relnotes-7.11.1.html revision b3aaafc74c15d6393b539651d2f8e4a490223cae
1<HTML> 2 3<head> 4<TITLE>Mesa Release Notes</TITLE> 5<link rel="stylesheet" type="text/css" href="mesa.css"> 6<meta http-equiv="content-type" content="text/html; charset=utf-8" /> 7</head> 8 9<BODY> 10 11<body bgcolor="#eeeeee"> 12 13<H1>Mesa 7.11.1 Release Notes / November 17, 2011</H1> 14 15<p> 16Mesa 7.11.1 is a bug fix release which fixes bugs found since the 7.11 release. 17</p> 18<p> 19Mesa 7.11 implements the OpenGL 2.1 API, but the version reported by 20glGetString(GL_VERSION) depends on the particular driver being used. 21Some drivers don't support all the features required in OpenGL 2.1. 22</p> 23<p> 24See the <a href="install.html">Compiling/Installing page</a> for prerequisites 25for DRI hardware acceleration. 26</p> 27 28 29<h2>MD5 checksums</h2> 30<pre> 31TBD 32</pre> 33 34 35<h2>New features</h2> 36<p>None.</p> 37 38<h2>Bug fixes</h2> 39 40<p>This list is likely incomplete.</p> 41 42<ul> 43<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=3165">Bug 3165</a> - texImage.IsCompressed and texImage.CompressedSize issues</li> 44 45<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=23525">Bug 23525</a> - Software rendering on QEMU guests badly broken</li> 46 47<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=28125">Bug 28125</a> - DRI2 prevents indirect glx</li> 48 49<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=34628">Bug 34628</a> - [ilk] skybox errors in quake4</li> 50 51<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=36371">Bug 36371</a> - r200: piglit readPixSanity failure</li> 52 53<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=36669">Bug 36669</a> - EmitNoMainReturn set to 1 doesn't make the GLSL compiler lower all the RET opcodes</li> 54 55<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=36939">Bug 36939</a> - multitexturing is messed up in quake wars (regression)</li> 56 57<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=37907">Bug 37907</a> - [swrast] SIGSEGV swrast/s_depth.c:569</li> 58 59<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=38163">Bug 38163</a> - Gnome Shell Display Bug</li> 60 61<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=38625">Bug 38625</a> - ast_to_hir.cpp:1761: const glsl_type* process_array_type(YYLTYPE*, const glsl_type*, ast_node*, _mesa_glsl_parse_state*): Assertion `dummy_instructions.is_empty()' failed.</li> 62 63<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=38729">Bug 38729</a> - [softpipe] sp_quad_depth_test.c:215:convert_quad_stencil: Assertion `0' failed.</li> 64 65<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=38863">Bug 38863</a> - [IVB]GPU hang when running 3D games like openarena</li> 66 67<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=39193">Bug 39193</a> - [llvmpipe and r600g] glCheckFramebufferStatusEXT segfaults in Gallium when checking status on a framebuffer bound to a texture that's bound to a pixmap</li> 68 69<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=39651">Bug 39651</a> - [glsl] Assertion failure when implicitly converting out parameters</li> 70 71<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=39991">Bug 39991</a> - [regression]GL_PALETTE8_RGBA8_OES format of glCompressedTexImage2D will cause err GL_INVALID_ENUM with GLES1.x</li> 72 73<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=40022">Bug 40022</a> - [i915] out-of-bounds write src/mesa/drivers/dri/i915/i915_fragprog.c:321</li> 74 75<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=40062">Bug 40062</a> - in etqw the strogg radar is black (regression)</li> 76 77<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=40324">Bug 40324</a> - [SNB] gpu hang in mesa 7.11</li> 78 79<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=40533">Bug 40533</a> - i915: piglit glean/readPixSanity: DRI2SwapBuffers: BadDrawable (invalid Pixmap or Window parameter)</li> 80 81<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=41096">Bug 41096</a> - [sandybridge-m-gt2+] GPU lockup render.IPEHR: 0x7a000002</li> 82 83<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=41969">Bug 41969</a> - The Mesa meta save/restore code doesn't always save the active program</li> 84 85<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=42175">Bug 42175</a> - RV730: Display errors in glxgears & WebGL</li> 86 87<li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=42268">Bug 42268</a> - [bisected] oglc pbo(negative.invalidOffsetValue) aborts on 7.11 branch</li> 88 89<!-- <li><a href="https://bugs.freedesktop.org/show_bug.cgi?id=">Bug </a> - </li> --> 90 91</ul> 92 93 94<h2>Changes</h2> 95<p>The full set of changes can be viewed by using the following GIT command:</p> 96 97<pre> 98 git log mesa-7.11..mesa-7.11.1 99</pre> 100 101<p>Adam Jackson (2): 102<ul> 103 <li>drisw: Remove cargo culting that breaks GLX 1.3 ctors</li> 104 <li>glx: Don't enable INTEL_swap_event unconditionally</li> 105</ul></p> 106 107<p>Alex Deucher (1): 108<ul> 109 <li>r600g: fix up vs export handling</li> 110</ul></p> 111 112<p>Ben Widawsky (1): 113<ul> 114 <li>intel: GetBuffer fix</li> 115</ul></p> 116 117<p>Brian Paul (15): 118<ul> 119 <li>docs: add 7.11 md5 sums</li> 120 <li>docs: news item for 7.11 release</li> 121 <li>st/mesa: Convert size assertions to conditionals in st_texture_image_copy.</li> 122 <li>softpipe: add missing stencil format case in convert_quad_stencil()</li> 123 <li>mesa: fix texstore addressing bugs for depth/stencil formats</li> 124 <li>mesa: add missing breaks for GL_TEXTURE_CUBE_MAP_SEAMLESS queries</li> 125 <li>swrast: don't try to do depth testing if there's no depth buffer</li> 126 <li>meta: fix/add checks for GL_EXT_framebuffer_sRGB</li> 127 <li>mesa: fix PACK_COLOR_5551(), PACK_COLOR_1555() macros</li> 128 <li>meta: fix broken sRGB mipmap generation</li> 129 <li>mesa: add _NEW_CURRENT_ATTRIB in _mesa_program_state_flags()</li> 130 <li>mesa: fix error handling for dlist image unpacking</li> 131 <li>mesa: generate GL_INVALID_OPERATION in glIsEnabledIndex() between Begin/End</li> 132 <li>mesa: fix incorrect error code in _mesa_FramebufferTexture1D/3DEXT()</li> 133 <li>mesa: fix format/type check in unpack_image() for bitmaps</li> 134</ul></p> 135 136<p>Carl Simonson (1): 137<ul> 138 <li>i830: Add missing vtable entry for i830 from the hiz work.</li> 139</ul></p> 140 141<p>Carl Worth (5): 142<ul> 143 <li>glcpp: Fix two (or more) successive applications of token pasting</li> 144 <li>glcpp: Test a non-function-like macro using the token paste operator</li> 145 <li>glcpp: Implement token pasting for non-function-like macros</li> 146 <li>glcpp: Raise error if defining any macro containing two consecutive underscores</li> 147 <li>glcpp: Add a test for #elif with an undefined macro.</li> 148</ul></p> 149 150<p>Chad Versace (5): 151<ul> 152 <li>glsl: Add method glsl_type::can_implicitly_convert_to()</li> 153 <li>glsl: Fix implicit conversions in non-constructor function calls</li> 154 <li>glsl: Remove ir_function.cpp:type_compare()</li> 155 <li>glsl: Fix conversions in array constructors</li> 156 <li>x86-64: Fix compile error with clang</li> 157</ul></p> 158 159<p>Chia-I Wu (3): 160<ul> 161 <li>glsl: empty declarations should be valid</li> 162 <li>intel: rename intel_extensions_es2.c to intel_extensions_es.c</li> 163 <li>intel: fix GLESv1 support</li> 164</ul></p> 165 166<p>Chris Wilson (1): 167<ul> 168 <li>i915: out-of-bounds write in calc_live_regs()</li> 169</ul></p> 170 171<p>Christopher James Halse Rogers (1): 172<ul> 173 <li>glx/dri2: Paper over errors in DRI2Connect when indirect</li> 174</ul></p> 175 176<p>David Reveman (1): 177<ul> 178 <li>i915g: Fix off-by-one in scissors.</li> 179</ul></p> 180 181<p>Eric Anholt (16): 182<ul> 183 <li>mesa: Don't skip glGetProgramEnvParam4dvARB if there was already an error.</li> 184 <li>mesa: Fix glGetUniform() type conversions.</li> 185 <li>mesa: Add support for Begin/EndConditionalRender in display lists.</li> 186 <li>mesa: Throw an error instead of asserting for condrender with query == 0.</li> 187 <li>mesa: Throw an error when starting conditional render on an active query.</li> 188 <li>mesa: Don't skip glGetProgramLocalParam4dvARB if there was already an error.</li> 189 <li>glsl: Allow ir_assignment() constructor to not specify condition.</li> 190 <li>glsl: Clarify error message about whole-array assignment in GLSL 1.10.</li> 191 <li>glsl: When assigning to a whole array, mark the array as accessed.</li> 192 <li>glsl: When assiging from a whole array, mark it as used.</li> 193 <li>i965/fs: Respect ARB_color_buffer_float clamping.</li> 194 <li>i965: Add missing _NEW_POLYGON flag to polygon stipple upload.</li> 195 <li>i965: Fix polygon stipple offset state flagging.</li> 196 <li>intel: Mark MESA_FORMAT_X8_Z24 as always supported.</li> 197 <li>mesa: Don't error on glFeedbackBuffer(size = 0, buffer = NULL)</li> 198 <li>glsl: Fix gl_NormalMatrix swizzle setup to match i965's invariants.</li> 199</ul></p> 200 201<p>Henri Verbeet (6): 202<ul> 203 <li>mesa: Also set the remaining draw buffers to GL_NONE when updating just the first buffer in _mesa_drawbuffers().</li> 204 <li>r600g: Support the PIPE_FORMAT_R16_FLOAT colorformat.</li> 205 <li>mesa: Check the texture against all units in unbind_texobj_from_texunits().</li> 206 <li>mesa: Allow sampling from units >= MAX_TEXTURE_UNITS in shaders.</li> 207 <li>mesa: Use the Elements macro for the sampler index assert in validate_samplers().</li> 208 <li>mesa: Fix a couple of TexEnv unit limits.</li> 209</ul></p> 210 211<p>Ian Romanick (17): 212<ul> 213 <li>mesa: Add utility function to get base format from a GL compressed format</li> 214 <li>mesa: Return the correct internal fmt when a generic compressed fmt was used</li> 215 <li>mesa: Make _mesa_get_compressed_formats match the texture compression specs</li> 216 <li>linker: Make linker_error set LinkStatus to false</li> 217 <li>linker: Make linker_{error,warning} generally available</li> 218 <li>mesa: Ensure that gl_shader_program::InfoLog is never NULL</li> 219 <li>ir_to_mesa: Use Add linker_error instead of fail_link</li> 220 <li>ir_to_mesa: Emit warnings instead of errors for IR that can't be lowered</li> 221 <li>i915: Fail without crashing if a Mesa IR program uses too many registers</li> 222 <li>i915: Only emit program errors when INTEL_DEBUG=wm or INTEL_DEBUG=fallbacks</li> 223 <li>mesa: Add GL_OES_compressed_paletted_texture formats to _mesa_is_compressed_format</li> 224 <li>mesa: Add GL_OES_compressed_paletted_texture formats to _mesa_base_tex_format</li> 225 <li>mesa: Refactor expected texture size check in cpal_get_info</li> 226 <li>mesa: Add _mesa_cpal_compressed_format_type</li> 227 <li>mesa: Refactor compressed texture error checks to work with paletted textures</li> 228 <li>mesa: Remove redundant compressed paletted texture error checks</li> 229 <li>mesa: Advertise GL_OES_compressed_paletted_texture in OpenGL ES1.x</li> 230</ul></p> 231 232<p>Jeremy Huddleston (3): 233<ul> 234 <li>apple: Silence some debug spew</li> 235 <li>apple: Use the correct (OpenGL.framework) glViewport and glScissor during init</li> 236 <li>apple: Implement applegl_unbind_context</li> 237</ul></p> 238 239<p>José Fonseca (1): 240<ul> 241 <li>docs: Update llvmpipe docs.</li> 242</ul></p> 243 244<p>Kenneth Graunke (12): 245<ul> 246 <li>glsl: Avoid massive ralloc_strndup overhead in S-Expression parsing.</li> 247 <li>mesa: In validate_program(), initialize errMsg for safety.</li> 248 <li>i965/gen5+: Fix incorrect miptree layout for non-power-of-two cubemaps.</li> 249 <li>i965: Use proper texture alignment units for cubemaps on Gen5+.</li> 250 <li>i965: Fix incorrect maximum PS thread count shift on Ivybridge.</li> 251 <li>i965: Emit depth stalls and flushes before changing depth state on Gen6+.</li> 252 <li>i965/fs: Allow SIMD16 with control flow on Ivybridge.</li> 253 <li>i965: Allow SIMD16 color writes on Ivybridge.</li> 254 <li>i965: Fix inconsistent indentation in brw_eu_emit.c.</li> 255 <li>intel: Depth format fixes</li> 256 <li>i965: Apply post-sync non-zero workaround to homebrew workaround.</li> 257 <li>mesa/get: Move MAX_LIGHTS from GL/ES2 to GL/ES1.</li> 258</ul></p> 259 260<p>Kristian Høgsberg (1): 261<ul> 262 <li>glx: Don't flush twice if we fallback to dri2CopySubBuffer</li> 263</ul></p> 264 265<p>Marc Pignat (1): 266<ul> 267 <li>drisw: Fix 24bpp software rendering, take 2</li> 268</ul></p> 269 270<p>Marcin Baczyński (2): 271<ul> 272 <li>configure: fix gcc version check</li> 273 <li>configure: allow C{,XX}FLAGS override</li> 274</ul></p> 275 276<p>Marcin Slusarz (3): 277<ul> 278 <li>nouveau: fix nouveau_fence leak</li> 279 <li>nouveau: fix crash during fence emission</li> 280 <li>nouveau: fix fence hang</li> 281</ul></p> 282 283<p>Marek Olšák (19): 284<ul> 285 <li>vbo: do not call _mesa_max_buffer_index in debug builds</li> 286 <li>winsys/radeon: fix space checking</li> 287 <li>r300/compiler: fix a warning that a variable may be uninitialized</li> 288 <li>r300/compiler: remove an unused-but-set variable and simplify the code</li> 289 <li>u_vbuf_mgr: cleanup original vs real vertex buffer arrays</li> 290 <li>u_vbuf_mgr: don't take per-instance attribs into acc. when computing max index</li> 291 <li>u_vbuf_mgr: fix max_index computation for large src_offset</li> 292 <li>u_vbuf_mgr: s/u_vbuf_mgr_/u_vbuf_</li> 293 <li>u_vbuf_mgr: remove unused flag U_VBUF_UPLOAD_FLUSHED</li> 294 <li>u_vbuf_mgr: rework user buffer uploads</li> 295 <li>u_vbuf_mgr: fix uploading with a non-zero index bias</li> 296 <li>configure.ac: fix xlib-based softpipe build</li> 297 <li>r600g: add index_bias to index buffer bounds</li> 298 <li>r300g: fix rendering with a non-zero index bias in draw_elements_immediate</li> 299 <li>Revert "r300g: fix rendering with a non-zero index bias in draw_elements_immediate"</li> 300 <li>pb_bufmgr_cache: flush cache when create_buffer fails and try again</li> 301 <li>r300g: don't return NULL in resource_from_handle if the resource is too small</li> 302 <li>r600g: set correct tiling flags in depth info</li> 303 <li>r300g: don't call u_trim_pipe_prim in r300_swtcl_draw_vbo</li> 304</ul></p> 305 306<p>Michel Dänzer (4): 307<ul> 308 <li>st/mesa: Finalize texture on render-to-texture.</li> 309 <li>glx/dri2: Don't call X server for SwapBuffers when there's no back buffer.</li> 310 <li>gallium/util: Add macros for converting from little endian to CPU byte order.</li> 311 <li>r300g: Fix queries on big endian hosts.</li> 312</ul></p> 313 314<p>Neil Roberts (1): 315<ul> 316 <li>meta: Fix saving the active program</li> 317</ul></p> 318 319<p>Paul Berry (18): 320<ul> 321 <li>glsl: Lower unconditional return statements.</li> 322 <li>glsl: Refactor logic for determining whether to lower return statements.</li> 323 <li>glsl: lower unconditional returns and continues in loops.</li> 324 <li>glsl: Use foreach_list in lower_jumps.cpp</li> 325 <li>glsl: In lower_jumps.cpp, lower both branches of a conditional.</li> 326 <li>glsl: Lower break instructions when necessary at the end of a loop.</li> 327 <li>glsl: improve the accuracy of the radians() builtin function</li> 328 <li>glsl: improve the accuracy of the atan(x,y) builtin function.</li> 329 <li>Revert "glsl: Skip processing the first function's body in do_dead_functions()."</li> 330 <li>glsl: Emit function signatures at toplevel, even for built-ins.</li> 331 <li>glsl: Constant-fold built-in functions before outputting IR</li> 332 <li>glsl: Check array size is const before asserting that no IR was generated.</li> 333 <li>glsl: Perform implicit type conversions on function call out parameters.</li> 334 <li>glsl: Fix type error when lowering integer divisions</li> 335 <li>glsl: Rework oversize array check for gl_TexCoord.</li> 336 <li>glsl: Remove field array_lvalue from ir_variable.</li> 337 <li>glsl hierarchical visitor: Do not overwrite base_ir for parameter lists.</li> 338 <li>glsl: improve the accuracy of the asin() builtin function.</li> 339</ul></p> 340 341<p>Tobias Droste (1): 342<ul> 343 <li>r300/compiler: simplify code in peephole_add_presub_add</li> 344</ul></p> 345 346<p>Tom Fogal (1): 347<ul> 348 <li>Only use gcc visibility support with gcc4+.</li> 349</ul></p> 350 351<p>Tom Stellard (1): 352<ul> 353 <li>r300/compiler: Fix regalloc for values with multiple writers</li> 354</ul></p> 355 356<p>Vadim Girlin (5): 357<ul> 358 <li>st/mesa: flush bitmap cache on query and conditional render boundaries</li> 359 <li>r600g: use backend mask for occlusion queries</li> 360 <li>r600g: take into account force_add_cf in pops</li> 361 <li>r600g: fix check_and_set_bank_swizzle</li> 362 <li>r600g: fix replace_gpr_with_pv_ps</li> 363</ul></p> 364 365<p>Yuanhan Liu (17): 366<ul> 367 <li>i965: fix the constant interp bitmask for flat mode</li> 368 <li>mesa: fix error handling for glEvalMesh1/2D</li> 369 <li>mesa: fix error handling for some glGet* functions</li> 370 <li>mesa: fix error handling for glTexEnv</li> 371 <li>mesa: fix error handling for glIsEnabled</li> 372 <li>mesa: fix error handling for glPixelZoom</li> 373 <li>mesa: fix error handling for glSelectBuffer</li> 374 <li>mesa: fix error handling for glMapBufferRange</li> 375 <li>mesa: fix error handling for glMaterial*</li> 376 <li>intel: fix the wrong code to detect null texture.</li> 377 <li>mesa: add a function to do the image data copy stuff for save_CompressedTex(Sub)Image</li> 378 <li>i965: setup address rounding enable bits</li> 379 <li>mesa: generate error if pbo offset is not aligned with the size of specified type</li> 380 <li>mesa: fix inverted pbo test error at _mesa_GetnCompressedTexImageARB</li> 381 <li>mesa: handle the pbo case for save_Bitmap</li> 382 <li>mesa: handle PBO access error in display list mode</li> 383 <li>intel: don't call unmap pbo if pbo is not mapped</li> 384</ul></p> 385 386</body> 387</html> 388