1Mesa 22.3.7 Release Notes / 2023-03-08 2====================================== 3 4Mesa 22.3.7 is a bug fix release which fixes bugs found since the 22.3.6 release. 5 6Mesa 22.3.7 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 22.3.7 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 17SHA256 checksum 18--------------- 19 20:: 21 22 894ce2f4a1c2e76177cdd2284620192d0da3066b243eec2fbb1d7cf37f13042c mesa-22.3.7.tar.xz 23 24 25New features 26------------ 27 28- None 29 30 31Bug fixes 32--------- 33 34- radv: A Plague Tale: Requiem black "flash" on 7900XTX 35- 7900 XTX: Graphical corruption / artifacts in Cyberpunk 36- radv: CmdCopyQueryPoolResults broken for VK_QUERY_TYPE_PRIMITIVES_GENERATED_EXT with queryCount > 1 37- rusticl over llvmpipe + ffmpeg's Opencl filter = error -51 38- System freeze when playing some h264 videos with VA-API on Rembrandt 39- radv: GTA IV graphical artifacts on 7900XTX 40- radv: Resident Evil Revelations 2 artifacts on 7900XTX with DCC 41- radv: Prototype 2 black textures on RDNA 3 when DCC is enabled 42- hasvk: Black pixels with 8xMSAA and fast clears on Intel(R) HD Graphics 4400 (HSW GT2) 43- ACO: dEQP-VK.binding_model.descriptor_buffer.multiple.graphics_geom_buffers1_sets3_imm_samplers hangs on NAVI10 44- Build failures with recent lld 45- [gen9atom] Vulkan tests cause gpu hang: dEQP-VK.memory_model.* 46- mesa: "Fragmented" dynamic lights in IronWail with \`r_fsaamode 1` on 47- GL_SHADER_BINARY_FORMAT_SPIR_V is not added to the list of GL_SHADER_BINARY_FORMATS even if GL_ARB_gl_spirv is supported. 48- [ANV/DG2] Vertex explosion in nvpro-samples/vk_raytracing_tutorial_KHR/ray_tracing_gltf 49- CUEtools FLACCL hit assert in rusticl 50- Assertion Failed on Intel HD 5500 with Linux / Mesa 22.3.1 / OpenGL 51 52 53Changes 54------- 55 56Caio Oliveira (2): 57 58- iris, crocus: Align workaround address to 32B 59- gallivm: Fix handling of nir_intrinsic_scoped_barrier 60 61Dave Airlie (1): 62 63- llvmpipe: fix compute address bits to return native pointer size. 64 65Dylan Baker (1): 66 67- util: rzalloc and free hash_table_u64 68 69Eric Engestrom (6): 70 71- docs/relnotes: add sha256sum for 22.3.6 72- .pick_status.json: Update to 41ae2d0725b3ecd2018a6105981a6db7b87c3e79 73- radv: split linker script for android since it requires different symbols 74- glapi/meson: drop duplicate line in deps 75- .pick_status.json: Mark 646cff13bca8a92b846984d782ef00e57d34d7a1 as denominated 76- .pick_status.json: Mark bb796f70a144e9c88ab46e6fb08be5895ca7a689 as denominated 77 78Friedrich Vock (1): 79 80- mesa: Report GL_SHADER_BINARY_FORMAT_SPIR_V as supported 81 82Gert Wollny (1): 83 84- r600/sfn: Fix atomic lowering 85 86Iván Briano (3): 87 88- vulkan: track the right value on CmdSetColorWriteMasks 89- anv: fix testing for dynamic color blend bits 90- anv: stop tracking color blend state in the pipeline 91 92Karol Herbst (2): 93 94- nir/deref: don't replace casts with deref_struct if we'd lose the stride 95- rusticl/kernel: Images arg sizes also have to match the host pointer size 96 97Lionel Landwerlin (7): 98 99- genxml: Fix STATE_BASE_ADDRESS::BindlessSurfaceStateSize field size 100- genxml: fix border color offset field on Gfx12+ 101- anv: fixup condition for Wa_14016118574 102- anv: remove pre hasvk split assert 103- anv: pull Wa_14016118574 out of some loop not changing state 104- anv: fix incorrect parameter 105- nir: fix nir_ishl_imm 106 107Marek Olšák (6): 108 109- amd: add missing gfx11 register definitions 110- radeonsi/gfx11: fix the CU_EN clear mask for RSRC4_GS 111- radeonsi/gfx11: don't set non-existent VGT_STRMOUT_BUFFER_CONFIG 112- radeonsi/gfx11: set CB_COLORi_INFO.MAX_COMP_FRAG on GFX1103_R2 113- radeonsi: disable Smart Access Memory because CPU access has large overhead 114- radeonsi: don't merge SET_* packets that have a different index in si_pm4_state 115 116Mike Blumenkrantz (9): 117 118- zink: fix shader read access removal for barrier generation 119- zink: avoid adding ubo/ssbo bindings multiple times for different bitsizes 120- zink: remove suspended queries from list before resuming 121- zink: disable queries for clear_texture() 122- zink: fix zink_resource_access_is_write() 123- zink: unbind fb on context destroy 124- zink: only add deferred barrier on fb unbind when layout needs to change 125- zink: fix descriptor update flagging on null ssbo set 126- zink: propagate valid_buffer_range when replacing buffer storage 127 128Pierre-Eric Pelloux-Prayer (1): 129 130- radeonsi: fix incorrect vgpr indices in the ps_prolog 131 132Rhys Perry (2): 133 134- aco: fix pathological case in LdsDirectVALUHazard 135- aco: always update orig_names in get_reg_phi() 136 137Ruijing Dong (1): 138 139- raseonsi/vcn: fix a h264 decoding issue 140 141Sam Edwards (1): 142 143- nouveau: Fix null dereference in nouveau_pushbuf_destroy 144 145Samuel Pitoiset (9): 146 147- radv: fix flushing non-coherent images inside secondaries on GFX9+ 148- radv: fix flushing non-coherent images in EndCommandBuffer() 149- radv: fix draw calls with 0-sized index buffers and robustness on NAVI10 150- amd,ac/rgp: fix SQTT memory types 151- radv: fix DCC decompress on GFX11 152- radv: fix incorrect stride for primitives generated query with GDS 153- radv: fix border color swizzle for stencil-only format on GFX9+ 154- radv: fix defining RADV_USE_WSI_PLATFORM 155- radv: disable DCC with signedness reinterpretation on GFX11 156 157Sil Vilerino (3): 158 159- d3d12: Fix VP9 Decode - Checking 0xFF instead of 0x7F for invalid frame_ref[i].Index7Bits 160- frontend/va: Keep track of some VP9 previous frame data for current frame use_prev_in_find_mvs_refs 161- d3d12: VP9 Decode - Fix use_prev_in_find_mvs_refs calculation 162 163Sviatoslav Peleshko (3): 164 165- iris: Avoid creating uncompressed view with unaligned tile offsets on BDW 166- anv: Handle all fields in VkAccelerationStructureBuildRangeInfoKHR 167- anv: Move WA MEDIA_VFE_STATE after stalling PIPE_CONTROL 168 169Tatsuyuki Ishi (2): 170 171- radeonsi: SDMA v4 size field is size - 1 172- radv: SDMA v4 size field is size - 1 173 174Väinö Mäkelä (2): 175 176- hasvk: Mark VK_IMAGE_LAYOUT_ATTACHMENT_OPTIMAL as stencil write optimal 177- hasvk: Disable non-zero fast clears for 8xMSAA images 178 179Yiwei Zhang (1): 180 181- venus: fix VK_EXT_image_view_min_lod feature query 182 183Yogesh Mohan Marimuthu (2): 184 185- wsi/display: check alloc failure in wsi_display_alloc_connector() 186- ac/surface: only adjust pitch if surf_pitch was modified 187