1Mesa 17.2.6 Release Notes / November 25, 2017 2============================================= 3 4Mesa 17.2.6 is a bug fix release which fixes bugs found since the 17.2.5 5release. 6 7Mesa 17.2.6 implements the OpenGL 4.5 API, but the version reported by 8glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) / 9glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being 10used. Some drivers don't support all the features required in OpenGL 114.5. OpenGL 4.5 is **only** available if requested at context creation 12because compatibility contexts are not supported. 13 14SHA256 checksums 15---------------- 16 17:: 18 19 a9ed76702ffb14ad674ad48899f5c8c7e3a0f987911878a5dfdc4117dce5b415 mesa-17.2.6.tar.gz 20 6ad85224620330be26ab68c8fc78381b12b38b610ade2db8716b38faaa8f30de mesa-17.2.6.tar.xz 21 22New features 23------------ 24 25None 26 27Bug fixes 28--------- 29 30- `Bug 100438 <https://bugs.freedesktop.org/show_bug.cgi?id=100438>`__ 31 - glsl/ir.cpp:1376: 32 ir_dereference_variable::ir_dereference_variable(ir_variable*): 33 Assertion \`var != NULL' failed. 34- `Bug 102177 <https://bugs.freedesktop.org/show_bug.cgi?id=102177>`__ 35 - [SKL] ES31-CTS.core.sepshaderobjs.StateInteraction fails 36 sporadically 37- `Bug 103115 <https://bugs.freedesktop.org/show_bug.cgi?id=103115>`__ 38 - [BSW BXT GLK] 39 dEQP-VK.spirv_assembly.instruction.compute.sconvert.int32_to_int64 40- `Bug 103519 <https://bugs.freedesktop.org/show_bug.cgi?id=103519>`__ 41 - wayland egl apps crash on start with mesa 17.2 42- `Bug 103529 <https://bugs.freedesktop.org/show_bug.cgi?id=103529>`__ 43 - [GM45] GPU hang with mpv fullscreen (bisected) 44- `Bug 103628 <https://bugs.freedesktop.org/show_bug.cgi?id=103628>`__ 45 - [BXT, GLK, BSW] KHR-GL46.shader_ballot_tests.ShaderBallotBitmasks 46- `Bug 103787 <https://bugs.freedesktop.org/show_bug.cgi?id=103787>`__ 47 - [BDW,BSW] gpu hang on 48 spec.arb_pipeline_statistics_query.arb_pipeline_statistics_query-comp 49 50Changes 51------- 52 53Adam Jackson (2): 54 55- glx/drisw: Fix glXMakeCurrent(dpy, None, ctx) 56- glx/dri3: Fix passing renderType into glXCreateContext 57 58Alex Smith (2): 59 60- spirv: Use correct type for sampled images 61- nir/spirv: tg4 requires a sampler 62 63Andres Gomez (14): 64 65- docs: add sha256 checksums for 17.2.5 66- cherry-ignore: intel/fs: Use a pure vertical stride for large 67 register strides 68- cherry-ignore: intel/nir: Use the correct indirect lowering masks in 69 link_shaders 70- cherry-ignore: intel/fs: Use the original destination region for int 71 MUL lowering 72- cherry-ignore: intel/fs: refactors 73- cherry-ignore: r600/shader: reserve first register of vertex shader. 74- cherry-ignore: anv/cmd_buffer: Advance the address when initializing 75 clear colors 76- cherry-ignore: anv/cmd_buffer: Take bo_offset into account in fast 77 clear state addresses 78- cherry-ignore: i965: Mark BOs as external when we export their handle 79- cherry-ignore: added 17.3 nominations. 80- cherry-ignore: glsl: Fix typo fragement -> fragment 81- cherry-ignore: egl: pass the dri2_dpy to the $plat_teardown functions 82- cherry-ignore: Revert "intel/fs: Use a pure vertical stride for large 83 register strides" 84- Update version to 17.2.6 85 86Anuj Phogat (2): 87 88- i965: Program DWord Length in MI_FLUSH_DW 89- i965/gen8+: Fix the number of dwords programmed in MI_FLUSH_DW 90 91Bas Nieuwenhuizen (2): 92 93- radv: Free syncobj with multiple imports. 94- radv: Free temporary syncobj after waiting on it. 95 96Dave Airlie (1): 97 98- r600: fix isoline tess factor component swapping. 99 100Derek Foreman (1): 101 102- egl/wayland: Add a fallback when fourcc query isn't supported 103 104Dylan Baker (1): 105 106- autotools: Set C++ visibility flags on Intel 107 108Emil Velikov (3): 109 110- targets/opencl: don't hardcode the icd file install to /etc/... 111- configure.ac: loosen --enable-glvnd check to honour egl 112- configure.ac: require xcb\* for the omx/va/... when using x11 113 platform 114 115George Barrett (1): 116 117- glsl: Catch subscripted calls to undeclared subroutines 118 119Jason Ekstrand (9): 120 121- intel/fs: Use ANY/ALL32 predicates in SIMD32 122- intel/fs: Use an explicit D type for vote any/all/eq intrinsics 123- intel/fs: Use a pair of 1-wide MOVs instead of SEL for any/all 124- intel/eu/reg: Add a subscript() helper 125- intel/fs: Fix MOV_INDIRECT for 64-bit values on little-core 126- intel/fs: Fix integer multiplication lowering for src/dst hazards 127- intel/fs: Mark 64-bit values as being contiguous 128- intel/fs: Rework zero-length URB write handling 129- i965: Add stencil buffers to cache set regardless of stencil 130 texturing 131 132Kenneth Graunke (5): 133 134- i965: properly initialize brw->cs.base.stage to MESA_SHADER_COMPUTE 135- i965: Make L3 configuration atom listen for TCS/TES program updates. 136- intel/tools: Fix detection of enabled shader stages. 137- i965: Implement another VF cache invalidate workaround on Gen8+. 138- i965: Upload invariant state once at the start of the batch on 139 Gen4-5. 140 141Matt Turner (2): 142 143- i965/fs: Fix extract_i8/u8 to a 64-bit destination 144- i965/fs: Split all 32->64-bit MOVs on CHV, BXT, GLK 145 146Neil Roberts (1): 147 148- glsl: Transform fb buffers are only active if a variable uses them 149 150Nicolai Hähnle (1): 151 152- ddebug: fix use-after-free of streamout targets 153 154Tim Rowley (2): 155 156- swr/rast: Use gather instruction for i32gather_ps on simd16/avx512 157- swr/rast: Faster emulated simd16 permute 158 159Timothy Arceri (3): 160 161- glsl: drop cache_fallback 162- glsl: use the correct parent when allocating program data members 163- mesa: rework how we free gl_shader_program_data 164