1Mesa 24.2.1 Release Notes / 2024-08-28 2====================================== 3 4Mesa 24.2.1 is a bug fix release which fixes bugs found since the 24.2.0 release. 5 6Mesa 24.2.1 implements the OpenGL 4.6 API, but the version reported by 7glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) / 8glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used. 9Some drivers don't support all the features required in OpenGL 4.6. OpenGL 104.6 is **only** available if requested at context creation. 11Compatibility contexts may report a lower version depending on each driver. 12 13Mesa 24.2.1 implements the Vulkan 1.3 API, but the version reported by 14the apiVersion property of the VkPhysicalDeviceProperties struct 15depends on the particular driver being used. 16 17SHA checksums 18------------- 19 20:: 21 22 SHA256: fc9a495f3a9af906838be89367564e10ef335e058f88965ad49ccc3e9a3b420b mesa-24.2.1.tar.xz 23 SHA512: 3b77e5faec51b67583131123b0cc010b52325ea308e4075323102aa999d9c9fbb65b873eb537ed4f577b5a0811e7f096e7e101510cb50326ea5c439b4b468380 mesa-24.2.1.tar.xz 24 25 26New features 27------------ 28 29- None 30 31 32Bug fixes 33--------- 34 35- Artifacts on Black Myth Wukong 36- DCS world radeon rx 7900 flicker 37- radv/aco: Scheduler bug in ACO somewhere in parallel-gs 38- Once human black artifacts consume the screen 39- [ANV] Breaking Limit benchmark does not run 40- Possible race condition when generating enums.h 41- [radeonsi] regression with bindless textures (bisected) 42- [anv] possible regression found on 982106e6768e6f69d71710e314325c813591452d 43- xe2: Non-compiler assertion failure replaying cp2077 fossil 44- Mesa doesn't build with the following config options 45- Intel: Enable Compression on depth buffers (Xe2) 46- Confidential issue #11711 47- Spectacle crashes recording h264 video with 24.2-rc3 48 49 50Changes 51------- 52 53Boris Brezillon (1): 54 55- panvk: Adjust RGB component order for fixed-function blending 56 57Connor Abbott (5): 58 59- tu: Fix off-by-one in UBO CP_LOAD_STATE size 60- ir3, tu: Use a UBO for VS primitive params on a750+ 61- Revert "tu/a750: Disable HW binning when there is GS" 62- tu: Fix passthrough D24S8 attachments 63- tu: Treat partially-bound depth/stencil attachments as passthrough 64 65Dave Airlie (3): 66 67- radv/video: fix reporting video format props for encode. 68- llvmpipe: handle stride properly on lvp udmabuf imports 69- llvmpipe: make sure to duplicate the fd handle before giving out 70 71David Heidelberg (3): 72 73- etnaviv: rename enums_h appropriately 74- etnaviv: build dependency for the etnaviv tests 75- bin/gen_release_notes: adjust checksums section 76 77David Rosca (3): 78 79- frontends/va: Fix use after free with EFC 80- radeonsi: Don't allow DCC for encode in is_video_target_buffer_supported 81- frontends/va: Fix leaks with multiple coded buffer segments 82 83Eric Engestrom (13): 84 85- docs: add release notes for 24.2.0 86- docs: add sha256sum for 24.2.0 87- .pick_status.json: Update to 2c47ad7774a7d0fe47cf870676c3e2390bca5b50 88- .pick_status.json: Mark 77f783462a9758b22e211c72a57ec7c36d6c09fd as denominated 89- .pick_status.json: Update to 2b2b66f497bf8c5f91067752995a5c1003255a6f 90- .pick_status.json: Update to 47a2ab6f3fb41bef9062182588f69c217d7e6541 91- .pick_status.json: Mark a33ef21106a3b34cb359f0272a103c8b8066fbe6 as denominated 92- .pick_status.json: Update to c5156257d93ae2a9e54ba5adc73b7342c9bef96e 93- .pick_status.json: Update to 81e3930ec0a4d586752e59c5172e08e2edf5e4d5 94- .pick_status.json: Update to 4a8f3181baa1eca48d44e5270962835040d0f743 95- .pick_status.json: Update to 7392e3306ef376121097cf03b388dc75bff127c0 96- vc4: Add missing libvc4_neon build dependencies 97- .pick_status.json: Update to 64ca0fd2f2896faa2ec85e454e2ecba2f68390c8 98 99Faith Ekstrand (6): 100 101- vulkan: Add null descriptor bits to vk_pipeline_robustness_state 102- nvk: Plumb the whole vk_pipeline_robustness_state through to nvk_ubo/ssbo_addr_format 103- nvk: Enable shader bounds checking when nullDescriptor is enabled 104- nouveau/mme: Fix add64 of immediates on Fermi 105- nvk: Disable conditional rendering around CopyQueryPoolResults 106- nil,nvk: Disable modifiers for B10G11R11_UFLOAT and E5B9G9R9_UFLOAT 107 108Francisco Jerez (1): 109 110- intel/brw/gfx12.5+: Fix IR of sub-dword atomic LSC operations. 111 112Friedrich Vock (1): 113 114- aco: Fix 1D->2D dispatch conversion on <gfx9 115 116GKraats (2): 117 118- i915g: Screen corruption with ENOBUFS caused by fence register shortage 119- i915g: fix count of buffers at i915_drm_batchbuffer_validate_buffers 120 121Guilherme Gallo (1): 122 123- ci/a618: Fix zink-tu-a618-full rules 124 125Ian Romanick (2): 126 127- anv: Protect against OOB access to anv_state_pool::buckets 128- anv: Larger memory pools for huge shaders 129 130Jianxun Zhang (2): 131 132- Revert "anv: Disable PAT-based compression on depth images (xe2)" 133- Revert "iris: Disable PAT-based compression on depth surfaces (xe2)" 134 135Job Noorman (4): 136 137- ir3: update merge set affinity in shared RA 138- ir3: fix clearing merge sets after shared RA 139- ir3: fix wrong dstn used in postsched 140- ir3/legalize: handle scalar ALU WAR hazards for a0.x 141 142José Roberto de Souza (3): 143 144- intel/isl/gfx20: Alow hierarchial depth buffer write through for multi sampled surfaces 145- anv/gfx20: Enable depth buffer write through for multi sampled images 146- iris/gfx20: Enable depth buffer write through for multi sampled images 147 148Karol Herbst (3): 149 150- rusticl/mem: do not check against image base alignment for 1Dbuffer images 151- rusticl/device: limit CL_DEVICE_IMAGE_MAX_BUFFER_SIZE more aggressively 152- vtn: ignore volatile on functions for now 153 154Kenneth Graunke (2): 155 156- intel/brw: Pass opcode to brw_swsb_encode/decode 157- intel/brw: Fix Xe2+ SWSB encoding/decoding for DPAS instructions 158 159Konstantin (2): 160 161- radv: Handle instruction encodings > 8 bytes when splitting disassembly 162- radv: Handle repeated instructions when splitting disassembly 163 164Lepton Wu (1): 165 166- egl/android: Fix wrong pipe format for RGB_565 167 168Lionel Landwerlin (10): 169 170- vulkan/runtime: fix GetBufferMemoryRequirements2 for maintenance4 171- anv: fix extended buffer flags usages 172- anv: only set 3DSTATE_CLIP::MaximumVPIndex once 173- anv: optimize CLIP::MaximumVPIndex setting 174- anv: move conditional render predicate after gfx_flush_state 175- anv: don't miss workaround for indirect draws 176- anv: explicitly disable BT pool allocations at device init 177- anv: always use workaround_address, not workaround_bo 178- nir/divergence: add missing load_constant_base_ptr 179- brw: switch mesh/task URB fence prior to EOT to GPU 180 181Mary Guillemard (2): 182 183- panvk: Fix NULL deref on model name when device isn't supported 184- panvk: Fix viewport calculation 185 186Matt Turner (1): 187 188- nir: Skip opt_if_merge when next_if has block ending in a jump 189 190Mauro Rossi (1): 191 192- nvk: Fix regression observed on Kepler 193 194Mike Blumenkrantz (6): 195 196- glx/dri2: strdup driver name 197- zink: bail on choose_pdev immediately if no devices are available 198- st/pbo: reject vs/fs pbo ops if rowstride < width 199- zink: don't skip cbuf store ops if resolve is set 200- tc: set resolve on renderpass info if blit terminates the renderpass 201- dril: add zink stub 202 203Nanley Chery (4): 204 205- anv: Add want_hiz_wt_for_image() 206- iris: Add and use want_hiz_wt_for_res 207- iris: Invalidate state cache for some depth fast clears 208- intel/isl: Fix packing of SINT formats 209 210Pavel Ondračka (1): 211 212- r300: fix RGB10_A2 CONSTANT_COLOR blending 213 214Rhys Perry (7): 215 216- aco: split selection_control_remove into rarely_taken and never_taken 217- aco: only remove branch jumping over SMEM/barrier if it's never taken 218- aco: ignore exec and literals when mitigating VALUMaskWriteHazard 219- aco: also consider VALU reads for VALUMaskWriteHazard 220- aco: don't consider sa_sdst=0 before SALU write to fix VALUMaskWriteHazard 221- aco: check SALU writing lanemask later for VALUMaskWriteHazard 222- aco: preserve bitsets after a lane mask is written 223 224Rob Clark (1): 225 226- nir/opt_loop: Don't peel initial break if loop ends in break 227 228Rohan Garg (5): 229 230- anv: program a custom byte stride on Xe2 for indirect draws 231- anv,iris: prefix the argument format with XI for a upcoming refactor 232- anv: refactor indirect draw support into it's own function 233- anv: dispatch indirect draws with a count buffer through the XI hardware on ARL+ 234- anv: migrate indirect mesh draws to indirect draws on ARL+ 235 236Sagar Ghuge (3): 237 238- intel/compiler: Ray query requires write-back register 239- intel/compiler: Adjust trace ray control field on Xe2 240- intel/compiler: Fix indirect offset in GS input read for Xe2+ 241 242Samuel Pitoiset (1): 243 244- aco: fix bogus assert in RT prolog on GFX11+ 245 246Sviatoslav Peleshko (3): 247 248- brw,elk: Fix opening flags on dumping shader binaries 249- anv: Release correct BO in anv_cmd_buffer_set_ray_query_buffer 250- anv: Add full subgroups WA for the shaders with barriers in Breaking Limit 251 252Tapani Pälli (1): 253 254- gbm: depend on libdrm indepedent of dri2 setting 255 256Timothy Arceri (4): 257 258- nir: create validate_tex_src_texture_deref() helper 259- nir: add nir_tex_src_{sampler,texture}_deref_intrinsic 260- glsl: make use of new tex src deref intrinsic 261- nir/glsl: set deref cast mode during function inlining 262 263Valentine Burley (2): 264 265- android: Extract version from llvm-project instead of hardcoding it 266- llvmpipe: Only use udmabuf with libdrm 267 268Yiwei Zhang (1): 269 270- venus: workaround cacheline overflush issue on Intel JSL 271 272bbhtt (1): 273 274- pipe_loader_drm: Fix virtgpu_drm header path 275