1Mesa 21.3.4 Release Notes / 2022-01-12 2====================================== 3 4Mesa 21.3.4 is a bug fix release which fixes bugs found since the 21.3.3 release. 5 6Mesa 21.3.4 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 21.3.4 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 77104fd4a93bce69da3b0982f8ee88ba7c4fb98cfc491a669894339cdcd4a67d mesa-21.3.4.tar.xz 23 24 25New features 26------------ 27 28- None 29 30 31Bug fixes 32--------- 33 34- i965: gen5 exposes EXT_texture_integer incorrectly 35- [radeonsi, regression, bisected]: Rendering issues with Factorio 36- mesa >= 21.x (with llvmpipe) will break the game JVGS 37- Panfrost - RK3399 - FACTORIO - glitches everywhere 38- Baldurs Gate 3 (Patch 6) - ribbon-like artifacts on textures 39- Game Starsector crashes under certain circumstances with mesa 21.3.+ 40- FreeSpace models incorrectly rendered on Polaris cards, causing system freeze 41- Incomplete evaluation of nested DEFINE macros 42- [r300g, bisected] piglit glsl-fs-discard-04 fails 43- Panfrost G52 Firefox terrible glitches on youtube playback 44 45 46Changes 47------- 48 49Alyssa Rosenzweig (3): 50 51- pan/bi: Fix load_const of 1-bit booleans 52- gallium/util: Add pixel->blocks box helper 53- lima,panfrost: Correct pixel vs block mismatches 54 55Bas Nieuwenhuizen (1): 56 57- radv: Set optimal copy alignment to 1. 58 59Boris Brezillon (1): 60 61- microsoft/compiler: Fix dxil_nir_create_bare_samplers() 62 63Connor Abbott (2): 64 65- ir3/ra: Fix logic bug in compress_regs_left 66- ir3: Bump type mismatch penalty to 3 67 68Daniel Schürmann (2): 69 70- aco: don't allow SDWA on VOP3P instructions 71- aco: validate VOP3P opsel correctly 72 73Danylo Piliaiev (1): 74 75- tu: fix workaround for depth bounds test without depth test 76 77Dave Airlie (2): 78 79- intel/genxml/gen4-5: fix more Raster Operation in BLT to be a uint 80- crocus: fail resource allocation properly. 81 82Emma Anholt (4): 83 84- freedreno/afuc: Disable the disassembler on 32-bit builds. 85- i915g: Turn off FP16 in the vertex shaders. 86- r300: Fix omod failing to increase the number of channels stored. 87- nir_to_tgsi: Fix a bug in TXP detection after backend lowering. 88 89Eric Engestrom (3): 90 91- .pick_status.json: Update to 8a78706643ecad8a1f303cc9358873abc29978b4 92- .pick_status.json: Mark 00bea38242d97e7ace1954f1bc7d32cbf0ce3ee0 as denominated 93- .pick_status.json: Mark 2a0253b9b5d21c7571555abe3a1d851468a18740 as denominated 94 95Filip Gawin (1): 96 97- r300: fix handling swizzle in transform_source_conflicts 98 99Henry Goffin (1): 100 101- intel/compiler/test: Fix build with GCC 7 102 103Ian Romanick (1): 104 105- i965: Disable EXT_texture_integer on Gfx4 and Gfx5 106 107Jason Ekstrand (1): 108 109- Revert "anv: Stop doing too much per-sample shading" 110 111Lionel Landwerlin (3): 112 113- anv: don't leave anv_batch fields undefined 114- anv: limit compiler valid color outputs using NIR variables 115- intel/dev: fixup chv workaround 116 117Lucas Stach (3): 118 119- etnaviv: initialize vertex attributes on context reset 120- etnaviv: drm: fix size limit in etna_cmd_stream_realloc 121- etnaviv: drm: properly handle reviving BOs via a lookup 122 123Michel Zou (1): 124 125- zink: fix -Warray-bounds warning 126 127Mike Blumenkrantz (4): 128 129- radv: fix xfb query copy param ordering 130- zink: always unset vertex shader variant key data when changing last vertex stage 131- zink: add extra synchronization for buffer descriptor binds 132- zink: use device-local heap for sparse backing allocations 133 134Pavel Ondračka (1): 135 136- r300: Remove broken optimization in rc_transform_KILL 137 138Pierre-Eric Pelloux-Prayer (4): 139 140- radeonsi/gfx8: use the proper dcc clear size 141- vbo/dlist: fix loopback crash 142- vbo/dlist: add vertices to incomplete primitives 143- radeonsi/gfx10: fix si_texture_get_offset for mipmapped tex 144 145Qiang Yu (1): 146 147- glapi: should not add alias function to static_data.py 148 149Rhys Perry (1): 150 151- aco: remove pack_half_2x16(a, 0) optimization 152 153Rohan Garg (1): 154 155- intel/fs: OpImageQueryLod does not support arrayed images as an operand 156 157Roman Stratiienko (1): 158 159- v3dv: Hotfix: Rename remaining V3DV_HAS_SURFACE->V3DV_USE_WSI_PLATFORM 160 161Samuel Pitoiset (1): 162 163- radv: add drirc radv_disable_htile_layers and enable it for F1 2021 164 165Tapani Pälli (3): 166 167- iris: unref syncobjs and free r/w dependencies array for slab entries 168- mesa: free idalloc storage for display lists 169- mesa: free vbo_save_vertex_list store prims 170 171Timothy Arceri (1): 172 173- glsl/glcpp: make sure to expand new token after concatenation 174 175Yiwei Zhang (1): 176 177- venus: subtract appended header size in vn_CreatePipelineCache 178 179satmandu (1): 180 181- Fix compilation on armv7l with gcc 11.2.0 182