1Mesa 10.2.5 Release Notes / August 2, 2014 2========================================== 3 4Mesa 10.2.5 is a bug fix release which fixes bugs found since the 10.2.4 5release. 6 7Mesa 10.2.5 implements the OpenGL 3.3 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 113.3. OpenGL 3.3 is **only** available if requested at context creation 12because compatibility contexts are not supported. 13 14SHA256 checksums 15---------------- 16 17:: 18 19 b4459f0bf7f4a3c8fb78ece3c9d2eac3d0e5bf38cb470f2a72705e744bd0310d MesaLib-10.2.5.tar.bz2 20 7b4dd0cb683f8c7dc48a3e7a315742bed58ddcd7b756c462aca4177bd1acdc79 MesaLib-10.2.5.tar.gz 21 6180565914fb238dd77ccdaff96b6155d9a6e1b3e981ebbf6a6851301b384fed MesaLib-10.2.5.zip 22 23New features 24------------ 25 26None 27 28Bug fixes 29--------- 30 31This list is likely incomplete. 32 33- `Bug 80991 <https://bugs.freedesktop.org/show_bug.cgi?id=80991>`__ - 34 [BDW]Piglit spec_ARB_sample_shading_builtin-gl-sample-mask_2 fails 35 36Changes 37------- 38 39Abdiel Janulgue (3): 40 41- i965/fs: Refactor check for potential copy propagated instructions. 42- i965/fs: skip copy-propate for logical instructions with negated src 43 entries 44- i965/vec4: skip copy-propate for logical instructions with negated 45 src entries 46 47Adel Gadllah (1): 48 49- i915: Fix up intelInitScreen2 for DRI3 50 51Anuj Phogat (2): 52 53- i965: Fix z_offset computation in intel_miptree_unmap_depthstencil() 54- mesa: Don't use memcpy() in \_mesa_texstore() for float depth texture 55 data 56 57Brian Paul (3): 58 59- mesa: fix geometry shader memory leaks 60- st/mesa: fix geometry shader memory leak 61- gallium/u_blitter: fix some shader memory leaks 62 63Carl Worth (6): 64 65- docs: Add sha256 checksums for the 10.2.3 release 66- Update VERSION to 10.2.4 67- Add release notes for 10.2.4 68- docs: Add SHA256 checksums for the 10.2.4 release 69- cherry-ignore: Ignore a few patches picked in the previous stable 70 release 71- Update version to 10.2.5 72 73Christian König (1): 74 75- radeonsi: fix order of r600_need_dma_space and r600_context_bo_reloc 76 77Eric Anholt (1): 78 79- i965: Generalize the pixel_x/y workaround for all UW types. 80 81Ian Romanick (2): 82 83- mesa: Don't allow GL_TEXTURE_BORDER queries outside compat profile 84- mesa: Don't allow GL_TEXTURE_{LUMINANCE,INTENSITY}_\* queries outside 85 compat profile 86 87Ilia Mirkin (5): 88 89- nv50/ir: retrieve shadow compare from first arg 90- nv50/ir: ignore bias for samplerCubeShadow on nv50 91- nvc0/ir: do quadops on the right texture coordinates for TXD 92- nvc0/ir: use manual TXD when offsets are involved 93- nvc0: make sure that the local memory allocation is aligned to 0x10 94 95Jason Ekstrand (2): 96 97- main/format_pack: Fix a wrong datatype in pack_ubyte_R8G8_UNORM 98- main/get_hash_params: Add GL_SAMPLE_SHADING_ARB 99 100Jordan Justen (1): 101 102- i965: Add auxiliary surface field #defines for Broadwell. 103 104José Fonseca (1): 105 106- st/wgl: Clamp wglChoosePixelFormatARB's output nNumFormats to 107 nMaxFormats. 108 109Kenneth Graunke (13): 110 111- i965: Don't copy propagate abs into Broadwell logic instructions. 112- i965: Set execution size to 8 for instructions with force_sechalf 113 set. 114- i965/fs: Set force_uncompressed and force_sechalf on samplepos setup. 115- i965/fs: Use WE_all for gl_SampleID header register munging. 116- i965: Add plumbing for Broadwell's auxiliary surface support. 117- i965: Drop SINT workaround for CMS layout on Broadwell. 118- i965: Hook up the MCS buffers in SURFACE_STATE on Broadwell. 119- i965: Add 2x MSAA support to the MCS allocation function. 120- i965: Enable compressed multisample support (CMS) on Broadwell. 121- i965: Add missing persample_shading field to brw_wm_debug_recompile. 122- i965/fs: Fix gl_SampleID for 2x MSAA and SIMD16 mode. 123- i965/fs: Fix gl_SampleMask handling for SIMD16 on Gen8+. 124- i965/fs: Set LastRT on the final FB write on Broadwell. 125 126Marek Olšák (14): 127 128- gallium: fix u_default_transfer_inline_write for textures 129- st/mesa: fix samplerCubeShadow with bias 130- radeonsi: fix samplerCubeShadow with bias 131- radeonsi: add support for TXB2 132- r600g: switch SNORM conversion to DX and GLES behavior 133- radeonsi: fix CMASK and HTILE calculations for Hawaii 134- gallium/util: add a helper for calculating primitive count from 135 vertex count 136- radeonsi: fix a hang with instancing on Hawaii 137- radeonsi: fix a hang with streamout on Hawaii 138- winsys/radeon: fix vram_size overflow with Hawaii 139- radeonsi: fix occlusion queries on Hawaii 140- r600g,radeonsi: switch all occurences of array_size to util_max_layer 141- radeonsi: fix build because of lack of draw_indirect infrastructure 142 in 10.2 143- radeonsi: use DRAW_PREAMBLE on CIK 144 145Matt Turner (8): 146 147- i965/vec4: Don't return void from a void function. 148- i965/vec4: Don't fix_math_operand() on Gen >= 8. 149- i965/fs: Don't fix_math_operand() on Gen >= 8. 150- i965/fs: Make try_constant_propagate() static. 151- i965/fs: Constant propagate into 2-src math instructions on Gen8. 152- i965/vec4: Constant propagate into 2-src math instructions on Gen8. 153- i965/fs: Don't use brw_imm\_\* unnecessarily. 154- i965/fs: Set correct number of regs_written for MCS fetches. 155 156Thorsten Glaser (1): 157 158- nv50: fix build failure on m68k due to invalid struct alignment 159 assumptions 160 161Tom Stellard (1): 162 163- clover: Call end_query before getting timestamp result v2 164