: Prepare things so we can push sysvals to our push uniform buffer
533 - panvk: Put dynamic uniform buffers at the end of the UBO array
534 - panvk: Move the dynamic SSBO descriptors to their own UBO
535 - panvk: Prepare dynamic buffer descriptors at bind time
536 - panvk: Lower sysvals to push uniforms
537 - panvk: Kill panvk_sysval_vec4
538 - panfrost: do not write outside num_wg_sysval
539 - panfrost: Add the BO containing fragment program descriptor to the batch
540 - pan/kmod: Fix a syncobj leak in the panthor backend
541 - pan/kmod: Make default allocator thread-safe
542 - panvk: clang-format fixups
543 - panvk: Kill panvk_queue_get_device()
544 - panvk: Get rid of panvk_descriptor_state::dirty
545 - panvk: Move panvk_cmd_state::batch to panvk_cmd_buffer::cur_batch
546 - panvk: Kill unused dynamic state bits
547 - panvk: Get rid of special attribute support
548 - panvk: Split the graphics and compute state at the cmd_buffer level
549 - panvk: Split compute/graphics pipeline objects
550 - panvk: Use memory pools to store pipeline shaders/descriptors
551 - panvk: Kill the panvk_pipeline_builder object
552 - panvk: Transition the graphics pipeline logic to vk_graphics_pipeline_state
553 - panvk: Fully transition to vk_viewport_state
554 - panvk: Fully transition to vk_rasterization_state
555 - panvk: Fully transition to vk_input_assembly_state
556 - panvk: Use vk_color_blend_state to fill our blend constant
557 - panvk: Fully transition to vk_depth_stencil_state
558 - panvk: Fully transition to vk_multisample_state
559 - panvk: Set unused attribute buffers descriptors to zero
560 - panvk: Rename non_vs_attribs into img_attribs
561 - panvk: Prevent re-emission of image attributes used in vertex shaders
562 - panvk: Move VS attribute/buffer state to panvk_cmd_graphics_state
563 - panvk: Emit VS-accessible image attributes at a fixed offset
564 - panvk: Leave holes in the attribute locations used by a shader
565 - panvk: Fully transition to vk_vertex_binding_state
566 - pan/blend: Move constant inlining out of pan_blend_create_shader()
567 - pan/blend: Expose pan_blend_create_shader()
568 - panvk: Add a blend library to deal blend shaders/descriptors
569 - panvk: Don't pass the stage to shader_create()
570 - panvk: Simplify shader initialization in the pipeline logic
571 - panvk: Fix/simplify the shader linking logic
572 - panvk: Replace the stages array in panvk_draw_info by vs/fs fields
573 - panvk: Move fs_rsd fields to an fs sub-struct
574 - panvk: Emit the fragment shader RSD dynamically
575 - panvk: Lower global memory IOs
576 - panvk: Clean Midgard leftovers in the cmd_close_batch() path
577 - panvk: Use vk_pipeline_shader_stage_to_nir()
578 - panvk: Kill cmd_get_tiler_context()
579 - panvk: Make sure we dump memory mappings before crashing
580 - pan/decode: Be robust to NULL texture payload
581 - pan/desc: Add missing format in translate_s_format()
582 - pan/jc: Drop unused pool argument passed to pan_jc_add_job()
583 - panvk: Add a render state to panvk_cmd_graphics_state
584 - panvk: Take VK_RENDERING_{RESUM,SUSPEND}ING_BIT flags into account
585 - panvk: Force a preload when the render area is not 32x32 aligned
586 - panvk: Skip depth/stencil attachments with non-matching aspect mask
587 - panvk: Fix dynamic rendering with images containing both depth and stencil
588 - panvk: Make sure we don't lose clear-only operations
589 - panvk: Make sure we run the fragment shader if alpha_to_coverage is enabled
590 - panvk: Make sure replay of command buffers containing Dispatch calls works
591 - panvk: Override the default GetRender[in]AreaGranularityKHR()
592 - egl: Use gbm_bo_create_with_modifiers2() when the surface has non-zero flags
593 - panvk: Fix formatting around OpaqueCaptureAddress implementation
594 - panvk/ci: Flag exact_sampling.*.edge_right test as fails
595 - pan/bi: Make sure global loads/stores don't exceed 16 bytes
596 - pan/bi: Fix dynamic indexing of push constants
597 - panvk: Fix Cube/2DArray/3D img -> buf copies
598 - panvk: Don't bail out when allocationSize is zero in AllocateMemory()
599 - panvk: Prepare for Valhall image views
600 - panvk: Prepare for Valhall buffer views
601 - panvk: Prepare things for compiling valhall source files
602 - panvk: Extend Valhall descriptor set implementation to support Bifrost
603 - panvk: Overhaul the Bifrost descriptor set implementation
604 - panvk: Refcount private BOs
605 - panvk: Store private BOs in lists instead of dynarrays
606 - panvk: Prepare panvk_mempool for shared device memory pools
607 - panvk: Use memory pools for internal GPU data attached to vulkan objects
608 - pan/desc: Extend pan_emit_fbd() to support multilayer rendering
609 - pan/desc: Prepare things for fragment job chaining
610 - pan/blitter: Let pan_preload_fb() callers queue the jobs to the job chain
611 - panvk: Use IDVS jobs when we can
612 - panvk: Add support for layered rendering
616 - intel/brw: Don't print IP as part of the dump
617 - intel/brw: Hide register pressure information in dumps
618 - intel/brw: Use \`vNN` instead of \`vgrfNN` when printing instructions
619 - intel/brw: Fix commas when dumping instructions
620 - spirv: Add MESA_SPIRV_DEBUG=values to dump all values
621 - intel/brw: Track the number of uses of each def in def_analysis
622 - intel/brw: Fix typo in DPAS emission code
623 - intel/brw: Add unit tests for scoreboard handling FIXED_GRF with stride
624 - intel/brw: Make component_size() consistent between VGRF and FIXED_GRF
625 - glsl: Fix warning related to tg4_offsets in release mode
626 - intel/brw: Print SWSB information when dumping instructions
627 - intel/brw: Reorganize lowering of LocalID/Index to handle Mesh/Task
628 - anv: Use brw_nir_lower_cs_intrinsics for lowering Mesh/Task LocalID
629 - intel/brw: Remove unused brw_reg related functions
630 - intel/brw: Remove RALLOC helper from fs_reg
631 - intel/brw: Remove unused variable from test
632 - intel/brw: Move fs_reg data members up to brw_reg
633 - intel/brw: Use public inheritance for fs_reg/brw_reg
634 - intel/brw: Move most member functions from fs_reg to brw_reg
635 - intel/brw: Remove conversion from fs_reg to brw_reg
636 - intel/brw: Replace some fs_reg constructors with functions
637 - intel/brw: Remove duplicated functions between fs_reg/brw_reg
638 - intel/brw: Rename brw_reg() helper to brw_make_reg()
639 - intel/brw: Make fs_reg an alias of brw_reg
640 - intel/brw: Replace uses of fs_reg with brw_reg
641 - intel/brw: Rename fs_reg_* helpers to brw_reg_*
642 - intel/brw: Move brw_reg helpers into brw_reg.h
643 - intel/brw: Don't set acc_wr_control for Xe2
644 - intel/brw: Use brw_inst_set_group() to set QtrCtrl and NibCtrl
645 - intel/brw: Account for reg_unit() in assembler
646 - intel/brw: Don't print extra newlines in assembler
647 - intel/brw: Split off assembler logic into library
648 - spirv: Don't warn about FPFastMathMode if not OpenCL
649 - intel/brw: Convert missing uses of ralloc to linear in fs_live_variables
650 - intel/elk: Convert missing uses of ralloc to linear in fs_live_variables
652 Chia-I Wu (8):
654 - gallium: add pipe_picture_desc::flush_flags
655 - frontends/va: track whether there are imported/exported surfaces
656 - frontends/va: set PIPE_FLUSH_ASYNC when possible
657 - radeonsi: prep for pipe_picture_desc::flush_flags
658 - radeonsi: respect pipe_picture_desc::flush_flags
659 - radv: check gs_copy_shader directly for executable props
660 - radv: make radv_pipeline_has_ngg static
661 - drm-shim: intercept access as well
665 - isaspec: Add method to get the displayname of BitSetEnumValue
666 - isaspec: Improve 'meta' handling
667 - etnaviv: isa: Drop capturing of python output
668 - etnaviv: isa: Add clang-format special comments
669 - etnaviv: isa: Print dst_full for ALU
670 - etnaviv: isa: Switch to enum isa_thread
671 - etnaviv: isa: Add more flags to etna_inst
672 - etnaviv: isa: Rework modeling of left shift for store/load
673 - etnaviv: isa: Add name for full writemask
674 - mr-label-maker: Add teflon marker
675 - etnaviv: isa: Do src swizzle with isaspec
676 - clc: Always use spir for 32 bit
677 - etnaviv: Zero init all srcs passed to etna_emit_alu(..)
678 - ci: uprev mold to 2.32.0
679 - gallium: Add vkms entrypoint
680 - nak: Move nak_optimize_nir declaration to nak_private.h
681 - meson: Update proc-macro2 subproject
682 - meson: Update syn subproject
683 - meson: Add pest rust dependencies
684 - meson: Add roxmltree rust dependency
685 - meson: Add indexmap rust dependencies
686 - etnaviv: isa: Add meta elements to instructions
687 - etnaviv: isa: Generate Rust FFI bindings for asm.h
688 - etnaviv: isa: Make header C++ safe
689 - etnaviv: isa: Add meson version check
690 - etnaviv: isa: Add IsaParser proc_macro_derive
691 - etnaviv: isa: Add struct etna_asm_result
692 - etnaviv: isa: Make etna_asm_result usable in Rust
693 - etnaviv: isa: Add EtnaAsmResultExt trait
694 - etnaviv: isa: Add parser module
695 - etnaviv: isa: Add C function impl
696 - etnaviv: isa: Add cli assembler
697 - etnaviv: isa: Extend disasm test
698 - ci/etnaviv: Drop shaders\@glsl-bug-110796 line
699 - etnaviv: isa: Drop 1:1 mapping of opc to bits
700 - etnaviv: isa: Add support for extended instructions
701 - nak: Update comment about explicit padding
702 - etnaviv: isa: Add support for bitset's displayname
703 - etnaviv: isa: Rework branch instruction
704 - nak: Set has_imad32 conditionally
705 - nak: Move imad late optimization to nir
706 - dri: fix driver names
710 - v3d: Move spec\@arb_texture_view\@rendering-formats, Crash in broadcom-rpi4-fails
711 - v3d: Move spec\@!opengl 1.1\@getteximage-formats, Fail in broadcom-rpi4-fails
712 - broadcom: fix issue of ‘addr’ is used uninitialized
716 - radv/video: don't truncate frame_num and POC to 32
717 - vulkan/video: generate profile_tier_level structure correctly
718 - vulkan/video: correctly set sub-layer ordering in H.265 VPS/SPS
722 - Uprev Piglit to 7aa7bc1b01d57b4b091c4fc82a94a6ff47f38ebf
723 - Uprev Piglit to 8a6ce9c6fc5c8039665655bca4904d5601c6dba0
724 - Uprev Piglit to e180f96239edba441f22f58dfc852cafb902844a
725 - Uprev Piglit to fdf3fc09deb6beecdf212e65a16c645112540b59
726 - Uprev Piglit to 647d0725024f72bc49bbc91c686c5f61168a1fe8
727 - Uprev Piglit to 582f5490a124c27c26d3a452fee03a8c85fa9a5c
731 - nir: Fix out-of-bounds access in ntt_emit_store_output()
735 - ir3: Add scan_clusters_macro to ir3_valid_flags()
736 - ir3: Add scan_clusters.macro to is_subgroup_cond_mov_macro()
737 - ir3: Validate tied sources better
738 - ir3/ra: Don't demote movmsk instructions to non-shared
739 - ir3: Rewrite postsched dependency handling
740 - ir3/legalize: Use define for register size
741 - ir3: Rewrite regmask implementation
742 - ir3/ra: Prepare for shared half-regs
743 - ir3/ra: Fix printing shared reg file
744 - ir3/ra: Prepare for shared phis
745 - ir3: Fix lowering shared parallel copies with immed src
746 - ir3/lower_pcopy: Fix immed/const flags for copy from shared
747 - ir3: Fix shared parallel copy validation
748 - ir3: Don't use swz with shared registers
749 - ir3/lower_copies: Handle HW bug with shared half-floats
750 - ir3/lower_copies: Fix "inaccessible" half reg lowering with shared regs
751 - ir3/ra: Use ra_reg_get_num() for validating num
752 - ir3: Use INVALID_REG in array store
753 - ir3: Reset num when creating parallel copies
754 - ir3: Validate that shared registers are in-bound
755 - ir3: Allow propagation of normal->shared copies
756 - ir3: Moves with shared destination are always legal
757 - ir3/legalize: Take (ss) into account in WaR hazards
758 - ir3/legalize: Remove bad (eq) micro-optimization
759 - ir3/legalize: any/all/getone are non-prefetch helper users
760 - ir3: Use correct category for OPC_PUSH_CONSTS_LOAD_MACRO
761 - ir3: Add support for "scalar ALU"
762 - ir3: Implement source restrictions for shared ALU
763 - ir3: Validate scalar ALU sources
764 - ir3: Immediate source for stc is invalid
765 - ir3: Don't emit single-source collects
766 - ir3/cp: Support swapping mad srcs for shared regs
767 - ir3/cf: Don't fold shared conversions
768 - ir3: Distinguish lowered shared->normal moves
769 - ir3: Add support for ldc.u
770 - ir3: Add builder support for shared immediates
771 - ir3: Create reduce identity directly
772 - ir3: Make type_flags() return a bitmask enum
773 - ir3: Support scalar ALU in the builder
774 - ir3: Add scalar ALU-specific passes
775 - ir3: Get sources before emitting scan_clusters.macro
776 - ir3: Rewrite shared reg handling when translating from NIR
777 - ir3: Directly use shared registers when possible
778 - ir3/nir: Fix imadsh_mix16 definition
779 - ir3: Use scalar ALU instructions when possible
780 - ir3: Don't scalarize all SSBO instructions
781 - ir3: Don't manually scalarize SSBO loads
782 - freedreno/a7xx: Add AQE-related registers from kgsl
783 - freedreno/a7xx: Add A7XX_HLSQ_DP_STR location from kgsl
784 - freedreno/crashdec: Initial a7xx support
785 - freedreno: Update HLSQ_*_CMD registers for a7xx
786 - docs/android: Fix example meson cross file again
787 - ir3: Put VS->TCS barrier after preamble
788 - ir3/legalize: Insert dummy bary.f after preamble
789 - freedreno,ir3: Add has_early_preamble
790 - tu: Workaround early preamble HW bug
791 - freedreno/a6xx: Workaround early preamble HW bug
792 - ir3: Add ir3_info::early_preamble
793 - tu: Implement early preamble
794 - freedreno/a6xx: Implement early preamble
795 - ir3: Enable early preamble
796 - tu: Use image aspects for feedback loops
797 - tu: Support VK_EXT_attachment_feedback_loop_dynamic_state
798 - tu: Use a7xx terminology for flushes
799 - freedreno, tu: Use CLEAN events on a7xx
800 - tu: Fix unaligned indirect command synchronization
801 - tu: Don't WFI after every dispatch
802 - freedreno/a7xx: Fix register file size
803 - ir3: Make sure constlen includes stc/ldc.k/ldg.k instructions
804 - freedreno: Disable early preamble on a6xx gen4
805 - ir3, tu, freedreno: Move early_preamble to ir3_shader
806 - tu: Add early preamble statistic
807 - ir3: Introduce elect_any_ir3
808 - ir3: Use elect_any_ir3 in preambles
809 - freedreno: Fix RBBM_NC_MODE_CNTL variants
810 - tu: Add support for aligned substreams
811 - ir3: Fix UBO size with indirect driver params
812 - tu: Make cs writeable for GMEM loads when FDM is enabled
813 - tu: Fix fdm_apply_load_coords patchpoint size
814 - tu: Support VK_EXT_fragment_density_map on a750
815 - tu: Support bufferDeviceAddressCaptureReplay on kgsl
816 - freedreno: Fix decoding primitive counter events on a7xx
817 - tu: Add VPC hardware workaround for a750
818 - ir3: Fix stg/ldg immediate offset on a7xx
819 - nir/instr_set: Return the matching instruction
820 - nir/instr_set: Don't remove matching instruction
821 - ir3: Split out bindless tex/samp encoding
822 - ir3: Don't consider r63.x as a GPR
823 - ir3: Plumb through descriptor prefetch intrinsics
824 - ir3: Make preamble rematerialization common code
825 - ir3: Expand preamble rematerialization
826 - ir3: Add descriptor prefetching optimization on a7xx
830 - pan/bi: fix 1D array tex coord lowering
831 - pan/bi: clean up tex coord lowering
832 - panfrost: report correct MAX_VARYINGS
833 - panvk: remove descriptor pool counters
834 - panvk: enable KHR and EXT BDA
838 - zink: Always call deinit_multi_pool_overflow when destroying zink_descriptor_pool_multi
839 - ci: Allow to override the virglrenderer render server
840 - venus: sync protocol for VK_KHR_maintenance5
841 - venus: enable VK_KHR_maintenance5
842 - venus/ci: add more recently found flakes
843 - wsi: Make sure to return a valid wayland id string
844 - venus/ci: Update expectations
845 - ci: Make sure to install libraries in the right directory on debian
849 - util: detect_os: add DETECT_OS_FUCHSIA and DETECT_OS_POSIX_LITE
850 - util: u_thread: add Fuchsia support
851 - util: os_misc: add Fuchsia support
852 - util: u_dl: add Fuchsia support
853 - util: os_time: add Fuchsia support
854 - vulkan/util: add missing dependencies
855 - meson: remove unnecessary line continuation
856 - vulkan/runtime: add spirv_info_h to vulkan_lite_runtime_header_gen_deps
860 - radeonsi: Fix unused variable when LLVM is not used for AMD.
864 - aco/ra: fix kill flags after renaming fixed Operands
865 - aco/ra: assert that the register file is empty after register allocation completed
866 - aco/lower_phis: simplify check for uniform predecessors
867 - aco: introduce aco_opcode::p_boolean_phi
868 - aco/vn: copy-propagate trivial phis
869 - aco/lower_phis: generalize init_state() so that it works with any scalar phis
870 - aco/lower_phis: implement SGPR phi lowering
871 - aco: use SGPR phi lowering for uniform phis in divergent merge blocks
872 - aco: use SGPR phi lowering for all loop header phis
873 - aco: use SGPR phi lowering for all scalar phis
874 - aco/optimizer: remove p_linear_phi handling from optimizer
875 - radv: mark nir_opt_loop() as not idempotent
876 - radv: move nir_opt_dead_cf() before nir_opt_loop()
877 - panfrost: skip gles-3.0-transform-feedback-uniform-buffer-object on Mali G52 and G57
878 - nir/loop_analyze: adjust negative (or huge) iteration count check for bit size
879 - nir/opt_if: don't split ALU of phi into otherwise empty blocks
880 - nir/opt_loop: add loop peeling optimization
881 - aco/ra: fix handling of killed operands in compact_relocate_vars()
882 - aco/ra: Fix array access when finding register for subdword variables
883 - aco/ra: refactor get_reg_simple() with increased stride.
884 - aco/ra: move can_write_m0() check into get_reg_specified()
885 - aco/ra: re-use registers from killed operands
886 - aco/ra: change heuristic to first fit
887 - aco/ra: use round robin register allocation
888 - aco/assembler: fix MTBUF opcode encoding on GFX11
889 - aco/assembler: slightly refactor MTBUF assembly for more readability
890 - aco/assembler: fix GFX67 MTBUF opcode encoding
891 - aco/scheduler: remove unused register_demand parameter
892 - aco: move live var information into struct Program
893 - aco/reindex_ssa: replace live_var parameter with boolean
894 - aco: make aco::monotonic_buffer_resource declaration visible for aco::IDSet
895 - aco: use aco::monotonic_allocator for IDSet
896 - spirv: make gl_HelperInvocation volatile if demote is being used
897 - radv: emit discard as demote by default
898 - nir: introduce discard_is_demote compiler option
899 - nir/opt_peephole_select: handle nir_terminate{_if}
900 - nir: remove nir_intrinsic_discard
901 - zink: pass zink_screen to nir_to_spirv().
902 - nir/shader_info: remove uses_demote
903 - spirv: workaround for tests assuming that OpKill terminates invocations or loops
904 - aco/scheduler: fix register_demand validation debug code
905 - aco/spill: Unconditionally add 2 SGPRs to live-in demand
906 - aco: calculate register demand per instruction as maximum necessary to execute the instruction
907 - aco: track and use the live-in register demand per basic block
908 - aco: remove get_demand_before()
909 - aco/live_var_analysis: slightly refactor handling of additional register demand for Operand copies
910 - aco/live_var_analysis: ignore dead phis
911 - aco/spill: don't remove spilled phis
912 - aco/ra: use live_in_demand in should_compact_linear_vgprs()
913 - aco: add RegisterDemand member to Instruction
914 - aco/util: skip empty blocks in IDSet::insert(IDSet)
915 - aco/live_var_analysis: refactor using ctx struct
916 - aco/live_var_analysis: ignore phi definition and operand demand at predecessors
917 - aco/live_var_analysis: inline block->register_demand updates
918 - aco/live_var_analysis: remove unused includes
919 - aco/live_var_analysis: use separate allocator for temporary live sets
920 - aco/ra: remove special-casing of p_logical_end
921 - nir: implement loop invariant code motion (LICM) pass
922 - radv: use NIR loop invariant code motion pass
923 - nir/opt_sink: ignore loops without backedge
924 - aco: compute live-in variables in addition to live-out variables
925 - aco/ra: use live-in variables directly rather than computing them
926 - aco/spill: use live-in variables directly rather than computing them
927 - aco/cssa: use live-in variables instead of live-out variables
928 - aco/validate: use live-in variables for RA validation
929 - aco/print_ir: print live-in instead of live-out variables
930 - aco: remove live-out variables from IR
931 - aco/spill: Don't add phi definitions to live-in variables
932 - util/disk_cache: enable Mesa-DB disk cache by default
936 - Revert "ci: disable g52"
937 - gbm: Support fixed-rate compression allocation
938 - venus/ci: Fix timeout
939 - venus/ci: Significantly reduce CTS fraction
940 - venus/ci: Temporarily disable jobs
941 - dri: Fix BGR format exclusion
942 - egl/surfaceless: Enable RGBA configs
943 - egl/gbm: Enable RGBA configs
944 - egl/dri2: Use createImageFromNames for DRM buffers
945 - dri: Remove old createImageWithModifiers
946 - dri: Remove createImageFromFds
947 - dri: Stop answering DRI_IMAGE_ATTRIB_FORMAT
948 - gallium/dri: Delete unused helper function
949 - gallium/dri: Drop mesa_format indirection for lookup
950 - loader/dri3: Use FourCC for create-image entrypoints
951 - egl/x11: Update to createImageFromNames
952 - dri: Delete createImageFromName
953 - dri: Unify createImage and createImageWithModifiers
954 - egl/x11: Remove __DRI_IMAGE_FORMAT remnants
955 - loader/dri3: Use FourCC for buffer allocations
956 - u_format: Rewrite format table to use YAML
957 - format: Generate endian-independent format aliases
958 - format: Generate sRGB<->linear conversions from table
959 - u_format: Reword introduction
960 - build: Check for PyYAML in Meson build
961 - dri: Allow INVALID for modifier-less drivers
962 - gbm/dri: Remove erroneous assert
966 - tu: Handle non-overlapping WaW hazard with buffer copy/fill/update
967 - tu/a7xx: Don't set FLUSH_PER_OVERLAP_AND_OVERWRITE for feedback loops
968 - tu/a750: Disable HW binning when there is GS
969 - freedreno/devices: Add support for Adreno A32 (G3x Gen 2)
970 - util/u_trace: Allow mixing of ArgStruct and Arg
971 - tu: Add more info to renderpass tracepoint
972 - vulkan/wsi: Make current_frame usable in all cases
973 - util/u_trace: Pass explicit frame_nr argument to delimit frames
974 - tu: Use current_frame from vk device to delimit u_trace frames
975 - anv: Use current_frame from vk device to delimit u_trace frames
976 - freedreno: Make fd_pps_driver.h usable without including other FD sources
977 - turnip/msm: Do rd dump only when there are commands in submission
978 - turnip: Implement VK_EXT_depth_clamp_zero_one
979 - freedreno/a7xx: Update TPL1_DBG_ECO_CNTL1 to fix UBWC corruption
980 - ir3/a7xx: Fix FS consts corruption when other FS has zero constlen
981 - tu: Add LRZ disable reason to renderpass tracepoint
982 - util/u_trace: Add support for fixed-length string params in tracepoints
983 - tu: Add attachments' UBWC info to renderpass tracepoint
984 - freedreno/rddecompiler: Make possible to use original shader
985 - freedreno/replay: Fix replaying without SET_IOVA
986 - freedreno/ir3: mova has special meaning for (r) flag
987 - ir3: Correctly assemble mova1 with (r) on const
988 - tu: Fix issues with render_pass tracepoint
989 - freedreno: Rename TPL1_DBG_ECO_CNTL1.UBWC_WORKAROUND into TP_UBWC_FLAG_HINT
990 - tu: Add enable_tp_ubwc_flag_hint feature to a7xx
991 - freedreno/devices: Turn off enable_tp_ubwc_flag_hint for a740 by default
992 - freedreno/devices: Fix magic regs for Adreno A32
993 - freedreno: Describe LRZ feedback mechanism
994 - freedreno/devices: Define and appropriately set has_lrz_feedback
995 - tu: Use LRZ feedback in gmem
996 - tu: Enable LRZ feedback in sysmem
997 - freedreno: Use LRZ feedback in gmem
998 - ir3: Print bindless samp/tex ids for tex prefetch
999 - ir3/tests: Make possible to specify raw instr value as uint64
1000 - ir3/tests: Make possible to add generated disasm tests
1001 - ir3: Fix decoding of stib.b/ldib.b with offset
1002 - turnip/kgsl: Support external memory via ION/DMABUF buffers
1003 - tu: Have single Flush/Invalidate memory entrypoints
1007 - radv/video/encode: fix quality params on v2 hw.
1008 - Revert "zink: use a slab allocator for zink_kopper_present_info"
1009 - nvk: Only enable WSI modifiers if the extension is supported.
1010 - draw/texture: handle mip_offset[0] being != 0 for layered textures.
1011 - nouveau/nvc0: increase overallocation on shader bo to 2K
1012 - nvidia: fixup classes import and import new classes.
1013 - nouveau/push: add support for m2mf/i2mf to dumper
1014 - nouveau/nvc0: add support for using common pushbuf dumper
1015 - radv/video: fix layered decode h264/5 tests.
1016 - radv/video: use vcn ip versions for encoder detection.
1017 - ac/radv/radeonsi: move av1 ctx/probs size/filling to common code.
1018 - ac/radv/radeon: move film grain init to common code.
1019 - st/mesa: drop u_simple_shaders.h include where not used.
1020 - gallivm: create a pass manager wrapper.
1021 - gallivm: move ppc denorm disable to inline
1022 - gallivm: split some code out from init module.
1023 - gallivm: make lp_bld_coro.h c++ include safe.
1024 - gallivm: export target init code for orc-jit to reuse
1025 - gallivm: split out generating LLVM Mattrs
1026 - llvmpipe: Introduce llvmpipe_memory_allocation
1027 - nvk: use 2k overallocation for shader heap.
1028 - anv/video: use correct offset for MPR row store scratch buffer.
1029 - radv/video: advertise mutable/extended for dst video images.
1030 - draw/orcjit: supply stub function for tcs coro
1031 - llvmpipe/cs/orcjit: add stub function name for coro
1032 - gallivm/sample: fix sampling indirect from vertex shaders
1033 - nvc0: fix null ptr deref on fermi due to debug changes.
1037 - radeonsi/vcn: set accurate size for dec header and index_codec
1038 - radeonsi/vcn: support DPB_MAX_RES on VCN5
1042 - turnip: rename tu_queue_submit struct to follow ODR
1043 - ci: fail pipeline for users who got access to restricted traces
1044 - ci/traces: majanes has no longer access to the restricted traces
1045 - ci/deqp: correct EGL_EXT_config_select_group detection
1046 - egl/x11: Move RGBA visuals in the second config selection group
1047 - mailmap: add Freya Gentz entry
1048 - etnaviv: migrate from piglit include to generic deqp and toml spec
1049 - freedreno/ci: move platform to the deqp toml file for a530
1050 - freedreno/ci: move the disabled jobs from include to the main file
1051 - freedreno/ci: Switch a306_* to deqp-runner
1052 - freedreno/ci: do not depend on single job rules for another jobs
1053 - freedreno/ci: switch a306 to weston
1054 - freedreno/ci: re-enable a306_piglit
1055 - ci/panfrost: disable G52 until machines gets fixed
1056 - ci: drop unused piglit-test and integrate it into piglit-traces-test
1057 - freedreno/ci: Drop duplicated include and add missing stages
1058 - freedreno/ci: Implement nightly piglit job for Adreno 630 and 618
1059 - ci/freedreno: update expectations from the nightly run
1060 - ci: bump ANGLE
1061 - ci: Revert "ci: update failures list with angle for jsl, tgl"
1062 - ci/intel: add new jsl flake
1063 - ci/panfrost: Revert "ci/panfrost: disable G52 until machines gets fixed"
1064 - ci/alpine: re-enable Mold linker
1065 - ci/etnaviv: add flakes from nightly runs
1066 - winsys/i915: depends on intel_wa.h
1067 - subprojects: uprev perfetto to v45.0
1068 - ci/r300: update flake list from nightly reports
1069 - ci/nouveau: move disabled jobs back from include into main gitlab-ci.yml
1070 - ci/nouveau: separate HW definition from SW
1071 - ci/nouveau: adjust and add DEVICE_TYPE
1072 - ci/freedreno: a3xx will never have Vulkan support
1073 - docs: correct svga3d redirected URLs
1074 - ci/radv: dEQP-GLES3.functional.polygon_offset.fixed16_render_with_units passes now
1075 - ci: re-enable shader-db for nouveau
1076 - ci: do not build Nine in debian-build-testing
1077 - ci/piglit: be explicit about what we building
1078 - ci/lava: enable Piglit OpenCL tests so we can test rusticl on the HW
1079 - ci/lava: do not build Vulkan for armhf images
1080 - ci/lava: move wayland-protocols to the main section
1081 - ci/freedreno: document new failure after piglit update
1082 - ci/etnaviv: skip Vulkan tests on GC2000
1083 - ci/etnaviv: remove duplicated line from skips
1084 - mailmap: update my email
1085 - ci/arm64: rustify the build
1086 - ci/lava: add support for RustiCL
1087 - ci/meson: reuse meson installation
1088 - ci: move (c)bindgen to own shell script
1089 - ci/radv: Document recent flake
1090 - ci/lava: the containers take sometimes more than 60m
1091 - ci: propagate RUSTICL_ENABLE and DEBUG variables to the DUTs
1092 - rusticl: add -cl-std only when it's not defined
1093 - ci/freedreno: some A306 tests now pass/skip since proper GL detection in Piglit
1094 - ci: introduce tool for comparing nightly runs
1095 - util: bump blake3 from 1.3.3 to 1.5.1, improve armv7 and aarch64 performance
1096 - build: pass licensing information in SPDX form
1097 - intel/debug: allow silencing CL warnings
1098 - llvmpipe: Silence "possibly uninitialized value" warning for ssbo_limit (cont)
1099 - ci/alpine: use llvm variables
1103 - radv/video: Set correct bit depth and format for 10bit input
1104 - radv/video: Check encode profiles and bit depth in capabilities query
1105 - radv/video: Report maxBitrate in encode capabilities
1106 - radeonsi/vcn: Allocate session buffer in VRAM
1107 - radeonsi/vcn: Fix 10bit HEVC VPS general_profile_compatibility_flags
1108 - radeonsi/vcn: Only enable VBAQ with rate control mode
1109 - frontends/va: Fix AV1 slice_data_offset with multiple slice data buffers
1110 - Revert "radeonsi/vcn: AV1 skip the redundant bs resize"
1111 - frontends/va: Only increment slice offset after first slice parameters
1112 - radeonsi: Update buffer for other planes in si_alloc_resource
1113 - frontends/va: Store slice types for H264 decode
1114 - radeonsi/vcn: Ensure DPB has as many buffers as references
1115 - radeonsi/vcn: Allow duplicate buffers in DPB
1116 - radeonsi/vcn: Ensure at least one reference for H264 P/B frames
1117 - frontends/va: Fix leak when destroying VAEncCodedBufferType
1118 - radeonsi/vcn: Avoid copy when resizing bitstream buffer
1119 - frontends/va: Send all bitstream buffers to driver at once
1120 - frontends/va: Fix crash in vaRenderPicture when decoder is NULL
1121 - radv/video: Add missing VCN 3.0.2 to decoder init switch
1122 - radeonsi: Make si_compute_clear_image work with 422 subsampled formats
1123 - gallium/vl: Init shaders on first use
1124 - frontends/va: Don't require exact match for packed headers
1125 - gallium: Add is_video_target_buffer_supported
1126 - radeonsi: Implement is_video_target_buffer_supported
1127 - frontends/va: Use is_video_target_buffer_supported for EFC
1128 - frontends/va: Rework EFC logic
1129 - frontends/va: Check if target buffer is supported in vlVaEndPicture
1130 - gallium: Remove PIPE_VIDEO_CAP_EFC_SUPPORTED
1131 - frontends/va: Simplify AV1 slice parameters handling
1132 - frontends/va: Move slice_data_offset to context
1133 - frontends/va: Rename slice_idx to have_slice_params and move to context
1134 - frontends/va: Support multi elements slice parameter buffers for H264/5
1135 - gallium: Remove pipe_h264_picture_desc.slice_parameter.slice_count
1136 - radeonsi/vcn: Limit size to target size in AV1 decode
1137 - radeonsi: Add debug option to enable low latency encode
1138 - radeonsi/vcn: Add low latency encode support
1139 - frontends/va: Support frame rate per temporal layer for AV1
1140 - radeonsi/vcn: Support 10bit RGB for EFC input
1141 - radeonsi/vcn: Add decode DPB buffers as CS dependency
1145 - ci/lava: Detect a6xx gpu recovery failures
1149 - wsi/wayland: refactor wayland dispatch
1150 - egl/wayland: Use loader_wayland_dispatch
1151 - perfetto: Add flows
1152 - wsi/wayland: Add perfetto flows to image acquisition and presentation
1153 - wsi/wayland: Add flow id to presentation feedback
1154 - wsi/wayland: Add timing debugging
1155 - perfetto: Add simple support for counters
1156 - wsi/wayland: Add latency information to perfetto profiling
1157 - perfetto: Add some functions for timestamped events
1158 - wsi/wayland: Add a perfetto track for image presentation
1159 - wsi/wayland: Add tracepoint in wsi_wl_swapchain_wait_for_present
1160 - wsi/wayland: Fix use after free from improperly stored VkAllocationCallbacks
1161 - wsi/wayland: Use different queue names for different queries
1165 - freedreno/registers: drop display-related register files
1169 - venus: make cross-device optional
1173 - xa: add missing stride setup in renderer_draw_yuv
1177 - treewide: Cleanup unused structs
1181 - meson: use glslang --depfile argument when possible
1182 - clc: remove check for null pointer that cannot be true in llvm_mod_to_spirv
1183 - compiler/glcpp: don't recalculate macro
1184 - intel/compiler: move predicated_break out of backend loop
1185 - anv/grl: add some validation that we're not going to overflow
1186 - egl/wayland: fix memory leak in error handling case
1187 - compilers/clc: Add missing break statements.
1188 - mesa: fix memory leak when using shader cache
1189 - util/glsl2spirv: fixup the generated depfile when copying sources
1190 - tgsi_to_nir: free disk cache value if the size is wrong
1191 - crocus: properly free resources on BO allocation failure
1192 - crocus: check for depth+stencil before creating resource
1196 - nvk: Add sha1_h as a dependency
1197 - d3d10umd: Use pipe_resource_usage enum in translate_resource_usage()
1198 - util: Fix the integer addition in os_time_get_absolute_timeout()
1202 - meson: create libglsl declared dependency to propagate order-only deps
1203 - meson: add various generated header dependencies as order-only deps
1207 - nir,panfrost,agx: Fix driver PIXEL_COORD_INTEGER setting and drop workaround.
1208 - dri: Fix a pasteo in dri2_from_names()
1209 - dri: Consistently use createImageWithModifiers2()
1210 - dri: Consistently use createImageFromFds2(), not createImageFromFds()
1211 - dri: Replace createImageFromDmaBufs() with createImageFromDmaBufs3()
1212 - dri: Drop old createImageFromRenderbuffer()
1213 - dri: Consistently use createImageFromDmabufs() not createImageFromFds()
1214 - dri: Drop createImageFromFds2() in favor of createImageFromDmaBufs()
1215 - dri: Move EGL image lookup/validate setup to dri_init_screen()
1216 - mesa: Drop some version checking around ValidateEGLImage
1217 - dri: Collapse dri2_validate_egl_image() into dri_validate_egl_image()
1218 - dri: Fold lookup_egl_image_validated into its one caller
1219 - dri: Drop the old lookupEGLImage wrapper function.
1223 - VERSION: bump to 20.2
1224 - docs: reset new_features.txt
1225 - docs: add release notes for 24.0.6
1226 - docs: update calendar for 24.0.6
1227 - docs: add an extra 24.0.x release
1228 - docs: add sha256sum for 24.0.6
1229 - docs: update calendar for 24.1.0-rc1
1230 - ci: fix container rules on release branches and tags
1231 - panvk/ci: add WSI testing to all the deqp-vk jobs
1232 - lavapipe/ci: add WSI testing to all the deqp-vk jobs
1233 - freedreno/ci: add flake
1234 - lavapipe/ci: add flakes
1235 - ci: pass MESA_VK_ABORT_ON_DEVICE_LOSS through to the DUT
1236 - rpi3/ci: drop duplicate comment without any corresponding actual skip line
1237 - v3dv/ci: skip all the WSI tests, they are way too flaky to be worth it
1238 - spirv: deduplicate default debug log level
1239 - v3dv/ci: add rpi5 failure
1240 - ci: mark microsoft farm as offline
1241 - meson: simplify \`-gsplit-dwarf` compiler argument check
1242 - egl+glx: fix two #ifdef that should be #if like the rest
1243 - meson: always set USE_LIBGLVND
1244 - meson: use bool.to_int() instead of manually converting
1245 - lavapipe/ci: drop fixed test from failures
1246 - lavapipe/ci: add the rest of the failures introduced by the 1.3.8.2 uprev
1247 - lavapipe/ci: skip another test that goes over the timeout
1248 - meson: move tsan-blacklist.txt to build-support with the other build support files
1249 - llvmpipe/ci: fix indentation
1250 - llvmpipe/ci: only run jobs when their corresponding files are changed
1251 - lavapipe/ci: fix indentation
1252 - lavapipe/ci: avoid running all lavapipe jobs when llvmpipe ci is changed
1253 - lavapipe/ci: only run jobs when their corresponding files are changed
1254 - docs: update calendar for 24.1.0-rc2
1255 - llvmpipe/ci: trigger jobs on draw & gallivm changes
1256 - lavapipe/ci: trigger jobs on draw & gallivm changes
1257 - lavapipe/ci: add flakes seen lately
1258 - lavapipe/ci: generalize flakes list to all formats for these flaky tests
1259 - lavapipe/ci: skip ray tracing tests that sometimes time out
1260 - vc4/ci: add fails seen overnight
1261 - ci: uprev mold to 2.31.0
1262 - lavapipe/ci: skip two more timing out ray query tests
1263 - ci: backport fix for gl_PointSize bug in CTS
1264 - lavapipe/ci: move a few skips out from under the "llvm jit" comment
1265 - mr-label-maker: fix yaml syntax
1266 - docs: add release notes for 24.0.7
1267 - docs: update calendar for 24.0.7
1268 - docs: add sha256sum for 24.0.7
1269 - docs: update calendar for 24.1.0-rc3
1270 - ci/debian-build-testing: drop extra nesting section
1271 - ci/shader-db: drop extra nesting section
1272 - rpi4/ci: use deqp-runner suite for vk job as well
1273 - rpi5/ci: use deqp-runner suite for vk job
1274 - microsoft/clc: fix incorrect changes that got through while the Windows CI was down
1275 - llvmpipe: wrap the push/pull in the ifdef as well
1276 - radv/ci: add navi21 flakes
1277 - zink: avoid designated initializers as they are not supported in C++ < 20
1278 - Revert "ci: fail pipeline for users who got access to restricted traces"
1279 - radeonsi/ci: document new crash (assert)
1280 - util/format: add missing null check in util_format_is_srgb()
1281 - ci: drop default VKD3D_PROTON_RESULTS file name
1282 - ci: hardcode \`-vkd3d` namespace for VKD3D_PROTON_RESULTS
1283 - amd/ci: track changes to VKD3D_PROTON_RESULTS files
1284 - mr-label-maker: mark \*-vkd3d.txt files as CI results expectations files
1285 - ci: reuse dead .vkd3d-proton-test to make vkd3d less radv-specific
1286 - ci: fix section_end in debian-build-testing
1287 - ci: rename debian version variable job to include the word "version"
1288 - ci: factor out all the deps to build the debian containers into .debian-container
1289 - ci: inherit the debian container building infra for test container images
1290 - ci/b2c: rename B2C_TIMEOUT_FIRST_* to B2C_TIMEOUT_FIRST_CONSOLE_ACTIVITY_*
1291 - ci/b2c: rename B2C_TIMEOUT_* to B2C_TIMEOUT_CONSOLE_ACTIVITY_*
1292 - ci/b2c: allow setting timeouts in seconds
1293 - ci: drop dead VK_CPU option
1294 - ci/piglit-traces: drop re-definition of VK_DRIVER_FILES
1295 - ci/init-stage2: set VK_DRIVER_FILES for both xorg and wayland
1296 - ci/vkd3d: un-hardcode architecture
1297 - ci/vkd3d: fix version sanity check
1298 - ci/vkd3d: fail job when failing to get driver version
1299 - ci/b2c: remove dead rules: that's always overwritten
1300 - ci/env: move dead-code-with-comment to the end of the list to make it clearer
1301 - zink/ci: rename .zink-lvp-venus-rules to .zink-venus-lvp-rules to match the rest of the names
1302 - README: update links to our own docs
1303 - docs: update calendar for 24.1.0-rc4
1304 - mailmap: add entry to unify Roman Stratiienko's contributions
1305 - nvk/ci: add nvk job on a GA106 (RTX 3060)
1306 - zink/ci: add zink+nvk glcts+piglit job on a GA106 (RTX 3060)
1307 - zink+nvk/ci: skip glx piglit tests as they all fail
1308 - zink+nvk/ci: skip timing out test
1309 - zink+nvk/ci: skip more tests that times out
1310 - zink+nvk/ci: document flakes seen during stress-testing
1311 - zink+nvk/ci: update expected failures
1312 - docs: add release notes for 24.0.8
1313 - docs: update calendar for 24.0.8
1314 - docs: add sha256sum for 24.0.8
1315 - docs: add release notes for 24.1.0
1316 - docs: add sha256sum for 24.1.0
1317 - docs: update calendar for 24.1.0
1318 - ci: fix build-kernel.sh -> download-prebuilt-kernel.sh
1319 - ci: drop dead variables (see previous commit)
1320 - ci: rename debian/arm*_test to debian/baremetal_arm*_test to be clear about which infra uses that
1321 - ci: prepare base debian test image for multi-arch
1322 - ci: prepare GL debian test image for multi-arch
1323 - ci: prepare VK debian test image for multi-arch
1324 - ci/image-tags: rename DEBIAN_X86_64_TEST_*_TAG to drop the x86 mention
1325 - ci: add debian/arm64_test images for gl & vk
1326 - zink/ci: rename zink-turnip collabora rule to make it unambiguous
1327 - ci/b2c: add aarch64 tests for gl & vk
1328 - turnip/ci: add vkcts jobs on the a750
1329 - turnip+zink/ci: add gl & gles CTS jobs on the a750
1330 - nvk/ci: adjust the regex for "dut is broken and needs to be rebooted"
1331 - nvk/ci: mark the job as failing in case of hangs, instead of silently rebooting
1332 - nvk/ci: add missing .test rules to avoid running nvk tests in post-merge pipeline
1333 - radv/ci: move amdgpu-specific kernel message warning to src/amd/ci/
1334 - ci/b2c: make B2C_JOB_WARN_REGEX optional
1335 - zink+nvk/ci: more KHR-GL46.packed_pixels.varied_rectangle.* flakes, so mark the group as flaky
1336 - zink+nvk/ci: add more flakes seen in nightly
1337 - zink+nvk/ci: spec\@ext_external_objects\@vk-vert-buf-reuse has been fixed
1338 - mr-label-maker: label src/vulkan/wsi/ as wsi
1339 - .mailmap: fix email address for \@cpmichael
1340 - v3dv/ci: fix typo in \`renderer_check`
1341 - ci: disable debian-build-testing until it can be fixed
1342 - vc4/ci: skip VK piglit tests
1343 - freedreno/a6xx: fix kernel -> compute handling
1344 - zink+nvk/ci: add flakes seen in latest nightly run
1345 - docs/calendar: add 24.2 branchpoint and release candidates schedule
1346 - panfrost/ci: add missing genxml trigger path
1347 - panfrost: mark tests as fixed
1348 - etnaviv/ci: skip VK piglit tests
1349 - radv/ci: document angle regressions from !29436 on stoney
1350 - zink+nvk/ci: add flakes seen in latest nightly run
1351 - docs/meson: replace deprecated pkgconfig with pkg-config
1352 - zink+nvk/ci: add flakes seen in latest nightly run
1353 - v3dv: add missing bounds check in VK_EXT_4444_formats
1354 - docs: add release notes for 24.1.1
1355 - docs: add sha256sum for 24.1.1
1356 - docs: update calendar for 24.1.1
1357 - turnip/ci: add a750 flakes seen in the latest nightly
1358 - radv/ci: fix manual rules
1359 - radv/ci: move radv manual rules into their own group
1360 - nvk+zink/ci: add another flake seen in nightly
1361 - docs: add release notes for 24.0.9
1362 - docs: update calendar for 24.0.9
1363 - docs: add sha256sum for 24.0.9
1364 - venus/ci: add flake that's been blocking MRs
1365 - v3d/drm-shim: emulate a rpi4 instead of a rpi3
1366 - nvk+zink/ci: add another flake seen in nightly
1367 - radv/ci: document navi31 regression from !29235
1368 - ci: set a common B2C_JOB_SUCCESS_REGEX with the message that's printed for all jobs
1369 - ci/deqp: uprev gl & gles cts
1370 - radeonsi/ci: mark a bunch of tests as fixed on vangogh
1371 - radv/ci: drop duplicate navi21-aco flakes line
1372 - radv/ci: drop duplicate navi31-aco flakes line
1373 - turnip+zink/ci: mark a dEQP-GLES(2|3).functional.rasterization.(fbo|primitives).line_(strip_|)wid…
1374 - turnip/ci: add a750 flakes seen in the latest nightly
1375 - panfrost/ci: remove duplicate path
1376 - nvk+zink/ci: mark KHR-GL46.sparse_texture2_tests.SparseTexture2* as fixed
1377 - nvk+zink/ci: add flakes seen in nightly pipeline
1378 - nvk+zink/ci: consider all the \`double` tests in spec\@glsl-4.00\@execution\@built-in-functions t…
1379 - freedreno/ci: disable mid-testing reboot on a750
1380 - driconf: drop param for setting default gpu vendor id in DRI_CONF_FORCE_VK_VENDOR()
1381 - egl: fix teardown when using xcb
1382 - egl: move android-specific code into an android branch
1383 - egl: ensure future platforms get their teardown implemented
1384 - egl/device: drop unnecessary intermediate variable
1385 - ci: fix meson install script
1386 - lavapipe/ci: update trace checksum following nir change
1387 - lavapipe/ci: document regression while it's being worked on
1388 - turnip+zink/ci: mark dEQP-GLES3.functional.fbo.depth.depth_test_clamp.* tests as fixed
1389 - bin/ci: escape literal url in regex
1390 - glx: fix build -D glx-direct=false
1391 - nvk+zink/ci: mark spec\@ext_image_dma_buf_import\@ext_image_dma_buf_import-refcount-multithread a…
1392 - nvk+zink/ci: add flakes seen over the last few nightlies
1393 - asahi/lib: generate git_sha1.h for agx_device.c
1394 - ci/vkd3d: deduplicate the diff between the expectation and the results
1395 - ci/vkd3d: print a message when the expected failures file is missing
1396 - ci/vkd3d: drop override of job artifacts
1397 - ci/vkd3d: fix error message printing
1398 - ci/vkd3d: stop ignoring errors in a block where errors can't happen
1399 - ci/vkd3d: don't ignore errors
1400 - ci/vkd3d: group version check lines together
1401 - ci/vkd3d: limit the vulkaninfo capture to the driverInfo line
1402 - ci/vkd3d: print a real error message when failing to get the list of failing tests
1403 - ci/vkd3d: rename vkd3d test log file to end in .txt
1404 - ci/vkd3d: print URL to the vkd3d-proton.log file to make it easier to access
1405 - ci/vkd3d: put \`then` on the same line as the \`if` to match the rest of the code style
1406 - ci/vkd3d: drop the "clear results folder without deleting the folder" logic
1407 - ci/vkd3d: drop \`quiet` wrapper
1408 - ci/vkd3d: drop redundant "vkd3d-proton execution: SUCCESS"
1409 - docs: add release notes for 24.1.2
1410 - docs: add sha256sum for 24.1.2
1411 - docs: update calendar for 24.1.2
1412 - venus/ci: fix indentation of list nested in a dict item
1413 - venus/ci: add manual/nightly venus-lavapipe-full
1414 - venus/ci: skip timed out test
1415 - nvk+zink/ci: add flakes seen over the last two nightly runs
1416 - nvk+zink/ci: catch more \`double` flakes
1417 - venus+zink/ci: drop fraction and add missing timeout on zink-venus-lvp
1418 - loader: use os_get_option() to allow android to set LIBGL_DRIVERS_PATH, GBM_BACKENDS_PATH, GALLIU…
1419 - gallium/hud: use os_get_option() to allow android to set GALLIUM_HUD and related vars
1420 - egl: use os_get_option() to allow android to set EGL_LOG_LEVEL
1421 - venus/ci: make sure nightly job doesn't get retried
1422 - venus/ci: drop fixed test from fails list
1423 - docs/ci: fix indentation of list nested in a dict item
1424 - docs/ci: merge test-docs and test-docs-mr
1425 - docs/ci: auto-run test-docs in fork pipelines
1426 - docs/ci: drop .no_scheduled_pipelines-rules from test-docs
1427 - ci: reorder alpine/x86_64_build rules to fix the nightly pipelines
1428 - drm-shim: stub syncobj_timeline_signal ioctl
1429 - llvmpipe/ci: add comment for later on weird-looking code
1430 - llvmpipe/ci: fix indentation of list nested in a dict item
1431 - llvmpipe/ci: set rusticl variables in deqp-runner instead of passing them down from the job
1432 - ci: include rusticl in the arm64 build
1433 - llvmpipe,rusticl/ci: move rusticl files rule out of llvmpipe
1434 - v3d/ci: add nightly job for rusticl testing
1435 - panfrost/ci: drop duplicate job rules
1436 - panfrost/ci: split gl & vk jobs rules
1437 - radeonsi/ci: mark test as fixed
1438 - lavapipe/ci: skip timing out test
1439 - broadcom/ci: disable auto-retry on manual jobs
1440 - docs/features: mark VK_KHR_maintenance7 as implemented on anv and lvp
1441 - docs: add release notes for 24.1.3
1442 - docs: update calendar for 24.1.3
1443 - docs: add sha256sum for 24.1.3
1444 - ci_run_n_monitor: add support for new \`canceling` job status
1445 - ci_run_n_monitor: be coherent about using sets for \`element in group` checks
1446 - ci_run_n_monitor: use COMPLETED_STATUSES in more places
1447 - ci_run_n_monitor: add RUNNING_STATUSES and use it where appropriate
1448 - bin/ci: allow bugfixes in requirements.txt
1449 - ci: split .no-auto-retry out of .scheduled_pipeline-rules
1450 - ci: simplify setting .no-auto-retry now that it isn't bundled with unrelated rules:
1451 - v3d/ci: include results of GL full run in expectations
1452 - v3d/ci: include results of CL run in expectations
1453 - zink+nvk/ci: ascii-sort fails
1454 - zink+nvk/ci: document regression from !30033
1455 - turnip+zink/ci: add two more CS related flakes
1456 - lvp+zink/ci: document a flake seen in a merge pipeline
1457 - v3d/ci: add disabled job for GL testing on the RPi5
1458 - v3d/ci: rename "rusticl on v3d" suite to \`v3d-rusticl`
1459 - v3d/ci: add disabled job for CL testing on the RPi5
1460 - features.txt: specify that VK_EXT_depth_clamp_zero_one is only supported on v3dv/vc7+
1461 - features.txt: specify that VK_EXT_depth_clip_enable is only supported on v3dv/vc7+
1462 - features.txt: specify that GL_ARB_depth_clamp is only supported on v3d/vc7+
1463 - docs: add release notes for 24.1.4
1464 - docs: update calendar for 24.1.4
1465 - docs: add sha256sum for 24.1.4
1466 - ci: replace gallium-drivers=swrast with gallium-drivers=llvmpipe,softpipe
1467 - bin/ci_run_n_monitor: explain that the 'Universal Recycling symbol' ♲ emoji means these jobs were…
1468 - bin/ci_run_n_monitor: add text labels next to the emojis
1469 - bin/ci_run_n_monitor: replace ♲ with �� to represent cancelled jobs
1470 - meson: fix filename printed when generating devenv files
1471 - meson/megadriver: fix install message to match the rest of meson
1472 - meson/megadriver: stop removing the "master" .so file
1473 - meson/megadriver: replace hardlinks with symlinks
1474 - ci/vkd3d: fix LD_LIBRARY_PATH
1475 - v3d/ci: mark spec\@amd_performance_monitor\@vc4 tests as fixed
1476 - llvmpipe/ci: mark spec\@!opengl 1.1\@gl_select tests as fixed
1477 - Revert "bin/ci_run_n_monitor: explain that the 'Universal Recycling symbol' ♲ emoji means these j…
1478 - VERSION: bump for 24.2.0-rc1
1479 - .pick_status.json: Update to 0cc23b652401600e57c278d8f6fe6756b13b9f6a
1480 - radeonsi/ci: skip timing out test
1481 - freedreno/ci: double job timeout for a306
1482 - freedreno/ci: document extra variants of failing tests on a618 and a630
1483 - anv+zink/ci: mark some tests as fixed
1484 - anv+zink/ci: document two tests, one failing and one crashing
1485 - anv+zink/ci: mark a couple of tests as flaky
1486 - venus/ci: skip timing out test
1487 - loader: gc loader_get_extensions_name() and __DRI_DRIVER_{GET\_,}EXTENSIONS defines
1488 - .pick_status.json: Update to 3b6867f53a6718de80bbff4acb84ffd5aca8a8c8
1489 - nak: fix meson typo
1490 - venus: initialize bitset in CreateDescriptorPool()
1491 - v3d/ci: mark spec\@amd_performance_monitor\@vc4 tests as flaky
1492 - meson: xcb & xcb-randr are needed by the loader whenever x11 is built
1493 - .pick_status.json: Update to c30e5d44b1027ed03a8fd542829df0055d3e1a96
1494 - .pick_status.json: Update to 6cd4372460b197fea98d257217328ddc3406e6ad
1495 - docs: add stub header for u_format_gen.h
1496 - .pick_status.json: Update to c33d2db06ac0ea4d3d5372caa93bee3bbbe028c7
1497 - VERSION: bump for 24.2.0-rc2
1498 - .pick_status.json: Update to ad90bf0500e07b1bc35f87a406f284c0a7fa7049
1499 - ci/baremetal: fix logic for retrying boot when it failed
1500 - meson: don't select the deprecated \`swrast` option ourselves
1501 - meson: improve wording of "incompatible llvm options" error
1502 - ci: remove llvmpipe in the job that disables llvm
1503 - .pick_status.json: Update to aa9745427b917bb0613b753ccd59c6c1e6f07584
1504 - VERSION: bump for 24.2.0-rc3
1505 - .pick_status.json: Update to 366e7e2ddc7d3b340bbf040eca1d3223219e6122
1506 - meson,ci: remove dead \`kmsro` option in \`gallium-drivers`
1507 - .pick_status.json: Mark 93f9afa1e039cbf681adcc6d170aec987d9f0f65 as denominated
1508 - .pick_status.json: Mark f427c9fe233e862bfa30d0c7441ce77592ce4654 as denominated
1509 - .pick_status.json: Update to d58f7a24d1be7b8b50ebdc0c1c3ce26bd65317a5
1510 - .pick_status.json: Update to d9849ac46623797a9f56fb9d46dc52460ac477de
1511 - .pick_status.json: Update to ef88af846761ca9e642f7ed46011db7d3d6b61fd
1512 - VERSION: bump for 24.2.0-rc4
1513 - .pick_status.json: Update to c90e2bccf756004e48f9e7e71e555db0d03c1b98
1514 - ci: pass MESA_SPIRV_LOG_LEVEL from job to the test
1515 - android: fix build in multiple ways
1516 - .pick_status.json: Update to 214b6c30406f844560bdf35a54ff8a51ee248709
1517 - .pick_status.json: Update to cc2dbb8ea5329b509d79eedb6c0cbb9a1903b5ad
1521 - panfrost: add a barrier when launching xfb jobs in CSF
1522 - get_color_read_type: make sure format/type combo is legal for gles
1523 - glsl: test both inputs when sorting varyings for xfb
1524 - glsl: make the xfb varying sort stable
1525 - panfrost: fix some omissions in valhall flow control
1526 - panfrost: change default rounding mode for samplers
1527 - panfrost: fix texture.border_clamp regression for valhall
1528 - panfrost: use RGB1 component ordering for R5G6B5 pixel formats
1532 - ci: lima farm maintenance
1533 - lima/ci: update piglit ci expectations
1534 - Revert "ci: lima farm maintenance"
1535 - lima: fix surface reload flags assignment
1536 - mesa/st: don't set lower_fdot in draw_nir_options
1537 - dri: fix sun4i-drm driver name
1539 Erik Faye-Lund (106):
1541 - panfrost: add PAN_MAX_TEXEL_BUFFER_ELEMENTS define
1542 - panfrost: clamp buffer-size to max-size
1543 - panfrost: remove nonsensical assert
1544 - panfrost: do not deref potentially null pointer
1545 - panfrost: check return-value from u_trim_pipe_prim
1546 - panfrost: assert that drmSyncobjWait returns 0
1547 - panfrost: check return-code of drmSyncobjWait
1548 - panfrost: correct first-tracking for signature
1549 - panvk: drop needless null-check
1550 - panvk: do not leak bindings
1551 - panvk: drop needless null-checks
1552 - panvk: avoid dereferencing a null-pointer
1553 - docs/panfrost: compact gpu-table
1554 - docs/panfrost: move details to separate articles
1555 - docs/panfrost: link to conformant products
1556 - panfrost: simplify panfrost_texture_num_elements
1557 - panfrost: explicitly loop over surfaces
1558 - panfrost: untangle faces from layers
1559 - util/format: correct a typo
1560 - mesa/main: rewrite mipmap generation code
1561 - mesa/main: remove unused function
1562 - mesa/main: rework GL_IMAGE_PIXEL_TYPE query
1563 - mesa/main: clean up _mesa_uncompressed_format_to_type_and_comps
1564 - mesa/main: clean up switch statement
1565 - mesa/main: do not return _REV format for uncompressed format
1566 - mesa/main: prefer non-suffixed enums
1567 - mesa/main: fixup indent
1568 - mesa/main: updates for EXT_texture_format_BGRA8888
1569 - docs: wrap long words instead of overflowing
1570 - meson: bump test-timeout
1571 - mesa/main: remove unused function
1572 - panfrost: lower maxVertexInputStride to match vulkan runtime
1573 - mesa/main: remove stale prototype
1574 - mesa/main: remove duplicate error-checks
1575 - mesa/main: require EXT_texture_integer for GL 3.0
1576 - mesa/main: do not allow RGBA_INTEGER et al in gles3
1577 - mesa/main: factor out format/type enum checking
1578 - mesa/main: use extension-helper
1579 - mesa/main: tighten rg/half-float interaction
1580 - mesa/main: use _mesa_is_gles1()-helper
1581 - mesa/main: remove needless check
1582 - mesa/main: simplify conditions
1583 - mesa/main: merge identical checks
1584 - panvk: move macro-definition to header
1585 - mailmap: invert tomeu's mapping
1586 - mailmap: merge Robert and Bob Beckett into one
1587 - mailmap: invert my mailmapping
1588 - mailmap: map collabora.co.uk to collabora.com
1589 - mailmap: move konstantin to the right sorted position
1590 - mailmap: use consistent spelling for constantine
1591 - mailmap: update rohan's primary email address
1592 - nir: fix utf-8 encoding-issue
1593 - Revert "docs: use html_static_path for static files"
1594 - docs: edgeflag -> edge flag
1595 - docs: zink -> Zink
1596 - docs: Anv -> ANV
1597 - docs: tgsi -> TGSI
1598 - docs: hw -> HW
1599 - docs: mooth -> smooth
1600 - docs: unify spelling of front/back-facing
1601 - docs: eg. -> e.g.
1602 - docs: url -> URL
1603 - docs: nabled -> enabled
1604 - docs: sommelier -> Sommelier
1605 - docs: remove apostrophe from uppercased
1606 - docs: utrace -> trace
1607 - docs: google -> Google
1608 - docs: Nvidia -> NVIDIA
1609 - docs: ssbo/ubo -> SSBO/UBO
1610 - docs: cpu -> CPU
1611 - docs: gpu -> GPU
1612 - docs: renderpass -> render pass
1613 - docs: spell out "stencil reference"
1614 - docs: submision -> submission
1615 - docs: Steamos -> SteamOS
1616 - docs: colour -> color
1617 - docs: occured -> occurred
1618 - docs: precidence -> precedence
1619 - docs: undifined behaviour -> undefined behavior
1620 - docs: debian -> Debian
1621 - docs: zink -> Zink
1622 - docs: vulkan -> Vulkan
1623 - docs: attachements -> attachments
1624 - docs: acress -> across
1625 - docs: pluggins -> plug-ins
1626 - docs: pusbuf -> pushbuf
1627 - docs: metadatas -> metadata
1628 - docs: use os.pardir
1629 - docs: allow out-of-tree docs build
1630 - meson: build html-docs
1631 - docs: automatically generate depfile
1632 - meson: error when missing hawkmoth
1633 - meson: allow specifying html-docs-path
1634 - ci: build docs using meson
1635 - panvk: support x11 wsi
1636 - vulkan/runtime: tne -> the
1637 - vulkan/runtime: initizlie -> initialize
1638 - vulkan/runtime: abreviation -> abbreviation
1639 - vulkan/runtime: multiesample -> multisample
1640 - vulkan/runtime: implementaiton -> implementation
1641 - docs: fix bootstrap-extension
1642 - docs/panfrost: fix numbered list
1643 - docs/panfrost: fix math-notation
1644 - docs/panfrost: use math-role more
1645 - docs/panfrost: use c:func-role for function
1646 - docs/panfrost: quote identifiers
1650 - nvk: Enable 8bit and 16bit access in VK_KHR_workgroup_memory_explicit_layout.
1654 - nak: Don't saturate depth writes
1655 - nvk: Only clip Z with the guardband
1656 - nouveau/class_parser.py: Fix the docs for --out-rs
1657 - nvk: Advertise VK_EXT_pipeline_robustness
1658 - nouveau/headers: Clean up the meson a bit
1659 - spirv: Auto-generate spirv_info.h
1660 - spirv: Update the JSON and headers
1661 - spirv: Better handle duplicated enums in the JSON parser
1662 - spirv: Generate a spirv_capabilities struct
1663 - spirv: Record capabilities rather than ad-hoc bools
1664 - mesa: Stop pretending to support SPV_AMD_gcn_shader in OpenGL
1665 - spirv: Move the old AMD extensions out of capabilities
1666 - spirv: Move the printf enable out of capabilities
1667 - spirv: Add supported_capabilities to vtn_builder
1668 - spirv: Use supported_capabilities for various checks
1669 - spirv: Drop the SubgroupUniformControlFlow check
1670 - spirv: Add a table of all implemented capabilities
1671 - spirv: Check capabilities using the supported_capabilities table
1672 - spirv: Add support for specifying caps through the new struct
1673 - spirv: Use spirv_capabilities in tests
1674 - mesa: Flip the script on SPIR-V extension enabling
1675 - mesa: Use the new spirv_capabilities struct
1676 - clover: Use the new spirv_capabilities struct
1677 - rusticl: Use the new spirv_capabilities struct
1678 - vulkan: Set SPIR-V caps from supported features
1679 - radv: Use vk_physical_device_get_spirv_capabilities()
1680 - intel/kernel: Use the new capabilities struct
1681 - asahi/clc: Use the new spirv_capabilities struct
1682 - zink: Use the new spirv_capabilities struct
1683 - anv: Use spirv_capabilities for the float64 shader
1684 - ir3: Use spirv_capabilities in ir3_cmdline
1685 - microsoft: Use spirv_capabilities for spirv_to_dxil
1686 - spirv: Get rid of the old caps struct
1687 - nvk: Re-emit sample locations when rasterization samples changes
1688 - nvk/meta: Restore set_sizes[0]
1689 - nvk: Get rid of sets_dirty
1690 - nvk: Don't rely on push_dirty for which push sets exist
1691 - nouveau/headers: Add a bool for whether or not to dump offsets
1692 - nvk/upload_queue: Only upload one line of data
1693 - nvk/upload_queue: Add some useful asserts
1694 - nvk/upload_queue: Add a _fill method
1695 - nvk: Use the upload queue for NVK_DEBUG=zero_memory
1696 - nvk: Improve the GetMemoryFdKHR error
1697 - nouveau/winsys: Take a reference to BOs found in the cache
1698 - nouveau/winsys: Make BO_LOCAL and BO_GART separate flags
1699 - nvk: Allow GART for dma-bufs
1700 - nil: Use the right PTE kind for Z32 pre-Turing
1701 - nvk: Set color/Z compression based on nil_image::compressed
1702 - nil: Default to NV_MMU_PTE_KIND_GENERIC_MEMORY on Turing+
1703 - nvk: Allow VK_IMAGE_ASPECT_MEMORY_PLANE_0_BIT
1704 - drm-uapi: Sync nouveau_drm.h
1705 - nouveau/winsys: Add back nouveau_ws_bo_new_tiled()
1706 - nvk: Support image creation with modifiers
1707 - nvk: Set tile mode and PTE kind on dedicated dma-buf BOs
1708 - nvk: Implement DRM format modifier queries
1709 - nvk: Advertise VK_EXT_queue_family_foreign
1710 - nvk: Advertise VK_EXT_image_drm_format_modifier
1711 - vulkan/wsi: Bind memory planes, not YCbCr planes.
1712 - nvk/wsi: Advertise modifier support
1713 - zink: Set workarounds.can_do_invalid_linear_modifier for NVK
1714 - nvk: Fix misc. whitespace and style issues
1715 - nvk: Go wide for query copies
1716 - nvk: Store descriptor set addresses in descriptor state
1717 - nvk: Add static asserts for nvk_buffer_address layout
1718 - nvk: Store an nvk_buffer_address for each set in the root table.
1719 - nvk: Advertise 32 descriptor sets
1720 - nvk: Move and better document set_dynamic_buffer_start
1721 - nvk: Add an NVK_MAX_SAMPLES #define
1722 - nvk: Refactor nvk_meta_begin() to use a desc helper
1723 - nvk/meta: Save and restore set_dynamic_buffer_start
1724 - nak: Emit !PT for carries on IADD3
1725 - nak: Add with -0 for fabs()
1726 - nak: Don't emit a plop3 for immediate shift sources
1727 - nak: Encode LDC directly
1728 - vulkan: Update XML and headers to 1.3.286
1729 - spirv: Update the JSON and headers
1730 - nir: Handle cmat types in lower_variable_initializers
1731 - spirv: Handle constant cooperative matrices in OpCompositeExtract
1732 - spirv: Assert that non-vector composites have the right length
1733 - spirv: Implement SPV_EXT_replicated_composites
1734 - nvk: Advertise VK_EXT_shader_replicated_composites
1735 - anv: Advertise VK_EXT_shader_replicated_composites
1736 - hasvk: Advertise VK_EXT_shader_replicated_composites
1737 - radv: Advertise VK_EXT_shader_replicated_composites
1738 - turnip: Advertise VK_EXT_shader_replicated_composites
1739 - lavapipe: Advertise VK_EXT_shader_replicated_composites
1740 - dozen: Advertise VK_EXT_shader_replicated_composites
1741 - nir/print: Improve divergence information
1742 - nak: Fix NAK_DEBUG=serial for warp barriers
1743 - nak: Only convert the written portion of the buffer in NirInstrPrinter
1744 - nak: Fix BasicBlock::phi*() for OpAnnotate
1745 - nak: BMov is always variable-latency
1746 - nak: Only copy-prop neg into iadd2/3 if no carry is written
1747 - nak: Get rid of OpINeg
1748 - nak: Expose a BasicBlock::map_instrs() helper
1749 - nak: Add some helpers for uniform instructions and registers
1750 - nak: Add OpR2UR
1751 - nak: Clean up bindless cbuf handles
1752 - nak/ra: Move an assert
1753 - nak: Make SSARef::file() return Option<RegFile>
1754 - nak: Drop BasicBlock::new()
1755 - nak: Add a concept of uniform blocks
1756 - nak/to_cssa: Resolve phi register file mismatches
1757 - nak/ra: Spill UGPRs and UPreds
1758 - nak/ra: Never move uniform regs in non-uniform blocks
1759 - nak: Support uniform regs in lower_copy_swap()
1760 - nak/sm70: Defer ALU src processing until encode_alu()
1761 - nak/sm70: Rework ALU source encode helpers
1762 - nak/sm70: Add support for encoding uniform ALU ops
1763 - nak/sm70: Fix encoding of fadd/fsetp and friends with UGPRs
1764 - nak/sm70: Implement a bunch of uniform ops on SM75+
1765 - nak/legalize: Fold immediate sources before instructions
1766 - nak/legalize: Drop some pointless plop3 logic
1767 - nak/legalize: Be more precise about shfl and out
1768 - nak/legalize: Fix imad and ffma legalization on SM50
1769 - nak/legalize: Patch a RegFile through to copy helpers
1770 - nak/legalize: Handle uniform sources in warp instructions
1771 - nak/legalize: Ensure all SSA values for a given ref are in the same file
1772 - nak/legalize: Copy uniform vectors in non-uniform control-flow
1773 - nak/legalize: Uniform instructions can't have cbuf sources
1774 - nak/legalize: Explicitly ignore OpPhiSrcs and OpPhiDsts
1775 - nak/calc_instr_deps: Rename a couple variables
1776 - nak/calc_instr_deps: Rewrite calc_delays() again
1777 - nak/calc_instr_deps: Add latencies for uniform instructions
1778 - nak: Add a opt_uniform_instrs() pass
1779 - nak/copy_prop: Rewrap a couple comments
1780 - nak/copy_prop: Don't propagate UBOs into uniform instructions
1781 - nak/lower_cf: Parent scopes are never NULL
1782 - nak/lower_cf: Track block divergence
1783 - nak: Convert to LCSSA before divergence analysis
1784 - nak/lower_cf: Flag phis as convergent when possible
1785 - nak/from_nir: Clean up phi annotations
1786 - nak: Add a UniformBuilder
1787 - nak/from_nir: Emit uniform instructions when !divergent
1788 - nak/sm70: Properly encode bindless cbufs
1789 - nak/dce: Account for bindless CBuf handles
1790 - nak/calc_instr_deps: Account for bindless CBufs
1791 - nak/bitset: Add an iterator
1792 - nak/ra: Handle bindless CBufs
1793 - nak/ra: Pull searching for unused/unpinned regs into a helper
1794 - nak/ra: Rename PinnedRegAllocator to VecRegAllocator
1795 - nak/ra: Add a concept of pinned registers to RegAllocator
1796 - nak: Add OpPin and OpUnpin
1797 - nak/legalize: Allow pinned uniform vectors in non-uniform blocks
1798 - nak/legalize: Bindless cbufs must be pinned in non-uniform blocks
1799 - nak/copy_prop: Don't propagate bindless cbufs into non-uniform blocks
1800 - nir: Add some new _nv intrinsics
1801 - nvk,nak: Switch to nir_intrinsic_ldc_nv
1802 - nak: Implement r2ur_nv
1803 - nak: Implement [un]pin_cx_handle_nv
1804 - nir: Add nir_foreach_block_in_cf_node_safe() iterators
1805 - nak: Lower non-uniform ldcx_nv to global loads
1806 - nak: Implement nir_intrinsic_ldcx_nv
1807 - nvk: Split SSBO and UBO address formats
1808 - nvk: Split write_[dynamic\_]buffer_desc into UBO and SSBO variants
1809 - nvk: Align buffer descriptors
1810 - nvk: Rename nvk_cmd_buffr_get_cbuf_descriptor()
1811 - nvk: Make nvk_min_cbuf_alignment() inline
1812 - nvk/lower_descriptors: Add a descriptor_type_is_ubo/ssbo() helper
1813 - nvk: Move the zero offset optimization to load_descriptor_for_idx_intrin()
1814 - nvk: Allow the cbuf optimization for VK_DESCRIPTOR_TYPE_MUTABLE_EXT
1815 - nvk/descriptor_set_layout: Record which dynamic buffers are UBOs
1816 - nvk: Use bindless cbufs on Turing+
1817 - nvk: Be much more conservative about rebinding cbufs
1818 - nvk: Use cbuf loads for variable pointers dynamic SSBO descriptors
1819 - nvk: s/draw_idx/draw_index/g
1820 - nvk: Pass the base workgroup and global size to flush_compute_state()
1821 - nvk: Use helper macros for accessing root descriptors
1822 - nvk: Pass the queue to draw/dispatch_state_init()
1823 - nvk: Use inline constant buffer updates for CB0
1824 - nvk: Only write draw parameters to cb0 when they change
1825 - nvk: Refactor build_cbuf_map()
1826 - nak,nir: Drop r2ur_nv in favor of as_uniform
1827 - nouveau: Fix a race in nouveau_ws_bo_destroy()
1828 - nvk: Use NVK_VK_GRAPHICS_STAGE_BITS in dirty_cbufs_for_descriprots()
1829 - nvk: Dirty cbufs in CmdPushDescriptorSetWithTemplate2KHR
1830 - intel/blorp: Set nir_shader::options up-front before building
1831 - util/format_pack: Fix packing of signed 1010102 SSCALED formats
1832 - util/format_pack: Also use iround for SCALED formats
1833 - util/format_pack: Clamp SNORM values to [-1, 1] when unpacking
1834 - util/format: Round to nearest even when converting to R11G11B20F
1835 - util/format: Handle denorms when converting to R11G11B10F
1836 - nir/format_convert: Smash NaN to 0 in pack_r9g9b9e5()
1837 - nir/format_convert: Use fmin/fmax to clamp R9G9B9E5 data
1838 - nir: Add a nir_intrinsic_use for unit tests
1839 - nir: Move most of nir_format_convert to a C file
1840 - nir: Support 0 and 32 bits in some format conversion helpers
1841 - util: Make format_srgb.h C++ safe
1842 - nir: Add a format pack helper and tests
1843 - nir: Add a format unpack helper and tests
1844 - nir/format_convert: Assert that UNORM formats are <= 16 bits
1845 - ci: Update trace SHAs
1846 - vulkan/meta: Use demote instead of discard
1847 - nvk: Fix whitespace issues around conservative rasterization
1848 - nvk: Re-order conservative rasterization checks
1849 - nvk: Don't emit conservative rasterization before Maxwell B
1850 - nvk: Silently fail to enumerate if not on nouveau
1851 - util/cnd_monotonic: Move the guts to a c file
1852 - util/cnd_monotonic: Use a void * on Windows
1853 - vulkan/wsi/wayland: Use mtx_t and u_cnd_monotonic
1854 - vulkan/wsi/x11: Use c11/threads for thread spawning
1855 - vulkan/wsi/x11: Use mtx_t and u_cnd_monotonic
1856 - vulkan/wsi/display: Use mtx_t and u_cnd_monotonic
1857 - vulkan/wsi/queue: Use mtx_t and u_cnd_monotonic
1858 - vulkan/wsi: Delete wsi_init_pthread_cond_monotonic
1859 - vulkan: Use u_cnd_monotonic for vk_sync_timeline
1860 - nvk: Why are nvk_image.c/h writeable?
1861 - nvk: Bump the sparse alignment requirement on buffers to 64K
1862 - nvk: Align sparse-bound images to the sparse binding size
1863 - nvk: Fetch debug flags from the physical device
1864 - nvk: Initialize the debug flags in nvk_instance
1865 - nvk: Add the start of a KMD abstraction
1866 - nvk/nvkmd: Implement dev and pdev for nouveau
1867 - nvk: Use the NVKMD interface for device enumeration
1868 - nvk/nvkmd: Add memory and virtual address interfaces
1869 - nvk/nvkmd: Implement the mem and va interfaces for nouveau
1870 - nvk: Add static wrappers for image/buffer binding
1871 - nvk: Use nvkmd_mem for nvk_device_memory
1872 - nvk: Use nvkmd_mem for nvk_image::linear_tiled_shadow_mem
1873 - nvk: Use nvkmd_mem for nvk_cmd_pool
1874 - nvk: Use nvkmd_mem for nvk_descriptor_pool
1875 - nvk: Use nvkmd_mem in nvk_upload_queue
1876 - nvk: Use nvkmd_mem for descriptor tables
1877 - nvk: Use nvkmd_mem for shader and event heaps
1878 - nvk: Use nvkmd_mem for query pools
1879 - nvk: Use an nvkmd_mem for the SLM area
1880 - nvk: Drop extra_bos from nvk_queue_submit_simple()
1881 - nvk: Use nvkmd_mem for the nvk_queue_state::push
1882 - nvk: Use nvkmd_mem for the zero page, VAB, and CB0
1883 - nvk/nvkmd: Add a context interface
1884 - nvk/nvkmd: Implement nvkmd_ctx for nouveau
1885 - nvk: Convert the upload queue to nvkmd_ctx
1886 - nvk: Use an nvkmd_ctx for sparse binding
1887 - nvk: Use nvkmd_ctx for queue submit
1888 - nvk: Remove the last vestages of nouveau/winsys from core NVK
1889 - nouveau/mme: Don't dereference an empty vector
1890 - nouveau/mme: Don't leak data_bo
1891 - nouveau/mme: Use fixed BO addresses in the MME test
1892 - nvk: Move Heaps and BO binding into nvkmd
1893 - nvk: Move debug flags int nvk_debug.h
1894 - nvk/nvmkd: Plumb parent pointers through everywhere
1895 - nvk/nvkmd: Re-implement NVK_DEBUG=vm
1896 - nvk: Do mem maps directly in nvkmd on nouveau
1897 - nvk/nvkmd: Add real mem<->bo flag translation
1898 - nvk/nvkmd: Flip the script on NO_SHARED
1899 - nvk: Drop nvk_buffer::is_local
1900 - nvk/nvkmd: Rework memory placement flags
1901 - nvk/nvkmd: Be more specific about memory alignments
1902 - nvk/nvkmd: Be a lot more pedantic about VA alignments
1903 - nvk: Put CB0 in VRAM
1904 - nvk: Put descriptors in VRAM
1905 - nouveau/push: Cache the last header DW to avoid read-back
1906 - nak/sph: Stop storing the shader model in ShaderProgramHeader
1907 - nak: Move encode_sm* to to sm*.rs
1908 - nak/sm50: Get rid of the hand-rolled align_up/down() helpers
1909 - nak: Plumb a ShaderModel trait through everywhere
1910 - nak/ra: Move the NAK_DEBUG=spill logic into RA
1911 - nak: Move RegFile::num_regs() into ShaderModel
1912 - nak: Move Instr::can_be_uniform() into ShaderModel
1913 - nak: Move instruction encoding into ShaderModel
1914 - nak/sm70: Move instruction encoding into a trait
1915 - nak/sm70: Re-organize the code a bit
1916 - nak/legalize: Move a bunch of helpers to a trait
1917 - nak/legalize: Handle OpBreak and OpBSSy specially
1918 - nak/legalize: Handle RA instructions up-front
1919 - nak/sm70: Move legalization into SM70Op
1920 - nak/sm50: Move instruction encoding into a trait
1921 - nak/sm50: Move legalization into SM50Op
1922 - nak: Add a legalize() method to ShaderModel
1923 - nak/sm50: Re-order all the ops
1924 - nak/sm50: Fix immediates for IMnMx
1925 - zink/kopper: Set VK_COMPOSITE_ALPHA_OPAQUE_BIT when PresentOpaque is set
1926 - nak: gather instr count explicitly
1927 - nvk/nvkmd: nouveau uses the OS page size
1928 - nvk: Drop the sparse alignment back down to 4096
1929 - nvk: Use the page size queried from NVKMD
1930 - nak/nir: Use an indirect load for sample locations
1931 - nak/copy_prop: Propagate OpSel with a selector of SrcRef::Zero
1932 - nak/copy_prop: Ignore the top 16 bits of OpPrmt::sel
1933 - nak: Don't print the destination of OpIpa twice
1934 - nir,nak: Add a nir_op_prmt_nv
1935 - nak/nir: Use prmt in texture lowering
1936 - nak/nir: Use prmt for barycentric offset lowering
1937 - nak/nir: Make interpolate_at_sample more efficient
1938 - nak: Add some helpers for working with OpPrmt selectors
1939 - nak: Optimize nested OpPrmt
1940 - nak: Add a pass macro for more consistent debug printing
1941 - nak: Run copy-prop again after opt_prmt and opt_lop
1942 - nvk: Fix indirect cbuf binds pre-Turing
1943 - nvk: Don't advertise sparse residency on Maxwell A
1944 - nvk: Reject sparse images on Maxwell A and earlier
1945 - nak/spill_values: Don't assume no trivial phis
1946 - meson/megadriver: Don't invoke the megadriver script with no drivers
1947 - nak: Sample locations are byte-aligned
1948 - nvk: Require color or depth/stencil attachment support for input attachments
1949 - nvk: Support STORAGE_READ_WITHOUT_FORMAT on buffers
1950 - zink: Align descriptor buffers to descriptorBufferOffsetAlignment
1954 - intel/brw/xe2+: Keep PS sample mask in the f1.0 register whether or not kill is used.
1955 - intel/brw: Don't emit Z coordinate interpolation if CPS isn't in use.
1956 - intel/brw/xe2+: Fix indirect extended descriptor setup for scratch space.
1957 - iris: Allocate fixed amount of space for blend state.
1958 - blorp: Allocate fixed amount of space for blend state.
1959 - intel/brw/xe2+: Don't use SEL peephole on 64-bit moves.
1960 - intel/brw/xe2+: Fix 64-bit subgroup scan intrinsics not to rely on SEL instructions.
1961 - intel/brw/xe2+: Lower 64-bit SHUFFLE and CLUSTER_BROADCAST.
1962 - intel/xe2+: Enable native 64-bit integer arithmetic.
1963 - nir: Add option to lower 64-bit uadd_sat.
1964 - intel/brw/xe2+: Lower 64-bit integer uadd_sat.
1965 - intel/brw/xe2+: Round up spill/unspill data size to nearest reg_size multiple.
1966 - intel/xe2+: Enable native 64-bit integer arithmetic.
1967 - iris,anv/xe2+: Enable the DX10/OGL border mode for YCrCb as per Wa_14014226147.
1968 - iris,anv/xe2+: Set tessellation redistribution regions per patch to recommended values.
1969 - iris,anv/xe2+: Use pipelined variant of 3DSTATE_DRAWING_RECTANGLE.
1970 - intel/brw/xe2+: Use active-thread-only barriers available since Xe2+.
1971 - iris/xe2+: Fix format of scratch space surface address in various 3DSTATE packets.
1972 - anv/xe2+: Fix format of scratch space surface address in various 3DSTATE packets.
1973 - intel/fs/gfx20+: Fix surface state address on extended descriptors for NIR scratch intrinsics.
1974 - intel/fs/xe2+: Ask driver for PS payload registers based on barycentric load intrinsics in use.
1975 - iris/gfx11+: Request PS payload fields for ALU-based interpolation via 3DSTATE_PS_EXTRA.
1976 - anv/gfx11+: Request PS payload fields for ALU-based interpolation via 3DSTATE_PS_EXTRA.
1977 - intel/fs/xe2+: Don't lower barycentric load offsets to fixed-point format on Xe2+.
1978 - intel/fs/xe2+: Add ALU-based implementation of barycentric interpolation at a per-channel offset.
1979 - intel/fs/xe2+: Add ALU-based implementation of barycentric interpolation at a per-channel sample.
1980 - intel/dev: Add GRF size information to the intel_device_info struct.
1981 - anv/xe2+: Align push constant ranges to GRF boundaries.
1982 - intel/brw: Implement null push constant workaround.
1983 - intel/dev: Add devinfo flag for TBIMR push constant workaround.
1984 - anv/gfx12.5: Pass non-empty push constant data to PS stage for TBIMR workaround.
1985 - iris/gfx12.5: Pass non-empty push constant data to PS stage for TBIMR workaround.
1986 - iris: Pin pixel hashing table BO from iris_batch submission instead of from iris_state.
1990 - aco/tests: Insert p_logical_start/end in reduce_temp tests
1991 - aco/spill: Insert p_start_linear_vgpr right after p_logical_end
1992 - radv: Use max_se instead of num_se where appropriate
1993 - radeonsi: Use max_se instead of num_se where appropriate
1994 - radv/rt: Fix memory leak when compiling libraries
1995 - aco/spill: Don't spill phis with all-undef operands
1996 - aco: Limit rt stages to 128 vgprs
2000 - i915g: fix generation of large mipmaps
2001 - i915g: fix mipmap-layout for npots
2002 - i915g: fix max_lod at mipmap-sampling
2006 - radeonsi: add GL_EXT_texture_filter_minmax extension
2007 - radeonsi: add GL_ARB_texture_filter_minmax extension
2008 - radeonsi: fix eptich on chips without image opcodes (e.g. gfx940)
2009 - amd/common: skip lane size determination for chips without image opcodes (e.g. gfx940)
2013 - aco/tests: don't use undef for descriptors
2014 - aco/tests/post_ra: fix various validation errors
2015 - aco/lower_to_hw: fix v_cvt_pk_u16_u32 instruction format
2016 - aco/lower_to_hw: fix 16bit p_insert on gfx8
2017 - aco/tests: validate before and after post-ra tests
2018 - spirv: preserve signed zero in modf
2019 - aco/lower_to_hw: don't use regClass to identify subdword reductions
2020 - aco: add a subdword lowering pass
2021 - aco: add tests for lower_subdword
2022 - aco/ra: remove gfx6/7 subdword paths
2023 - aco/lower_to_hw: remove gfx6/7 subdword paths
2024 - ac/nir: explicitly use pack_half_2x16_rtz
2025 - radv, radeonsi: don't use D16 for f2f16_rtz
2026 - radv: always run nir_opt_16bit_tex_image
2027 - nir/opt_16bit_tex_image: pass options to opt_16bit_dest
2028 - nir/opt_16bit_tex_image: optimize packed conversions too
2029 - aco/gfx11+: use v_cvt_pk_u8_f32 for 8bit constant copies
2030 - aco/gfx10: use v_add_u16 with literal for constant copies
2031 - aco/tests: simplify small constant copy test
2032 - aco/gfx11+: optimize v_fma_mix throughput
2033 - zink: use bitcasts instead of pack/unpack double opcodes
2034 - aco/gfx11: use v_swap_b16
2035 - aco/optimizer: remove ineffective vcc opt
2036 - aco/optimizer: remove ineffective undef opt
2037 - aco: remove perfwarn
2038 - aco: don't pass program to emit_bpermute
2039 - aco/lower_to_hw: add copy_constant_sgpr
2040 - aco: small constant copy optimizations
2041 - aco/lower_to_hw: use copy_constant_sgpr for masks
2042 - aco/lower_to_hw: optimize split 64bit constant copies
2043 - aco/optimizer: use p_create_vector to create mask when a copy can't be used
2044 - nir: remove unpack_half_flush_to_zero
2045 - nir/opt_uniform_atomics: handle inverse_ballot when detecting single lane ifs
2046 - aco: optimize branching sequence with p_create_vector exec producer
2047 - nir: sink/move inverse_ballot like moves
2048 - ac: set has_pack_32_4x8
2049 - nir: lower pack_uvec4_to_uint to pack_32_4x8 if supported
2050 - nir/opt_algebraic: alternative 8bit pack_[us]norm_4x8 lowering
2051 - aco: rework how affinities for acc operands are determined
2052 - aco: add affinities for possible sopk optimizations
2053 - aco/gfx11+: fix inline constants for v_pk_fmac_f16
2054 - aco: move literal unswizzle opt to RA
2055 - aco/ra: use a switch to check vop2acc instruction support
2056 - aco: move s_add_u32 -> s_addk_i32 optimization fully to ra
2057 - amd/common: set COMPUTE_STATIC_THREAD_MGMT_SE2-3 correctly on gfx10-11
2058 - aco: add more anonymous namespaces
2059 - aco: make local functions static in files without anonymous namespace
2060 - radv: inline partial push constant loads
2061 - nir: add ford, funord, fneo, fequ, fltu, fgeu
2062 - aco: implement ford, funord, fneo, fequ, fltu, fgeu
2063 - ac/llvm: implement ford, funord, fneo, fequ, fltu, fgeu
2064 - ac/nir: enable ford, funord, fneo, fequ, fltu, fgeu
2065 - nir/opt_algebraic: look through fabs/fneg when matching fmulz/ffmaz
2066 - nir/optimize cmp(a, -0.0)
2067 - nir/opt_algebraic: optimize cmp(fneg(a), #b) and feq with fabs
2068 - nir/opt_algebraic: add various unordered/ordered patterns from aco
2069 - aco: remove ordered/unordered optimizations
2070 - aco/ir: remove unused vopc helpers
2071 - iris/ci: update trace checksums
2072 - aco/ra: fix affinity for s_addk
2073 - aco: fix s_delay_alu with salu and trans dependency
2074 - aco,nir: add dpp16_shift_amd intrinsic
2075 - radv/nir: add a pass to optimize shuffle/booleans dependent only on tid/consts
2076 - radv: use radv_nir_opt_tid_function for shuffles
2077 - radv: use radv_nir_opt_tid_function to create inverse_ballot
2078 - aco/gfx12: use trans s_delay_alu for pseudo scalar
2079 - aco/gfx12: don't allow vgpr operands for pseudo scalar
2080 - aco/gfx11.5: select s_cvt_[ui]32_f32
2081 - aco/gfx11.5: select s_(ceil|floor|trunc|rndne)
2082 - aco: add aco_opcode::p_s_cvt_f16_f32_rtne
2083 - aco/gfx11.5: select SALU float conversions
2084 - aco/gfx11.5: fix s_fmac acc to definition
2085 - aco/gfx11.5: select SOP2 float instructions
2086 - aco/gfx11.5: select SOPC float instructions
2087 - aco/gfx11.5: select SALU fsat
2088 - aco/gfx11.5: select SALU fsign
2089 - aco/gfx11.5+: allow sgpr dst for trans ops and use pseudo scalar ops on gfx12
2090 - aco/gfx11.5: select SALU fneg/fabs
2091 - aco/gfx11.5: select SALU fquantize2f16
2092 - aco: micro optimize VALU fquantize2f16
2093 - aco: handle clustered uniform reductions correctly
2094 - nir: constant fold inverse_ballot
2095 - aco: remove optimize_cmp_subgroup_invocation
2096 - spirv: ignore more function param decorations
2097 - aco/optimizer: update temp_rc when converting to uniform bool alu
2098 - aco/gfx11+: don't use VOP3 v_swap_b16
2099 - nir/lower_int64: replace uadd_sat with ior for find_lsb64 and ufind_msb64
2100 - aco/gfx10+: set lateKill for sgprs used by wave64 VALU writing a mask
2104 - zink/kopper: Wait for last QueuePresentKHR to finish before acquiring for readback
2105 - mesa/st: don't use base shader serialization when uniforms are not packed
2106 - r600/sfn: Set bit size for newly created store intrinsic
2107 - zink: limit minSampleShading to a maxium value of 1.0
2111 - ci: Add S3 id_token for all jobs
2112 - ci: Use id_tokens for JWT auth
2113 - ci/lava: Fix cmdline for UART/fastboot devices
2115 Hans-Kristian Arntzen (5):
2117 - vulkan: Update XML and headers to 1.3.285.
2118 - ac/surface: Add surface flags to prefer 4K and 64K alignment.
2119 - radv: Implement VK_MESA_image_alignment_control
2120 - wsi/common: Do not update present mode with MESA_VK_WSI_PRESENT_MODE.
2121 - wsi/x11: Bump maximum number of outstanding COMPLETE events.
2125 - zink: remove workaround of FB modifiers forcing present state
2129 - v3dv: fix VK_KHR_vertex_attribute_divisor
2130 - v3d,v3dv: stop hard-coding max attrib divisor
2131 - broadcom/compiler: assert on array overflow
2132 - v3d: fix array_len when precompiling outputs for shader-db
2133 - broadcom/compiler: fix num_textures for precompiled shaders
2134 - broadcom/compiler: don't read excess channels on image loads
2135 - broadcom/compiler: simplify v3d_vir_emit_tex
2136 - broadcom/cle: fix up shader record for V3D 7.1.10 / 2712D0
2137 - v3d: support 2712D0
2138 - v3dv: support 2712D0
2139 - broadcom/compiler: make add_node return the node index
2140 - broadcom/compiler: don't assign payload registers to spilling setup temps
2141 - broadcom/compiler: apply payload conflict to spill setup before RA
2142 - broadcom/compiler: check if vertex shader writes point size
2143 - v3dv: only flag 'shader writes point size' if the shader actually writes it
2144 - v3dv: emit a default point size when drawing points
2145 - v3dv: drop unused stride field from v3dv_pipeline_vertex_binding
2146 - v3dv: fix incorrect index buffer size
2147 - v3dv: use pSizes paramater in vkCmdBindVertexBuffers2
2148 - v3dv: implement vkCmdBindIndexBuffer2KHR
2149 - v3dv: handle VkBufferUsageFlags2CreateInfoKHR
2150 - v3dv: handle VkPipelineCreateFlags2CreateInfoKHR
2151 - v3dv: lower maxVertexInputBindingStride to match vulkan runtime
2152 - v3dv: shader modules are deprecated with VK_KHR_maintenance5
2153 - v3dv: implement vkGetImageSubresourceLayout2KHR
2154 - v3dv: refactor create_image
2155 - v3dv: add a get_image_subresource_layout helper
2156 - v3dv: implement vkGetDeviceImageSubresourceLayoutKHR
2157 - v3dv: implement vkGetRenderingAreaGranularityKHR
2158 - v3dv: fix pipeline leaks when meta pipeline cache is disabled
2159 - v3dv: fix a few asserts that check layerCount instead of array_layers
2160 - v3dv: allow VK_REMAINING_ARRAY_LAYERS in VkImageSubresourceLayers
2161 - v3dv: remove blit shader restriction on depth/stencil not being linear
2162 - v3dv: disable some TLB paths for cases of linear depth/stencil stores
2163 - v3dv: support VK_FORMAT_A1B5G5R5_UNORM_PACK16_KHR
2164 - v3dv: add more checks for device loss
2165 - v3dv: fix handling of pipeline flags when pipeline init fails
2166 - v3dv: expose VK_KHR_maintenance5
2167 - broadcom/compiler: initialize payload_conflict for all initial nodes
2168 - v3dv: don't call wsi_device_init too early
2169 - broadcom/compiler: don't spill in between multop and umul24
2170 - broadcom/compiler: fix per-quad spilling
2171 - broadcom/compiler: validate rtop + thrsw hazard
2172 - broadcom/compiler: drop multop if we dce umul24
2173 - broadcom/compiler: add missing signal compatibilities for V3D 7.x
2174 - broadcom/compiler: add new float32 unpack modifiers in V3D 7.x
2175 - broadcom/compiler: disallow copy propagation of FMOV exclusive modifiers
2176 - broadcom/compiler: implement nir_op_fsat
2177 - v3d: don't lower fsat on V3D 7.x
2178 - v3dv: make nir helpers receive nir compiler options from caller
2179 - v3dv: don't lower fsat on V3D 7.x
2180 - v3d: skip tlb loads when emitting clears with a draw call
2181 - v3d: rename job->clear to job->clear_tlb
2185 - intel/brw: Fix optimize_extract_to_float for i2f of unsigned extract
2186 - intel/brw: Avoid optimize_extract_to_float when it will just be undone later
2187 - intel/elk: Fix optimize_extract_to_float for i2f of unsigned extract
2188 - nir/algebraic: Optimize some extract_* expressions
2189 - spirv: Use fp16 fp_fast_math settings when lowering fp16 asin and acos
2190 - intel/brw: Remove dsign optimization
2191 - intel/elk: Remove dsign optimization
2192 - intel/brw: Use fs_inst::resize_sources in brw_fs_opt_algebraic
2193 - intel/brw: Add support for fcsel opcodes
2194 - intel/brw: Handle fsign optimization in a NIR algebraic pass
2195 - intel/brw: Update CSEL source type validation
2196 - intel/brw: Combine constants and constant propagation for CSEL
2197 - intel/brw: Algebraic optimizations for CSEL
2198 - intel/brw: Implement more strictly correct fsign lowering
2199 - intel/brw: Use range analysis to optimize fsign
2200 - nir/algebraic: Add nir_lower_int64_options::nir_lower_iadd3_64
2201 - nir/search: Fix is_16_bits for vectors
2202 - nir/search: Refactor is_16_bits
2203 - nir/algebraic: More patterns to generate iadd3
2204 - nir/algebraic: intel/fs: Optimize some patterns before lowering 64-bit integers
2205 - intel/brw: Temporarily disable result=float16 matrix configs
2206 - intel/brw: Major rework of lower_cmat_load_store
2207 - intel/brw/xe2+: Catch invalid uses of writes_accumulator earlier
2208 - intel/brw/xe2+: Adjust size_read() for DPAS
2209 - intel/brw/xe2+: Scale size_written by reg_unit for DPAS
2210 - intel/brw/xe2+: Adjust DPAS lowering to DP4A to accommodate larger GRF and SIMD16
2211 - intel/brw/xe2+: Allow vec16 for cooperative matrix
2212 - nir: dpas_intel second source can have different number of components
2213 - intel/brw/xe2+: Add LNL cooperative matrix configurations
2214 - intel/tools: Advertise I915_PARAM_HAS_EXEC_TIMELINE_FENCES
2215 - intel/brw: Test corner case CSE of ADD3 instructions
2216 - intel/brw: Don't propagate saturate to an instruction that writes flags
2217 - intel/elk: Don't propagate saturate to an instruction that writes flags
2221 - llvmpipe: add shader cache support for ORCJIT implementation
2222 - gallivm: orcjit: use a mutex to protect symbol looking up
2223 - util: detect LoongArch architecture
2224 - gallivm: add LoongArch support to the mattrs setting code
2225 - llvmpipe: add LoongArch support in ORCJIT
2226 - gallivm: orcjit: keep the ownership of tm for LPJit
2227 - gallivm: orcjit: use atexit to release LPJit singleton at exit
2231 - nir: add {load,store}_global_etna intrinsics
2235 - compiler: reorder FLOAT_CONTROLS enums
2236 - nir: track some float controls bits per instruction
2237 - spirv: gather some float controls bits per instruction
2238 - nir: check inf/nan/sz preserve per-instruction
2239 - nir/algebraic: support float controls conditions per instruction
2240 - nir/algebraic: move float control conditions to be per instruction
2241 - vtn: support float controls2
2242 - anv: enable VK_KHR_shader_float_controls2
2243 - anv: check requirements for VK_IMAGE_USAGE_FRAGMENT_SHADING_RATE
2244 - anv: fix casting to graphics_pipeline_base
2245 - anv: consolidate DestroyPipeline for graphics and graphics_lib
2246 - intel/brw: fix subgroup size of geometry stages for lnl+
2247 - anv: check cmd_buffer is on a transfer queue more properly
2248 - intel/brw: add fetch_viewport_index function
2249 - intel/brw: always read LAYER/VIEWPORT from the FS payload
2250 - vulkan/runtime: pColorAttachmentInputIndices is allowed to be NULL
2251 - vulkan/properties: handle LayeredApiPropertiesListKHR
2252 - anv: enable VK_KHR_maintenance7
2253 - anv: get scratch surface from the correct pool
2254 - anv: set MOCS for protected memory when needed
2255 - intel/rt: fix terminateOnFirstHit handling
2259 - meson: Fix invalid kwarg name
2263 - radv: Return hang status from radv_check_gpu_hangs()
2267 - nir_opt_algebraic: Add a couple optimizations for lowered unpack(pack())
2268 - wgl: Delete pixelformat support query
2269 - wgl: Fix flag check for GDI compat
2270 - nir_range_analysis: Use fmin/fmax to fix NAN handling
2271 - d3d12: Use GetResourceAllocationInfo instead of GetCopyableFootprints for residency sizes
2272 - nir: Remove assert-only variable by inlining its single use
2273 - zink: Add ASSERTED to assert-only local variable
2274 - mesa: Add ASSERTED to assert-only local variable
2275 - subprojects: Use depth=1 in the git wrap files
2276 - blake3: fix Windows ARM64 build and detect ARM64EC as ARM64
2277 - ci/windows: Disable zlib in LLVM
2278 - ci/windows: Specify numpy < 2.0 to prevent breaking changes
2279 - microsoft/clc: Split struct copies before vars_to_ssa in pre-inline optimizations
2280 - meson: Add an error message for llvmpipe without llvm draw support
2284 - Revert "meson: Do not require libdrm for DRI2 on hurd"
2285 - Revert "meson: fix with_dri2 definition for GNU Hurd"
2286 - meson: egl: Build egl_dri2 driver even for plain DRI
2290 - intel/isl: Allow multi-sample on depth aux usage (xe2)
2291 - isl: Add a heading 4KB to MCS surface (xe2)
2292 - isl: Add AUX MCS encoding into aux modes (xe2)
2293 - blorp: Scaledown rectangle of MSAA fast clear (xe2)
2294 - blorp: Fix offset when ambiguating MCS buffer (xe2)
2295 - isl: Clone from isl_gfx12.* files (xe2)
2296 - isl: Update isl_gfx20 code (xe2)
2297 - isl: Add isl_gfx20 into build (xe2)
2298 - isl: Add dispatching in isl.c (xe2)
2299 - isl: Implement a part of WA_22018390030 (xe2)
2300 - isl: Remove code for Xe2 from isl_gfx12.c
2301 - isl: Update render CMF mapping (xe2)
2302 - isl: Don't set clear values or their address (xe2)
2303 - blorp: Get fast clear rectangle of non-MSAA surfaces (xe2)
2304 - blorp: Pass down fast clear color value (xe2)
2305 - intel/genxml,blorp,common: Update 3DSTATE_PS command (xe2)
2306 - iris: Update aux state for color fast clears (xe2)
2307 - iris: Limit FCV_CCS_E to platforms that enable it
2308 - anv: Don't enable compression with modifiers (xe2)
2309 - iris: Add more restrictions on compression (Xe2)
2310 - anv: Don't enable compression on external bos (xe2)
2311 - iris: Disable PAT-based compression on depth surfaces (xe2)
2312 - anv: Disable PAT-based compression on depth images (xe2)
2313 - iris: Update synchronization of fast clear (xe2)
2314 - iris: Workaround: Don't allocate compressed bo from cache (xe2)
2315 - isl: Remove restriction of CCS_E support on formats (xe2)
2316 - blorp: Don't convert ccs_e formats for copy (xe2)
2317 - isl: Initialize the last usage in isl_encode_aux_mode[] (xe2)
2318 - anv: Update synchronization of fast clear (xe2)
2319 - iris: Disable predraw resolve (xe2)
2320 - blorp: Ensure MSAA fast clear in correct modes (xe2)
2321 - intel/dev: Select a compressed PAT entry (xe2)
2322 - isl: Add some formats not covered in CMF table (xe2)
2323 - anv: Disable tracking fast clear and aux state (xe2)
2324 - anv: Fix Vulkan CTS failure related to MCS (xe2)
2325 - anv: Support arbitrary fast-clear value on all layouts (xe2)
2326 - anv: Disable tracking of clear color on color attachment
2327 - intel/common: Ensure SIMD16 for fast-clear kernel (xe2)
2328 - intel/common: Remove blank lines in intel_set_ps_dispatch_state() (xe2)
2329 - anv: Fix assertion failures on BMG (xe2)
2330 - iris: Fix an assertion failure with compressed format
2331 - anv: Disable compression on legacy modifiers (xe2)
2332 - anv: Disable legacy CCS setup in binding (xe2)
2336 - ir3: simplify cat5 parsing
2337 - ir3: add encoding for isam.v
2338 - ir3: use isam.v for multi-component SSBO loads
2339 - ir3: add encoding of ldib/stib offsets
2340 - ir3: lower SSBO access imm offsets
2341 - nir/opt_offsets: add callback for max base offset
2342 - nir/opt_offsets: add option to allow offset wrapping
2343 - nir/opt_offsets: add load/store_ssbo_ir3
2344 - ir3: use nir_opt_offsets for SSBO accesses
2345 - ir3: optimize SSBO offset shifts for nir_opt_offsets
2346 - ir3: remove spilled splits in shared RA
2347 - ir3: set wrmask for spilled splits in shared RA
2348 - ir3: print sharedness/halfness of merge set regs
2349 - ir3: print intervals when dumping merge sets
2350 - ir3: print dst_offset of spill.macro
2351 - ir3: debug print limit pressure and post-spill max pressure
2352 - ir3: set current instruction before all validation asserts
2353 - ir3: fix crash in try_evict_regs with src reg
2354 - ir3: fix handling of early clobbers in calc_min_limit_pressure
2355 - ir3: set offset on splits created while spilling
2356 - ir3: correctly set wrmask for reload.macro
2357 - ir3: don't remove intervals for non-killed tex prefetch sources
2358 - ir3: don't remove collects early while spilling
2359 - ir3: expose instruction indexing helper for merge sets
2360 - ir3: make indexing instructions optional in ir3_merge_regs
2361 - ir3: index instructions before fixing up merge sets after spilling
2362 - ir3: move liveness recalculation inside ir3_ra_shared
2363 - ir3: restore interval_offset after liveness recalculation in shared RA
2364 - ir3: add ir3_cursor/ir3_builder helpers
2365 - ir3: refactor ir3_spill.c to use the ir3_cursor/ir3_builder API
2366 - ir3: only add live-in phis for top-level intervals while spilling
2367 - ir3: print rounding mode for cov
2368 - ir3: set rounding mode for all floating point conversions
2372 - blorp: Update programming for XY_FAST_COLOR_BLT on xe2
2373 - intel/genxml: Add XY_FAST_COLOR_BLT for xe2
2374 - intel/genxml: Update 3DSTATE_BTD for xe2
2375 - intel/dev: Allow setting FORCE_PROBE for intel PCI IDs
2376 - intel/dev: Support INTEL_FORCE_PROBE env-var
2377 - docs: Document INTEL_FORCE_PROBE env-var
2378 - intel/dev: Add LNL device info
2379 - pci_ids/intel: Add LNL PCI IDs (with FORCE_PROBE set)
2380 - anv/grl: Set INTEL_FORCE_PROBE=* when running intel_clc
2381 - intel/brw: Simplify enabling brw_fs_test_dispatch_packing
2382 - intel/brw: Allow xe2 in brw_stage_has_packed_dispatch()
2383 - intel/brw: Fix SSBO/shared load offset register size for Xe2
2384 - anv/grl: Build for xe2
2385 - Revert "anv: Disable Ray Tracing on xe2 until our compiler supports Xe2 RT"
2386 - intel/dev/mesa_defs.json: Update LNL WA entries
2387 - intel/dev: Add INTEL_PLATFORM_BMG enum, BMG WA info
2388 - intel/dev: Add BMG device info
2389 - intel/dev: Add BMG PCI IDs (with FORCE_PROBE set)
2390 - intel/dev: Silence INTEL_FORCE_PROBE warning for intel_clc
2391 - intel/dev: If building the driver, always allow getting device info
2392 - Revert "anv/grl: Set INTEL_FORCE_PROBE=* when running intel_clc"
2393 - intel/compiler: Don't set size written in brw_lower_logical_sends.cpp
2394 - intel/tools: Fix intel_dev_info --hwconfig switch
2395 - isl: Move isl_get_render_compression_format in isl_genX_helpers.h
2396 - isl: Implement isl_get_render_compression_format for xe2
2397 - intel/brw: Retype some regs to BRW_TYPE_UD for Xe2 indirect accesses
2398 - intel/perf/xe: Fix free pointer location in xe_add_config()
2399 - intel/dev: Enable LNL PCI IDs without INTEL_FORCE_PROBE
2400 - anv/generated_indirect_draws: Adjust xe2 simd32 sends_count_expectation
2401 - intel/dev: Disable LNL PCI IDs on Mesa 24.2 (require INTEL_FORCE_PROBE)
2402 - intel/brw/validate: Simplify grf span validation check by not using a mask
2403 - intel/brw/validate: Update dst grf crossing check for Xe2
2404 - intel/brw/validate: Convert access mask to be grf based
2408 - Revert "ci: mark microsoft farm as offline"
2412 - v3d: fix CLE MMU errors avoiding using last bytes of CL BOs.
2413 - v3dv: fix CLE MMU errors avoiding using last bytes of CL BOs.
2414 - v3d: Increase alignment to 16k on CL BO on RPi5
2415 - v3dv: Increase alignment to 16k on CL BO on RPi5
2416 - v3dv: V3D_CL_MAX_INSTR_SIZE bytes in last CL instruction not needed
2417 - v3dv/ci: Add more dEQP-VK subgroups that are currently skipped
2418 - v3dv: Emit stencil draw clear if needed for GFXH-1461
2419 - v3dv: really fix CLE MMU errors on 7.1HW Rpi5
2420 - v3d: really fix CLE MMU errors on 7.1HW Rpi5
2424 - radv: Fix crash when using SQTT and NO_COMPUTE
2425 - radv: Add \`RADV_PROFILE_PSTATE` envvar
2426 - radv: Fix shader mask for SQ_WGP SPM counters
2430 - meson: Update proc_macro2 meson.build patch
2431 - llvmpipe: Init eglQueryDmaBufModifiersEXT num_modifiers
2435 - intel/perf: Nuke platform_supported
2436 - intel/perf: Remove i915_drm.h include from gen_perf.py
2437 - intel/perf: Fix the error check of i915_add_config()
2438 - intel/perf: Change oa_format to uint64_t
2439 - intel/perf: Store pointer intel_device_info to in intel_perf_config
2440 - intel/perf: Add intel_perf_free()
2441 - intel/perf: Add intel_perf_free_context()
2442 - intel/ds: Free perf config and context
2443 - intel/ds: Nuke ralloc_ctx and ralloc_cfg
2444 - anv: Free intel_perf_config when destroying physical device
2445 - hasvk: Free intel_perf_config when destroying physical device
2446 - iris: Free intel_perf_config and intel_perf_context
2447 - crocus: Free intel_perf_config and intel_perf_context
2448 - intel/perf: Add and use a function to return platform OA format
2449 - intel/perf: Add function to open perf stream
2450 - intel/perf: Fix return of read_oa_samples_until()
2451 - anv: Nuke perf_query_pass from anv_execbuf
2452 - intel/perf: Replace I915_OA_FORMAT_* usage by platform check
2453 - intel/perf: Move code that will be shared by both KMDs
2454 - intel/perf: Move i915 specific code from common code
2455 - intel/perf: Move i915 specific code to load configurations to i915 file
2456 - intel/perf: Allocate sseu in heap memory
2457 - intel/perf: Replace drm_i915_perf_record_header by intel_perf_record_header
2458 - intel/perf: Add a macro with header + sample length
2459 - intel/perf: Add intel_perf_stream_read_samples()
2460 - intel/dev: Add LNL stepping mapping
2461 - intel/dev: Add BMG stepping mapping
2462 - intel: Move slm functions from brw_compiler.h to intel_compute_slm.c/h
2463 - intel/common: Implement Xe2 SLM encode
2464 - intel/common: Implement preferred SLM encode
2465 - intel/dev: Use topology variables to calculate strides in Xe KMD
2466 - intel/dev: Add function to get the number of EUs per subslice
2467 - intel: Set preferred SLM allocation size >= than SLM size for Xe2
2468 - anv: Set maxComputeSharedMemorySize value for Xe2 platforms
2469 - intel: Compute the optimal preferred SLM size per subslice
2470 - anv: Initialize variable to fix static analyzer warning
2471 - intel/genxml/gfx20: Sync POSTSYNC_DATA struct with spec
2472 - anv/xe2: Enable compute walker and BTD thread preemption
2473 - anv/xe2: Add STATE_COMPUTE_MODE individual masks
2474 - anv: Remove block promoting non CPU mapped bos to coherent
2475 - intel/isl: Set dummy_aux_address to implement Wa_14019708328
2476 - anv: Implement Wa_14019708328
2477 - iris: Implement Wa_14019708328
2478 - anv: Implement Wa_14019857787
2479 - iris: Implement Wa_14019857787
2480 - intel/dev: Add compressed PAT entry
2481 - anv: Add support for compressed images allocation in Xe2
2482 - anv: Give apps the choice of compressed or uncompressed but cpu visible images
2483 - iris: Add support for compressed images allocation in Xe2
2484 - anv: Fix assert in xe_gem_create()
2485 - intel/perf: Change order of if blocks
2486 - intel/perf: Add assert to check if allocated enough query fiels
2487 - intel/dev: Add engine_class_supported_count to intel_device_info
2488 - intel/perf: Add LNL OA XML
2489 - intel/perf: Add INTEL_PERF_QUERY_FIELD_TYPE_SRM_OA_PEC
2490 - intel/perf: Extend intel_perf_query_result_read_gt_frequency() to gfx 20
2491 - intel: Sync xe_drm.h
2492 - intel/perf: Implement function that returns OA format for Xe KMD
2493 - intel/perf: Add function to check if OA/perf is supported by Xe KMD
2494 - intel/perf: Replace i915_perf_version and i915_query_supported by a feature bitmask
2495 - intel/perf: Refactor and add Xe KMD support to add and remove configs
2496 - intel/perf: Add Xe KMD perf stream open function
2497 - intel/perf: Refactor and add Xe KMD support to enable and disable perf stream
2498 - intel/perf: Refactor and add Xe KMD support to change stream metrics id
2499 - tool/pps: Add Xe KMD support
2500 - intel/perf: Remove i915_drm.h includes from common code
2501 - intel/perf: Implement Xe KMD perf stream read
2502 - anv: Implement Xe KMD query pools
2503 - intel/perf: Enable perf on Xe KMD
2504 - intel/perf: Implement intel_perf_query_result_accumulate() for gfx 20+
2505 - intel/perf: Add support for LNL OA sample format size
2506 - intel/perf: Return LNL OA sample format
2507 - intel/perf: Do not add INTEL_PERF_QUERY_FIELD_TYPE_SRM_OA_PEC
2508 - intel/perf: Adjust EU count for Xe2+
2509 - intel/dev: Replace intel_device_info::apply_hwconfig by a gfx version check
2510 - intel: Rename XE_PERF to XE_OBSERVATION
2511 - anv: Fix return of PAT index for compressed bos for discrete GPUs
2512 - intel/dev: Drop DG1 PAT entries
2513 - intel/dev: Add documentation about intel_device_info_pat_entry::mmap
2514 - intel/dev: Drop coherency from intel_device_info_pat_entry
2515 - intel/dev: Add comment documenting the PAT entries
2516 - intel/dev: Use GPU WB PAT for Xe2 writecombining
2517 - intel/dev: Drop writeback_incoherent from Xe2
2518 - isl: Fix Xe2 protected mask
2519 - anv: Propagate protected information to blorp_batch_isl_copy_usage()
2520 - intel: Sync xe_drm.h
2521 - intel/dev: Support new topology type with SIMD16 EUs
2525 - vc4/ci: update results
2526 - vc4/v3d/ci: update expected list
2527 - vc4: set src type on storing sample mask
2528 - broadcom/compiler: remove unused parameters in vpm read
2529 - broadcom/compiler: do not run lowering I/O for FS
2530 - v3d/vc4/ci: set full renderer version check
2531 - nir,v3d: rename tlb_color_v3d intrinsic
2532 - vc4: use tlb_color_brcm intrinsic
2533 - .gitignore: add .cache folder
2534 - vc4: use IO semantics for location
2535 - v3d: use BITSET for the masks
2536 - v3d: remove handled cases for devices <= 42
2537 - ci: define SNMP base interface on runner
2538 - v3d: use screen name in disk cache
2539 - v3d,v3dv: add compatibility revision in GPU name
2540 - broadcom/ci: update expected results
2541 - v3dv/ci: add expected failure
2542 - v3dv/ci: fix spurious line in expected
2543 - v3dv/ci: add new timeouts
2544 - dri: cast constant to uint for bitshift
2545 - util: do not access member of a NULL structure
2546 - util: use unsigned types when performing bitshift
2547 - vulkan: do not access member of a NULL structure
2548 - nir: fix overflow when negating maxint in constant expressions
2549 - nir: use unsigned types when performing bitshifting
2550 - glsl: fix downcasting addresses to wrong object types
2551 - egl: do not access member of a NULL structure
2552 - mesa: use unsigned types when performing bitshifting
2553 - mesa: do not pass NULL pointer to function not expecting NULLs
2554 - ci: disable Igalia farm
2555 - broadcom/compiler: use unsigned types when performing bitshifting
2556 - v3dv: do not access member of a NULL structure
2557 - v3dv: do not pass NULL pointer to function not expecting NULLs
2558 - v3dv: restrict to channels when encoding border color
2559 - v3dv: fix misalignment in descriptor layout structure
2560 - v3d: do not access member of a NULL structure
2561 - v3d: do not pass NULL pointer to function not expecting NULLs
2562 - vc4: use unsigned types when performing bitshifting
2563 - vc4: do not access member of a NULL structure
2564 - vc4: do not pass NULL pointer to function not expecting NULLs
2565 - vc4: do not create 0-bytes variable length arrays
2566 - vc4: fix out-of-bounds access to array
2567 - Revert "ci: disable Igalia farm"
2568 - v3d: use original enabled_mask on setting vertex buffers
2569 - broadcom/ci: read 32-bit kernel from arm32 path
2570 - broadcom/ci: remove arch from hardware name
2571 - vc4/ci: run tests in 64-bits
2572 - broadcom/ci: run some GL tests in arm32 arch
2573 - broadcom/qpu: clean all versions not supported
2574 - broadcom: follow version naming convention
2575 - broadcom/ci: add more jobs to test with rpi5
2576 - broadcom/ci: update traces for rpi4
2577 - v3d/ci: update expected list
2578 - v3dv: adversise VK_EXT_depth_clamp_zero_one
2579 - v3d: expose ARB_depth_clamp in V3D 7.x
2580 - v3dv: free temp image created when copying with blit
2581 - v3dv: don't leak cache key
2585 - egl/wayland: ignore unsupported driver configs
2589 - venus: refactor out image requirements helpers
2590 - venus: extend image cache to vkGetDeviceImageMemoryRequirements
2591 - sync protocol for VkRingPriorityInfoMESA
2592 - venus: forward nice priority when creating ring
2593 - zink: disable cpu_storage for PIPE_USAGE_STREAM
2594 - venus: add missing sTypes for vk_set_physical_device_properties_struct
2595 - venus: sync protocol for conditionally ignored dyn arrays
2596 - anv/android: set ANV_BO_ALLOC_EXTERNAL for imported AHW
2600 - ir3: Don't set saturation on \`flat.b`
2601 - zink: Add missing currentExtent special value handling
2602 - turnip: Remove workaround for CTS bug zero-sized inline uniform block
2603 - mailmap: Add Karmjit Mahil
2604 - freedreno/isa: Fix isaspec map for a3xx-ld
2605 - tu: Set \`TU_ACCESS_CCHE_READ` for transfer ops with read access
2609 - nir: add SYSTEM_VALUE_BASE_WORKGROUP_ID
2610 - nir/divergence_analysis: handle load_base_global_invocation_id
2611 - intel/compiler: lower workgoup id to index only for mesh shaders
2612 - v3d: call nir_lower_compute_system_values to get rid of base intrinsics
2613 - lavapipe: lower base_workgroup_id to zero
2614 - mesa/st: lower base invoc and workgroup id
2615 - nir: remove global_invocation_id_zero_base
2616 - nir: remove workgroup_id_zero_base
2617 - nir: document base_global_invocation_id and base_workgroup_id
2618 - core/kernel: skip validating unique kernel signatures
2619 - rusticl/program: Arc the stored KernelInfo
2620 - rust/program: remove Program::kernels
2621 - nouveau: fix potential double-free in nouveau_drm_screen_create
2622 - nir: fix nir_shader_get_function_for_name for functions without names.
2623 - rusticl: use stream uploader for cb0 if prefered
2624 - rusticl/kernel: properly handle grid and offsets being usize
2625 - rusticl: lower huge grids
2626 - rusticl: add RUSTICL_MAX_WORK_GROUPS
2627 - rusticl/event: use Weak refs for dependencies
2628 - rusticl/icd: remove CLObject
2629 - rusticl/spirv: enable more caps
2630 - Revert "rusticl/event: use Weak refs for dependencies"
2631 - event: break long dependency chains on drop
2632 - rusticl/device: add DeviceCaps and move timestamp stuff into it
2633 - rusticl/device/caps: move enough for has_images
2634 - rusticl/device: properly handle devices with no support for images
2635 - rusticl/mesa/context: flush context before destruction
2636 - rusticl: merge rusticl_nir and rusticl_mesa_bindings_inline_wrapper targets
2637 - rusticl: move mesa_version_string out of the inline wrapper
2638 - rusticl: bump bindgen req to 0.65
2639 - rusticl: bump meson req to 1.4
2640 - rusticl: make use of new \`output_inline_wrapper` meson.rust.bindgen feature
2641 - nir/lower_cl_images: set binding also for samplers
2642 - nouveau: import nvif/ioctl.h file from libdrm_nouveau
2643 - gallium/vl: stub vl_video_buffer_create_as_resource
2644 - gallium/vl: remove stubs which are defined in mesa_util
2645 - meson: centralize galliumvl_stub handling
2646 - rusticl: link against libgalliumvl_stub
2647 - wgl: link against libgalliumvl_stub
2648 - gallium/drivers: do not link against libgalliumvl directly
2649 - rusticl/event: fix deadlock when calling clGetEventProfilingInfo inside callbacks
2650 - iris: fix PIPE_RESOURCE_PARAM_STRIDE for buffers
2651 - rusticl/icd: make sure returned function pointers are of the right type
2652 - rusticl/kernel/launch: fix mapping usize types to GPU pointer sizes
2653 - rusticl/kernel/launch: remove useless upload of the input
2654 - rusticl/kernel: move most of the code in launch inside the closure
2655 - rusticl/kernel/launch: move allocation of resources vec
2656 - rusticl/kernel/launch: rework how the printf buffer is allocated
2657 - rusticl/kernel/launch: get rid of Arc clones for global resources
2658 - rusticl/kernel/launch: add helper to bind global buffers
2659 - broadcom/compiler: handle load_workgroup_size
2660 - v3d: add support for load_workgroup_size
2661 - rusticl/spirv: do not pass a NULL pointer to slice::from_raw_parts
2662 - rusticl/memory: copies might overlap for host ptrs
2663 - gallium: reduce pipe_resource.usage to 4 bits
2664 - gallium: properly type pipe_resource.usage with the enum
2665 - gallium: properly type fields of pipe_resource.usage
2666 - nir_lower_mem_access_bit_sizes: support unaligned store_scratch
2667 - nir: add global_atomic_2x32 variants to nir_get_io_offset_src_number
2668 - broadcom/compiler: support global load/store intrinsics
2669 - broadcom/compiler: use nir_lower_mem_access_bit_sizes for memory lowering
2670 - broadcom/compiler: convert 2x32 global operations to scalar variants
2671 - broadcom/compiler: only handle load_uniform explicitly in v3d_nir_lower_load_store_bitsize
2672 - broadcom/compiler: rework scratch lowering
2673 - rusticl/meson: add build root dir to the include dirs of rusticl_c
2674 - rusticl: depend on the spirv_info target
2675 - util/u_printf: properly handle %%
2676 - rusticl/memory: assume minimum image_height of 1
2677 - rusticl/memory: fix clFillImage for buffer images
2678 - rusticl: add new CL_INVALID_BUFFER_SIZE condition for clCreateBuffer
2679 - rusticl: add bsymbolic to linker flags
2680 - rusticl/icd: rename all entry points to the actual correct name
2681 - radeonsi: set bo_size for user memory allocations
2682 - rusticl/queue: gracefully stop the worker thread
2683 - rusticl/queue: run rustfmt
2684 - nir/lower_alu: support 8 and 16 bit bit_count
2685 - nir/opt_sink: add load_kernel_input
2686 - gallium: add PIPE_CAP_TEXTURE_SAMPLER_INDEPENDENT
2687 - rusticl/device: require PIPE_CAP_TEXTURE_SAMPLER_INDEPENDENT for image support
2688 - rusticl/mesa/context: handle clear_buffer not set by driver
2689 - rusticl/mesa/screen: handle get_timestamp not set by driver
2690 - rusticl/kernel/launch: fix global work offsets for 32 bit archs again
2691 - broadcom/compiler: add generated v3d_nir_lower_algebraic
2692 - broadcom/compiler: handle fp16 conversion ops
2693 - broadcom/compiler: fix iu2f32 for 8 and 16 bit inputs
2694 - broadcom/compiler: try handling 8/16 bit alu operations
2695 - broadcom/compiler: handle up to vec16 load_uniforms
2696 - broadcom/compiler: abort on unknown intrinsics
2697 - broadcom/compiler: implement load_kernel_input
2698 - broadcom/compiler: call nir_lower_64bit_phis
2699 - broadcom/compiler: handle variable shared memory
2700 - v3d: implement gallium APIs for OpenCL support
2701 - v3d: treat SHADER_KERNEL as SHADER_COMPUTE
2702 - v3d: lower CL alus
2703 - v3d: lower 64 bit ALUs
2704 - v3d: support variable shared memory
2705 - v3d: fix MAX_GLOBAL_SIZE and MAX_MEM_ALLOC_SIZE
2706 - v3d: never replace a mapped bo
2707 - rusticl: enable v3d
2708 - nir/schedule: add write dep also for shared_atomic
2709 - meson: rename with_gallium_opencl to with_gallium_clover
2710 - rusticl/program: move binary parsing into its own function
2711 - rusticl/program: make binary API not crash on errors
2712 - rusticl/program: use blob.h to parse binaries
2713 - rusticl/program: update binary format
2714 - rusticl/program: use default in more places
2715 - Revert "rusticl/queue: run rustfmt"
2716 - Revert "rusticl/queue: gracefully stop the worker thread"
2717 - rusticl/buffer: harden bound checks against overflows
2718 - rusticl/context: move SVM pointer tracking into own type
2719 - rusticl/ptr: add a few APIs to TrackedPointers
2720 - rusticl/memory: complete rework on how mapping is implemented
2721 - rusticl: remove unused interfaces to simplify code
2722 - rusticl/mesa: remove ResourceType::Cb0
2723 - rusticl/memory: optimize sw_copy when the row_pitch matches the height
2724 - rusticl/mesa: make PipeResource repr(transparent)
2725 - v3d: support unnormalized coords
2726 - rusticl/spirv: support more caps
2727 - rusticl/device: fix image_3d_write_supported for embedded
2728 - rusticl/device: turn image_3d_write_supported into a cap
2729 - rusticl/device: fix advertizement of 3d write images support
2730 - rusticl: require PIPE_CAP_IMAGE_STORE_FORMATTED for image support.
2731 - rusticl/event: make set_status handle error status properly
2732 - rusticl/queue: do not overwrite event error states
2733 - rusticl/queue: properly check all dependencies for an error status
2734 - rusticl/event: properly implement CL_EXEC_STATUS_ERROR_FOR_EVENTS_IN_WAIT_LIST
2735 - rusticl/queue: properly implement in-order queue error checking
2736 - rusticl/event: return execution errors when doing a blocking enqueue
2737 - rusticl/mesa: handle failures with u_upload_data
2738 - rusticl/mesa: set take_ownership to true in set_constant_buffer_stream
2739 - rusticl/event: fix outdated comment in call
2740 - rusticl/queue: format file
2741 - zink: fix OpenCL read_write images
2742 - rusticl: support read_write images
2743 - spirv: generate info for FunctionParameterAttribute
2744 - spirv: initial parsing of function parameter decorations
2745 - spirv: handle function parameters passed by value
2746 - nak: allow clippy::not_unsafe_ptr_arg_deref lints
2747 - clc: force linking of spirvs with mismatching pointer types in signatures
2748 - rusticl: fix clippy lint having bounds defined in multiple places
2749 - rusticl/program: protect against 0 length in slice::from_raw_parts
2750 - rusticl/api: protect against 0 length in slice::from_raw_parts
2751 - rusticl/spirv: protect against 0 length in slice::from_raw_parts
2752 - nouveau: handle realloc failure inside cli_kref_set
2753 - mesa: check for enabled extensions for \*UID enums
2754 - nouveau/winsys: fix handling of NV_DEVICE_TYPE_IGP
2755 - nouveau: use nv_devince_info and fill in PCI and type information
2756 - nouveau: add nv_device_uuid
2757 - nouveau: implement driver_uuid and device_uuid
2758 - nvk: use nv_device_uuid
2759 - zink: lower 64 bit find_lsb, ufind_msb and bit_count
2760 - zink: lower 8/16 bit alu ops vk spirv doesn't allow
2761 - rusticl/kernel: properly respect device thread limits per dimension
2762 - rusticl/memory: Fix memory unmaps after rework
2763 - rusticl/image: take pitches into account when allocating memory for maps
2764 - rusticl/image: properly sync mappings content for 1Dbuffer images
2765 - rusticl/queue: add clSetCommandQueueProperty
2766 - util/u_printf: do not double print format string with unused arugments
2767 - rusticl/memory: fix sampler argument size check
2771 - isl: Set MOCS to uncached for Gfx12.0 blitter sources/destinations
2772 - intel/brw: Delete gfx10 table for align1 3src type encoding
2773 - intel/brw: Drop NF type support
2774 - intel/brw: Rework BRW_REGISTER_TYPE's representation semantics
2775 - intel/brw: Stop using long BRW_REGISTER_TYPE enum names
2776 - intel/brw: Reindent after shortening BRW_REGISTER_TYPE_* to BRW_TYPE_*
2777 - intel/brw: Use newer brw_type_is_* shorter names
2778 - intel/brw: Replace brw_reg_type_from_bit_size by brw_type_with_size
2779 - intel/brw: Replace type_sz and brw_reg_type_to_size with brw_type_size_*
2780 - intel/brw: Combine a1/a16 3src type encoding functions
2781 - intel/brw: Combine a1/a16 3src type decoding functions
2782 - intel/brw: Rename brw_reg_type_to_hw_type to brw_type_encode
2783 - intel/brw: Don't use inst return value when it isn't needed
2784 - intel/brw: Make a helper for finding the largest of two types
2785 - intel/brw: Add builder helpers for math functions
2786 - intel/brw: Add builder helpers that allocate temporary destinations
2787 - intel/brw: Use new builder helpers that allocate a VGRF destination
2788 - intel/brw: Print W/UW immediates correctly
2789 - intel/brw: Do not create empty basic blocks when removing instructions
2790 - intel/brw: Support CSE on more ops
2791 - intel/brw: Don't include unnecessary undefined values in texture results
2792 - intel/brw: Add a new VEC() helper.
2793 - intel/brw: Use VEC for load_const
2794 - intel/brw: Use VEC for gl_FragCoord
2795 - intel/brw: Use VEC for TCS/TES/GS input/output loads
2796 - intel/brw: Use VEC for FS outputs
2797 - intel/brw: Use VEC for output stores
2798 - intel/brw: Use VEC for NIR vec*() sources
2799 - intel/brw: Use VEC for emit_unzip()
2800 - intel/nir: Set src_type on TCS quads workaround store_output
2801 - intel/brw: Blockify convergent load_shared on Gfx11-12 as well
2802 - intel/brw: Recreate GS output registers after EmitVertex
2803 - intel/brw: Skip fs_nir_setup_outputs for compute shaders
2804 - intel/brw: Handle scratch address swizzling of constants
2805 - intel/brw: Add a idom_tree::dominates(a, b) helper.
2806 - intel/brw: Make brw_reg::bits publicly accessible from fs_reg
2807 - intel/brw: Update instructions_match() to compare more fields
2808 - intel/brw: Drop compiler parameter from try_constant_propagate()
2809 - intel/brw: Drop BRW_OPCODE_IF from try_constant_propagate
2810 - intel/brw: Refactor try_constant_propagate()
2811 - intel/brw: Refactor code to commute immediates into legal positions
2812 - intel/brw: Delete SAD2 and SADA2 opcodes
2813 - intel/brw: Make VEC() perform a single write to its destination.
2814 - intel/brw: Make gl_SubgroupInvocation lane index loading SSA
2815 - intel/brw: Skip LOAD_PAYLOADs after every texture instruction if possible
2816 - intel/brw: Add a new def analysis pass
2817 - intel/brw: Print defs in dump_instructions
2818 - intel/brw: Write a new global CSE pass that works on defs
2819 - intel/brw: Switch to the new defs-based global CSE pass
2820 - intel/brw: Delete old local common subexpression elimination pass
2821 - intel/brw: Introduce a new SSA-based copy propagation pass
2822 - intel/brw: Use the defs-based copy propagation along with the old one
2823 - intel/brw: Make opt_copy_propagation_defs clean up its own trash
2824 - intel/brw: Build the scratch header on the fly for pre-LSC systems
2825 - intel/brw: Skip discarding the interference graph
2826 - intel/brw: Delay liveness calculations in saturate propagation
2827 - intel/brw: Make an alu2 builder helper
2828 - intel/brw: Make bld.ADD(x, 0) emit no instructions and return x directly
2829 - intel/brw: Support CSE of ADD3
2830 - intel/brw: Add a lower_csel pass and allow building it for all types
2831 - intel/nir: Don't needlessly split u2f16 for nir_type_uint32
2832 - intel/brw: Don't mix types for unary extended math instructions
2833 - intel/brw: Disallow scalar byte to float conversions on DG2+
2837 - anv: Properly fetch partial results in vkGetQueryPoolResults
2838 - anv: Properly handle cases for different query types in copy_query_results_with_shader
2839 - intel/genxml: add task/mesh shader statistics registers
2840 - anv: Update pipeline statistics mask for task/mesh shader invocations
2841 - anv: implement mesh shader queries
2842 - anv: toggle meshShaderQueries based on whether we support mesh_shader or not
2846 - amd: Include missing llvm IR header Module.h
2850 - docs: Add documentation about debugging GPU hangs on RADV
2851 - ac/debug,radv: Read UMR wave dumps into memory before parsing
2852 - radv: Use a struct for the trace_bo layout
2853 - radv: Trace indirect dispatch sizes
2857 - radv: Handle all dependencies of CmdWaitEvents2
2858 - nir/print: Do not access invalid indices of load_uniform
2859 - radv: Fix radv_shader_arena_block list corruption
2860 - radv: Remove arenas from capture_replay_arena_vas
2861 - radv: Zero initialize capture replay group handles
2862 - radv/ci: Add back pipeline library flakes
2863 - radv/ci: Document recent flakes
2864 - gitlab: Reference hang debugging documenttion
2865 - radv: Remove radv_cmd_dirty_dynamic_bits
2866 - llvmpipe: Use a second LLVMContext for compiling sample functions
2867 - radv: Add locking to radv_replay_shader_arena_block
2868 - radv: Replace is_rt_shader with RADV_SHADER_TYPE_RT_PROLOG
2869 - radv: Remove uses_dynamic_rt_callable_stack
2870 - radv/rt: Track ray_launch_id reads
2871 - radv/rt: Track ray_launch_size reads
2872 - radv/rt: Remove load_rt_dynamic_callable_stack_base_amd
2873 - radv: Return a block from radv_replay_shader_arena_block
2874 - ac/llvm: Fix DENORM_FLUSH_TO_ZERO with exact instructions
2875 - ac/llvm: Enable helper invocations for vote_all/any
2876 - radv/ci: Bring back vkcts-navi21-llvm-valve
2877 - khronos-update: Add ANDROID guards to vk_android_native_buffer.h
2878 - zink: Always include renderdoc_app.h
2879 - zink: Blit using one triangle for nearest filtering
2880 - llvmpipe: Lock shader access to sample_functions
2881 - llvmpipe: Stop using a sample_functions pointer as cache key
2882 - llvmpipe: Only evict cache entries if a fence is available
2883 - lavapipe: Always call finish_fence after lvp_execute_cmd_buffer
2884 - radv: Clean up pipeline barrier handling
2885 - radv: Remove dead access bits
2886 - radv/meta: Use READ access for dst_access_flush
2887 - radv/rra: Detect BVHs with back edges
2888 - radv/rra: Move some code into handle_accel_struct_write
2889 - radv/rra: Fix disabling the ray history
2890 - radv/rra: Fix reporting the isec invocations
2891 - radv/rra: Bump rt_driver_interface_version to 8.0
2892 - radv/rra: Reduce the memory requirement of copy_after_build
2893 - radv/rra: Rework calculating the ray history size
2894 - radv/rra: Enable RADV_RRA_TRACE_COPY_AFTER_BUILD by default
2895 - util: Add a helper for querying sparse tile sizes
2896 - lavapipe: Do not allocate 0 sized buffers for descriptor sets
2897 - gallium: Add a memory range parameter to resource_bind_backing
2898 - llvmpipe: Use an anonymous file for memory allocations
2899 - lavapipe: Implement sparse buffers and images
2900 - lavapipe: Implement shaderResourceResidency
2901 - venus: Refactor hiding sparse features and properties
2902 - venus: Disable sparse binding on lavapipe
2903 - vulkan: Handle group stages in vk\_.*_access2_for_pipeline_stage_flags2
2904 - vulkan: Add vk_expand_(dst|src)_access_flags2
2905 - radv: Use vk_expand_(src|dst)_access_flags2
2906 - radv: Remove no-op access flag handling
2907 - radv: Remove handling for expanded access flags
2908 - radv: Remove write access handling from radv_dst_access_flush
2909 - radv: Handle AS access bits like shader storage access bits
2910 - radv: Refactor radv_(dst|src)_access_flush
2911 - radv: Fix smooth lines with dynamic polygon mode and topology
2912 - radv: Always use dynamic line smoothing
2913 - nir: Stop using "capture : true" for nir_opt_algebraic
2914 - nir: Add FLOAT_CONTROLS\_.*_PRESERVE
2915 - aco: print s_delay_alu INSTSKIP>3 correctly
2919 - ac/surface: add GFX12 256B tile mode for video
2920 - ac/surface/tests: add the test for ADDR3_256B_2D
2921 - radeon/vcn: use pipe video buffers for dpb
2922 - radeon/vcn: enable dpb to use pipe video buffer with swizzle mode
2926 - anv: disable dual source blending state if not used in shader
2927 - anv: reuse embedded samplers across shaders
2928 - anv: simplify multisampling check
2929 - anv: fixup indentation
2930 - anv: factor out wm_prog_data get in runtime flush
2931 - intel/brw: fixup wm_prog_data_barycentric_modes()
2932 - intel/fs: decouple alphaToCoverage from per sample dispatch
2933 - intel/brw: add min_sample_shading value in wm_prog_data
2934 - anv: track sample shading enable & min sample shading
2935 - anv: add dirty tracking of fs_msaa_flags in runtime
2936 - anv: move 3DSTATE_WM::BarycentricInterpolationMode programming to runtime
2937 - anv: move more PS_EXTRA programming to runtime
2938 - anv: move 3DSTATE_PS to partial packing
2939 - anv: move 3DSTATE_MULTISAMPLE to partial emission
2940 - anv: remove fs_msaa_flags from the graphics pipeline
2941 - anv: enable EDS3 AlphaToCoverageEnable & RasterizationSamples
2942 - anv: fixup alloc failure handling in reserved_array_pool
2943 - anv: fix leak of custom border colors
2944 - anv: avoid requirement to put flush_data as first field
2945 - anv: move device initialization as the last step of vkCreateDevice
2946 - anv: move empty_vs_input to physical device
2947 - anv: VK_EXT_legacy_vertex_attributes
2948 - docs: update anv features
2949 - anv: fix ycbcr plane indexing with indirect descriptors
2950 - intel/hang_replay: use newer API of i915 execbuffer
2951 - intel/hang_replay: use hw image param
2952 - intel/tools: add README file
2953 - brw: add more condition for reducing sampler simdness
2954 - intel: move debug identifier out of libintel_dev
2955 - brw: drop dependency on libintel_common
2956 - anv: fix push constant subgroup_id location
2957 - nir/divergence: add missing load_printf_buffer_address
2958 - nir: add a base offset for printf indexing
2959 - nir: add ptr_bit_size parameter to nir_lower_printf
2960 - nir: add a low level printf emission helper
2961 - intel/nir: remove unused prototypes
2962 - intel/nir: add reloc delta to load_reloc_const_intel intrinsic
2963 - intel/compiler: store u_printf_info in prog_data
2964 - intel/nir: add printf lowering
2965 - anv: add debug shader printf support
2966 - intel/clc: enable printfs support
2967 - anv: shader printf example
2968 - anv: switch to vk_device::mem_cache field for default cache
2969 - anv: use weak_ref mode for global pipeline caches
2970 - anv: fix shader identifier handling
2971 - intel/brw: ensure find_live_channel don't access arch register without sync
2972 - anv: fix utrace compute walker timestamp captures
2973 - anv: fix timestamp copies from secondary buffers
2974 - anv: move last compute command pointers to the state structure
2975 - u_trace: extend tracepoint end_of_pipe bit into flags
2976 - anv: optimize POSTSYNC_DATA rewrites in timestamp emissions
2977 - intel: fix HW generated local-id with indirect compute walker
2978 - brw: use a single virtual opcode to read ARF registers
2979 - brw: limit dependencies on SR register
2980 - brw: better model READ_ARF_REG opcode
2981 - anv: fix Gfx9 fast clears on srgb formats
2982 - anv: rewrite Wa_18019816803 tracking to be more like state
2983 - anv: factor out some more gpu_memcpy setup
2984 - anv: fix pipeline flag fields
2985 - anv: expose VK_MESA_image_alignment_control
2986 - anv: support setting CFE_STATE::StackIDControl per application
2987 - anv: limit aux invalidations to primary command buffers
2988 - anv: ensure completion of surface state copies before secondaries
2989 - anv: simplify TRTT initialization
2990 - anv: reuse setup_execbuf_fence_params for utrace submissions
2991 - anv: rework utrace submission
2992 - anv: move trtt submissions over to the anv_async_submit
2993 - anv: use reserved array pool for legacy custom border colors
2994 - anv: make device initialization more asynchronous
2995 - mi-builder: rename relocated api
2996 - mi-builder: c++ warning fix
2997 - mi-builder: make instruction pointer manipulation more obvious
2998 - mi-builder: add missing write completion check
2999 - mi-builder: add relocated register/memory writes
3000 - mi-builder: add a write check parameter
3001 - anv: centralize mi_builder setup
3002 - anv: use the new relocated write mi-builder api
3003 - anv: move more MI_SDI to mi_builder
3004 - anv: use default mocs for memory bits only touched by CS
3005 - anv: set query mi-builder mocs only once
3006 - anv: use new mi-builder write check API to avoid stalls
3007 - genxml: add MI_MEM_FENCE for Gfx20
3008 - mi-builder: add read/write memory fencing support on Gfx20+
3009 - intel/fs: fix lower_simd_width for MOV_INDIRECT
3010 - anv: add custom mi write fences
3011 - anv: emit conditional after gfx state flushing
3012 - anv: factor out STATE_BASE_ADDRESS filling to helper function
3013 - anv: predicate emission of STATE_BASE_ADDRESS
3014 - anv: reuse device local variable
3015 - anv: avoid initalizing TRTT stuff without sparseBinding
3016 - anv: fix vkCmdWaitEvents2 handling
3017 - anv: don't apply descriptor array bound checking
3018 - brw: add missing break
3019 - brw: factor out source extraction for rematerialization
3020 - brw: improve rematalization of surface/sampler handles
3021 - brw: bound the amount of rematerialized NIR instructions
3022 - brw: remove rematerialization assert
3023 - brw: remove some brackets
3024 - brw: enable rematerialization of non 32bit uniforms
3025 - brw: always use new registers for load address increments
3026 - brw: annotation send instructions with surface handles generated with exec_all
3027 - brw: avoid Wa_1407528679 in uniform cases
3028 - brw: blockify load_global_const_block_intel
3029 - brw: enable A64 loads source rematerialization
3030 - anv: limit vertex fetch invalidation on indirect read
3031 - anv: add a protected scratch pool
3032 - anv: prepare 2 variants of all shader instructions
3033 - anv: allocate compute scratch using the right scratch pool
3034 - anv: emit the right shader instruction for protected mode
3035 - anv: workaround flaky xfb query results on Gfx11
3036 - anv: fix u_trace on < Gfx12.0
3037 - intel/ds: remove duplicate arguments
3038 - hasvk: move cmd_emit_timestamp initialization to genX
3039 - hasvk: pass anv_address to predicate helper
3040 - brw: fix uniform rebuild of sources
3041 - anv: get rid of the second dynamic state heap
3042 - isl: account for protection in base usage checks
3043 - anv: properly flag image/imageviews for ISL protection
3044 - anv: propagate protected information for blorp operations
3045 - anv: fix check on pipeline mode to track buffer writes
3046 - vulkan/runtime: allow null/empty debug names
3047 - anv: reuse object string for RMV token
3048 - anv: add missing MEDIA_STATE_FLUSH for internal shaders
3049 - anv/blorp: force CC_VIEWPORT reallocation when programming 3DSTATE_VIEWPORT_STATE_POINTERS_CC
3050 - brw/rt: fix ray_object_(direction|origin) for closest-hit shaders
3052 Louis-Francis Ratté-Boulianne (20):
3054 - dri_interface: add interface for EGL_EXT_surface_compression
3055 - gallium: add interface for fixed-rate surface/texture compression
3056 - egl/wayland: factor out common part of DRI image creation
3057 - egl: wire up EGL_EXT_surface_compression extension
3058 - st/dri2: add support for fixed-rate compression interface
3059 - egl/dri2: add support for EGL_EXT_surface_compression
3060 - mapi: add EXT_texture_storage_compression extension
3061 - mesa/st: add compression parameter to st_texture_create
3062 - mesa: implement EXT_texture_storage_compression extension
3063 - mesa: implement EXT_EGL_image_storage_compression extension
3064 - panfrost: Add AFRC overlay in v10 xml specification
3065 - panfrost: add device querying for AFRC support
3066 - panfrost: add utils for AFRC fixed-rate support
3067 - panfrost: encode component order as an inverted swizzle (v10)
3068 - panfrost: add support for AFRC textures
3069 - panfrost: add support for AFRC render targets
3070 - panfrost: add support for AFRC modifiers
3071 - panfrost: add translation between modifier and compression rates
3072 - panfrost: add support for fixed-rate compression
3073 - panfrost: add PAN_AFRC_RATE env var to force a compression rate
3077 - loader: silence implicit-load zink error by the loader
3078 - gallium: properly propagate the usage of resource
3079 - gallium: inline trivial needs_pack()
3080 - meson: Build pipe-loader when build-tests is true
3084 - llvmpipe: query winsys support for dmabuf mapping
3085 - u_gralloc/fallback: Set fd from handle directly
3086 - egl/x11/sw: Implement swapbuffers with damage
3087 - vulkan/wsi: Update sw x11 wsi to only copy damage regions
3088 - egl/x11/sw: Implement shm support
3089 - egl/x11: Remove force software check for exporting SBWD
3090 - lp: only map dt buffer on import from dmabuf
3094 - etnaviv: drm: don't skip flush when there are active PMRs
3095 - etnaviv: always flush pending queries on get_query_result
3099 - nir: Handle texop_*_nv in nir_tex_instr_is_query
3100 - nak: Add minimum bindgen requirement
3104 - svga: Retry DRM_VMW_SYNCCPU ioctl on failure.
3105 - svga: Replace shared surface flag and simplify surface creation
3109 - intel/genxml/xe2: update MESH_CONTROL
3110 - anv,intel/compiler/xe2: fill MESH_CONTROL.VPandRTAIndexAutostripEnable
3114 - ac/gpu_info: set tcc_rb_non_coherent only if number of TCCs != number of RBs
3115 - ac/surface: disable DCC for 3D textures on gfx9 to improve performance
3116 - ac/surface: enable thick tiling for 3D textures for better perf on gfx6-8
3117 - radeonsi: don't invalidate L2 for internal compute without DCC stores
3118 - radeonsi: fix KHR-GL46.texture_lod_bias.texture_lod_bias_all on gfx10-11
3119 - radeonsi: validate IO semantics in scan_io_usage
3120 - radeonsi: add workarounds for DCC MSAA for gfx9-10
3121 - radeonsi: enable DCC for MSAA on gfx10-10.3
3122 - radeonsi: check for FMASK correctly in gfx10_get_bin_sizes
3123 - amd/ci: 17 piglit failures are fixed for raven
3124 - nir: add ACCESS_CP_GE_COHERENT_AMD
3125 - nir: add nir_atomic_op_ordered_add_gfx12_amd
3126 - nir: add streamout intrinsics for AMD GFX12
3127 - nir: add sleep intrinsics for AMD
3128 - nir: validate src_type of store_output intrinsics, require bit_size >= 16
3129 - nir: add more build helpers
3130 - nir: add shader_info::use_aco_amd
3131 - nir/lower_tex: support FMASK loads with a 16-bit sample index
3132 - nir/lower_image: support FMASK loads with a 16-bit sample index
3133 - drm-uapi: update amdgpu_drm.h and drm_fourcc.h for gfx12
3134 - amd: import gfx12 addrlib
3135 - amd: add gfx12 register definitions
3136 - amd: add gfx12 register definitions into the register header generator
3137 - amd: add initial common code for gfx12
3138 - ac/nir: update ac_nir_lower_resinfo for gfx12
3139 - ac/nir,llvm: add GS VGPR changes for gfx12
3140 - ac/llvm: use new s_wait instructions and split the existing ones for gfx12
3141 - ac/llvm: add new cache flags for gfx12
3142 - ac/llvm: add CS SGPR changes for gfx12
3143 - ac/llvm: update inline assembly for buffer_load_format_xyzw with TFE for gfx12
3144 - ac/nir: add ac_nir_sleep and handle the intrinsics
3145 - ac/nir: add gfx12 streamout NIR code
3146 - ac/llvm: handle nir_atomic_op_ordered_add_gfx12_amd
3147 - ac/llvm: implement nir_intrinsic_ordered_xfb_counter_add_gfx12_amd
3148 - ac/llvm: add a workaround for nir_intrinsic_load_constant for LLVM on gfx12
3149 - ac/surface: add gfx12
3150 - ac/surface/tests: add gfx12 tests
3151 - radeonsi: add gfx12
3152 - util: shift the mask in BITSET_TEST_RANGE_INSIDE_WORD to be relative to b
3153 - ac/llvm: improve/simplify/fix load_ssbo
3154 - radeonsi: serialize shader disassembly string to fix asm dumps for ACO
3155 - radeonsi: fix the size of the query result SSBO
3156 - radeonsi: validate the buffer range in si_set_shader_buffer
3157 - radeonsi: remove GDS tests
3158 - radeonsi: set flags directly instead of having needs_db_flush
3159 - radeonsi/gfx11: use a lighter workaround for Navi31 dEQP failures
3160 - radeonsi: get NIR options from si_screen instead of calling get_compiler_options
3161 - radeonsi: minor simplifications of clear/copy_buffer shaders
3162 - radeonsi: simplify the complex clear/copy_buffer shader
3163 - radeonsi: use set_work_size for all internal compute dispatches
3164 - radeonsi: replace the clear_12bytes_buffer shader with the DMA compute shader
3165 - radeonsi: remove slow code from si_msaa_resolve_blit_via_CB
3166 - radeonsi/ci: fix caselists for vk-gl-cts/main
3167 - radeonsi/ci: update failures for all generations
3168 - radeonsi/ci: remove some gfx11 flakes
3169 - radeonsi: constify struct pipe_vertex_buffer *
3170 - nir/algebraic: eliminate pack+unpack and unpack+pack pairs
3171 - ac: move radv_mem_vectorize_callback to common code
3172 - ac/llvm: global stores should have no holes in the writemask
3173 - radeonsi: call nir_lower_int64 later to fix ACO failure with Tomb Raider
3174 - radeonsi: vectorize load/stores and shrink stores
3175 - amd: update addrlib
3176 - amd: add more gfx11 APUs
3177 - amd: enable 32B minimum DCC block size for gfx1151
3178 - ac/llvm: fix incorrect parameter type in llvm.amdgcn.s.nop
3179 - radeonsi: vectorize loads/store after ABI lowering and optimizations
3180 - radeonsi/gfx12: fix the alpha ref value
3181 - radeonsi/gfx12: fix incorrect condition for when to do clear_buffer via compute
3182 - radeonsi/gfx12: disable CU1 instead of CU0 for GS due to SQTT
3183 - radeonsi/gfx12: fix a regression in si_set_mutable_tex_desc_fields
3184 - radeonsi/gfx12: fix depth bounds register values
3185 - radeonsi/gfx12: fix a regression in si_init_depth_surface
3186 - radeonsi: don't lower UBO/SSBOs to descriptors if they are already lowered
3187 - radeonsi: lower NIR resource srcs to descriptors last
3188 - ac/descriptors: fix gfx12 regressions
3189 - ac/nir/lower_ngg: use global_atomic_amd to fix gfx12 streamout
3190 - ac/nir/lower_ngg: use voffset in global_atomic_add for xfb
3191 - ac: add gfx12 DCC shared code
3192 - radeonsi/gfx12: fix GPU deadlocks due to query result incoherency
3193 - radeonsi: assume si_set_ring_buffer is only used by gfx6-10.3
3194 - radeonsi: remove cp_to_L2 and L2_to_cp, inline the values
3195 - radeonsi: remove RADEON_FLAG_READ_ONLY
3196 - radeonsi: allow RADEON_HEAP_BIT_GL2_BYPASS for VRAM
3197 - radeonsi: remove leftover comment of non-existent RADEON_FLAG_MALL_NOALLOC
3198 - radeonsi/gfx12: add DCC
3199 - ac/surface: pass the correct addrlib handle to Addr3GetPossibleSwizzleModes
3200 - amd: update addrlib
3201 - ac/nir/lower_ngg: don't use gfx12 xfb defs outside their basic block on gfx11
3202 - radeonsi/gfx12: fix stencil corruption
3203 - gallium/u_blitter: add option to override fragment shader for util_blitter_blit
3204 - radeonsi: don't declare 3D coordinates in the compute blit if they aren't needed
3205 - radeonsi: use better workgroup sizes for compute blits to improve perf
3206 - radeonsi: ignore PIPE_SWIZZLE_1 for 40% VGPR usage reduction for compute blits
3207 - radeonsi: remove fp16_rtz from the compute blit
3208 - radeonsi: use MIMG D16 (16-bit data) for image instructions in compute blits
3209 - radeonsi: optimize unaligned compute blits
3210 - radeonsi: fix sample0_only for the compute blit
3211 - radeonsi: reject unsupported parameters as the first thing in the compute blit
3212 - radeonsi: don't use si_can_use_compute_blit in the compute blit
3213 - radeonsi: don't fail due to DCC when using the compute blit on compute queues
3214 - radeonsi/gfx11: enable MSAA image stores in the compute blit
3215 - radeonsi: document better how X/Y flipping in the compute blit works
3216 - radeonsi: cosmetic and robustness changes for the compute blit
3217 - radeonsi: extend the compute blit to do image clears as well
3218 - radeonsi: switch compute image clears to the compute blit shader
3219 - radeonsi: rename si_compute_blit "testing" parameter to "fail_if_slow"
3220 - radeonsi: rename si_compute_copy_image -> si_compute_copy_image_old
3221 - radeonsi: add a new version of si_compute_copy_image using the compute blit
3222 - radeonsi: switch the old compute image copy to the new one using the blit
3223 - radeonsi: remove the old si_compute_copy_image
3224 - radeonsi: convert the compute blit shader hash table to u64 keys
3225 - radeonsi: split xy_clamp_to_edge to separate X and Y flags for the compute blit
3226 - radeonsi: restructure (rewrite) the compute blit shader
3227 - radeonsi: adds flags parameter into si_compute_blit to replace fail_if_slow
3228 - radeonsi: change the compute blit to clear/blit multiple pixels per lane
3229 - radeonsi: extend NIR compute helpers to allow returning 16-bit results
3230 - radeonsi: use MIMG A16 (16-bit image coordinates) in compute blits
3231 - radeonsi: print the compute shader blit key for AMD_DEBUG
3232 - radeonsi: use shader_info::use_aco_amd to determine whether to use ACO
3233 - radeonsi: add use_aco into CS blit shader key
3234 - radeonsi: clear color buffers via compute for special tiling cases
3235 - radeonsi: add a custom MSAA resolving pixel shader
3236 - radeonsi: add fail_if_slow parameter into si_msaa_resolve_blit_via_CB
3237 - radeonsi: add a new blit microbenchmark
3238 - radeonsi: add decision code to select when to use CB_RESOLVE for performance
3239 - radeonsi: add decision code to select when to use compute blit for performance
3240 - ac/nir: import the MSAA resolving pixel shader from radeonsi
3241 - ac/nir: import the universal compute clear/blit shader
3242 - ac/nir: import the dispatch logic for the universal compute clear/blit shader
3243 - Revert "radeonsi: fix initialization of occlusion query buffers for disabled RBs"
3244 - radeonsi/ci: update gfx10.3 failures
3245 - nir/lower_io_to_scalar: add new_component temporary variable
3246 - nir/lower_io_to_scalar: don't create output stores that have no effect
3247 - nir: add nir_opt_vectorize_io, vectorizing lowered IO
3248 - glsl/linker: vectorize lowered IO
3249 - nir: add a NIR option flag nir_io_prefer_scalar_fs_inputs
3250 - ac/nir/cdna: allow 16-bit coordinates
3251 - ac/nir/cdna: ignore image_descriptor intrinsics
3252 - ac/nir/cdna: don't use image_descriptor intrinsics if the src is a descriptor
3253 - mesa: switch remaining shader functions from SHA1 to BLAKE3
3254 - radeonsi: replace shader SHA1 hashes with BLAKE3
3255 - radeonsi: don't use CP DMA on GFX940
3256 - nir: rename ordered_xfb_counter_add_gfx12_amd -> ordered_add_loop_gfx12_amd
3257 - ac/nir: remove sleeps from gfx12 streamout code
3258 - ac/llvm: remove s_nop from ordered_add_loop_gfx12_amd
3259 - ac/llvm: fix inline assembly register constraints for ordered_add_loop_gfx12_amd
3260 - as/llvm: add s_nops before the ordered add loop and s_wait_alu workaround
3261 - radeonsi: implement nir_intrinsic_load_ssbo_address
3262 - radeonsi: expose internal buffer bindings to compute shaders
3263 - radeonsi/gfx12: always set BO metadata, not just during export
3264 - radeonsi/gfx12: fix compute register settings for global_atomic_ordered_add
3265 - ac/surface: finish display DCC for gfx11.5
3266 - ac/surface: finish display DCC for gfx12
3267 - radeonsi: add fail_if_slow parameter into compute_clear/copy_buffer
3268 - radeonsi: use a hash_table and define a shader key for the DMA compute shader
3269 - radeonsi: add dwords_per_thread parameter into si_compute_clear_copy_buffer
3270 - radeonsi: clear buffers with a 12B clear value by clearing 4 dwords per thread
3271 - radeonsi: rewrite the clear/copy_buffer microbenchmark
3272 - radeonsi/ci: update gfx11 failures
3273 - radeonsi: replace si_shader::scratch_bo with scratch_va, don't set it on gfx11+
3274 - radeonsi: don't update compute scratch if the compute shader doesn't use it
3275 - ac: add radeon_info::has_scratch_base_registers
3276 - radeonsi: lock a mutex when updating scratch_va for compute shaders
3277 - util: make util_idalloc_exists private
3278 - util: don't use variable names that can appear in args of idalloc foreach macros
3279 - util: add util_idalloc_sparse, solving the excessive virtual memory usage
3280 - mesa: switch ID allocation to util_idalloc_sparse to reduce virtual memory usage
3281 - nir/opt_algebraic: use fmulz for fpow lowering to fix incorrect rendering
3282 - radeonsi/gfx12: fix a GPU hang due to an invalid packet with window rectangles
3283 - radeonsi: ensure TC_L2_dirty is set if we don't sync after internal SSBO blits
3284 - radeonsi: fix buffer coherency issues on gfx6-8,12 due to missing PFP->ME sync
3285 - radeonsi/gfx12: fix register programming to fix GPU hangs
3286 - radeonsi/gfx12: fix VS output corruption with streamout
3287 - ac/surface/gfx12: turn off HiZ for pre-production samples
3291 - gallivm: Fix compilation errors when using LLVM 13.
3295 - vdrm: Add fixed VA parameter for mapping memory
3296 - tu: Handle VkDeviceMemory BO unmapping in VkUnmapMemory
3297 - tu: Implement VK_EXT_map_memory_placed
3298 - docs/features: Add VK_EXT_map_memory_placed
3299 - tu/shader: Allow LRZ when write pos with explicit early frag test
3300 - tu/lrz: Emit GRAS_LRZ_CNTL2 on A7XX
3301 - tu/lrz: Use actual CHIP rather than hardcoding A6XX
3302 - fd/a7xx: Initialize magic register 8C34 to 0
3303 - fd/a7xx: Initialize magic register 8008 to 0
3304 - tu: Allow LRZ on A7XX
3305 - tu/lrz: Add structure for LRZ FC layout
3306 - tu: Update LRZ FC allocation for A7XX layout
3307 - tu: Update LRZ FC dirty clear for A7XX
3308 - tu: Specify LRZ FC depth clear value on A7XX
3309 - tu: Enable LRZ fast-clear for A7XX
3310 - fd/a7xx: Document \`LRZ_FLIP_BUFFER` event
3311 - docs/freedreno: Add documentation on A7XX LRZ
3312 - tu: Emit GRAS_LRZ_DEPTH_BUFFER_INFO correctly
3313 - tu/kgsl: Spin unti KGSL reports queue timestamp during profiling
3314 - tu/kgsl: Fix profiling buffer GPU IOVA
3315 - fd/meson: Only build 'ds' when system has DRM
3319 - svga: convert license block to SPDX
3320 - svga: update timespan in copyright message
3324 - ci/b2c: Reduce the length of the kernel cmdline
3325 - nvk+zink/ci: rename the ga106 jobs to be more in line with RADV
3326 - nvk+zink/ci/ga106: make the expectations codename-specific
3327 - nvk+zink/ci: document more flakes in the ga106
3328 - turnip/ci: document a missing flake from the a750_vk job
3329 - turnip/ci: bump the a750_vk timeout
3330 - turnip+zink/ci: add more flakes to the expectations
3331 - radv+zink/ci: document recent flakes
3332 - radv/ci: add a bunch of flakes
3336 - nak: Pass has_mod to all form of src2 requiring it
3337 - panvk: Ensure we lower load_base_workgroup_id to 0
3338 - panfrost: Skip new failure from VKCTS 1.3.8.x
3339 - nvk, nak: Wire up conservative rasterization underestimate
3340 - docs/features: Add EXT_conservative_rasterization for NVK
3341 - agx: speed-up dce
3342 - panvk: Only clear UBOs descriptors when set isn't present
3343 - nouveau: nvidia_header: Add AMPERE_B class generation
3344 - nak: Set SPH version to 4 on SM75+
3345 - nak: Migrate sph.rs to use SPH headers defintion
3346 - bi: Reformat code
3347 - midgard: Reformat code
3348 - bi: Alloc replacement array once in opt_cse
3349 - pan/lib, panvk: Ensure data_size is on 64 bits
3350 - panvk: Fix shader destruction when vk_shader_module_to_nir fail
3351 - panvk: Remove panvk_lower_blend
3352 - panvk: Remove dynarray from panvk_shader
3353 - panvk: Keep panvk_shader alive in panvk_pipeline_shader
3354 - panvk: Upload shader in panvk_shader
3355 - panvk: Upload copy tables in panvk_shader
3356 - panvk: Upload render state in panvk_shader
3357 - panvk: Move the linking bits to panvk_shader
3358 - panvk: Kill panvk_pipeline_shader and use panvk_shader directly
3359 - panvk: Link shaders at draw time
3360 - panvk: Move compile logic out of shader_create
3361 - panvk: Move NIR lower logic out of shader_create
3362 - panvk: Move preprocess logic out of shader_create
3363 - panvk: Implement vk_shader
3364 - panvk: Remove panvk_pipeline
3365 - pan/va: Ensure no clash with other defs in disassembler
3366 - bi: Make disassembler take a const void*
3367 - midgard: Make disassembler take a const void*
3368 - bi: Move bi_disasm definitions to their own header
3369 - panfrost: Add pan_shader_disassemble
3370 - panvk: Implement executable IR reporting
3371 - panvk: Advertise VK_KHR_pipeline_executable_properties
3372 - panvk: Generate proper device and driver UUIDs
3373 - panvk: Advertise VK_EXT_pipeline_creation_cache_control and VK_EXT_pipeline_creation_feedback
3374 - panvk: Advertise VK_EXT_shader_module_identifier
3375 - panvk: Advertise VK_KHR_pipeline_library and VK_EXT_graphics_pipeline_library
3376 - panvk: Enable pipeline library in CI for Mali-G52
3377 - docs: Update features.txt to add panvk for BDA extensions
3378 - panvk: Advertise VK_KHR_device_group and VK_KHR_device_group_creation
3379 - panvk: Reorder extensions by name
3380 - panvk: Advertise VK_KHR_maintenance3
3381 - panvk: Add missing null check in DestroyCommandPool
3382 - panvk: Add missing clean up in blend_shader_cache_init
3383 - panvk: Make mempool detect NULL BOs
3384 - panvk: Check for maxBufferSize in panvk_CreateBuffer
3385 - panvk: Make panvk_kmod_zalloc use correct allocation scope on non-transient
3386 - panvk: Ensure to unref transient bo in reset for mempools
3387 - panvk: Fix device mempool leaks
3388 - panvk: Add more allocation checks in create_device
3389 - panvk: Implement CmdDispatchBase
3390 - panvk: Enable device_init, null_handle and object_management in CI for Mali-G52
3391 - panvk: Advertise shaderModuleIdentifier feature
3392 - panvk: Report correct min value for discreteQueuePriorities
3393 - panvk: Enable dEQP-VK.info tests in CI for Mali-G52
3394 - panvk: Clamp viewport scissor to valid range
3395 - panvk: Enable offscreen_viewport tests in CI for Mali-G52
3396 - panvk: Skip dispatch on empty workgroup
3397 - panvk: Report proper workgroup invocation and size
3398 - panvk: Enable compute pipeline in CI for Mali-G52
3399 - panvk: Advertise VK_EXT_private_data
3400 - panvk: Do not emit blend shader when color_mask is 0
3401 - panvk: Run nir_lower_io_to_vector for fragment shader
3402 - panvk: Enable glsl.440.linkage in CI for Mali-G52
3403 - panvk: Implement and advertise anisotropy support
3404 - panvk: Advertise VK_KHR_sampler_mirror_clamp_to_edge
3405 - panvk: Enable texture filtering in CI for Mali-G52
3406 - pan/kmod: Avoid deadlock on VA allocation failure on panthor
3407 - panfrost: Handle context_init errors correctly
3408 - panfrost: Handle gracefully resource BO alloc failures
3409 - ci/panfrost: Update t760 fails
3410 - rusticl: Add panthor when panfrost is present in RUSTICL_ENABLE
3411 - bi: Clean up mem_access_size_align_cb
3412 - bi: Enable lower_pack_64_4x16
3413 - bi: Lower pack_32_4x8_split and pack_32_2x16_split in algebraic
3414 - bi: Enable lower_pack pass in compiler
3415 - bi: Implement basic 8-bit vec support
3416 - panfrost: Rewrite set_global_binding to make resources truly global
3417 - panfrost: Do not recreate bo if already mapped
3418 - panfrost: Increase address space to 48-bit
3419 - panfrost: Fetch available system memory
3420 - panvk: Fix image support in vertex jobs
3421 - panvk: Pass attrib_buf_idx_offset to desc_copy_info
3425 - gallivm: Call StringMapIterator from llvm:: scope
3429 - docs: List VK_EXT_debug_utils
3433 - intel: Build float64 shader only for Vulkan
3434 - intel/clc: Free parsed_spirv_data
3435 - intel/clc: Free disk_cache
3436 - intel/brw: Use REG_CLASS_COUNT
3437 - intel/elk: Use REG_CLASS_COUNT
3438 - docs: Drop references to LIBGL_DRIVERS_PATH
3439 - util: Add ATTRIBUTE_OPTIMIZE(flags)
3440 - util: Force emission of stack frame in stack unit test
3444 - intel/common: fix building error in intel_common.c
3448 - v3dv: Use errno when logging an error to stderr
3449 - drm-uapi: Update v3d_drm.h
3450 - broadcom/common: Add maximum number of perf counters to v3d_device_info
3451 - v3dv: Use DRM_IOCTL_V3D_GET_COUNTER to get perfcnt information
3452 - v3d: Use DRM_IOCTL_V3D_GET_COUNTER to get perfcnt information
3453 - broadcom/simulator: Add DRM_V3D_PARAM_MAX_PERF_COUNTERS parameter support
3454 - broadcom/simulator: Add DRM_IOCTL_V3D_GET_COUNTER to simulator
3458 - wsi/wayland: Dispatch event queue in wsi_wl_swapchain_queue_present
3459 - wsi: Call drmSyncobjQuery only once for all images
3460 - egl/dri: Use packed pipe_format
3461 - dri: Go back to hard-coded list of RGBA formats
3465 - glthread: check for invalid primitive modes in DrawElementsBaseVertex
3466 - zink: reconstruct features pnext after determining extension support
3467 - zink: prune zink_shader::programs under lock
3468 - zink: fully wait on all program fences during ctx destroy
3469 - kopper: fix bufferage/swapinterval handling for non-window swapchains
3470 - zink: slightly better swapinterval failure handling
3471 - kopper: don't set drawable buffer age
3472 - zink: handle swapchain currentExtent special value
3473 - zink: clean up accidental debug print
3474 - dri: rename 'implicit' param from earlier series
3475 - tu: support VK_EXT_legacy_vertex_attributes
3476 - llvmpipe: add KHR-Single-GL45.arrays_of_arrays_gl.AtomicUsage skip
3477 - ci: disable lavapipe-vk-asan job
3478 - lavapipe: VK_EXT_legacy_vertex_attributes
3479 - zink: clamp buffer_indices_hashlist resets to used region
3480 - zink: delete GS conditional in update_so_info
3481 - zink: use zink_shader_key_optimal unions for pipeline state asserts
3482 - zink: use info.fs.uses_sample_qualifier instead of manual scan
3483 - zink: simplify confusing return in rewrite_tex_dest
3484 - zink: simplify flagging legacy shadow samplers
3485 - zink: rename zink_shader variable in create functions
3486 - zink: break out shadow sampler scanning
3487 - zink: always block the precompile threads when pruning shaders
3488 - zink: more effectively synchronize separate shader program precompiles
3489 - zink: use zink_shader type directly in zink_create_gfx_shader_state()
3490 - zink: split shader create into 2-stage functions
3491 - zink: reorder precompile_separate_shader_job() in file
3492 - zink: split generated tcs creation into 2-stage functions
3493 - zink: move gfx shader init to thread
3494 - zink: reorder some code in zink_create_gfx_program()
3495 - zink: reorder fencing in zink_create_gfx_program()
3496 - zink: split gfx program creation into 2-stage functions
3497 - zink: precompile_job() -> gfx_program_precompile_job()
3498 - zink: move blocking gfx program init functions to thread
3499 - ci: disable g52
3500 - egl/x11: disable dri3 with LIBGL_KOPPER_DRI2=1 as expected
3501 - zink: add a batch ref for committed sparse resources
3502 - u_blitter: stop leaking saved blitter states on no-op blits
3503 - freedreno/replay: use inttypes format string for 64bit
3504 - zink: delete unused zink_batch struct member
3505 - zink: move in_rp to zink_context
3506 - zink: move ref_lock from zink_batch to zink_batch_state
3507 - zink: move has_work from zink_batch to zink_batch_state
3508 - zink: rename last_was_compute -> last_work_was_compute
3509 - zink: move last_work_was_compute from zink_batch to zink_context
3510 - zink: move work_count from zink_batch to zink_context
3511 - zink: move swapchain from zink_batch to zink_context
3512 - zink: rename zink_batch::state -> zink_batch::bs
3513 - zink: delete all zink_batch uses from zink_query.c
3514 - zink: remove zink_batch usage from zink_clear.c
3515 - zink: remove all uses of zink_batch from zink_batch.c
3516 - zink: remove all zink_batch usage from zink_resource.h
3517 - zink: remove all zink_batch usage from zink_draw.cpp
3518 - zink: remove all zink_batch usage from zink_render_pass.c
3519 - zink: remove all zink_batch usage from zink_context.c
3520 - zink: delete zink_batch
3521 - zink: zink_batch_state::has_barriers -> has_reordered_work
3522 - zink: reset all the has_work flags in the same place
3523 - zink: check all has_work flags for flushes
3524 - zink: rely on zink_get_cmdbuf() to set has_work flags
3525 - zink: flag has_work in a few more places
3526 - zink: stop flagging has_work on batch tracking
3527 - zink: don't submit main cmdbuf if has_work is not set
3528 - frontends/dri: only release pipe when screen init fails
3529 - frontends/dri: always init opencl_func_mutex in InitScreen hooks
3530 - zink: use u_minify for sparse calcs
3531 - zink: always commit full miptails
3532 - zink: refcount miptails
3533 - zink: clean up semaphore arrays on batch state destroy
3534 - zink: add a batch array for tracked semaphores
3535 - zink: stop leaking sparse semaphores
3536 - zink: rework sparse semaphore waits
3537 - ci: bump VVL to snapshot-2024wk19
3538 - zink: hook up VK_EXT_legacy_vertex_attributes
3539 - zink: set all spirv caps for the vvl vtn pass
3540 - ci: bump VVL to v1.3.285
3541 - zink: make unassigned io variables unreachable
3542 - zink: minor tweaks to shader io assignment
3543 - zink: outdent assign_producer_var_io()
3544 - zink: outdent assign_consumer_var_io()
3545 - zink: pass a struct through io assignment functions
3546 - zink: track masks of io locations used during linking
3547 - zink: unify io assignment
3548 - zink: move 'reserved' into io assign struct
3549 - zink: split slot map between regular varyings and patch
3550 - zink: ci updates
3551 - egl/dri2: fix error returns on dri2_initialize_x11_dri3 fail
3552 - nir/lower_aaline: fix for scalarized outputs
3553 - nir/linking: fix nir_assign_io_var_locations for scalarized dual blend
3554 - lavapipe: split out DGC into separate file
3555 - lavapipe: plumb print_cmds through NV DGC
3556 - lavapipe: lvp_indirect_command_layout -> lvp_indirect_command_layout_nv
3557 - zink: remove dgc debug mode
3558 - zink: add atomic image ops to the ms deleting pass
3559 - build/amd: add amd-use-llvm build option
3560 - ir3: flag progress from nir_lower_io_to_scalar
3561 - ir3: assert that no further optimizations can be done if !progress
3562 - gallium: add drawid_offset to draw_mesh_tasks interface
3563 - gallium: stop dropping drawid_offset param with util_draw_indirect
3564 - vulkan: Update XML and headers to 1.3.287
3565 - zink: add HKP to tiler mode switch
3566 - lavapipe: fix mesh+task binding with shader objects
3567 - mesa/st: fix zombie shader handling for non-current programs
3568 - zink: null check pipe loader config before use
3569 - zink: split out msaa replication
3570 - zink: implement msaa replication with dynamic rendering
3571 - radeonsi: enable compute pbo blits
3572 - ci: kill filament trace globally
3573 - zink: add a driver workaround to disable 2D_VIEW_COMPATIBLE+sparse
3574 - zink: free sparse page for miptail on uncommit
3575 - zink: remove adreno from broken_cache_semantics driver workaround
3576 - egl: deduplicate MESA_image_dma_buf_export enablement
3577 - egl: only enable MESA_image_dma_buf_export with PIPE_CAP_DMABUF
3578 - lavapipe: maint7
3579 - st/pbo: fix MESA_COMPUTE_PBO=spec crash on shutdown
3580 - st/pbo_compute: special case stencil extraction from Z24S8
3581 - mesa/st: use compute pbo download for readpixels
3582 - ci: bump vvl to v1.3.289
3583 - zink: add an a750 skip
3584 - zink: enable compute pbos for turnip
3585 - aux/tc: update docs to indicate replaced buffers have multiple pipe_resources
3586 - zink: don't lower fpow
3587 - zink: propagate valid buffer range to real buffer when mapping staging
3588 - zink: track the "real" buffer range from replacement buffers
3589 - zink: modify some buffer mapping behavior for buffer replacement srcs
3590 - zink: move all driverID checks to a helper function
3591 - zink: hook up maintenance7
3592 - zink: use maint7 to capture venus driver and more accurately use workarounds
3593 - mesa/st: load state params for feedback draws with allow_st_finalize_nir_twice
3594 - egl/x11/sw: fix partial image uploads
3595 - egl/x11/sw: plug in swap_buffers_with_damage handling
3596 - winsys/radeon: take the full winsys struct in radeon_get_drm_value()
3597 - winsys/radeon: wrap fd access with util function
3598 - winsys/radeon: switch to rendernode when card node doesn't work
3599 - winsys/radeon: revert recent changes
3600 - glx: directly link to gallium
3601 - egl: link with libgallium directly
3602 - gbm: link directly with libgallium
3603 - loader: delete loader_open_driver()
3604 - loader/dri3: check xfixes version in loader_dri3_open()
3605 - loader/dri3: avoid killing the xcb connection if dri3 not found
3606 - loader/glx: move multibuffers check to loader
3607 - egl: use loader's multibuffer check to deduplicate lots of code
3608 - vl/dri3: use loader's dri3 init code and delete everything else
3609 - zink: move image aoa access to nir pass
3610 - zink: use PIPE_CAP_NIR_SAMPLERS_AS_DEREF
3611 - gallium: install gallium-$version.so to libdir
3612 - ci: prune dri from LD_LIBRARY_PATH
3613 - dril: rework config creation
3614 - llvmpipe: handle vma allocation failure
3615 - llvmpipe: only use vma allocations on linux
3616 - dri: fix kmsro define
3617 - Revert "vl/dri3: use loader's dri3 init code and delete everything else"
3618 - glx: include src/gallium for apple
3619 - dri: link with libloader
3620 - kopper: check swapchain size after possible loader image resize
3621 - pipe-loader: fix driconf memory management
3622 - dril: always take the egl init path
3623 - egl: fix zink init
3624 - dri: fix kms_swrast screen fail
3625 - egl/wayland: bail on zink init in non-sw mode if extension check fails
3626 - zink: fix partial update handling
3630 - radeonsi,aco: Run ac_nir_lower_global_access pass
3631 - ac/llvm: Remove global access ops handling
3635 - meson: set default drivers for ppc, ppc64
3636 - meson: set default Vulkan drivers for ppc, ppc64
3640 - nil: Add a nil_image::compressed bit
3641 - nil: Add some helpers for DRM format modifiers
3642 - nil: Support creating images with DRM modifiers
3643 - nvk: enable rendering to DRM_FORMAT_MOD_LINEAR images
3647 - egl/wayland: Fix sRGB format look up for config
3648 - ci/lima: expect fail of window_8888_colorspace_srgb on wayland
3652 - intel/isl: Add and use _isl_surf_info_supports_ccs
3653 - intel/isl: Reduce halign for disabled CCS on XeHP
3654 - intel/isl: Update quote for XeHP's CCS halign rule
3655 - intel/isl: Allow sampling from 3D HIZ_CCS_WT
3656 - intel/blorp: Factor bpb into the fast-clear rect
3657 - intel/blorp: Allow gfx12 fast-clears without CCS surf
3658 - intel/isl: Add and use ISL_DRM_CC_PLANE_PITCH_B
3659 - anv: Refactor modifier plane layout queries
3660 - intel/aux_map: Add and use INTEL_AUX_MAP_MAIN_PITCH_SCALEDOWN
3661 - intel/aux_map: Add and use INTEL_AUX_MAP_META_ALIGNMENT_B
3662 - intel/aux_map: Add and use INTEL_AUX_MAP_MAIN_SIZE_SCALEDOWN
3663 - intel/isl: Add and use ISL_MAIN_TO_CCS_SIZE_RATIO_XE
3664 - intel/isl: Add and use multi-engine surf usage bits
3665 - iris: Simplify bo import in memobj_create_from_handle
3666 - intel/isl: Assert alignments of surface addresses
3667 - anv: Rely on the primary surf usage to disable aux
3668 - anv,hasvk: Drop anv_get_isl_format_with_usage
3669 - anv: Support multiple aspects in anv_formats_ccs_e_compatible
3670 - anv: Rely more on ISL_SURF_USAGE_DISABLE_AUX_BIT
3671 - anv: Restrict CCS ISL surface creation to gfx9-11
3672 - iris: Add and use comp_ctrl_surf_offset on gfx12
3673 - intel/isl: Drop support for the gfx12 CCS ISL surf
3674 - intel/isl: Add and use isl_drm_modifier_needs_display_layout
3675 - iris,anv: Disable gfx12.0 fast-clears with unaligned pitch
3676 - intel/isl: Consolidate some tiling checks for CCS
3677 - intel/isl: Require display flag for 512B pitch alignment
3678 - intel/isl: Pad the pitch on gfx12.0 for fast-clears
3679 - anv+zink/ci: Change sparse test result from crash to fail
3680 - intel/isl: Enable Tile4 for CPB surfaces
3684 - nir/opt_varyings: reduce stack usage
3688 - svga: Retrieve stride info from hwtnl->cmd.vdecl for swtnl draws
3689 - dri: fix macro name check to detect svga driver
3693 - vulkan/properties: support Android in the property generator
3694 - v3dv: constify arguments of vendor/device id getters
3695 - v3dv: Use common runtime vk_properties
3696 - vulkan/properties: Document RENAMED_PROPERTIES in the property generator
3697 - anv: Move completely over to common runtime GetPhysicalDeviceProperties2
3698 - hasvk: switch to use runtime physical device properties infrastructure
3699 - vulkan: add a property struct setter function
3700 - venus: Use common physical device properties
3704 - gallium/auxiliary/vl: fix typo which negatively impacts the src_stride initialization
3705 - clover: fix pipe_box update regression
3706 - clover: fix memory leak related to optimize
3707 - r600: fix vertex state update clover regression
3708 - mesa/main: fix stack overflow related to the new mipmap code
3709 - radeonsi: fix assert triggered on gfx6 after the tessellation update
3710 - clover: fix meson opencl-spirv option
3711 - st/pbo_compute: fix async->nir memory leak
3715 - isl: add ISL_TILING_64_XE2 to isl_tiling_to_name()
3716 - anv/sparse: add the MSAA block shape tables
3717 - anv/sparse: we can't do multi-sampled depth/stencil sparse images
3718 - anv/sparse: properly reject sample counts we don't support
3719 - anv/sparse: reject all sample flags that non-sparse doesn't support
3720 - anv/sparse: fix block_size_B when the image is multi-sampled
3721 - anv/sparse: exclude Xe2's Tile64's non-standard block shapes
3722 - anv/sparse: flush the tile cache when resolving sparse images
3723 - anv/sparse: enable MSAA for Sparse when applicable
3724 - anv: check for VK_RENDERING_SUSPENDING_BIT once at CmdEndRendering
3725 - anv+zink/ci: add failures related to multi-sampled sparse binding
3726 - anv/sparse: assert a format can't be standard and non-standard
3727 - anv/xe: fix declaration of memory flags for integrated non-LLC platforms
3728 - anv/sparse: reject 1D sparse residency images
3729 - anv/sparse: fix the image property sizes for multi-sampled images
3730 - anv/sparse: fix reporting of VK_SPARSE_IMAGE_FORMAT_SINGLE_MIPTAIL_BIT
3731 - intel/isl: pass struct isl_tile_info to choose_image_alignment_el()
3732 - anv/sparse: dump info about opaque binds when DEBUG_SPARSE
3733 - anv/sparse: fix TR-TT page table bo size and flags
3734 - anv/sparse: remove obsolete linear tiling code path
3735 - anv/sparse: unify and rework tile size calculation
3736 - anv/sparse: use ANV_SPARSE_BLOCK_SIZE instead of tile_size when possible
3737 - anv: properly store the engine_class_supported_count values
3738 - anv: LNL+ doesn't need the special flush for sparse
3739 - anv: reimplement the anv_fake_nonlocal_memory workaround
3740 - iris: fix iris_xe_wait_exec_queue_idle() on release builds
3741 - anv/trtt: fix the process of picking device->trtt.queue
3742 - anv/xe: try harder when the vm_bind ioctl fails
3743 - anv: don't expose the compressed memory types when DEBUG_NO_CCS
3744 - anv: disable CCS for Source2 games on Xe2
3745 - intel: fix compute SLM sizes on Xe2 and newer
3749 - r300: replace constant size field with usemask
3750 - r300: move dead constants pass earlier for vertex shaders
3751 - r300: switch to a new constant remap table format
3752 - r300: compact scalar uniforms into empty slots
3753 - r300: better packing for immediates
3754 - r300/ci: fails list update
3755 - r300: fix cycles counting for KIL
3756 - r300: fix writemask rewrite when converting to omod
3757 - r300: fix for ouput modifier and DDX/DDX
3758 - r300: fix RC_OMOD_DIV_2 modifier
3759 - r300: missing whitespace in shader stats
3760 - r300: vectorization tweaks for R300/R400
3761 - r300: bias presubtract fix
3765 - etnaviv: drm: Suppress get-param error message for non-existent core
3766 - etnaviv: drm: Stop after model query failure
3767 - etnaviv: Pass npu to etna_screen_create in a separate parameter
3768 - etnaviv: Add a separate NPU pipe
3769 - etnaviv: Allow collecing both GPU and NPU specs
3770 - etnaviv/nn: Pipe through input/accumulation buffer depth from hwdb
3771 - etnaviv: update headers from rnndb
3773 Pierre-Eric Pelloux-Prayer (34):
3775 - radeonsi/sqtt: use ac_sqtt_get_shader_mask for spm counters
3776 - radeonsi/sqtt: cleanup si_sqtt_add_code_object a bit
3777 - radeonsi/sqtt: support sqtt buffer auto-resizing
3778 - radeonsi: add new si_shader_binary_upload_at method
3779 - radeonsi/sqtt: use si_shader_binary_upload_at to reupload shaders
3780 - radeonsi: allocate sqtt and spm buffers in GTT
3781 - radeonsi: add testmemperf mem bandwidth test
3782 - radv/sqtt: use radeon_check_space before emit_spm_*
3783 - radeonsi: use the common SQTT implementation
3784 - radeonsi/sqtt: add AMD_THREAD_TRACE_INSTRUCTION_TIMING
3785 - ac/sqtt: make VA helpers static
3786 - ac/llvm: implement WA in nir to llvm
3787 - radeonsi: store the total binary size in si_shader
3788 - radeonsi: handle DBG(TEX) after tc_compatible_htile is set
3789 - radeonsi/tests: don't match gfx10_3 baseline for gfx10 family
3790 - radeonsi/tests: add a shortcut to re-run only failing tests
3791 - ac/surface: reject modifiers with retile_dcc and bpe != 32
3792 - radeonsi: add gfx11 workaround for upgraded_depth
3793 - ac/nir: don't use the compute blit for PIPE_FORMAT_R5G6B5_UNORM
3794 - radeonsi/tests: update tests baseline
3795 - radeonsi/tests: clarify the output when results changes
3796 - radeonsi: fix buffer_size in si_compute_shorten_ubyte_buffer
3797 - Revert "ac, radeonsi: remove has_syncobj, has_fence_to_handle"
3798 - ac/info: remove has_syncobj
3799 - winsys/radeon: fill lds properties
3800 - radeonsi: fix crash in si_update_tess_io_layout_state for gfx8 and earlier
3801 - radeonsi/tests: correctly parse the family name
3802 - radeonsi: fix ac_create_shadowing_ib_preamble parameter
3803 - radeonsi, radv: bump libdrm_amdgpu version requirement
3804 - ci: bump Fedora and Android libdrm2 to 2.4.122
3805 - radeonsi: fix si_get_dmabuf_modifier_planes for gfx12
3806 - frontends/dri: add error logs to dri2_create_image_from_fd
3807 - amd: use a valid size for ac_pm4_state allocation
3808 - egl,gbm,glx: fix log message spam
3812 - glsl: respect GL_EXT_shader_image_load_formatted when image is embedded in a struct
3813 - radeonsi: add missing nir_intrinsic_bindless_image_descriptor_amd
3814 - nir: fix lower array to vec metadata preserve
3815 - nir: fix clip cull distance lowering metadata preserve
3816 - nir: add filter parameter to nir_lower_array_deref_of_vec
3817 - nir: nir_vectorize_tess_levels support indirect access
3818 - nir: consider more deref types when fixup deref
3819 - glsl: fix indirect tess factor access for compact_arrays=false drivers
3823 - panvk: Add jm and bifrost dirs
3824 - panvk: Add push_uniform/constant helpers
3825 - panvk: Make helper functions panvk_cmd_buffer agnostic
3826 - panvk: Move panvk_descriptor_state to bifrost subdir
3827 - panvk: Move vkCmdDraw* functions to their own file
3828 - panvk: Move vkCmdDispatch* functions to their own file
3829 - panvk: Move vkCmd*Event functions to their own file
3830 - panvk: Add Valhall DescriptorSetLayout implementation
3831 - panvk: Add Valhall Descriptor{Set,Pool} implementations
3835 - anv: Attempt to compile all pipelines even after errors
3836 - intel/perf: Move sysmacros.h include from header to implementation
3840 - aco/tests: add tests for hidden breaks/continues
3841 - aco/tests: add tests for divergent merge phi with undef
3842 - nir/dead_cf: stop reindexing blocks for each non-block cf node
3843 - aco/stats: fix s_waitcnt parsing
3844 - aco/stats: don't use VS counter pre-GFX10
3845 - aco/waitcnt: fix DS/VMEM ordered writes when mixed
3846 - aco: make wait_imm indexable
3847 - aco/waitcnt: add target_info
3848 - aco/waitcnt: refactor for indexable wait_imm
3849 - aco/stats: refactor for indexable wait_imm
3850 - aco: add wait_imm::unpack and wait_imm::max
3851 - radv: keep track of unaligned dynamic vertex access
3852 - aco: form hard clauses in VS prologs
3853 - aco: copy VS prolog constants after loads
3854 - aco: support VS prologs with unaligned access
3855 - aco/util: improve small_vec assertion
3856 - radv: advertise VK_EXT_legacy_vertex_attributes
3857 - aco: don't count certain pseudo towards VMEM_STORE_CLAUSE_MAX_GRAB_DIST
3858 - aco/tests: support GFX12
3859 - aco: add SFPU/ValuPseudoScalarTrans instr class
3860 - aco: add GFX11.5+ opcodes
3861 - aco: support GFX12 in assembler
3862 - aco/tests: add GFX12 assembler tests
3863 - aco: don't change prefetch mode on GFX11.5+
3864 - aco/gfx12: disable s_cmpk optimization
3865 - aco: add GFX12 wait counters
3866 - aco/waitcnt: support GFX12 in waitcnt pass
3867 - aco/stats: support GFX12 in collect_preasm_stats()
3868 - aco: update VS prolog waitcnt for GFX12
3869 - aco/lower_phis: create loop header phis for non-boolean loop exit phis
3870 - aco: create lcssa phis for continue_or_break loops when necessary
3871 - aco: use scalar phi lowering for lcssa workaround
3872 - aco: remove nir_to_aco
3873 - aco/lower_phis: don't create boolean loop header phis in some situations
3874 - radv: malloc graphics pipeline stages
3875 - aco: support GFX12 in insert_NOPs
3876 - aco/gfx12: implement subgroup shader clock
3877 - aco/gfx12: implement workgroup barrier
3878 - aco/gfx12: sign-extend s_getpc_b64
3879 - aco/gfx12: don't create v_fmac_legacy_f32
3880 - aco/gfx12: use ttmp9/ttmp7 for workgroup id
3881 - radv/gfx12: don't add workgroup id shader args
3882 - aco/gfx12: remove MIMG vector affinity
3883 - aco/gfx12: decrease max_nsa_vgprs for VSAMPLE
3884 - aco/gfx12: disallow SCC and most constants for BUF SOFFSET
3885 - aco: fix fddx/y with uniform inf/nan input
3886 - meson: remove --depfile for aco_tests
3887 - ac/llvm: implement load_subgroup_id
3888 - aco/gfx12: implement load_subgroup_id
3889 - ac/nir: skip subgroup_id/local_invocation_index lowering for gfx12
3890 - aco/gfx12: fix s_wait_event immediate
3891 - aco: don't combine vgpr into writelane src0
3892 - aco: implement nir_atomic_op_ordered_add_gfx12_amd
3893 - aco: implement nir_intrinsic_nop_amd and nir_intrinsic_sleep_amd
3894 - ac/nir: support lowering of sub-dword push constants
3895 - radv: lower sub-dword push constants
3896 - ac/llvm: remove support for sub-dword push constants
3897 - aco: remove support for sub-dword push constants
3898 - aco/gfx6: set glc for buffer_store_byte/short
3899 - aco: inline store_vmem_mubuf/emit_single_mubuf_store
3900 - aco: use ac_hw_cache_flags
3901 - aco: use GFX12 scope/temporal-hint
3902 - ac: stop using radeon_info for ac_get_hw_cache_flags
3903 - aco: use ac_get_hw_cache_flags()
3904 - aco: remove some missing label resets
3905 - nir/opt_loop: rematerialize derefs instead of creating phis
3906 - nir/opt_loop: fix formatting
3907 - aco: insert s_nop before discard early exit sendmsg(dealloc_vgpr)
3908 - radv: lower push constants in NIR
3909 - ac/llvm: remove push constants
3910 - aco: remove push constants
3911 - aco/insert_exec_mask: ensure top mask is not a temporary at loop exits
3912 - vtn: ensure TCS control barriers have a large enough memory scope
3913 - aco: use 1.5x vgprs for gfx1151 and gfx12
3914 - aco: skip continue_or_break LCSSA phis when not needed
3915 - aco: use s_pack_ll_b32_b16 for pack_32_2x16_split
3916 - aco: combine extracts into s_pack_ll_b32_b16
3917 - aco: use s_pack_*_b32_b16 more in p_insert/p_extract lowering
3918 - aco: turn split(vec()) into p_parallelcopy instead of p_create_vector
3919 - aco: add missing isConstant()/isTemp() checks
3920 - aco: fix follow_operand with combined label_extract and label_split
3921 - aco: use alignment information in visit_load_constant()
3922 - aco: fix wmma raw hazard
3923 - aco: replace constant v_bfrev_b32 with v_mov_b32 to create vopd
3924 - aco/gfx11: don't use v_bfrev_b32 with wave64
3925 - glsl: always lower non-TCS outputs to temporaries
3926 - gallium: remove PIPE_CAP_SHADER_CAN_READ_OUTPUTS
3927 - nir/linking_helpers: remove special case for read mesh outputs
3928 - nir/linking_helpers: remove varying accesses in nir_remove_unused_io_vars
3929 - nir/linking_helpers: remove nested IF
3930 - radv: remove unecessary nir_remove_unused_varyings cleanup passes
3931 - aco/gfx11.5: workaround export priority issue
3932 - aco: fix validation of v_s\_ opcodes
3933 - docs: update ACO_DEBUG documentation for scheduler options
3934 - docs: update ACO_DEBUG documentation for perfwarn
3938 - tu: Add missing error path cleanup
3939 - tu: Fix a6xx lineWidthGranularity
3940 - freedreno/ir3: Skip DAG validation on release builds
3941 - llvmpipe: Fix build error with clang-18
3942 - freedreno/ci: Switch a618_piglit to deqp-runner
3943 - vulkan/android: Add helper to probe AHB support
3944 - vulkan: Don't request Ycbcr conversion for rgb
3945 - vulkan: Add helper to resolve Android external format
3946 - tu: Skip YUV conversion for RGB formats
3947 - tu: Support VkExternalFormatANDROID
3948 - freedreno/ci: Remove some skips
3949 - freedreno/ci: Remove some obsolete skips
3950 - freedreno/ci: Refactor out common a6xx skips list
3951 - freedreno/ci: Skip unsupported legacy gl stuff
3952 - freedreno/ci: Skip max-texture-size
3953 - freedreno/ci: Add a common skips file to a618_piglit.
3954 - freedreno/ci: Skip built-in-functions VS/GS tests
3955 - freedreno/ci: Skip some slow tests
3956 - freedreno/ci: Increase a630/a618 piglit fraction
3957 - freedreno/ir3: Fix ldg/stg offset
3958 - egl/android: Fix sRGB visuals
3959 - docs/features: Add missing AHB for tu
3960 - tu: Don't advertise AHB handle time on non-android
3961 - freedreno: Namespace DEFINE_CAST()
3962 - virgl: Update headers
3963 - loader: Add better support for virtgpu nctx driver loading
3964 - freedreno/loader: Switch over to probe_nctx
3965 - vulkan/android: Fix suggestedYcbcrModel with !mapper4
3966 - tu: Fix imageview + ahb
3967 - vulkan/android: Fix YcbcrRange for !mapper4
3968 - ir3: Add some more missing progress accumulation
3969 - gallium/tc: Add optional buffer replacement limit
3970 - freedreno: Use buffer replacement limit
3971 - gallium/tc: Allow replacement if replacing valid_range
3972 - freedreno/drm: Add rd dumper support
3973 - st/mesa/pbo: Set src type on image_store
3974 - freedreno: Handle non-null cb with null buffer
3975 - u_blitter+d3d12: Move stencil fallback clear to caller
3976 - freedreno/a6xx: Implement S8 support
3977 - freedreno: Implement stencil blit fallback
3978 - freedreno: Use LINEAR for staging resources
3979 - freedreno/a6xx: Drop 16b packed image formats
3980 - freedreno/bc: Rework flush order
3981 - freedreno/a6xx: Tweak blitter traces
3982 - freedreno/a6xx: Skip blitter for L/A conversions
3983 - freedreno/a6xx: Add more format swizzles
3984 - freedreno/a6xx: Allow blit based transfers
3985 - freedreno: Enable the X1-85
3986 - tu: Fix issues with 16k (or larger) page sizes
3987 - freedreno/drm/virtio: Fix issues with 16k (or larger) page sizes
3988 - freedreno/a6xx: Implement reg stomper support
3989 - freedreno/a7xx: Fix GRAS_UNKNOWN_80F4 writes
3990 - freedreno/cffdec: Fix a7xx CP_EVENT_WRITE decoding
3991 - tu/drm/virtio: Add missing a7xx case
3992 - freedreno/drm: Handle a7xx case
3993 - freedreno: Move GENX/CALLX magic to common
3994 - freedreno: Extract out common UBWC helper
3995 - freedreno: Extract out shared LRZFC layout helpers
3996 - freedreno/a6xx: Allocate lrcfc when needed for direction tracking
3997 - freedreno/a6xx: Refactor CP_EVENT_WRITE emit
3998 - freedreno/a6xx: Rework CCU_CNTL emit for a7xx
3999 - freedreno/a6xx: Initial a7xx support
4000 - gallium: Add option to not add version to libgallium filename
4004 - egl: Implement EGL_EXT_config_select_group
4005 - egl: Implement EGL_MESA_x11_native_visual_id
4006 - egl/x11: Allow all RGB visuals to match 32-bit RGBA EGLConfigs
4010 - anv: formatting fix when printing pipe controls
4011 - anv: allocate space for generated indirect draw id's using the temporary allocation helper
4012 - intel/brw: update Xe2 max SIMD message sizes
4013 - Revert "iris: slow clear higher miplevels on single sampled 8bpp resources that have TILE64"
4014 - intel/eu/xe2+: Fix src1 length bits of SEND instruction with UGM target.
4015 - intel/brw: Advertise fp64 atomic add's when we have 64 bit float support and a LSC
4016 - intel/brw: We no longer have atomic fmin/fmax ops for fp64 in xe2
4017 - intel/genxml: add the new state byte stride instruction
4018 - intel/genxml: update 3DSTATE_CPSIZE_CONTROL_BUFFER for xe2+
4019 - isl: enable compression for CPS buffers on xe2+
4020 - intel/genxml: update CFE_STATE for LNL
4021 - intel/genxml: Update XY_BLOCK_COPY_BLT
4022 - intel/genxml: update MI_SEMAPHORE_WAIT for Xe2
4023 - intel/genxml: Update STATE_COMPUTE_MODE for Xe2
4024 - anv: 3D stencil surfaces have fewer layers for higher miplevels
4025 - isl: disable CCS for 3D depth/stencil surfaces when WA is applicable
4026 - isl: Enable volumetric STC_CCS,HiZ+CCS on gfx12.0
4027 - intel/genxml: Add RESOURCE_BARRIER for xe2
4028 - intel/compiler: fix shuffle generation on LNL
4029 - anv: flag WSI images as scanout images for ISL
4030 - anv: reuse existing macro to query for flushes
4034 - lavapipe: add option to enable snorm blending
4038 - glxext: don't try zink if not enabled in mesa
4042 - vulkan/android: Add basic u_gralloc support
4043 - vulkan/android: Add common vkGetSwapchainGrallocUsage{2}ANDROID
4044 - vulkan/android: Add android buffer classification to vk_image
4045 - vulkan/android: Add common helpers for the ANB extension
4046 - vulkan/android: Add common helpers for the AHB extension
4047 - vulkan/android: Add common vkGetAndroidHardwareBufferPropertiesANDROID
4048 - turnip/android: Migrate to common ANB code
4049 - v3dv/android: Migrate ANB and AHB to use common helpers
4050 - u_gralloc/fallback: Extract modifier from QCOM native_handle
4051 - turnip/android: Use DETECT_OS_ANDROID in tu_device
4052 - turnip/android: Use DETECT_OS_ANDROID in freedreno_rd_output
4056 - intel/brw: allocate large table in the heap instead of the stack
4060 - radeonsi/vcn: add vcn5 encoding interface change
4061 - radeonsi/vcn: add vcn5.0 for h264 enc only
4062 - radeonsi/vcn: add hevc support for vcn5
4063 - radeonsi/vcn: enable decoding in vcn5.
4064 - radeonsi/vcn: correct tile_size_bytes_minus1
4065 - radeonsi/vcn: add cdef modes for vcn5 encoding
4066 - radeonsi/vcn: apply cdef mode to vcn5
4067 - radeonsi/vcn: share functions between vcn4/vcn5
4068 - frontends/va: parsing uniform_tile_spacing flag
4069 - radeonsi/vcn: add header files for vcn5 av1 tile
4070 - radeonsi/vcn: enable av1 encoding in vcn5
4071 - radeonsi/vcn: enable roi feature for vcn5
4072 - radeonsi/vcn: remove tile_config_flag
4073 - radesonsi/vcn: update vcn4 tile processing logic
4077 - venus: reclaim signal semaphore feedback resources for wasteful clients
4078 - venus: sync headers for VK_EXT_external_memory_acquire_unmodified
4079 - venus: enable VK_EXT_external_memory_acquire_unmodified
4080 - venus: factor image memory barrier fixes to common implementation
4081 - venus: refactor image memory barrier fix storage and conventions
4082 - venus: skip barrier fixes as early as possible
4083 - venus: chain VkExternalMemoryAcquireUnmodifiedEXT for wsi ownership transfers
4087 - zink: Add VKAPI_PTR specifier to zink_stub_function_not_loaded.
4088 - zink: Add VKAPI_PTR specifier to generated stub functions.
4092 - intel/compiler: Fix destination type for CMP/CMPN
4093 - intel/disasm: Fix cache load/store disassembly for URB messages
4094 - iris: Load 32-bit MMIO PREDICATE register from buffer
4095 - intel/compiler: No need to re-type the destination register
4096 - intel/fs: Adjust destination register size for untyped atomic on Xe2+
4097 - intel/fs: Adjust destination register size for global atomic on Xe2+
4098 - intel/compiler: Don't use half float param for sample_b
4099 - intel/compiler: Add indirect mov lowering pass
4103 - radv: fix image format properties with fragment shading rate usage
4104 - docs: Add an alternative way to debug GPU hangs with RADV
4105 - radv/rt: add radv_ray_tracing_state_key
4106 - radv/rt: pass radv_ray_tracing_state_key to radv_rt_pipeline_compile()
4107 - radv/rt: rework the helper that hashes a ray tracing pipeline
4108 - radv/ci: add more flakes
4109 - radv: simplify DB_Z_INFO.NUM_SAMPLES with null ds target on GFX11
4110 - radv: remove bogus VkShaderCreateInfoEXT::flags being 0 assert for compute
4111 - radv: simplify radv_emit_primitive_restart_enable()
4112 - radv: inline radv_get_pa_su_sc_mode_cntl() in radv_emit_culling()
4113 - radv: remove useless DB_Z_INFO.NUM_SAMPLES when emitting the MSAA state
4114 - radv: pre-compute VGT_TF_PARAM.DISTRIBUTION_MODE
4115 - radv: use the bound GS copy shader when emitting shader objects
4116 - radv: add GS copy shader BO to the cmdbuf BO list at bind time
4117 - radv: add RT prolog BO to the cmdbuf BO list at bind time
4118 - radv: add shaders BO to the cmdbuf BO list at bind time
4119 - radv: emit compute pipelines directly from the cmdbuf
4120 - radv: precompute compute/task shader register values
4121 - radv: clear unwritten color attachments for monolithic PS earlier
4122 - radv: compact SPI_SHADER_COL_FORMAT as late as possible
4123 - radv: rename col_format_non_compacted to spi_shader_col_format
4124 - radv: store cb_shader_mask for fragment shaders and epilogs
4125 - radv: add a new dirty state for emitting the color output state
4126 - radv/ci: document a recent regression on GFX6-8
4127 - radv: split cmdbuf dirty flags into dirty/dirty_dynamic
4128 - radv: precompute existing legacy GS register values later
4129 - radv: precompute fragment shader register values
4130 - radv: precompute mesh shader register values
4131 - radv: precompute legacy GS register values
4132 - radv: precompute vertex shader register values
4133 - radv: precompute DB_SHADER_CONTROL for fragment shaders later
4134 - vulkan: Update XML and headers to 1.3.284
4135 - aco: add support for remapping color attachments
4136 - radv: implement VK_KHR_dynamic_rendering_local_read
4137 - radv: advertise VK_KHR_dynamic_rendering_local_read
4138 - radv: add a new mechanism for tracking registers per cmdbuf
4139 - radv: move common registers between VS/GS and NGG
4140 - radv: precompute NGG register values
4141 - radv: remove unused parameter to radv_pipeline_emit_pm4()
4142 - radv: stop recomputing the last VGT API stage when emitting graphics shaders
4143 - radv: do not emit non-context registers to radv_pipeline::ctx_cs
4144 - radv: track and bind more VRS states from the graphics pipeline
4145 - radeonsi: remove the _unused parameter in all radeon_xxx macros
4146 - radv: remove gfx10_emit_ge_pc_alloc()
4147 - radv: do not emit VGT_GS_OUT_PRIM_TYPE to ctx_cs on GFX11
4148 - radv: simplify radv_emit_hw_ngg() slightly
4149 - radv: simplify radv_emit_hw_vs() slightly
4150 - radv: simplify radv_emit_hw_gs() slightly
4151 - radv: fix the late scissor workaround for GFX9 since a recent refactoring
4152 - radv: make radv_conv_gl_prim_to_gs_out() a non-static function
4153 - radv: emit graphics pipelines directly from the cmdbuf
4154 - radv: add graphics shaders context registers that need to be tracked
4155 - radv: add more radeon_opt_set_xxx variants
4156 - radv: track all graphics shaders context registers
4157 - radv: simplify radv_emit_ps_inputs() slightly
4158 - radv: stop using radv_physical_device for radeon helpers
4159 - radv: introduce radeon_set_reg_seq()
4160 - radv: remove redundant radeon_set_perfctr_reg() helper
4161 - radv: rename radeon perfctr uconfig helpers
4162 - radv: add a helper to configure ring buffer descriptors
4163 - radv: only enable VK_MESA_image_alignment_control on GFX9-11.5
4164 - radv: reject unsupported buffer formats earlier
4165 - ac,radv,radeonsi: add a helper to translate buffer numformat
4166 - ac,radv,radeonsi: add a helper to translate buffer dataformat
4167 - radv: simplify radv_emit_default_sample_locations()
4168 - radv: pass radv_physical_device to radv_emit_default_sample_locations()
4169 - radv: use float instead of double for viewport zscale/ztranslate
4170 - radv: add more helpers to emit viewports
4171 - radeonsi: refactor si_translate_border_color()
4172 - ac,radv,radeonsi: introduce a helper to build a sampler descriptor
4173 - radv: stop checking the output value of radv_translate_tex_numformat
4174 - radv: use PIPE_FORMAT in radv_translate_tex_numformat()
4175 - ac,radv,radeonsi: add a function to translate tex numformat
4176 - radv: use PIPE_FORMAT in radv_translate_colorswap()
4177 - ac,radv,radeonsi: add a function to translate colorswap
4178 - radv: use PIPE_FORMAT in radv_translate_dbformat()
4179 - ac,radv,radeonsi: add a function to translate db format
4180 - ac,radv,radeonsi: add a function to get the color format endian swap
4181 - radv: allow 3d views with VK_IMAGE_CREATE_2D_VIEW_COMPATIBLE_BIT_EXT
4182 - radv: simplify creating gfx10 texture descriptors for sliced 3d/2d view of 3d
4183 - radv: remove redundant check for VK_FORMAT_E5B9G9R9_UFLOAT_PACK32 on GFX6-10
4184 - radv: stop checking the return value of ac_get_cb_number_type()
4185 - radv: add radv_is_colorbuffer_format_blendable()
4186 - amd/common: move some format related helpers to ac_formats.c
4187 - ac,radv,radeonsi: add helper to know if a format is supported by CB
4188 - ac,radv,radeonsi: add helper to know if a format is supported by DB
4189 - ac,radv,radeonsi: introduce a helper to build a FMASK descriptor
4190 - radeonsi: reject some texture formats but only on GFX8/CARRIZO
4191 - radv: mark some formats as unsupported on GFX8/CARRIZO
4192 - radv: set image view descriptors as buffer for non-graphics GPU
4193 - ac,radv,radeonsi: add a helper to get the tile mode index
4194 - radv: tidy up swizzle in radv_set_mutable_tex_desc_fields()
4195 - radv: tidy up custom pitch for gfx10.3 in radv_set_mutable_tex_desc_fields()
4196 - radv: tidy up meta_va in radv_set_mutable_tex_desc_fields()
4197 - radv: set ITERATE_256 for GFX10+ in radv_set_mutable_tex_desc_fields()
4198 - radv: stop clearing unnecessary bitfields in radv_set_mutable_tex_desc_fields()
4199 - ac,radv,radeonsi: add a helper to set mutable tex desc fields
4200 - ac,radv,radeonsi: add a function for getting border color swizzle
4201 - radv: only set ALPHA_IS_ON_MSB if the image has DCC on GFX6-9
4202 - radv: adjust determining if alpha should be on MSB
4203 - ac,radv,radeonsi: add function to determine if alpha should be on MSB
4204 - ac,radv,radeonsi: add a common helper for translating swizzle
4205 - amd/common: only pass gfx_level to ac_get_gfx10_format_table()
4206 - amd/common: add new helpers to build buffer descriptors
4207 - radv: use new common helpers for building buffer descriptor
4208 - aco: use new common helpers for building buffer descriptors
4209 - radv: remove unused radv_translate_buffer_dataformat()
4210 - radv: remove useless check about FIXED formats
4211 - radv: simplify radv_is_vertex_buffer_format_supported()
4212 - radv: replace vk_to_non_srgb_format() by vk_format_no_srgb()
4213 - radv: fix setting a custom pitch for CB on GFX10_3+
4214 - radv: replace db_{z,stencil}_{read,write}_base by db_{depth,stencil}_base
4215 - radv: tidy up radv_initialise_ds_surface()
4216 - radv: separate non-mutable vs mutable fields for ds surface
4217 - amd/common: add a function to initialize ds surface
4218 - radv: use the common helper for initializing DS surfaces
4219 - radeonsi: use the common helper for initializing DS surfaces
4220 - radv: only enable DB_STENCIL_INFO.ITERATE_FLUSH when necessary
4221 - radv: do not check image usage for ITERATE256 with TC-compat HTILE
4222 - ac,radv,radeonsi: add function to get the number of ZPLANES
4223 - ac,radv,radeonsi: a function that sets mutable DS surface fields
4224 - ac,radv,radeonsi: move ZRANGE_PRECISION to mutable DS fields
4225 - amd/common: add ac_gpu_info::has_etc_support
4226 - radv: use PIPE_FORMAT in radv_translate_tex_dataformat()
4227 - ac,radv,radeonsi: add a function to translate texture data format
4228 - amd/common: add a helper to set the third word of buffer descriptor
4229 - ac,radv: add a stride parameter to ac_build_attr_ring_descriptor()
4230 - radeonsi: use common build buffer descriptor helpers
4231 - radv: align DCC control settings to RadeonSI for GFX8
4232 - amd/common: add a function to initialize cb surface
4233 - radv: use the common helper for initializing CB surfaces
4234 - radeonsi: use the common helper for initializing CB surfaces
4235 - ac,radv,radeonsi: a function that sets mutable CB surface fields
4236 - radv: apply the workaround for no PS inpputs and LDS on GFX11 only
4237 - radv: apply the SQ_THREAD_TRACE_WPTR workaround on GFX11 only
4238 - radv: fix flushing DB meta cache on GFX11.5
4239 - radv: only emit streamout enable for legacy streamout
4240 - amd/common: define SDMA v7.0 for GFX12
4241 - ac,radv,radeonsi: add ac_gpu_info::has_tc_compatible_htile
4242 - radv: update NUM_THREAD_FULL bitfields
4243 - radv: enable GS_FAST_LAUNCH_2 by default on GFX12
4244 - radv: do not emit non-existent registers on GFX12
4245 - radv: update configuring sample locations on GFX12
4246 - radv: update configuring viewport/scissor on GFX12
4247 - radv: update configuring PS states on GFX12
4248 - radv: update configuring NGG states on GFX12
4249 - radv: update configuring VGT states on GFX12
4250 - radv: update configuring DB states on GFX12
4251 - radv: update configuring rasterization states on GFX12
4252 - radv: update configuring some CB states on GFX12
4253 - radv: update configuring occlusion query state on GFX12
4254 - radv: update configuring MSAA state on GFX12
4255 - radv: update configuring GFX preamble on GFX12
4256 - radv: update configuring tess rings on GFX12
4257 - radv: update binning settings on GFX12
4258 - radv: update emitting discard rectangles on GFX12
4259 - radv: update shader input arguments for GS stage on GFX12
4260 - aco: adjust loading local invocation ID for GS on GFX12
4261 - radv: do not emulate clear state for shadowed regs on GFX12
4262 - radv: update cache flush emission on GFX12
4263 - radv: update emitting stipple line on GFX12
4264 - radv: disallow merging multiple draws into one wave on GFX12
4265 - radv: emit SQ_NON_EVENT packets after drawing with streamout on GFX12
4266 - radv: update configuring the number of patch control points on GFX12
4267 - radv: update configuring VGT_SHADER_STAGES_EN on GFX12
4268 - radv: enable GE_CNTL.DIS_PG_SIZE_ADJUST_FOR_STRIP on GFX12
4269 - radv: update NUM_THREAD_FULL bitfields on GFX12
4270 - radv: update global graphics shader pointers on GFX12
4271 - radv: update SDMA resource type on GFX12
4272 - radv: update VS input VGPRs on GFX12
4273 - radv: do not enable HTILE for depth/stencil storage images
4274 - radv: allow STORAGE for depth formats
4275 - radv: fix configuring the number of patch control points on GFX6
4276 - radv: configure DB_Z_INFO.NUM_SAMPLES on GFX12
4277 - radv: configure DB_RENDER_CONTROL to zero on GFX12
4278 - radv: do not enable MEM_ORDERED on GFX12
4279 - radv: update configuring the attribute ring on GFX12
4280 - radv: do not flush L2 metadata on GFX12
4281 - radv: mark all images coherent with TC L2 on GFX12
4282 - radv: update configuring SPI_SHADER_PGM_LO_LS on GFX12
4283 - radv: update configuring SPI_SHADER_PGM_LO_ES on GFX12
4284 - radv: update configuring SPI_SHADER_PGM_RSRC4_{HS,GS,PS} on GFX12
4285 - radv: update configuring GE_CNTL.PRIM_GRP_SIZE_GFX11 on GFX12
4286 - radv: update configuring SPI_PS_IN_CONTROL on GFX12
4287 - radv: configure PA_SC_HISZ_CONTROL on GFX12
4288 - radv: configure SPI_SHADER_GS_OUT_CONFIG_PS on GFX12
4289 - radv: update configuring GS_VGPR_COMP_CNT on GFX12
4290 - radv: do not set DX10_CLAMP on GFX12
4291 - radv: fix VRS subpass attachments with mipmaps
4292 - ac,radeonsi: set COLOR_SW_MODE for mutable CB surfaces on GFX12
4293 - radv: configure PA_SC_SAMPLE_PROPERTIES on GFX12
4294 - radv: update number of input VGPRs for VS on GFX12
4295 - radv: update configuring color buffers on GFX12
4296 - radv: update configuring depth stencil buffers on GFX12
4297 - radv: update configuring PA_SC_WINDOW_SCISSOR on GFX12
4298 - radv: do not emit SPI_SHADER_PGM_RSRC3_GS on GFX12
4299 - radv: fix configuring NGG registers on GFX12
4300 - radv: do not set VGT_PRIMITIVEID_EN.PRIMITIVEID_EN on GFX12
4301 - radv: cleanup radv_precompute_registers_hw_{ngg,fs}
4302 - radv: assert that GDS/GDS OA buffers can't be created on GFX12
4303 - radv: only set valid bitfields for CB/DS surfaces address
4304 - radv: only emit VGT_GS_MAX_PRIMS_PER_SUBGROUP on GFX9
4305 - radv: only emit SQ_PERFCOUNTER_MASK on GFX7-9
4306 - radv: do not set VGT_SHADER_STAGES_EN.DYNAMIC_HS on GFX9
4307 - radv: only emit SPI_SHADER_PGM_SRC3_GS on GFX7+
4308 - radv: only emit CB_COLOR0_DCC_CONTROL on GFX8
4309 - radv: use pipe_format when building image view descriptors
4310 - ac,radv,radeonsi: add a function to build texture descriptors
4311 - amd/common: add MIN_LOD for texture descriptors on GFX12
4312 - Revert "radv/ci: Bring back vkcts-navi21-llvm-valve"
4313 - radv: update configuring depth clamp enable on GFX12
4314 - radv: update configuring COVERAGE_TO_SHADER_SELECT on GFX12
4315 - radv: fix emitting VGT_PRIMITIVEID_RESET in the GFX preamble on GFX12
4316 - radv: only set valid bitfields for CB/DS surfaces address on GFX12
4317 - radv: add a helper to get image VA
4318 - ac,radeonsi import PM4 state from RadeonSI
4319 - ac,radeonsi: add a function to initialize compute preambles
4320 - radv: initialize compute preambles with the common helper
4321 - radv: fix creating unlinked shaders with ESO when nextStage is 0
4322 - radv: pass a radv_shader to radv_get_compute_pipeline_metadata()
4323 - radv: don't assume that TC_ACTION_ENA invalidates L1 cache on gfx9
4324 - ac,radv: add a helper for SQTT control register
4325 - ac,radv,radeonsi: add more parameters to ac_sqtt
4326 - amd: allow to emit privileged config registers in PM4
4327 - amd: mark more registers that need RESET_FILTER_CAM in PM4
4328 - amd: add a common implementation for SQTT using PM4
4329 - radv: emit more consecutive registers for SQTT on GFX8-9
4330 - radv: use the common SQTT implementation
4331 - radv: update VGT_TESS_DISTRIBUTION.ACCUM_ISOLINE value
4332 - radv: do not set registers set by CLEAR_STATE in the preamble on GFX10-11.5
4333 - radv: emit SPI_GS_THROTTLE_CNTL1 when the attr ring is emitted
4334 - radv: fix incorrect buffer_list advance for multi-planar descriptors
4335 - radv: use BDA in the DGC prepare shader
4336 - radv: remove dynamic uniform/storage buffers support with DGC
4337 - radv: do not use nir_pkt3() when the packet len is constant with DGC
4338 - radv: add new macros for emiting packets in DGC
4339 - radv: remove redundant nir_builder param in some DGC helpers
4340 - radv: add a helper to load the pipeline VA for DGC
4341 - radv: store a pointer to the logical device in dgc_cmdbuf
4342 - radv: allow VK_NV_device_generated_commands_{compute} with LLVM
4343 - radv: always save/restore all shader objects for internal operations
4344 - radv: update configuring WALK_ALIGN8_PRIM_FITS_ST on GFX12
4345 - ac/surface: add NBC view support on GFX12
4346 - radv: declare a new user SGPR for the streamout state buffer on GFX12
4347 - radv/nir: lower nir_intrinsic_load_xfb_state_address_gfx12_amd
4348 - radv: implement streamout on GFX12
4349 - radv: force using indirect descriptor sets for indirect compute pipelines
4350 - radv: emit indirect sets for indirect compute pipelines with DGC
4351 - radv: fix emitting indirect descriptor sets in the DGC prepare shader
4352 - radv: cleanup getting AC_UD_TASK_RING_ENTRY for mesh shader
4353 - radv: use radv_shader_info::user_data_0 for task shaders
4354 - radv: remove dead mesh shader code for indirect draws
4355 - radv: remove useless masking in radv_cs_emit_indirect_mesh_draw_packet()
4356 - radv: remove useless draw_id to radv_emit_userdata_task()
4357 - radv: add the DGC preprocess BO to the cmdbuf BO list
4358 - radv/amdgpu: allow cs_execute_ib() to pass a VA instead of a BO
4359 - radv/amdgpu: use the non-IB path for dumping CS with external IBs
4360 - ac/parse_ib: dump PKT3_DISPATCH_{TASKMESH_GFX,TASKMESH_DIRECT_ACE}
4361 - radv/amdgpu: fix chaining CS with external IBs on compute queue
4362 - radv: add a helper to execute a DGC IB
4363 - radv: add support for computing the DGC ACE IB size
4364 - radv: prepare for DISPATCH_TASKMESH_GFX emission in the DGC shader
4365 - radv: prepare for DISPATCH_TASKMESH_DIRECT_ACE emission in the DGC shader
4366 - radv: refactor some DGC helpers in preparation for the ACE IB
4367 - radv: add a helper to pad DGC IB
4368 - radv: add support for preparing the ACE IB in DGC
4369 - radv: add support for executing the DGC ACE IB
4370 - radv: fix incorrect cache flushes before decompressing DCC on compute
4371 - radv: improve clarity of DGC offset computations
4372 - radv: pre-compute the base upload offset in radv_prepare_dgc()
4373 - radv: add a helper that determines if DGC uses task shaders
4374 - radv: split allocating and emitting push constants with DGC
4375 - radv: rework emitting push constants with DGC
4376 - radv: reserve space for push constants in the DGC ACE IB
4377 - radv: adjust the base upload offset when DGC uses task shaders
4378 - radv: emit push constant for task shaders with DGC
4379 - radv: disable conditional rendering with DGC and task shaders
4380 - radv: fix a synchronization issue with non-preprocessed DGC with task shader
4381 - radv: enable task shaders support with NV DGC
4382 - radv: suspend user conditional rendering when DGC has task shaders
4383 - radv: rename radv_get_user_sgpr() to radv_get_user_sgpr_info()
4384 - radv: add radv_get_user_sgpr{_loc}() helpers
4385 - radv: use radv_get_user_sgpr_loc() for the GS copy shader too
4386 - radv: remove unused parameter to dgc_emit_draw_mesh_tasks_ace()
4387 - radv: do not emit compute userdata for empty dispatches
4388 - radv: cleanup using vtx_base_sgpr for userdata with DGC
4389 - radv: use radv_dgc_with_task_shader() more
4390 - radv: move radv_CmdPreprocessGeneratedCommandsNV() to radv_cmd_buffer.c
4391 - radv: use the graphics pipeline from the DGC info
4392 - radv: use radv_get_user_sgpr() more in DGC
4393 - vulkan: Update XML and headers to 1.3.289
4394 - radv: advertise VK_KHR_maintenance7
4395 - ci: bump vkd3d-proton to 3d46c082906c77544385d10801e4c0184f0385d9
4396 - radv: remove unused parameter to radv_pipeline_import_retained_shaders()
4397 - radv: simplify importing libraries with retained shaders
4398 - radv: remove unused get_vs_output_info() function
4399 - radv: remove unnecessary radv_pipeline_has_ngg() function
4400 - radv: move radv_hash_shaders() to radv_graphics_pipeline.c
4401 - radv: simplify determining when the rasterization primitive is unknown
4402 - radv: simplify determining when a VS prolog is needed
4403 - radv: stop passing a pipeline to some graphics related helpers
4404 - radv: rework generating all graphics state for compiling pipelines
4405 - radv: remove radv_descriptor_set_layout::shader_stages
4406 - radv: use blake3 for hashing descriptor set layouts
4407 - radv: use blake3 for hashing pipeline layouts
4408 - radv: disable VK_EXT_sampler_filter_minmax on TAHITI and VERDE
4409 - ac,radeonsi: add ac_is_reduction_mode_supported()
4410 - radv: use ac_is_reduction_mode_supported()
4411 - radv: fix marking RADV_DYNAMIC_COLOR_ATTACHMENT_MAP as dirty
4412 - nir/gather_info: handle uses_fbfetch_output for sparse image loads
4413 - nir/gather_info: handle uses_fbfetch_output for texture operations
4414 - radv: destroy the perf counter BO in radv_device_finish_perf_counter()
4415 - radv: add radv_device_init_perf_counter()
4416 - radv: add helpers for init/deinit device memory cache
4417 - radv: add helpers for init/deinit RGP
4418 - radv: simplify keeping shader info for GPU hangs debugging
4419 - radv: add radv_device_init_trap_handler()
4420 - radv: add helpers for init/deinit device fault detection
4421 - radv: add radv_device_init_rmv()
4422 - radv: regroup all tools initialization in one helper
4423 - radv: use zero allocation for the device queues
4424 - radv/meta: remove non-valuable comments
4425 - radv/meta: remove unnecessary blit2d_dst_temps struct
4426 - radv/meta: remove redundant check for hw resolve pipelines
4427 - radv/meta: remove unused number of rectangles for internal operations
4428 - radv/meta: remove useless checks for NULL handles before destroying
4429 - radv/meta: add a helper to create compute pipeline
4430 - radv/meta: add a helper to create pipeline layout
4431 - radv/meta: add a helper to create descriptor set layout
4432 - zink/ci: skip arb_shader_image_load_store also on NAVI31/VANGOGH
4433 - zink/ci: remove redundant arb_shader_image_load_store skips on POLARIS10
4434 - radv: do not expose ImageFloat32AtomicMinMax on GFX11_5
4435 - radv: fix programming DB_RENDER_CONTROL for NULL depth/stencil on GFX11_5
4436 - radv: expose BufferFloat32AtomicMinMax on GFX11_5
4437 - radv: disable SPM trace on GFX11_5
4438 - ac/rgp: assume GFX11_5 use the same SQTT/RGP versions as GFX11
4439 - radv: allow to capture with RGP on GFX11_5
4440 - radv/meta: fix potential race condition when creating the copy VRS pipeline
4441 - radv/meta: rework creating the VRS copy HTILE pipeline
4442 - radv/meta: remove the depth resummarize operation
4443 - radv/meta: avoid potential NULL deref with the gfx depth decompress pipeline
4444 - radv/meta: move locking around the gfx depth decompress pipeline
4445 - radv/meta: remove unused parameter to radv_get_depth_pipeline()
4446 - radv/meta: rework creating the gfx depth decompress pipeline
4447 - radv/meta: create the compute depth decompress pipeline on-demand
4448 - radv/meta: cleanup creating the compute depth decompress pipeline
4449 - radv/meta: separate creating the fill/copy pipelines
4450 - radv/meta: create the fill/copy pipelines on-demand
4451 - radv/meta: cleanup radv_device_init_meta_blit_{color,depth,stencil]()
4452 - radv/meta: move the locking around creating blit pipelines
4453 - radv/meta: cleanup meta_emit_blit()
4454 - radv/meta: rework creating blit pipelines
4455 - radv/meta: create fmask expand layouts regardless on-demand
4456 - radv/meta: rework creating FMASK expand pipelines
4457 - radv/meta: create fmask copy layouts regardless on-demand
4458 - radv/meta: rework creating copy expand pipelines
4459 - radv/meta: fix potential race condition when creating DCC retile pipelines
4460 - radv/meta: fix potential memleak when creating DCC retile pipelines
4461 - radv/meta: rework creating DCC retile pipelines
4462 - radv/meta: remove useless memset when destroying DCC retile state
4463 - radv/meta: rework creating GFX depth/stencil resolve pipelines
4464 - radv/meta: rework creating GFX color resolve pipelines
4465 - radv/meta: rework creating compute color resolve pipelines
4466 - radv/meta: rework creating compute depth/stencil resolve pipelines
4467 - radv/meta: cleanup creating HW resolve pipelines
4468 - radv/meta: rework creating HW resolve pipelines
4469 - radv/meta: rework creating DCC decompress compute pipelines
4470 - radv/meta: rework creating clear HTILE mask pipeline
4471 - radv/meta: create clear HTILE mask pipeline on-demand when needed
4472 - radv/meta: create DCC comp-to-single pipelines on-demand when needed
4473 - radv/meta: add a helper to create itob pipelines
4474 - radv/meta: create itob pipelines on-demand when needed
4475 - radv/meta: add a helper to create btoi pipelines
4476 - radv/meta create btoi pipelines on-demand when needed
4477 - radv/meta: add a helper to create btoi r32g32b32 pipeline
4478 - radv/meta: create btoi r32g32b32 pipeline on-demand when needed
4479 - radv/meta: update the helper that creates itoi pipelines
4480 - radv/meta: create itoi pipelines on-demand when needed
4481 - radv/meta: add a helper to create itoi r32g32b32 pipeline
4482 - radv/meta: create itoi r32g32b32 pipelines on-demand when needed
4483 - radv/meta: update the helper that creates clear pipelines
4484 - radv/meta: create clear pipeliones on-demand when needed
4485 - radv/meta: add a helper to create clear r32g32b32 pipeline
4486 - radv/meta: create clear r32g32b32 pipelines on-demand when needed
4487 - radv: fix shaders cache corruption with indirect pipeline binds
4488 - radv/meta: stop checking that creating NIR shaders failed
4489 - radv/meta: remove unnecessary goto
4490 - radv/meta: stop creating similar pipeline layouts for depth decompress
4491 - radv/meta: create the layouts for blit pipelines on-demand
4492 - radv/meta: create the layouts for FS resolve pipelines on-demand
4493 - radv/meta: create the layouts for depth decompress on-demand
4494 - radv/meta: create the layouts for FMASK copy on-demand
4495 - radv/meta: create the layouts for FMASK expand on-demand
4496 - radv/meta: create the layouts for compute resolve on-demand
4497 - radv/meta: create the louts for DCC comp-to-single clear on-demand
4498 - radv/meta: rework getting clear color pipelines
4499 - radv/meta: create the layout for clear color on-demand
4500 - radv/meta: rework getting depth stencil clear pipelines
4501 - radv/meta: create the layout for clear depth/stencil on-demand
4505 - mesa: Add functions to print blake3
4506 - mesa: replace shader_info::source_sha1
4510 - util/format: add planar3 y8_u8_v8_440 pipe format
4511 - frontends/va,gallium/vl: add support for yuv440 format
4512 - radeonsi/vcn: enable yuv440 jpeg decode
4516 - vulkan/wsi/wayland: refactor wsi_wl_swapchain_wait_for_present
4520 - mr-label-maker: specialize CI labels
4521 - ci: kernel stored in a different s3 bucket
4522 - ci: identify and label S3 buckets
4523 - ci: disable Collabora's farm due to maintance
4524 - Revert "ci: disable Collabora's farm due to maintance"
4525 - ci: fix stress counter in run'n'monitor
4526 - ci: disable Collabora's farm due to maintenance
4527 - Uprev Piglit to cf8daaf5ba90fc9b8a0e144355026e2a14c79944
4528 - Revert "ci: disable Collabora's farm due to maintenance"
4529 - ci: disable Collabora's farm due to runners maintenance
4530 - Revert "ci: disable Collabora's farm due to runners maintenance"
4531 - ci: continue stress run'n'monitor
4532 - ci: Fix parse GitLab pipeline url
4533 - ci: run_n_monitor, collect and summarize
4534 - ci: disable Collabora's farm due to maintenance
4535 - Revert "ci: disable Collabora's farm due to maintenance"
4536 - ci: run_n_monitor, arguments review and unicode
4537 - ci: run_n_monitor, pretty duration with padding
4538 - ci: run_n_monitor, listing job names with a padding
4539 - ci: run_n_monitor, sort by name when listing jobs
4540 - ci: fix run_n_monitor single execution
4544 - d3d12: Fix static analysis issues due to bad parenthesis closing
4545 - nir: Mark variable as ASSERTED to fix unused variable warning treated as error
4546 - d3d12: Video Encode - Fix inputs for older OS support query cap
4547 - d3d12: Add missing case for CQP in d3d12_video_encoder_disable_rc_qualitylevels
4548 - Revert "d3d12: Video Encode - Remove PIPE_VIDEO_PROFILE_MPEG4_AVC_BASELINE as not supported" This…
4552 - glapi: fix param type in TexGenxOES
4556 - drm-uapi: Sync i915_drm.h with a78313bb206e
4557 - anv/drirc: add option to provide low latency hint
4558 - anv: Fix I915_PARAM_HAS_CONTEXT_FREQ_HINT check
4559 - intel/clflush: Utilize clflushopt in intel_invalidate_range
4563 - anv: Fix descriptor sampler offsets assignment
4564 - anv,driconf: Add fake non device local memory WA for Total War: Warhammer 3
4565 - intel/brw: Actually retype integer sources of sampler message payload
4566 - intel/elk: Actually retype integer sources of sampler message payload
4567 - mesa: Fix PopAttrib not restoring states that changed on deeper stack level
4571 - iris: change stream uploader default size to 2MB
4572 - anv: skip gfx push constants alloc optimization on gfx9/11
4573 - iris: ForceZeroRTAIndexEnable if last geom stage does not write layer id
4574 - vulkan/runtime: add a subpass bit for legacy dithering
4575 - anv: VK_EXT_legacy_dithering support
4576 - docs/features: add VK_EXT_legacy_dithering
4577 - ci: update failures list with angle for jsl, tgl
4578 - anv/android: enable emulated astc for applications
4579 - anv: implement WA 14018283232
4580 - mesa: remove some conditions in mipmap code
4581 - isl: fix condition for enabling sampler route to lsc
4582 - isl/iris/anv: provide drirc toggle intel_sampler_route_to_lsc
4583 - anv: move some pc was to batch_emit_pipe_control_write
4584 - anv: fix a cmd_buffer reference in simple shader
4588 - radv: Remove radv_queue::device again
4589 - vk_entrypoints_gen: Add missing ATTR_WEAK for instance and physdev entrypoints
4590 - vk_entrypoints_gen: Rework ATTR_WEAK to unify Unix and MinGW
4591 - vk_entrypoints_gen: Apply hidden visibility to generated symbols
4592 - vk_cmd_queue_gen: Exclude CmdDispatchGraphAMDX
4596 - nvk: advertise EXT_depth_range_unrestricted
4597 - nvk/upload_queue: fix the _fill method
4601 - amd: add GFX v11.5.2 support
4602 - amd/vpelib: support VPE IP v6.1.3
4606 - glsl: wrap nir_opt_loop in NIR_PASS()
4607 - glsl: use hash table when serializing resource data
4608 - glsl: move geom input array sizing to nir linker
4609 - lima: drop unrequired opt from standalone compiler
4610 - glsl: remove unused detect_recursion_linked()
4611 - lima: remove the standalone compiler
4612 - glsl: add support for glsl es 310/320 to standalone compiler
4613 - nir: clarify and update loop conditional instruction
4614 - nir: more aggressively remove in loop during partial unroll
4615 - nir: support more loop unrolling for logical operators
4616 - nir: add merge loop terminators optimisation
4617 - nir: add test for opt_loop_merge_terminators
4618 - nir: correctly track current loop in nir_opt_loop()
4619 - nir: test opt_loop_merge_terminators() skips unhandled loops
4620 - nir: add additional opt_loop_merge() test of deref handling
4621 - glsl: drop dump-builder support from standalone compiler
4622 - glsl: remove Par-linking from the standalone linker
4623 - glsl: remove do_function_inlining()
4624 - glsl: make glsl_to_nir() more generic
4625 - glsl: remove unused symbol table functionality
4626 - glsl: remove out of date TODO
4627 - glsl: move call to create explicit ifc layout out of glsl_to_nir
4628 - glsl: drop glsl ir optimisation from the standalone compiler
4629 - glsl: make warning tests pass linking
4630 - glsl/mesa: remove UniformHash field
4631 - glsl/standalone: init EmptyUniformLocations
4632 - glsl/tests: fix test_gl_lower_mediump
4633 - mesa: remove _mesa_get_log_file() wrapper
4634 - util/mesa: move mesa/main log code to util
4635 - mesa: add unreachable to _mesa_shader_stage_to_subroutine_prefix()
4636 - glsl: set how_declared to hidden for compiler temps
4637 - glsl: fix cross validate globals
4638 - glsl: remove out of date comment
4639 - nir: set disallow_undef_to_nan for legacy ARB asm programs
4640 - glsl: fix glsl to nir support for lower precision builtins
4641 - glsl: always copy bindless sampler packing constructors to a temp
4645 - ac/nir/esgs: Slightly refactor emitting IO loads and stores.
4646 - ac/nir/tess: Slightly refactor emitting LS outputs.
4647 - ac/nir: Add helper macros for emitting IO code.
4648 - ac/nir/esgs: Implement packed 16-bit ES->GS I/O using helper macros.
4649 - ac/nir/tess: Implement packed 16-bit LS->HS I/O using helper macros.
4650 - ac/nir/tess: Implement packed 16-bit HS->TES I/O using helper macros.
4651 - aco: Add missing nir_builder include.
4652 - ac/nir: Move some helpers to new file.
4653 - ac/nir: Add helper for pre-rasterization output info.
4654 - ac/nir/ngg: Use new pre-rasterization output info helper.
4655 - ac/nir/legacy: Use new pre-rasterization output info helper.
4656 - nir: Add nir_opt_load_store_update_alignments.
4657 - radv: Add TES num_linked_patch_inputs.
4658 - radv: Add shader stats for inputs and outputs.
4659 - radv: Fix TCS -> TES I/O linking typo of VARYING_SLOT vs. BIT.
4660 - nir/opt_varyings: Print FS VEC4 type when debugging relocate_slot.
4661 - nir/opt_varyings: Don't promote flat inputs when moving post-dominator.
4662 - ac/nir/tess: Adjust TCS->TES output mapping for linked shaders.
4663 - radv: Properly link TCS->TES IO again.
4664 - nir/lower_io: Add option to implement mediump as 32-bit.
4665 - radv: Ignore mediump IO flag.
4666 - ac/nir/tess: Only write tess factors that the TES reads.
4667 - ac/nir/tess: Fix per-patch output LDS mapping.
4668 - ac/nir/tess: Fix per-patch output VRAM mapping.
4669 - radv: Use number of TES inputs for TCS-TES linking.
4673 - etnaviv/nn: Make parallel jobs disabled by default
4674 - etnaviv: handle missing alu conversion opcodes
4678 - radv: Optimize memcpy in write_image_descriptor
4682 - intel/hang_replay: fix batch address
4686 - docs: Update VK_EXT_legacy_vertex_attributes entries
4687 - tu: Add missing VK_EXT_legacy_vertex_attributes feature
4688 - tu: Change commas to semicolons in VK_EXT_map_memory_placed features
4689 - drm-shim: Stub syncobj reset ioctl
4690 - tu: Expose VK_EXT_nested_command_buffer
4691 - freedreno/devices: Fix indentation for Adreno A32
4692 - freedreno/ci: Update expectations
4693 - wsi: Guard DRM-dependent function implementations with HAVE_LIBDRM
4694 - tu: Add support for VkBindMemoryStatusKHR
4695 - tu: Add support for NULL index buffer
4696 - tu: Add support for version 2 of all descriptor binding commands
4697 - tu: Advertise VK_KHR_maintenance6
4698 - tu: Move event related related code to tu_event.cc/h
4699 - tu: Handle all dependencies of CmdWaitEvents2
4700 - mr-label-maker: Update nouveau directories
4701 - mr-label-maker: Separate freedreno and turnip labels
4702 - tu: Handle the new sync2 flags
4703 - tu: Remove declaration of unused update_stencil_mask function
4704 - tu: Switch to vk_ycbcr_conversion
4705 - tu: Use vk_sampler
4706 - tu: Use device->vk.enabled_features instead of iterating twice
4707 - tu: Move sampler related code to tu_sampler.cc/h
4708 - tu: Drop tu_init_sampler helper function
4709 - tu: Advertise VK_KHR_shader_float_controls2
4710 - tu: Use the common version of vkGetBufferMemoryRequirements2
4711 - tu: Move buffer related code to tu_buffer.cc/h
4712 - tu: Use the common version of vkQueueBindSparse
4713 - tu: Use vk_buffer_view
4714 - tu: Drop tu_buffer_view_init helper function
4715 - tu: Move buffer view related code to tu_buffer_view.cc/h
4716 - tu: Rename tu_query.cc/h to tu_query_pool.cc/h
4717 - tu: Use the common versions of vkBegin/EndQuery()
4718 - tu: Use vk_query_pool
4719 - tu: Don't disable 2 10-bit formats
4720 - freedreno,tu,ir3: Move threadsize_base and max_waves to fd_dev_info
4721 - freedreno/ci: Use the common a6xx-skips on a750
4722 - tu: Enable VK_KHR_shader_subgroup_uniform_control_flow
4723 - tu/kgsl: Remove unused variable
4724 - vulkan/wsi: Refactor can_present_on_device
4725 - tu: Always report that we can present on kgsl
4729 - virtio/ci: separate hiden jobs to -inc.yml files
4730 - ci: add farm variable for devices in collabora farm
4731 - ci/lava: add farm in structured log files
4735 - panvk: Remove duplicate variable src_idx
4736 - panvk: Fix assert
4740 - pvr: Handle VK_STRUCTURE_TYPE_EXPORT_MEMORY_ALLOCATE_INFO
4741 - pvr: Handle VK_STRUCTURE_TYPE_IMAGE_FORMAT_LIST_CREATE_INFO
4745 - meson: Force use of LLVM ORCJIT for hosts without MCJIT support
4746 - meson: Additionally probe -mtls-dialect=desc for TLSDESC support
4750 - anv/anroid: Query gralloc for tiling mode
4754 - egl/haiku: fix double free of BBitmap
4755 - egl/haiku: fix synchronization problems, add missing header
4759 - venus: avoid client allocators for ring internals
4760 - venus: silence a stack array false alarm
4761 - venus: workaround excessive dma-buf import failure on turnip
4762 - venus: fix to destroy all pipeline handles on early error paths
4763 - meson: disallow Venus debug + LTO build via GCC
4764 - turnip: msm: clean up iova on error path
4765 - turnip: msm: fix racy gem close for re-imported dma-buf
4766 - venus: drop the workaround for excessive dma-buf import oom on turnip
4767 - turnip: virtio: fix error path in virtio_bo_init
4768 - turnip: virtio: fix iova leak upon found already imported dmabuf
4769 - turnip: virtio: fix racy gem close for re-imported dma-buf
4770 - vulkan: cast to avoid -Wswitch for Android struct beyond VkStructureType
4771 - venus: directly use vk drm and pci props in renderer info
4772 - venus: move custom props fill from GPDP2 to props init
4773 - venus: move props sanitization to a separate helper
4774 - venus: define VN_SET_VK_PROPS(_EXT) to simplify vk props init
4775 - vulkan: drop redundant core props query and copy helpers
4776 - venus: drop internal memory pools
4777 - venus: allow non-wsi image alias path to passthrough upon bind memory
4778 - ci/venus: skip a timeout test
4779 - anv: use os_get_option instead of getenv
4780 - venus: defer qfb buffer init upon query being used
4781 - venus: refactor vn_android_image_from_anb
4782 - venus: refactor to add vn_android_image_from_anb_internal
4783 - venus: support VK_ANDROID_NATIVE_BUFFER_SPEC_VERSION 8
4784 - vulkan: properly ignore unsupported feature structs
4785 - venus: tentative fix for test flakiness from invalid ring wait
4786 - venus: simplify cached mem type emulation
4787 - venus: clarify wsi image ownership
4788 - venus: fix a race condition between gem close and gem handle tracking
4789 - Revert "meson: disallow Venus debug + LTO build via GCC"
4793 - radeonsi: remove si_query_hw_ops table and call func directly
4794 - radeonsi: use reseults_end instead of unprepared to init query buffer
4795 - radeonsi: rename query_hw_ops to hw_query_ops match sw
4796 - radeonsi: add more comments in si_query.c
4800 - util: Rename DETECT_OS_UNIX to DETECT_OS_POSIX
4801 - gallivm: add lp_context_ref for combine usage of LLVMContextSetOpaquePointers
4805 - llvmpipe: add gallivm_add_global_mapping
4806 - llvmpipe: make unnamed global have internal linkage
4807 - util: detect RISC-V architecture
4808 - gallivm: add riscv support to the mattrs setting code
4809 - llvmpipe: add function name to gallivm_jit_function
4810 - llvmpipe/tests: add a new test for multiple symbols for orc jit testing
4811 - llvmpipe: add an implementation with llvm orcjit
4815 - nouveau: Fix crash when destination or source screen fences are null
4816 - nouveau/headers: Make nvk_cl**** turn to nv_push_cl****
4817 - nvk: remove NVK_MME_COPY_QUERIES
4818 - zink/query: begin time elapsed queries even if we arent in a rp
4819 - nvc0/vbo: wrap draw_vbo for multidraw performance
4820 - nv50/vbo: wrap draw_vbo to avoid ovehead from multidraw
4821 - aux/draw: Use the draw info we get passed in instead of our own
4825 - intel/brw: update comment to accurately reflect intended behavior
4826 - intel/brw: update Wa_1805992985 to use workarounds mechanism
4830 - gles1: fix GL_OES_vertex_array_object
4831 - gles1: fix glBufferSubData()
4835 - fdperf: use snprintf instead of asprintf
4836 - fdperf: select_counter() should work with a countable value
4837 - fdperf: prettify logic around the reserved CP counter
4838 - fdperf: improve reads of counter values
4839 - fdperf: simplify counter value output
4840 - freedreno: add a7xx perfcounter support
4841 - tu: fix ZPASS_DONE interference between occlusion queries and autotuner
4842 - tu: avoid memory polling in occlusion query endings using ZPASS_DONE
4843 - tu: use either the 16-bit or 32-bit descriptor
4844 - ir3_nir_opt_preamble: handle 8-bit preamble loads and stores
4845 - ir3: rework TYPE_S8 as TYPE_U8_32
4846 - tu: support KHR_8bit_storage
4847 - tu: add format feature flag checks for VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT
4848 - freedreno/drm: add mesautil dependency
4852 - nvk: Clean up unused header from libdrm_nouveau
4856 - amd/vpelib: Resolve mismatch with shader
4857 - amd/vpelib: Remove linear_0_125 TF
4858 - amd/vpelib: Remove gamma cached table
4859 - amd/vpelib: Remove support for non-linear FP16
4860 - amd/vpelib: adding blend gamma bypass
4861 - amd/vpelib: Remove checks for pitch alignment
4862 - amd/vpelib: Fix Color Adjustment Failing Test Cases
4863 - amd/vpelib: Fix blndgam bypass flag assignment
4864 - amd/vpelib: Bypass de/regam on HLG
4868 - egl/wayland: Fix direct scanout with EGL_EXT_present_opaque
4872 - frontends/va: add support for A2RGB10/X2RGB10/A2BGR10/X2BGR10
4876 - turnip: Change tu_image to use common initialization helpers
4877 - turnip: Convert tu_device_memory to use vk_device_memory
4878 - turnip: Split tu_image_init to use layout setting logic separately
4879 - turnip: Support AHardwareBuffer