1Mesa 22.1.2 Release Notes / 2022-06-16 2====================================== 3 4Mesa 22.1.2 is a bug fix release which fixes bugs found since the 22.1.1 release. 5 6Mesa 22.1.2 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 22.1.2 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 0971226b4a6a3d10cfc255736b33e4017e18c14c9db1e53863ac1f8ae0deb9ea mesa-22.1.2.tar.xz 23 24 25New features 26------------ 27 28- None 29 30 31Bug fixes 32--------- 33 34- Random texture flickering in Minecraft 35- panfrost: Broken interaction between CRC and tile size 36- [anv, bisected, regression] Pipeline statistics query segfaults 37- d3d12: point coord regression due to NIR IO changes 38- Gallium driver crashes due to fd double close in iris driver 39- Mesa currently FTBFS on x32 architecture 40- Runescape randomly crashes. malloc_consolidate(): unaligned fastbin chunk 41 42 43Changes 44------- 45 46Alyssa Rosenzweig (7): 47 48- panfrost: Fix XML for AFBC header on v9 49- pan/bi: Fix clper_xor on Mali-G31 50- pan/mdg: Scalarize with 64-bit sources 51- pan/mdg: Remove uppercase write masks 52- panfrost: Fix inverted indirect dispatch condition 53- panfrost: Inline pan_fbd_has_zs_crc_ext 54- panfrost: Disable CRC at <16x16 tile sizes 55 56Boris Brezillon (5): 57 58- microsoft/spirv_to_dxil: Fix push_constants type declaration 59- microsoft/spirv_to_dxil: Fix the push_constant UBO size calculation 60- nir: Fix flat new_var assignment in create_new_io_vars() 61- vulkan/util: Fill VkPhysicalDeviceIDProperties::deviceNodeMask 62- microsoft/compiler: Images are no longer reprensented by uniform vars 63 64Dave Airlie (1): 65 66- llvmpipe: handle timespec overflow on fence waits. 67 68Dylan Baker (7): 69 70- docs: Add sh256sums for 22.1.1 71- .pick_status.json: Update to 1c63a0beafefccec599ddbe355ec5a8b61df6438 72- .pick_status.json: Update to ad5c84999b9eb2522cdf78bb0890939450280cf3 73- .pick_status.json: Update to 49234585772f49626435d3d9324ad144e27fc453 74- .pick_status.json: Update to 35dd4ac886283444294238ae0be1a7edcc776c8a 75- .pick_status.json: Update to 2ef6b0aab1c7b99b85f2d88ced44a63f28cfca6b 76- .pick_status.json: Update to ea33e44ce4f95ce45729b75dee34be50df749f09 77 78Emma Anholt (1): 79 80- nouveau/nv30: Make sure fsat is lowered in the VS. 81 82Erik Faye-Lund (6): 83 84- nir: introduce and use nir_component_mask 85- freedreno: remove stale makefile 86- intel: remove stale makefile 87- ci/windows: add back build-error detection 88- dzn: handle stencil-attachment-optimal layout 89- gallium: remove stale comment 90 91Georg Lehmann (1): 92 93- zink: Use VK_USE_64_BIT_PTR_DEFINES to check for 64bit platforms. 94 95Iago Toral Quiroga (1): 96 97- broadcom/compiler: disable flags optimization for loop conditions 98 99Ian Romanick (1): 100 101- anv: Remove FS executables when applying the null FS optimization 102 103Jason Ekstrand (2): 104 105- nir/builder: Add a nir_trim_vector helper 106- compiler/types: Don't place members in the previous substruct's end padding 107 108Jesse Natalie (2): 109 110- nir: Consider PNTC to be a varying 111- d3d12: Properly set HS input control point count 112 113Juan A. Suarez Romero (1): 114 115- v3d: save only required states in blitter 116 117Kenneth Graunke (1): 118 119- intel/compiler: Fix uncompaction of signed word immediates on Tigerlake 120 121Kyle Russell (1): 122 123- etnaviv: don't halve max_instructions for GC400 124 125Marek Olšák (2): 126 127- ac/gpu_info: always retile DCC on gfx10 and newer chips 128- gallium/u_threaded: fix buffer busyness tracking 129 130Mike Blumenkrantz (31): 131 132- st/pbo_compute: make compute download conditional in shader slightly more readable 133- st/pbo_compute: fix z coords for compute pbos 134- zink: remove first_frame stalling 135- lavapipe: fix dynamic patchControlPoints 136- zink: force queries to start/end out of renderpass 137- egl/wayland: manually swap backbuffer when using zink 138- egl/wayland: skip buffer creation on zink 139- kopper: use get_drawable_info path for non-x11 drawables 140- zink: fix dmabuf plane layout struct scoping 141- format_utils: properly parenthesize macro params 142- zink: always create descriptor templates 143- zink: fix ntv partial stores 144- zink: fix variable locations in manual xfb emission 145- zink: scalarize when rewriting explicit 64bit io 146- zink: use separate pointer for swapchain presents 147- zink: keep a count of async presents occuring for a given swapchain 148- zink: defer old swapchain destruction 149- zink: fix framebuffer attachment usage asserts for dummy attachments 150- zink: only update layout when doing mixed zs attachment renderpass check 151- zink: add implicit sync workaround for non-mesa drivers 152- zink: unset res->swapchain upon killing a swapchain 153- zink: fix up KILL to a more sensible log message 154- zink: fix generated tcs update path 155- zink: unset generated tcs pointer from tes on deletion 156- zink: fix generated tcs deletion 157- zink: avoid uninit values in renderpass state 158- zink: don't print VK_EXT_shader_atomic_float warning for nir_intrinsic_image_deref_atomic_exchange 159- zink: fix cubemap lowering for array samplers 160- zink: fix cubemap lowering bit size 161- zink: flag optimization pass after inlining uniforms 162- zink: fix 32bit bo rewriting 163 164Nagappa Koppad, Basanagouda (1): 165 166- iris:Duplicate DRM fd internally instead of reuse. 167 168Pierre-Eric Pelloux-Prayer (5): 169 170- st/mesa: use mutex in st_get_texture_sampler_view_from_stobj 171- gallium: add size attribute to winsys_handle 172- va/surface: set the correct size in vaExportSurfaceHandle 173- radeonsi: prevent recursion in si_decompress_dcc 174- radeonsi: add helper to use si_screen::aux_context 175 176Qiang Yu (1): 177 178- mesa/program: fix nir output reg overflow 179 180Rhys Perry (2): 181 182- aco: fix SMEM load_global_amd with non-zero offset 183- aco: fix SMEM load_global with VGPR address and non-zero offset 184 185Samuel Pitoiset (1): 186 187- radv: enable radv_zero_vram for Hammerting 188