1Mesa 21.2.6 Release Notes / 2021-11-24 2====================================== 3 4Mesa 21.2.6 is a bug fix release which fixes bugs found since the 21.2.5 release. 5 6Mesa 21.2.6 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 21.2.6 implements the Vulkan 1.2 API, but the version reported by 14the apiVersion property of the VkPhysicalDeviceProperties struct 15depends on the particular driver being used. 16 17SHA256 checksum 18--------------- 19 20:: 21 22 1e7e22d93c6e8859fa044b1121119d26b2e67e4184b92ebb81c66497dc80c954 mesa-21.2.6.tar.xz 23 24 25New features 26------------ 27 28- None 29 30 31Bug fixes 32--------- 33 34- NIR validation failed after nir_copy_prop 35- lima: Corrupted Android-12 UI on Allwinner A64 36- RADV/ACO: Rendering glitches in Forza Horizon 5 windshields 37- dEQP-GLES31.*imulextended* compiling fp64 glsl 4.00 shader 38- Regression/Bisected: Crash in Godot games when steam overlay enabled 39- RADV: IsHelperInvocationEXT query is not considered volatile in ACO 40- Final Fantasy V (Old version): Random text characters are not displayed 41- Diagonal rendering artifacts in Tomb Raider 42- [hsw][bisected][regression] dEQP-VK.reconvergence.*nesting* failures 43 44 45Changes 46------- 47 48Connor Abbott (2): 49 50- ir3/lower_pcopy: Fix shr.b illegal copy lowering 51- ir3/lower_pcopy: Fix bug with "illegal" copies and swaps 52 53Danylo Piliaiev (1): 54 55- ir3/ra: Check register file upper bound when updating preferred_reg 56 57Dave Airlie (2): 58 59- llvmpipe: disable 64-bit integer textures. 60- llvmpipe: fix compressed image sizes. 61 62Dylan Baker (9): 63 64- docs: add sha256 sums for 21.2.5 65- .pick_status.json: Update to 8bfa146b8056d5dded6fdd2c4d8524aa0c438058 66- .pick_status.json: Update to 79f57f68930ce4fcf59bdca82357e19b4f3d62e6 67- .pick_status.json: Update to c356f3cfce9459dc1341b6a2a0fd5336a9bdcc3c 68- .pick_status.json: Mark a large number of zink patches as denominated 69- .pick_status.json: Update to 78b613db23c10dc257247d99d8b2b1ef6c96bebb 70- .pick_status.json: Update to 09bb4dbe60b1bb233c5c0102446ffbc142a29246 71- .pick_status.json: Update to e0de7aa4d7717bbd79bbc5bfc660f941f680402b 72- .pick_status.json: Update to f8009d3db25512b5dc62d0272c979fb79f8c7075 73 74Eric Engestrom (1): 75 76- meson: drop duplicate addition of surfaceless & drm to the list of platforms 77 78Greg V (1): 79 80- util: make util_get_process_exec_path work on FreeBSD w/o procfs 81 82Ian Romanick (1): 83 84- glsl/nir: Don't build soft float64 when it cannot be used 85 86Jason Ekstrand (1): 87 88- anv: Fix FlushMappedMemoryRanges for odd mmap offsets 89 90Jesse Natalie (2): 91 92- util/hash_table: Clear special 0/1 entries for u64 hash table too 93- d3d12: Fix Linux fence wait return value 94 95Kenneth Graunke (1): 96 97- iris: Fix MOCS for buffer copies 98 99Lionel Landwerlin (6): 100 101- intel/dev: fix HSW GT3 number of subslices in slice1 102- intel/dev: don't forget to set max_eu_per_subslice in generated topology 103- intel/perf: fix perf equation subslice mask generation for gfx12+ 104- intel/devinfo: fix wrong offset computation 105- intel: remove 2 preproduction pci-id for ADLS 106- intel/fs: fix shader call lowering pass 107 108Liviu Prodea (1): 109 110- aco: Work around MSVC restrict in c99_compat.h 111 112Marek Olšák (4): 113 114- radeonsi: print the border color error message only once 115- radeonsi: fix 2 issues with depth_cleared_level_mask 116- radeonsi: fix a typo preventing a fast depth-stencil clear 117- driconf: disallow 10-bit pbuffers for viewperf2020/maya due to X errors 118 119Mike Blumenkrantz (10): 120 121- zink: add better handling for CUBE_COMPATIBLE bit 122- zink: always use explicit lod for texture() when legal in non-fragment stages 123- zink: be more permissive for injecting LOD into texture() instructions 124- zink: inject LOD for sampler version of OpImageQuerySize 125- zink: don't clamp cube array surfacess to cubes 126- zink: don't clamp 2D_ARRAY surfaces to 2D 127- zink: add more int/float types to cast switching in ntv 128- zink: force float dest types on some alu results 129- aux/primconvert: handle singular incomplete restarts 130- zink: clamp PIPE_SHADER_CAP_MAX_INPUTS for xfb 131 132Pierre-Eric Pelloux-Prayer (3): 133 134- radeonsi: treat nir_intrinsic_load_constant as a VMEM operation 135- ac/surface: don't validate DCC settings if DCC isn't possible 136- mesa: always call _mesa_update_pixel 137 138Rhys Perry (2): 139 140- aco/spill: use spills_entry instead of spills_exit to kill linear VGPRs 141- spirv: run nir_copy_prop before nir_rematerialize_derefs_in_use_blocks_impl 142 143Roman Stratiienko (1): 144 145- android.mk: Add missing variables to the make target 146 147Thomas H.P. Andersen (1): 148 149- svga: fix bitwise/logical and mixup 150 151Timothy Arceri (1): 152 153- mesa: fix buffer overrun in SavedObj texture obj array 154 155Timur Kristóf (1): 156 157- aco: Fix how p_is_helper interacts with optimizations. 158 159Vadym Shovkoplias (1): 160 161- intel/fs: Fix a cmod prop bug when cmod is set to inst that doesn't support it 162