1 Mesa 13.0.6 Release Notes / March 20, 2017
2 ==========================================
4 Mesa 13.0.6 is a bug fix release which fixes bugs found since the 13.0.5
7 Mesa 13.0.6 implements the OpenGL 4.4 API, but the version reported by
8 glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) /
9 glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being
10 used. Some drivers don't support all the features required in OpenGL
11 4.4. OpenGL 4.4 is **only** available if requested at context creation
12 because compatibility contexts are not supported.
19 1076590f29103f022a2cd87e6dff6ae77072013745603d06b0410c373ab2bb1a mesa-13.0.6.tar.gz
20 29ef104a7fc082d352b1599bd6cb1d040be424ccd22f5e0eb7ee9b0e9acd3597 mesa-13.0.6.tar.xz
30 - `Bug 68504 <https://bugs.freedesktop.org/show_bug.cgi?id=68504>`__ -
31 9.2-rc1 workaround for clover build failure on ppc/altivec: cannot
32 convert 'bool' to '__vector(4) \__bool int' in return
33 - `Bug 97102 <https://bugs.freedesktop.org/show_bug.cgi?id=97102>`__ -
34 [dri][swr] stack overflow / infinite loop with GALLIUM_DRIVER=swr
35 - `Bug 98869 <https://bugs.freedesktop.org/show_bug.cgi?id=98869>`__ -
36 Electronic Super Joy graphic artefacts (regression,bisected)
37 - `Bug 99401 <https://bugs.freedesktop.org/show_bug.cgi?id=99401>`__ -
38 [g33] regression: piglit.spec.!opengl 1_0.gl-1_0-beginend-coverage
39 - `Bug 99456 <https://bugs.freedesktop.org/show_bug.cgi?id=99456>`__ -
40 Firefox crashing when opening about:support with WebGL2 enabled
41 - `Bug 99677 <https://bugs.freedesktop.org/show_bug.cgi?id=99677>`__ -
42 heap-use-after-free in glsl
43 - `Bug 99715 <https://bugs.freedesktop.org/show_bug.cgi?id=99715>`__ -
44 Don't print: "Note: Buggy applications may crash, if they do please
46 - `Bug 99850 <https://bugs.freedesktop.org/show_bug.cgi?id=99850>`__ -
47 Tessellation bug on Carrizo
48 - `Bug 100049 <https://bugs.freedesktop.org/show_bug.cgi?id=100049>`__
49 - "ralloc: Make sure ralloc() allocations match malloc()'s
50 alignment." causes seg fault in 32bit build
57 - radv: Emit pending flushes before executing a secondary command
59 - radv: Flush before copying with PKT3_WRITE_DATA in CmdUpdateBuffer
63 - glsl: fix heap-buffer-overflow
65 Bas Nieuwenhuizen (8):
67 - radv: Pass CMASK alignment to application.
68 - radv: Pass DCC alignment to application.
69 - radv: Never try to create more than max_sets descriptor sets.
70 - radv: Reset emitted compute pipeline when calling secondary cmd
72 - radv: Only use PKT3_OCCLUSION_QUERY when it doesn't hang.
73 - radv: Use correct size for availability flag.
74 - radv: Disable HTILE for textures with multiple layers/levels.
75 - radv: Emit cache flushes before CP DMA.
79 - gallivm: Improve debug output (V2)
80 - gallivm: Override getHostCPUName() "generic" w/ "pwr8" (v4)
81 - gallivm: Reenable PPC VSX (v3)
85 - egl/dri3: implement query surface hook
89 - swr: Prune empty nodes in CalculateProcessorTopology.
93 - anv: fix Get*MemoryRequirements for !LLC
97 - radv: program a default point size.
98 - radv: handle transfer_write as a dst flag.
99 - radv/ac: handle nir irem opcode.
100 - radv/ac: implement txs for buffer textures.
101 - radv/ac: correctly size shared memory usage.
102 - radv/ac: avoid the fmask path when doing txs.
103 - radv: pass FMASK alignment to application
104 - tgsi: fix memory leak in tgsi sanity check
105 - radv: fix depth format in blit2d.
106 - radv: fix txs for sampler buffers
107 - radv: drop Z24 support.
108 - radv: disable mip point pre clamping.
109 - radv: setup llvm target data layout
113 - docs: add sha256 checksums for 13.0.5
114 - Revert "get-pick-list.sh: Require explicit "13.0" for nominating
116 - cherry-ignore: don't pick nir_op_pack_double optimisation fix
117 - i965: move brw_define.h ifndef guard to the top
118 - cherry-ignore: add ANV fast clears related fixes
119 - Update version to 13.0.6
123 - radv: fix the dynamic buffer index in vkCmdBindDescriptorSets
124 - radv/ac: fix multiple descriptor sets with dynamic buffers
128 - swr: Align query results allocation
130 Grazvydas Ignotas (3):
132 - r300g: only allow byteswapped formats on big endian
133 - gallium/u_queue: fix a crash with atexit handlers
134 - gallium/u_queue: set num_threads correctly if not all threads start
138 - glapi: fix typo in count_scale
142 - mesa: Don't advertise GL_OES_read_format in core profile
146 - nvc0: increase number of ubo binding points
147 - nvc0/ir: fix robustness guarantees for constbuf loads on kepler+
149 - nvc0/ir: fix ubo max clamp, reset file index
150 - gm107/ir: fix address offset bitfield for ATOMS
151 - nvc0: set the render condition in the compute object
152 - st/mesa: don't pass compare mode for stencil-sampled textures
153 - nvc0: take extra pushbuf space into account for pushbuf_space calls
154 - nvc0: increase alignment to 256 for texture buffers on fermi
158 - vulkan/wsi: Improve the DRI3 error message
162 - i965: Use a better guardband calculation.
163 - intel/blorp: Swizzle clear colors on the CPU
164 - i965/fs: Remove the inline pack_double_2x32 optimization
165 - anv: Add an invalidate_range helper
166 - anv/query: clflush the bo map on non-LLC platforms
167 - genxml: Make MI_STORE_DATA_IMM more consistent
168 - anv/query: Perform CmdResetQueryPool on the GPU
169 - blorp/exec: Use uint32_t for copying varying data
170 - intel/blorp: Explicitly flush all allocated state
171 - anv: Accurately advertise dynamic descriptor limits
172 - anv: Properly handle destroying NULL devices and instances
176 - ralloc: Make sure ralloc() allocations match malloc()'s alignment.
178 Jose Maria Casanova Crespo (1):
180 - glsl: non-last member unsized array on SSBO must fail compilation on
185 - i965: Fix fast depth clears for surfaces with a dimension of 16384.
186 - i965: Use a UW source type for CS_OPCODE_CS_TERMINATE.
187 - i965: Fix check for negative pitch in can_do_fast_copy_blit().
188 - i965: Support the force_glsl_version driconf option.
189 - i965: Combine the Gen6 SF and Clip viewport atoms.
190 - mesa: Do (TCS && !TES) draw time validation in ES as well.
191 - egl: Ensure ResetNotificationStrategy matches for shared contexts.
193 Lionel Landwerlin (3):
195 - spirv: don't assert with location decorations on non i/o variables
196 - anv: wsi: report presentation error per image request
197 - i965/fs: fix uninitialized memory access
201 - glsl: correct compute shader checks for memoryBarrier functions
205 - st/mesa: destroy pipe_context before destroying st_context (v2)
206 - radeonsi: don't invoke DCC decompression in
207 update_all_texture_descriptors
208 - radeonsi: fix UNSIGNED_BYTE index buffer fallback with non-zero start
210 - gallium/util: remove unused u_index_modify helpers
211 - gallium/u_index_modify: don't add PIPE_TRANSFER_UNSYNCHRONIZED
213 - gallium/u_queue: fix random crashes when the app calls exit()
214 - st/mesa: reset sample_mask, min_sample, and render_condition for PBO
216 - st/mesa: set blend state for PBO readbacks
217 - radeonsi: fix broken tessellation on Carrizo and Stoney
218 - radeonsi: mark all bound shader buffer ranges as initialized
222 - clover: Work around build failure with AltiVec.
226 - mesa/main: fix meta caller of \_mesa_ClampColor
227 - radeonsi: fix texture gather on stencil textures
228 - glsl: split DIV_TO_MUL_RCP into single- and double-precision flags
229 - glx/dri3: handle NULL pointers in loader-to-DRI3 drawable conversion
230 - glx/dri3: guard in_current_context against a disappeared drawable
231 - glx: guard swap-interval functions against destroyed drawables
232 - dri/common: clear the loaderPrivate pointer in driDestroyDrawable
233 - winsys/amdgpu: reduce max_alloc_size based on GTT limits
234 - radeonsi: handle MultiDrawIndirect in si_get_draw_start_count
235 - radeonsi: fix UINT/SINT clamping for 10-bit formats on <= CIK
236 - st/glsl_to_tgsi: avoid iterating past the head of the instruction
238 - st/mesa: inform the driver of framebuffer changes before compute
241 Samuel Iglesias Gonsálvez (6):
243 - glsl: fix heap-use-after-free in ast_declarator_list::hir()
244 - i965/fs: mark last DF uniform array element as 64 bit live one
245 - i965/fs: detect different bit size accesses to uniforms to push them
247 - i965/fs: fix indirect load DF uniforms on BSW/BXT
248 - i965/fs: fix source type when emitting MOV_INDIRECT to read ICP
250 - i965/fs: emit MOV_INDIRECT with the source with the right register
255 - winsys/amdgpu: avoid potential segfault in amdgpu_bo_map()