1Mesa 7.10 Release Notes / January 7, 2011 2========================================= 3 4Mesa 7.10 is a new development release. People who are concerned with 5stability and reliability should stick with a previous release or wait 6for Mesa 7.10.1. 7 8Mesa 7.10 implements the OpenGL 2.1 API, but the version reported by 9glGetString(GL_VERSION) depends on the particular driver being used. 10Some drivers don't support all the features required in OpenGL 2.1. 11 12See the :doc:`Compiling/Installing page <../install>` for 13prerequisites for DRI hardware acceleration. 14 15MD5 checksums 16------------- 17 18:: 19 20 0a70c15c135561824bdcae92bf232e43 MesaLib-7.10.tar.gz 21 33fb94eccc02cbb4d8d1365615e38e46 MesaLib-7.10.tar.bz2 22 5cafdc0eda0f9bf370b95c98df3338fa MesaLib-7.10.zip 23 bc644be551ed585fc4f66c16b64a91c9 MesaGLUT-7.10.tar.gz 24 5c2677a155672352d62b177e4f0f92e8 MesaGLUT-7.10.tar.bz2 25 2ce5001f74496d1ba719ef74d910a5cf MesaGLUT-7.10.zip 26 27New features 28------------ 29 30- GL_ARB_explicit_attrib_location extension (Intel and software 31 drivers). 32- GL_ARB_texture_rg (Intel, software drivers, gallium drivers). 33- GL_EXT_separate_shader_objects extension (Intel and software 34 drivers). 35- GL_NV_primitive_restart extension (Gallium softpipe, llvmpipe). 36- New fragment shader back-end for i965-class hardware. 37- Support for Sandybridge chipset in i965 DRI driver. 38 39Bug fixes 40--------- 41 42This list is likely incomplete. 43 44- `Bug 28800 <https://bugs.freedesktop.org/show_bug.cgi?id=28800>`__ - 45 [r300c, r300g] Texture corruption with World of Warcraft 46- `Bug 29420 <https://bugs.freedesktop.org/show_bug.cgi?id=29420>`__ - 47 Amnesia / HPL2 RendererFeatTest - not rendering correctly 48- `Bug 29946 <https://bugs.freedesktop.org/show_bug.cgi?id=29946>`__ - 49 [swrast] piglit valgrind glsl-array-bounds-04 fails 50- `Bug 30261 <https://bugs.freedesktop.org/show_bug.cgi?id=30261>`__ - 51 [GLSL 1.20] allowing inconsistent invariant declaration between two 52 vertex shaders 53- `Bug 30632 <https://bugs.freedesktop.org/show_bug.cgi?id=30632>`__ - 54 [softpipe] state_tracker/st_manager.c:489: 55 st_context_notify_invalid_framebuffer: Assertion \`stfb && 56 stfb->iface == stfbi' failed. 57- `Bug 30694 <https://bugs.freedesktop.org/show_bug.cgi?id=30694>`__ - 58 wincopy will crash on Gallium drivers when going to front buffer 59- `Bug 30771 <https://bugs.freedesktop.org/show_bug.cgi?id=30771>`__ - 60 [r600g] vert-tex glsl demo 61- `Bug 30787 <https://bugs.freedesktop.org/show_bug.cgi?id=30787>`__ - 62 Invalid asm shader does not generate draw-time error when used with 63 GLSL shader 64- `Bug 30974 <https://bugs.freedesktop.org/show_bug.cgi?id=30974>`__ - 65 [llvmpipe] SIGABRT src/gallium/drivers/llvmpipe/lp_state_fs.c:779 66- `Bug 30993 <https://bugs.freedesktop.org/show_bug.cgi?id=30993>`__ - 67 getFramebufferAttachmentParameteriv wrongly generates error 68- `Bug 31101 <https://bugs.freedesktop.org/show_bug.cgi?id=31101>`__ - 69 [glsl2] abort() in ir_validate::visit_enter(ir_assignment \*ir) 70- `Bug 31193 <https://bugs.freedesktop.org/show_bug.cgi?id=31193>`__ - 71 [regression] aa43176e break water reflections 72- `Bug 31194 <https://bugs.freedesktop.org/show_bug.cgi?id=31194>`__ - 73 The mesa meta save/restore code doesn't ref the current GLSL program 74- `Bug 31371 <https://bugs.freedesktop.org/show_bug.cgi?id=31371>`__ - 75 glslparsertest: ir.cpp:358: ir_constant::ir_constant(const 76 glsl_type*, const ir_constant_data*): Assertion \`(type->base_type >= 77 0) && (type->base_type <= 3)' failed. 78- `Bug 31439 <https://bugs.freedesktop.org/show_bug.cgi?id=31439>`__ - 79 Crash in glBufferSubData() with size == 0 80- `Bug 31495 <https://bugs.freedesktop.org/show_bug.cgi?id=31495>`__ - 81 [i965 gles2c bisected] OpenGL ES 2.0 conformance 82 GL2Tests_GetBIFD_input.run regressed 83- `Bug 31514 <https://bugs.freedesktop.org/show_bug.cgi?id=31514>`__ - 84 isBuffer returns true for unbound buffers 85- `Bug 31560 <https://bugs.freedesktop.org/show_bug.cgi?id=31560>`__ - 86 [tdfx] tdfx_tex.c:702: error: ‘const struct gl_color_table’ has no 87 member named ‘Format’ 88- `Bug 31617 <https://bugs.freedesktop.org/show_bug.cgi?id=31617>`__ - 89 Radeon/Compiz: 'failed to attach dri2 front buffer', error case not 90 handled 91- `Bug 31648 <https://bugs.freedesktop.org/show_bug.cgi?id=31648>`__ - 92 [GLSL] array-struct-array gets assertion: \`(size >= 1) && (size <= 93 4)' failed. 94- `Bug 31650 <https://bugs.freedesktop.org/show_bug.cgi?id=31650>`__ - 95 [GLSL] varying gl_TexCoord fails to be re-declared to different size 96 in the second shader 97- `Bug 31673 <https://bugs.freedesktop.org/show_bug.cgi?id=31673>`__ - 98 GL_FRAGMENT_PRECISION_HIGH preprocessor macro undefined in GLSL ES 99- `Bug 31690 <https://bugs.freedesktop.org/show_bug.cgi?id=31690>`__ - 100 i915 shader compiler fails to flatten if in Aquarium webgl demo. 101- `Bug 31832 <https://bugs.freedesktop.org/show_bug.cgi?id=31832>`__ - 102 [i915] Bad renderbuffer format: 21 103- `Bug 31841 <https://bugs.freedesktop.org/show_bug.cgi?id=31841>`__ - 104 [drm:radeon_cs_ioctl] \*ERROR\* Invalid command stream ! 105- `Bug 31894 <https://bugs.freedesktop.org/show_bug.cgi?id=31894>`__ - 106 Writing to gl_PointSize with GLES2 corrupts other varyings 107- `Bug 31909 <https://bugs.freedesktop.org/show_bug.cgi?id=31909>`__ - 108 [i965] brw_fs.cpp:1461: void 109 fs_visitor::emit_bool_to_cond_code(ir_rvalue*): Assertion 110 \`expr->operands[i]->type->is_scalar()' failed. 111- `Bug 31934 <https://bugs.freedesktop.org/show_bug.cgi?id=31934>`__ - 112 [gallium] Mapping empty buffer object causes SIGSEGV 113- `Bug 31983 <https://bugs.freedesktop.org/show_bug.cgi?id=31983>`__ - 114 [i915 gles2] "if (expression with builtin/varying variables) discard" 115 breaks linkage 116- `Bug 31985 <https://bugs.freedesktop.org/show_bug.cgi?id=31985>`__ - 117 [GLSL 1.20] initialized uniform array considered as "unsized" 118- `Bug 31987 <https://bugs.freedesktop.org/show_bug.cgi?id=31987>`__ - 119 [gles2] if input a wrong pname(GL_NONE) to glGetBoolean, it will not 120 case GL_INVALID_ENUM 121- `Bug 32035 <https://bugs.freedesktop.org/show_bug.cgi?id=32035>`__ - 122 [GLSL bisected] comparing unsized array gets segfault 123- `Bug 32070 <https://bugs.freedesktop.org/show_bug.cgi?id=32070>`__ - 124 llvmpipe renders stencil demo incorrectly 125- `Bug 32273 <https://bugs.freedesktop.org/show_bug.cgi?id=32273>`__ - 126 assertion fails when starting vdrift 2010 release with shaders 127 enabled 128- `Bug 32287 <https://bugs.freedesktop.org/show_bug.cgi?id=32287>`__ - 129 [bisected GLSL] float-int failure 130- `Bug 32311 <https://bugs.freedesktop.org/show_bug.cgi?id=32311>`__ - 131 [965 bisected] Array look-ups broken on GM45 132- `Bug 32520 <https://bugs.freedesktop.org/show_bug.cgi?id=32520>`__ - 133 [gles2] glBlendFunc(GL_ZERO, GL_DST_COLOR) will result in 134 GL_INVALID_ENUM 135- `Bug 32825 <https://bugs.freedesktop.org/show_bug.cgi?id=32825>`__ - 136 egl_glx driver completely broken in 7.9 branch [fix in master] 137 138Changes 139------- 140 141Adam Jackson (2): 142 143- i965: Update renderer strings for sandybridge 144- drivers/x11: unifdef XFree86Server 145 146Alex Deucher (30): 147 148- r600c: fix mipmap stride on evergreen 149- r600c: add reloc for CB_COLOR0_ATTRIB 150- r600c: pull over 6xx/7xx vertex fixes for evergreen 151- r600c: fix segfault in evergreen stencil code 152- r100: revalidate after radeon_update_renderbuffers 153- r600c: add missing radeon_prepare_render() call on evergreen 154- r600c: properly align mipmaps to group size 155- egl_dri2: Add radeon chip ids 156- r600c/evergreen: texture align is group_bytes just like 6xx/7xx 157- r600g: fix buffer alignment 158- r600g: All EVENT_WRITE packets need the EVENT_INDEX field 159- r600g: translate ARR instruction for evergreen 160- r600g: use meaningful defines for chiprev 161- r600g: use full range of VS resources for vertex samplers 162- r600g: fix additional EVENT_WRITE packet 163- r600g: fix some winsys functions to deal properly with evergreen 164- r600c: add Ontario Fusion APU support 165- r600g: add support for ontario APUs 166- r600c: fix VC flush on cedar and palm 167- gallium/egl: fix r300 vs r600 loading 168- r600c: fix some opcodes on evergreen 169- r600c: bump texture limits to hw limits 170- r600g: bump texture/cb limits appropriately for evergreen 171- radeon: bump mip tree levels to 15 172- r600g: fix rendering with a vertex attrib having a zero stride 173- r600g: remove useless switch statements 174- r600g: add support for NI (northern islands) asics 175- r600c: add support for NI asics 176- r600g: support up to 64 shader constants 177- r600c: fix up SQ setup in blit code for Ontario/NI 178 179Andre Maasikas (3): 180 181- r600c: fix buffer height setting in dri2 case 182- r600g: break alu clause earlier 183- r600g: fix evergreen interpolation setup 184 185Andrew Randrianasulu (2): 186 187- dri/nv04: Don't expose ARB_texture_env_combine/dot3. 188- dri/nv04: Enable eng3dm for A8/L8 textures. 189 190Aras Pranckevicius (2): 191 192- glsl: fix crash in loop analysis when some controls can't be 193 determined 194- glsl: fix matrix type check in ir_algebraic 195 196Bas Nieuwenhuizen (3): 197 198- r600g: set ENABLE_KILL in the shader state in the new design 199- r600g: set ENABLE_KILL on evergreen too 200- r600g: use dirty list to track dirty blocks 201 202Ben Skeggs (3): 203 204- nv50: DST 205- nv50: DPH 206- nv50: silence some unknown get_param warnings 207 208Benjamin Franzke (2): 209 210- st/egl image: multiply drm buf-stride with blocksize 211- r600g: implement texture_get_handle (needed for 212 eglExportDRMImageMESA) 213 214Brian Paul (296): 215 216- glx: add const qualifiers to \__indirect_glMultiDrawArraysEXT() 217- glsl2: fix signed/unsigned comparison warning 218- llvmpipe: cast to silence warning 219- llvmpipe: s/boolean/unsigned/ in bitfield to silence warning 220- nv50: use unsigned int for bitfields to silence warnings 221- tgsi: fix incorrect usage_mask for shadow tex instructions 222- gallivm: expand AoS sampling to cover all filtering modes 223- gallivm: fix incorrect vector shuffle datatype 224- gallivm: move i32_vec_type inside the #ifdef 225- mesa: include mfeatures.h in formats.c 226- gallivm: fix wrong return value in bitwise functions 227- tgsi/sse: fix aos_to_soa() loop to handle num_inputs==0 228- gallivm: added missing case for PIPE_TEXTURE_RECT 229- gallium: better docs for pipe_rasterizer_state::sprite_coord_enable 230- gallium: rework handling of sprite_coord_enable state 231- gallium/docs: added new pipeline.txt diagram 232- mesa: don't call valid_texture_object() in non-debug builds 233- glsl2: silence compiler warnings in printf() calls 234- docs: remove old broken link 235- docs: mark as obsolete, remove dead links 236- llvmpipe: fix query bug when no there's no scene 237- gallivm: remove debug code 238- llvmpipe: maintain fragment shader state for draw module 239- llvmpipe: indentation fix 240- llvmpipe: reformatting, remove trailing whitespace, etc 241- llvmpipe: clean-up, comments in setup_point_coefficient() 242- llvmpipe: rename sprite field, add sprite_coord_origin 243- llvmpipe: implement sprite coord origin modes 244- draw: fix test for using the wide-point stage 245- llvmpipe: check bitshift against PIPE_MAX_SHADER_OUTPUTS 246- draw: check bitshift against PIPE_MAX_SHADER_OUTPUS 247- Merge branch 'sprite-coord' 248- draw: new draw_fs.[ch] files 249- glsl2: fix typo in error msg 250- gallivm: fix lp_build_sample_compare() 251- softpipe: add missing calls to set draw vertex samplers/views 252- mesa: don't advertise bogus GL_ARB_shading_language_120 extension 253- configs: remove egl-swrast target from linux-dri config 254- llvmpipe: fix sprite texcoord setup for non-projective texturing 255- mesa: fix assertions to handle srgb formats 256- st/mesa: add missing MESA_FORMAT_S8 case in 257 st_mesa_format_to_pipe_format() 258- st/mesa: use the wrapped renderbuffer in CopyPixels() 259- llvmpipe: make min/max lod and lod bias dynamic state 260- llvmpipe: make texture border_color dynamic state 261- softpipe: fix repeat() function for NPOT textures 262- gallivm: fix repeat() function for NPOT textures 263- swrast: update comments for REMAINDER() macro 264- softpipe: rename sp_state_fs.c -> sp_state_shader.c 265- softpipe: make shader-related functions static 266- softpipe: make blend/stencil/depth functions static 267- softpipe: make sampler state functions static 268- softpipe: make vertex state functions static 269- softpipe: make rasterizer state functions static 270- softpipe: make stream out state functions static 271- softpipe: make clip state functions static 272- softpipe: minor asst. clean-ups 273- softpipe: allocate tile data on demand 274- llvmpipe: fix swizzling of texture border color 275- softpipe: fix swizzling of texture border color 276- draw: pass sampler state down to llvm jit state 277- gallivm: check for level=0 case in lp_build_minify() 278- gallivm: added some comments 279- draw: check for null sampler pointers 280- swrast: fix choose_depth_texture_level() to respect mipmap filtering 281 state 282- st/mesa: replace assertion w/ conditional in framebuffer invalidation 283- glsl2: fix signed/unsigned comparison warning 284- st/xlib: add some comments 285- ir_to_mesa: assorted clean-ups, const qualifiers, new comments 286- mesa: remove assertion w/ undeclared variable texelBytes 287- gallivm: remove newlines 288- draw/llvmpipe: replace DRAW_MAX_TEXTURE_LEVELS with 289 PIPE_MAX_TEXTURE_LEVELS 290- mesa: reformatting, comments, code movement 291- x11: fix breakage from gl_config::visualType removal 292- gallivm: work-around trilinear mipmap filtering regression with LLVM 293 2.8 294- mesa: remove post-convolution width/height vars 295- gallivm: add compile-time option to emit inst addrs and/or line 296 numbers 297- llvmpipe: code to dump bytecode to file (disabled) 298- gallivm: added lp_build_print_vec4() 299- gallivm: added lp_build_load_volatile() 300- glsl: add ir_unop_round_even case to silence unhandled enum warning 301- st/mesa: fix regressions in glDrawPixels(GL_STENCIL_INDEX) 302- st/mesa: reformatting in st_cb_drawpixels.c 303- st/mesa: use GLuint to avoid problem w/ uint not defined on mingw32 304- st/mesa: update function name, comments 305- gallivm: use util_snprintf() 306- llvmpipe: remove lp_setup_coef*.c files from Makefile 307- mesa: fix mesa version string construction 308- gallivm: fix incorrect type for zero vector in emit_kilp() 309- llvmpipe/draw: always enable LLVMAddInstructionCombiningPass() 310- draw: use float version of LLVM Mul/Add instructions 311- draw: fix typo in comment 312- mesa: add GL_RG case to \_mesa_source_buffer_exists() 313- mesa: add missing cases for packing red/green images 314- st/mesa: added cases for GL_COMPRESSED_RED/RG in st_choose_format() 315- docs: update texture red/green support in GL3.txt 316- docs: add GL_ARB_texture_rg to release notes 317- mesa: driver hook for primitive restart 318- mesa: set/get primitive restart state 319- mesa: API spec for primitive restart 320- mesa: regenerated files with primitive restart 321- mesa: plug in primitive restart function 322- vbo: support for primitive restart 323- gallium: new CAP, state for primitive restart 324- st/mesa: support for primitive restart 325- draw: implement primitive splitting for primitive restart 326- softpipe: enable primitive restart 327- llvmpipe: enable primitive restart 328- docs: added GL_NV_primitive_restart extension 329- Merge branch 'primitive-restart-cleanup' 330- winsys/xlib: formatting fixes 331- winsys/xlib: use Bool type for shm field 332- winsys/xlib: fix up allocation/dealloction of XImage 333- winsys/xlib: rename xm->xlib 334- galahad: silence warnings 335- mesa: move declaration before code 336- docs: updated GL3 status for primitive restart 337- mesa: 80-column wrapping 338- mesa: simplify fbo format checking code 339- mesa: split up the image.c file 340- mesa: add pixel packing for unscaled integer types 341- mesa: \_mesa_ClearColorIuiEXT() and \_mesa_ClearColorIiEXT() 342- mesa: \_mesa_is_format_integer() function 343- mesa: minor reformatting, clean-ups 344- mesa: added \_mesa_is_fragment_shader_active() helper 345- mesa: new glDrawPixels error check for integer formats 346- softpipe: added some texture sample debug code (disabled) 347- mesa: added new gl_extensions::EXT_gpu_shader4 field 348- mesa: added new gl_framebuffer::_IntegerColor field 349- mesa: added glGet query for GL_RGBA_INTEGER_MODE_EXT 350- mesa: compute \_IntegerColor field in 351 \_mesa_test_framebuffer_completeness() 352- mesa: added cases for GL_EXT_texture_integer formats 353- mesa: added cases for GL_EXT_texture_integer 354- st/mesa: add format selection for signed/unsigned integer formats 355- mesa: simplify target_can_be_compressed() function 356- glapi: GL_EXT_texture_integer API 357- glapi: include/build EXT_texture_integer.xml 358- mesa: regenerated API files for GL_EXT_texture_integer 359- mesa: plug in GL_EXT_texture_integer functions 360- mesa: display list support for GL_EXT_texture_integer 361- st/mesa: be smarter choosing texture format for glDrawPixels() 362- softpipe: remove >32bpp color restriction 363- mesa: silence enum comparison warning 364- mesa: fix uninitialized var warning 365- xlib: silence unused var warning 366- util: use pointer_to_func() to silence warning 367- rtasm: use pointer_to_func() to silence warning 368- translate: use function typedefs, casts to silence warnings 369- translate: remove unused prototypes 370- mesa: additional glReadPixels error checks for GL_EXT_texture_integer 371- mesa: additional switch cases for GL_EXT_texture_integer 372- mesa: additional teximage error checks for GL_EXT_texture_integer 373- mesa: do integer FB / shader validation check in 374 \_mesa_valid_to_render() 375- mesa: call \_mesa_valid_to_render() in glDrawPixels, glCopyPixels, 376 glBitmap 377- mesa: remove the unused \_mesa_is_fragment_shader_active() function 378- mesa: fix bug in \_mesa_is_format_integer() 379- mesa: rename function to \_mesa_is_format_integer_color() 380- mesa: remove 'normalized' parameter from 381 \_mesa_VertexAttribIPointer() 382- vbo: re-indent file 383- glapi: xml spec file for GL_EXT_gpu_shader4 384- glapi: include EXT_gpu_shader4.xml 385- glapi: regenerated API files 386- mesa: plug in stubs for glBindFragDataLocation(), 387 glGetFragDataLocation() 388- mesa: add glGetUniformuiv(), plug in uint glUniform funcs 389- mesa: plug in more GL_EXT_gpu_shader4 functions 390- mesa: add new GLvertexformat entries for integer-valued attributes 391- mesa: implement integer-valued vertex attribute functions 392- mesa: add gl_client_array::Integer field and related vertex array 393 state code 394- mesa: consolidate glVertex/Color/etcPointer() code 395- mesa: state/queries for GL_MIN/MAX_PROGRAM_TEXEL_OFFSET_EXT 396- mesa: glArrayElement support for integer-valued arrays 397- mesa: clean-up array element code 398- mesa: add extension table entry for GL_EXT_gpu_shader4 399- mesa: remove obsolete comment 400- mesa: fix incorrect type in \_mesa_texstore_rgba_int16() 401- mesa: fix integer cases in \_mesa_is_legal_format_and_type() 402- mesa: add const qualifier to \_mesa_is_legal_format_and_type() 403- mesa: additional integer formats in \_mesa_bytes_per_pixel() 404- mesa: pixel transfer ops do not apply to integer-valued textures 405- mesa: remove dead code 406- osmesa: fix renderbuffer memleak in OSMesaMakeCurrent() 407- mesa: use GLubyte for edge flag arrays 408- mesa: move the gl_config struct declaration 409- dri/util: add a bunch of comments 410- mesa: remove always-false conditional in check_compatible() 411- mesa: fix aux/accum comment and error message mixups 412- llvmpipe: assign context's frag shader pointer before using it 413- llvmpipe: add a cast 414- mesa: silence new warnings in texobj.c 415- egl/gdi: fix typo: xsurf->gsurf 416- mesa: code to unpack RGBA as uints 417- gallivm: implement scatter stores into temp register file 418- gallivm: add some LLVM var labels 419- gallivm: added debug code to dump temp registers 420- gallivm: add pixel offsets in scatter stores 421- gallivm: added lp_elem_type() 422- gallivm: implement execution mask for scatter stores 423- tgsi: remove unused function 424- llvmpipe: added some debug assertions, but disabled 425- gallivm: alloca() was called too often for temporary arrays 426- gallivm: add const qualifiers, fix comment string 427- softpipe: disable vertex texturing with draw/llvm 428- mesa: consolidate pixel packing/unpacking code 429- mesa: rename vars in pixel pack/unpack code 430- mesa: implement uint texstore code 431- mesa: remove stray GL_FLOAT case in \_mesa_is_legal_format_and_type() 432- mesa: make fixed-pt and byte-valued arrays a runtime feature 433- softpipe: can't no-op depth test stage when occlusion query is 434 enabled 435- mesa: no-op glBufferSubData() on size==0 436- mesa: #include mfeatures.h in enums.h 437- mesa: improve error message 438- mesa: add missing formats in \_mesa_format_to_type_and_comps() 439- mesa: handle more pixel types in mipmap generation code 440- mesa: make glIsBuffer() return false for never bound buffers 441- mesa: fix glDeleteBuffers() regression 442- tdfx: s/Format/_BaseFormat/ 443- mesa: consolidate assertions in teximage code 444- radeon: set gl_texture_image::TexFormat field in 445 radeonSetTexBuffer2() 446- r600: set gl_texture_image::TexFormat field in r600SetTexBuffer2() 447- r200: set gl_texture_image::TexFormat field in r200SetTexBuffer2() 448- r300: set gl_texture_image::TexFormat field in r300SetTexBuffer2() 449- evergreen: set gl_texture_image::TexFormat field in 450 evergreenSetTexBuffer() 451- st/mesa: fix glDrawPixels(depth/stencil) bugs 452- glsl: fix assorted MSVC warnings 453- mesa: add more work-arounds for acoshf(), asinhf(), atahf() 454- glsl: remove opt_constant_expression.cpp from SConscript 455- mesa: fix error messages and minor reindenting 456- mesa: whitespace cleanups 457- mesa: 80-column wrapping 458- mesa: reorder texture_error_check() params 459- mesa: minor clean-ups in context code 460- mesa: upgrade to glext.h version 66 461- mesa: pass gl_format to \_mesa_init_teximage_fields() 462- mesa: fix error msg typo 463- glapi: rename GL3.xml to GL3x.xml as it covers all GL 3.x versions 464- mesa: hook up GL 3.x entrypoints 465- docs: update some GL 3.0 status 466- mesa: fix get_texture_dimensions() for texture array targets 467- swrast: init alpha value to 1.0 in opt_sample_rgb_2d() 468- glsl: fix off by one in register index assertion 469- glsl: use gl_register_file in a few places 470- mesa: rename, make \_mesa_register_file_name() non-static 471- mesa: \_mesa_valid_register_index() to validate register indexes 472- mesa: replace #defines with new gl_shader_type enum 473- mesa: use gl_shader_type enum 474- glsl: better handling of linker failures 475- glsl: start restoring some geometry shader code 476- mesa: add assertion and update comment in \_mesa_format_image_size() 477- mesa: added \_mesa_format_image_size64() 478- x11: remove test_proxy_teximage() function 479- st/mesa: fix mapping of zero-sized buffer objects 480- gallivm/llvmpipe: squash merge of the llvm-context branch 481- mesa: raise max texture sizes to 16K 482- softpipe: increase max texture size to 16K 483- mesa: replace large/MAX_WIDTH stack allocations with heap allocations 484- mesa: replace large/MAX_WIDTH stack allocations with heap allocations 485- swrast: avoid large stack allocations in blend code 486- swrast: avoid large stack allocations in tex combine code 487- st/mesa: avoid large stack allocations in readpixels code 488- mesa: replace more MAX_WIDTH stack allocations with heap allocations 489- gallivm/llvmpipe: remove lp_build_context::builder 490- gallivm: fix null builder pointers 491- mesa: fix GL_FRAMEBUFFER_ATTACHMENT_OBJECT_NAME query 492- mesa: return GL_FRAMEBUFFER_DEFAULT as FBO attachment type 493- llvmpipe: fix broken stencil writemask 494- mesa: consolidate some compiler -D flags 495- swrast: allow GL_RG format in glDrawPixels() 496- swrast: fix indentation 497- swrast: accept GL_RG in glReadPixels() 498- swrast: restructure some glReadPixels() code 499- mesa: make glGet*(GL_NONE) generate GL_INVALID_ENUM 500- mesa: remove unneeded cast 501- mesa: update comments, remove dead code 502- st/mesa: new comment about updating state vars 503- mesa: add error margin to clip mask debug/check code 504- gallium/util: minor formatting fixes 505- mesa/llvm: use llvm-config --cppflags 506- st/mesa: fix mipmap generation bug 507- mesa: test for cube map completeness in glGenerateMipmap() 508- mesa: set gl_texture_object::_Complete=FALSE in incomplete() 509- mesa: consolidate glTexImage1/2/3D() code 510- mesa: simplify proxy texture code in texture_error_check() 511- mesa: consolidate the glTexSubImage1/2/3D() functions 512- mesa: consolidate glCopyTexImage1/2D() code 513- mesa: consolidate glCopyTexSubImage1/2/3D() functions 514- mesa: consolidate glCompressedTexImage1/2/3D() functions 515- mesa: make \_mesa_test_proxy_teximage() easier to read 516- configure: use llvm-config --cppflags instead of --cflags 517- mesa: revamp error checking for compressed texture images 518- mesa: simplify target checking for TexImage functions 519- draw/llvm: don't flush in vs_llvm_delete() 520- tnl: Initialize gl_program_machine memory in run_vp. 521- tnl: a better way to initialize the gl_program_machine memory 522- mesa, st/mesa: disable GL_ARB_geometry_shader4 523- mesa/meta: fix broken assertion, rename stack depth var 524- glsl: new glsl_strtod() wrapper to fix decimal point interpretation 525- st/mesa: fix renderbuffer pointer check in st_Clear() 526 527Brian Rogers (1): 528 529- mesa: Add missing else in do_row_3D 530 531Chad Versace (25): 532 533- intel_extensions: Add ability to set GLSL version via environment 534- glsl: Add glsl_type::uvecN_type for N=2,3 535- glsl: Add lexer rules for uint and uvecN (N=2..4) 536- glsl: Changes in generated file glsl_lexer.cpp 537- glsl: Add lexer rules for << and >> in GLSL 1.30 538- glsl: Change generated file glsl_lexer.cpp 539- glsl: Implement ast-to-hir for binary shifts in GLSL 1.30 540- glsl: Implement constant expr evaluation for bitwise-not 541- glsl: Implement constant expr evaluation for bit-shift ops 542- glsl: Implement constant expr evaluation for bitwise logic ops 543- glsl: Fix ir validation for bit logic ops 544- glsl: Define shift_result_type() in ast_to_hir.cpp 545- glsl: Implement ast-to-hir for bit-shift-assignment 546- glsl: Define bit_logic_result_type() in ast_to_hir.cpp 547- glsl: Implement ast-to-hir for bit-logic ops 548- glsl: Fix lexer rule for ^= 549- glsl: Commit generated file glsl_lexer.cpp 550- glsl: Fix ast-to-hir for ARB_fragment_coord_conventions 551- mesa: Fix C++ includes in sampler.cpp 552- glsl: Fix ir_expression::constant_expression_value() 553- glsl: Fix erroneous cast in ast_jump_statement::hir() 554- glsl: Fix Doxygen tag file in recently renamed files 555- glsl: Improve usage message for glsl_compiler 556- glsl: Fix linker bug in cross_validate_globals() 557- glsl: In ast_to_hir, check sampler array indexing 558 559Chia-I Wu (149): 560 561- glapi: Fix build errors for ES. 562- glapi: Fix ES build errors again. 563- mesa: Update ES APIspec.xml. 564- st/xlib: Notify the context when the front/back buffers are swapped. 565- targets/egl: Use C++ compiler to link GL/ES state trackers. 566- libgl-xlib: Remove unused st_api_create_OpenGL. 567- st/egl: Split modeset code support to modeset.c. 568- st/egl: Rename kms backend to drm. 569- st/egl: s/kms/drm/ on the drm backend. 570- egl: Enable drm platform by default. 571- egl: Check extensions. 572- st/egl: Skip single-buffered configs in EGL. 573- mapi: Fix compiler warnings. 574- st/egl: Drop context argument from egl_g3d_get_egl_image. 575- targets/egl: Fix linking with libdrm. 576- st/vega: Fix version check in context creation. 577- egl: Use attribute names as the \_EGLConfig member names. 578- egl: Access config attributes directly. 579- st/egl: Access \_EGLConfig directly. 580- st/egl: Do not finish a fence that is NULL. 581- mesa: Remove unused vtxfmt_tmp.h. 582- egl_dri2: Drop the use of \_egl[SG]etConfigKey. 583- egl_glx: Drop the use of [SG]ET_CONFIG_ATTRIB. 584- egl_glx: Fix borken driver. 585- egl: Move attributes in \_EGLImage to \_EGLImageAttribs. 586- egl: Parse image attributes with \_eglParseImageAttribList. 587- egl: Move fallback routines to eglfallbacks.c. 588- egl: Drop dpy argument from the link functions. 589- egl: Minor changes to the \_EGLConfig interface. 590- egl: Minor changes to the \_EGLScreen interface. 591- egl: Fix \_eglModeLookup. 592- st/egl: Fix native_mode refresh mode. 593- egl: Add reference count for resources. 594- egl: Use reference counting to replace IsLinked or IsBound. 595- egl: Fix a false negative check in \_eglCheckMakeCurrent. 596- st/egl: Use resource reference count for egl_g3d_sync. 597- egl_dri2: Fix a typo that make glFlush be called at wrong time. 598- glapi: Do not use glapidispatch.h. 599- glapi: Move glapidispatch.h to core mesa. 600- glapi: Do not use glapioffsets.h. 601- glapi: Merge glapioffsets.h into glapidispath.h. 602- vbo: Use CALL\_\* macros. 603- mesa: Remove unnecessary glapitable.h includes. 604- autoconf: Better client API selection. 605- docs: Update egl and openvg docs. 606- autoconf: Update configuration info. 607- Merge branch 'glapi-reorg' 608- targets: Add missing quotes to Makefile.xorg. 609- autoconf: st/vega requires --enable-openvg. 610- st/mesa: Unreference the sampler view in st_bind_surface. 611- autoconf: Tidy configure output for EGL. 612- targets/egl: Fix a warning with --disable-opengl build. 613- egl: Rework \_eglGetSearchPath. 614- mesa: Select FEATURE_remap_table when multiple APIs are enabled. 615- mesa: Allow contexts of different APIs to coexist. 616- egl: Set up the pthread key even TLS is used. 617- st/egl: Add native_surface::present callback. 618- st/egl: Use native_surface::present callback. 619- d3d1x: Use native_surface::present. 620- st/egl: Remove flush_frontbuffer and swap_buffers. 621- st/egl: Add support for swap interval and swap behavior. 622- st/egl: Add support for EGL_MATCH_NATIVE_PIXMAP. 623- st/egl: Add extern "C" wrapper to native.h. 624- st/egl: Add native_display_buffer interface. 625- st/egl: Use native_display_buffer for EGL_MESA_drm_image. 626- autoconf: Add --enable-gallium-egl. 627- docs: Update egl docs. 628- st/dri: Add support for surfaceless current contexts. 629- egl_dri2: Fix \__DRI_DRI2 version 1 support. 630- st/vega: Do not wait NULL fences. 631- gallium: Add st_api::name. 632- gallium: Add st_context_iface::share to st_api. 633- st/wgl: Use st_context_iface::share for DrvShareLists. 634- st/glx: Replace MESA_VERSION_STRING by xmesa_get_name. 635- mesa: Clean up core.h. 636- scons: Define IN_DRI_DRIVER. 637- tgsi: Add STENCIL to text parser. 638- st/vega: vegaLookupSingle should validate the state. 639- st/vega: Set wrap_r for mask and blend samplers. 640- st/vega: Fix vgReadPixels with a subrectangle. 641- egl_dri2: Fix one context, multiple surfaces. 642- auxiliary: util_blit_pixels_tex should restore the viewport. 643- st/vega: Fix a crash with empty paths. 644- st/vega: Masks and surfaces should share orientation. 645- st/vega: No flipping in vg_prepare_blend_surface. 646- st/vega: Fix a typo in EXTENDED_BLENDER_OVER_FUNC. 647- llvmpipe: Fix build errors on x86. 648- st/vega: Overhaul renderer with renderer states. 649- st/vega: Add DRAWTEX renderer state. 650- st/vega: Add SCISSOR renderer state. 651- st/vega: Add CLEAR renderer state for vgClear. 652- st/vega: Add FILTER renderer state for image filtering. 653- st/vega: Use the renderer for vgMask. 654- st/vega: Add POLYGON_STENCIL and POLYGON_FILL renderer state. 655- st/vega: Delay fb state update to vg_validate_state. 656- st/vega: Use st_framebuffer for fb width/height. 657- st/vega: Move g3d states to renderer. 658- st/vega: Make shader_bind call into the renderer. 659- st/vega: vg_manager should care about only the color buffer. 660- st/vega: Clean up vg_context fields and functions. 661- st/vega: Clean up renderer fields and functions. 662- st/vega: vg_copy_texture and vg_copy_surface should share code. 663- st/vega: Get rid of renderer_copy_texture. 664- st/vega: Update to latest headers. 665- st/vega: Fix image sampler views for alpha-only formats. 666- st/vega: Make path_render and path_stroke take a matrix. 667- st/vega: Make image_draw take a matrix. 668- st/vega: Add primitive text support. 669- st/vega: Revive mask layer support. 670- st/vega: More flexible shader selection. 671- st/vega: Add color transformation support. 672- st/vega: Bump version to 1.1. 673- st/vega: Fix paint coordinates transformations. 674- st/vega: Fix negated logic in image_draw. 675- st/vega: Fix degenerate paints. 676- st/vega: Simplify radial gradient. 677- st/vega: Remove st_inlines.h. 678- st/vega: Delay blend texture creation until needed. 679- st/vega: Create drawing surface mask as needed. 680- st/vega: Initialize pipe states with renderer. 681- st/vega: Avoid unnecessary constant bufer upload. 682- st/vega: Destroy the pipe context with vg_context. 683- st/vega: polygon_array requires a deep free. 684- st/egl: Set pipe_resource::array_size to 1. 685- st/vega: Set pipe_resource::array_size to 1. 686- st/vega: Move vertex transformation to shader. 687- st/vega: Add a missing break. 688- st/vega: Add some comments to pipeline shaders. 689- st/vega: Refactor blend shaders. 690- st/vega: Move masking after blending. 691- st/vega: Add support for per-channel alpha. 692- st/vega: Blending should use premultiplied alpha. 693- st/vega: Fix VG_BLEND_MULTIPLY. 694- st/vega: Add blend shaders for all blend modes. 695- st/vega: Fix pipe blend state for various blend modes. 696- egl: \_eglFilterArray should not allocate. 697- mapi: Rewrite mapi_abi.py to get rid of preprocessor magic. 698- vbo: Drop second ATTR macro. 699- vbo: Fix GLES2 glVertexAttrib. 700- mesa: Do not advertise GL_OES_texture_3D. 701- mesa: Fix GL_FIXED arrays. 702- mesa: Fix glTexCoordPointer with type GL_FIXED. 703- st/egl: Plug pbuffer leaks. 704- st/egl: Fix eglCopyBuffers. 705- st/egl: Assorted fixes for dri2_display_get_configs. 706- docs/egl: Update ../egl.html. 707- st/egl: Fix eglChooseConfig when configs is NULL. 708- docs: Add an example for EGL_DRIVERS_PATH. 709- autoconf: Fix --with-driver=xlib --enable-openvg. 710 711Chris Wilson (2): 712 713- i915g: Fix closure of full batch buffers 714- intel: Check for unsupported texture when finishing using as a render 715 target 716 717Christoph Bumiller (80): 718 719- nv50: import new compiler 720- nouveau: update nouveau_class.h 721- nv50: introduce the big formats table 722- nv50: don't produce MOV immediate to output reg in store opt 723- nv50: change back accidentally swapped UNORM,SNORM vertex type 724- nv50: add/fix some license headers 725- nv50: simple reload elimination and local CSE 726- nv50: fix constant_operand opt mul by 2 case 727- nv50: permit usage of undefined TGSI TEMPs 728- nv50: add missing 2nd source for POW multiplication 729- nv50: add signed RGTC1 to format table, allow 2_10_10_10 for vbufs 730- nv50: fix for empty BBs 731- nv50: insert MOVs also for PHI sources from dominating block 732- nv50: explicitly set src type for SET ops 733- nv50: fixes for nested IFs 734- nv50: don't eliminate loads to dedicated values 735- nv50: fix constbuf validation 736- nv50: build proper phi functions in the first place 737- nv50: fix reg count 738- nv50: begin implementing loops 739- nv50: more constant folding 740- nv50: loops part 2 741- nv50: flatten simple IF/ELSE/ENDIF constructs 742- nv50: fix thinko in store to output reg possible check 743- nv50: generate JOINs for outermost IF clauses 744- nv50: more TGSI opcodes (SIN, SCS, ARL, RET, KILP) 745- nv50: fix PSIZ and PRIMID mapping 746- nv50: check dst compatibility in CSE 747- nv50: initialize edgeflag input index 748- nv50: emit predicate for interp 749- Merge remote branch 'origin/master' into nv50-compiler 750- nv50: DP2, fix ARL 751- nv50: yet another case we need a nop.exit 752- nv50: fix check for sprite/point coord enable 753- nv50: handle TEXTURE_SWIZZLE and GEOMETRY_SHADER4 caps 754- nv50: set the FragDepth output index 755- nv50: turn off verbose debug output by default 756- nv50: attempt at making more complicated loops work 757- nv50: SSG 758- nv50: make FrontFacing -1 or +1 759- nv50: re-add proper TEXBIAS sequence 760- nv50: make use of TGSI immediate type 761- nv50: must join SELECT inputs before MOV inputs 762- nv50: fix XPD, was negated 763- nv50: fix find_dom_frontier 764- nv50: fix build-predicate function 765- Merge remote branch 'origin/master' into nv50-compiler 766- nv50: load address register before using it, not after 767- nv50: save tgsi instructions 768- nv50: prepare for having multiple functions 769- nv50: don't parse again in tgsi_2_nc 770- nv50: use actual loads/stores if TEMPs are accessed indirectly 771- nv50: create value references with the right type 772- nv50: duplicate interps in load_proj_tex_coords 773- nv50: address regs are 16 bit 774- nv50: fix can_load check for 3rd source 775- nv50: reduce bb_reachable_by runtime from pot to linear 776- nv50: minor compiler fixes and cleanups 777- nv50: cannot move from local mem to output reg directly 778- nv50: newlines in shader bincode printing 779- nv50: match TEMP limit with nv50 ir builder 780- nv50: handle TGSI EXP and LOG again 781- nv50: check for immediates when turning MUL ADD into MAD 782- nv50: interp cannot write flags reg 783- nv50: MOV TEMP[0], -CONST[0] must be float32 negation 784- nv50: fix indirect CONST access with large or negative offsets 785- nv50: fix TXP depth comparison value 786- nv50: consider address register in reload elimination 787- nv50: improve and fix modifier folding optimization 788- nv50: put low limit on REG_ALLOC_TEMP and FP_RESULT_COUNT 789- Merge remote branch 'origin/nv50-compiler' 790- nv50: don't segfault on shaders with 0 instructions 791- nv50: get shader fixups/relocations into working state 792- nv50: add relocs for stack and local mem buffers 793- nv50: emit constbuf relocs before uploading constants 794- nv50: fix typo in fifo packet length limit 795- nv50: use formats table in nv50_surface.c 796- nv50: use CLEAR_BUFFERS for surface fills 797- nv50: fix/handle a few more PIPE_CAPs 798- nv50: fix GP state bind and validate 799 800Corbin Simpson (8): 801 802- r600g: Use align() instead of handrolled code. 803- r600g: Trivially deobfuscate r600_hw_states. 804- r600g: Deobfuscate and comment a few more functions in 805 r600_hw_states. 806- r600g: Clean up some indentation and \|= vs. \| usage. 807- r600g: Fix false and true. 808- r600g: "tmp" is such a bad name for a texture. 809- r600g: Clean up PS setup. 810- r600g: Cleanup viewport floats. 811 812Daniel Lichtenberger (1): 813 814- radeon: fix potential segfault in renderbuffer update 815 816Daniel Vetter (21): 817 818- r200: revalidate after radeon_update_renderbuffers 819- i915g: rip out ->sw_tiled 820- i915g: s/hw_tiled/tiling 821- i915g: add pineview pci ids 822- i915g: kill RGBA/X formats 823- i915g: kill buf->map_gtt 824- i915g: kill idws->pool 825- i915g: drop alignment parameter from iws->buffer_create 826- i915g: add winsys function to create tiled buffers 827- i915g: switch to tiled allocations, kill set_fence 828- i915g: prepare winsys/batchbuffer for execbuf2 829- i915g: return tiling in iws->buffer_from_handle 830- i915g: implement unfenced color&depth buffer using tiling bits 831- i915g: implement unfenced relocs for textures using tiling bits 832- i915g: postpone mipmap/face offset calculation 833- i915g: don't pot-align stride for tiled buffers 834- i915g: enable X-tiling for textures 835- i915g: switch rendering to mipmapped textures to (x,y) offsets 836- i915g: enable x-tiling for render targets 837- i915g: assert(depth_surface->offset == 0) 838- i915g: track TODO items 839 840Dave Airlie (182): 841 842- r300g: fix buffer reuse issue caused by previous commit 843- r600g: pull r600_draw struct out into header 844- r600g: use index min/max + index buffer offset. 845- r600g: add vgt dma src defines 846- r600g: fixup texture state on evergreen. 847- r600g: fix texture bos and avoid doing depth blit on evergreen 848- r600g: hide radeon_ctx inside winsys. 849- r600g: attempt to abstract kernel bos from pipe driver. 850- r600g: move constant buffer creation behind winsys abstraction. 851- r600g: use malloc bufmgr for constant buffers 852- r600g: add support for kernel bo 853- r600g: add winsys bo caching. 854- r600g: add upload manager support. 855- r600g: fixup map flushing. 856- r600g: use calloc for ctx bo allocations 857- r600g: oops got the use_mem_constant the wrong way around. 858- r600g; add uses waterfall to asm cf for r6xx. 859- r600g: only emit uses waterfall on r6xx hw. 860- util/r300g: split the r300 index buffer modifier functions out to 861 util 862- r600g: modify index buffers for sizes the hw can't deal with. 863- r600g: send correct surface base update for multi-cbufs 864- r600g: fix fbo-drawbuffers-maxtargets 865- r600g: clean up valgrind issues on maxtargets test. 866- r600g: drop debugging that snuck in 867- r600g: fix tiling support for ddx supplied buffers 868- r600g: add z16 to color setup 869- r600g: add color/texture support for more depth formats. 870- r600g: fix r700 cube map sizing. 871- r600g: fixup r700 CB_SHADER_CONTROL register. 872- r600g: add missing BC_INST wrapper for evergreen 873- r600g: only flush for the correct colorbuffer, not all of them. 874- r600g: deal with overflow of VTX/TEX CF clauses. 875- r600g: set back to correct codepaths. 876- r600g: fixup evergreen miptree setup. 877- r600g: fix eg texture borders. 878- r600g: fix typo in struct member name 879- r600g: cleanup some of the DB blit code 880- r600g: make stencil readback work 881- r600g: disable dirty handling on texture from depth code. 882- r600g: use floats instead of hex for blit vbo 883- r600g: fix depth readback on rv610 and other quirky variants. 884- r600g: fix typo in evergreen register list 885- u_blitter: add a custom blitter call passing a dsa cso 886- r600g: use blitter to do db->cb flushing. 887- r600g: fix warnings since last commit. 888- egl: fix build since 17eace581d25a626a7d75d9d1205d012cbb14a6e 889- r300g: fix point sprite coord. 890- r600g: add vert support for 16/16 and 16/16/16 floats. 891- r600g: add some more vertex format support. 892- r600g: some more vertex formats 893- r600g: fix draw-elements and draw-elements-base-vertex 894- r600g: drop index_offset parameter to index buffer translate. 895- r600g: fixup tex wrapping. 896- r600g: fixup VP->FP output->input routing. 897- r600g: fix typo in r700 alu emit 898- r600g: fixup sprite coord enable. 899- r600g: fix polygon mode 900- mesa/mipmap: fix warning since 901 1acadebd6270d3604b026842b8a21360968618a0 902- r600g: add eg poly mode code. 903- r600g: make index bias fix for evergreen 904- r600g: add eg db count control register. 905- r300g: fix glsl-fs-pointcoord 906- r600g: add evergreen texture resource properly. 907- r600g: fix db flush breaking config state 908- r600g: on evergreen the centroid isn't set in this register. 909- r600g: add back evergreen name. 910- r600g: add evergreen texture border support to new path 911- r600g: move radeon.h members around to add back map flushing. 912- r600g: add initial vertex translate support. 913- r600g: remove old assert from new codepath 914- Revert "r600g: add initial vertex translate support." 915- r600g: port r300g fix for X\* formats in texformat code 916- r600g: add L8A8 unorm. 917- r600g: clean up some code from move to new paths. 918- r600g: return string for chip family 919- r600g: use Elements macro instead of manual sizeofs 920- r600g: fix evergreen depth flushing. 921- r600g: add winsys support for CTL constants. 922- r600g: drop depth quirk on evergreen 923- r600g: add reloc for evergreen color attrib 924- r600g: realign evergreen code with r600 code. 925- r600g: add assembler support for other vtx fetch fields. 926- r600g: fixup vertex format picking. 927- r600g: sync vertex/texture cache on resources on evergreen 928- r600g: add cb flushing for extra buffers + depth buffer on 929 r600/evergreen 930- r600g: fix evergreen draw-buffers 931- r600g: flush SH cache on constant change on evergreen 932- r600g: only set the Z export if shader exports it. 933- r600g: setup basic loop consts on r600 + evergreen. 934- mesa/st: initial attempt at RG support for gallium drivers 935- r600g: break out of search for reloc bo after finding it. 936- r600g: the code to check whether a new vertex shader is needed was 937 wrong 938- r600g: fix wwarning in bo_map function 939- r600g: TODO domain management 940- r600g: add bo fenced list. 941- pb: don't keep checking buffers after first busy 942- r600g: add bo busy backoff. 943- r600g: drop mman allocator 944- r600g: drop use_mem_constant. 945- r600g: avoid unneeded bo wait 946- pb: fix numDelayed accounting 947- r600g: add evergreen stencil support. 948- r600g: use format from the sampler view not from the texture. 949- r600g: fix Z export enable bits. 950- r600g: add some RG texture format support. 951- r600g: drop width/height per level storage. 952- r600g: fix input/output Z export mixup for evergreen. 953- r600g: evergreen has no request size bit in texture word4 954- r600g: enable vertex samplers. 955- r600g: add TXL opcode support. 956- r600g: don't run with scissors. 957- r600g: fix typo in vertex sampling on r600 958- gallium/tgsi: add support for stencil writes. 959- gallium/format: add support for X24S8 and S8X24 formats. 960- gallium/format: add X32_S8X24_USCALED format. 961- gallium/util: add S8 tile sampling support. 962- mesa: add support for FRAG_RESULT_STENCIL. 963- mesa: improve texstore for 8/24 formats and add texstore for S8. 964- softpipe: add support for shader stencil export capability 965- st/mesa: add option to choose a texture format that we won't render 966 to. 967- st/mesa: use shader stencil export to accelerate shader drawpixels. 968- r600g: add support for S8, X24S8 and S8X24 sampler formats. 969- r600g: add shader stencil export support. 970- glsl: add support for shader stencil export 971- st/mesa: enable stencil shader export extension if supported 972- r600g: fix depth0 setting 973- r600g: fix scissor/cliprect confusion 974- r600g: store samplers/views across blit when we need to modify them 975- r600g: reduce size of context structure. 976- r600g: the vs/ps const arrays weren't actually being used. 977- r600g: add copy into tiled texture 978- r600g: split out miptree setup like r300g 979- r600g: use common texture object create function 980- r600g: rename pitch in texture to pitch_in_bytes 981- r600g: remove bpt and start using pitch_in_bytes/pixels. 982- r600g: fix transfer stride. 983- r600g: drop all use of unsigned long 984- r600g: use blitter for hw copy region 985- r600g: evergreen add stencil export bit 986- r600g: add missing eg reg definition 987- r600g: fix stencil export for evergreen harder 988- r600g: drop unused context members 989- r600g: only pick centroid coordinate when asked. 990- r600g: fixup pos/face ena/address properly 991- r600g: fixup typo in macro name 992- r600g: select linear interpolate if tgsi input requests it 993- r300g: clean up warning due to unknown cap. 994- tgsi: add scanner support for centroid inputs 995- r600g: evergreen interpolation support. 996- r600g: add evergreen ARL support. 997- r600g: switch to a common formats.h file since they are in different 998 regs 999- r600g: add defines for tiling 1000- r600g: get tiling info from kernel 1001- r600g: set tiling bits in hw state 1002- r600g: do proper tracking of views/samplers. 1003- r600g: fix typo in tiling setup cb code. 1004- r600g: depth needs to bound to ds 1005- r600g: attempt to cleanup depth blit 1006- r600g: fix transfer function for tiling. 1007- r600g: retrieve tiling info from kernel for shared buffers. 1008- r600g: all non-0 mipmap levels need to be w/h aligned to POT. 1009- r600g: move to per-miplevel array mode. 1010- r600g: start adding hooks for aligning width/height for tiles. 1011- r600g: add r600 surface to store the aligned height. 1012- r600g: introduce a per-driver resource flag for transfers. 1013- r600g: add texture tiling alignment support. 1014- r600g: add texture tiling enable under a debug option. 1015- r600g: initial translate state support 1016- r600g: start splitting out common code from eg/r600. 1017- r600g: not fatal if we can't get tiling info from kernel 1018- r600g: merge more of the common r600/evergreen state handling 1019- r600g: drop more common state handling code 1020- r600g: fix magic 0x1 ->flat shade ena 1021- r600g: add assembler support for all the kcache fields. 1022- gallium/noop: report GL 2.1 1023- r600g: pick correct color swap for A8 fbos. 1024- r300g/r600g: bump cache manager timeouts to 1s 1025- r600g: it looks like r600 can handle dword offsets in the indices. 1026- r300g: try and use all of vertex constant space 1027- r300g: fixup rs690 tiling stride alignment calculations. 1028- r600g: fix evergreen segfaults. 1029- r600g: hack around property unknown issues. 1030 1031Eric Anholt (300): 1032 1033- glsl: Add definition of gl_TextureMatrix inverse/transpose builtins. 1034- i965: Share the KIL_NV implementation between glsl and non-glsl. 1035- i965: Also enable CC statistics when doing OQs. 1036- i965: Track the windowizer's dispatch for kill pixel, promoted, and 1037 OQ 1038- glsl: Rework assignments with write_masks to have LHS chan count 1039 match RHS. 1040- glsl: Fix copy'n'wasted ir_noop_swizzle conditions. 1041- ir_to_mesa: Only compare vector_elements present for 1042 any_nequal/all_equal 1043- i965: Fix the vector/expression splitting for the write_mask change. 1044- i965: When splitting vector variable assignment, ignore unset 1045 channels. 1046- i965: Update expression splitting for the vector-result change to 1047 compares. 1048- i965: Warning fix for vector result any_nequal/all_equal change. 1049- mesa: Remove the non-required ARB_imaging extension. 1050- mesa: Remove EXT_histogram. 1051- mesa: Remove SGI_color_table. 1052- mesa: Remove SGI_color_matrix. 1053- mesa: Remove EXT_convolution. 1054- intel: Remove disabled stencil drawpixels acceleration. 1055- intel: Remove unnecessary minimum pitch alignment to 32 bytes. 1056- intel: Replace my intel_texture_bitmap code with \_mesa_meta_Bitmap. 1057- radeon: Remove copied minimum pitch alignment code. 1058- unichrome: Mostly revert my convolution removal changes. 1059- intel: Remove dead intelIsTextureResident(). 1060- i915: Remove a dead if (0) block. 1061- intel: Dead comment removal. 1062- intel: Corresponding FinishRenderTexture debug to BeginRenderTexture. 1063- i965: Add support for rendering to SARGB8 FBOs. 1064- intel: Fix segfault on INTEL_DEBUG=fbo with unsupported framebuffers. 1065- intel: Add fallback debug to glGenerateMipmap. 1066- intel: More reverting of the sw fallback for depth texture border 1067 color. 1068- intel: Improve some of the miptree debugging. 1069- mesa: Fix type typo in glGenerateMipmap handling of GL_UNSIGNED_INT 1070 data. 1071- glsl: Fix broadcast_index of lower_variable_index_to_cond_assign. 1072- glsl: Add validation that a swizzle only references valid channels. 1073- i965: Fix up writemasked assignments in the new FS. 1074- i965: Remove swizzling of assignment to vector-splitting 1075 single-channel LHS. 1076- i965: Handle all_equal/any_nequal in the new FS. 1077- i965: Fix vector splitting RHS channel selection with sparse 1078 writemasks. 1079- i965: Add support for dFdx()/dFdy() to the FS backend. 1080- i965: Add support for attribute interpolation on Sandybridge. 1081- i965: Set up inputs to the fragment shader according to FP 1082 InputsRead. 1083- i965: Add support for POW in gen6 FS. 1084- i965: Fix negation in the new FS backend. 1085- i965: Actually track the "if" depth in loop in the new FS backend. 1086- i965: Apply the same set of lowering passes to new FS as to Mesa IR. 1087- i965: Fix valgrind complaint about base_ir for new FS debugging. 1088- i965: Fix up the FS backend for the variable array indexing pass. 1089- i965: Set the variable type when dereferencing an array. 1090- i965: Add support for dereferencing structs to the new FS backend. 1091- i965: Add support for struct, array, and matrix uniforms to FS 1092 backend. 1093- i965: Fix all non-snb regression in the snb attribute interpolation 1094 commit. 1095- i965: Fix up part of my Sandybridge attributes support patch. 1096- i965: Add support for gl_FrontFacing to the new FS backend. 1097- i965: Subtract instead of adding when computing y delta in new FS 1098 backend. 1099- mesa: Pull ir_to_mesa's sampler number fetcher out to shared code. 1100- i965: Set up sampler numbers in the FS backend. 1101- i965: Add support for non-color render target write data to new FS 1102 backend. 1103- i965: Add support for MRT to the new FS backend. 1104- i965: Add support for ir_loop counters to the new FS backend. 1105- i965: Add support for ARB_fragment_coord_conventions to the new FS 1106 backend. 1107- glsl: Also update implicit sizes of varyings at link time. 1108- i965: Do interpolation for varying matrices and arrays in the FS 1109 backend. 1110- i965: Don't try to emit interpolation for unused varying slots. 1111- i965: Fix array indexing of arrays of matrices. 1112- i965: Clean up obsolete FINISHME comment. 1113- mesa: Move the list of builtin uniform info from ir_to_mesa to shared 1114 code. 1115- i965: Add support for builtin uniforms to the new FS backend. 1116- i965: Fix use of undefined mem_ctx in vector splitting. 1117- i956: Make new FS discard do its work in a temp, not the null reg! 1118- i965: Clean up the virtual GRF handling. 1119- ra: First cut at a graph-coloring register allocator for mesa. 1120- i965: First cut at register allocation using graph coloring. 1121- i965: Add live interval analysis and hook it up to the register 1122 allocator. 1123- i965: Remove my "safety counter" code from loops. 1124- i965: Fix whole-structure/array assignment in new FS. 1125- mesa: Don't reference a W component in setting up a vec3 uniform 1126 component. 1127- i965: Fix new FS handling of builtin uniforms with packed scalars in 1128 structs. 1129- glsl: Add a lowering pass for texture projection. 1130- i965: Use the lowering pass for texture projection. 1131- i965: Split the gen4 and gen5 sampler handling apart. 1132- i965: Add gen6 attribute interpolation to new FS backend. 1133- i965: Fix the gen6 jump size for BREAK/CONT in new FS. 1134- i965: Also increment attribute location when skipping unused slots. 1135- i965: Pre-gen6, map VS outputs (not FS inputs) to URB setup in the 1136 new FS. 1137- i965: Add real support for pre-gen5 texture sampling to the new FS. 1138- i965: Fix up copy'n'pasteo from moving coordinate setup around for 1139 gen4. 1140- i965: Restore the forcing of aligned pairs for delta_xy on chips with 1141 PLN. 1142- i965: When producing a single channel swizzle, don't make a 1143 temporary. 1144- i965: Add a sanity check for register allocation sizes. 1145- i965: Fix off-by-ones in handling the last members of register 1146 classes. 1147- i965: Don't try to emit code if we failed register allocation. 1148- i965: Add support for EXT_texture_swizzle to the new FS backend. 1149- i965: Set up swizzling of shadow compare results for 1150 GL_DEPTH_TEXTURE_MODE. 1151- i965: Fix glean/texSwizzle regression in previous commit. 1152- i965: Be more conservative on live interval calculation. 1153- i965: Add trivial dead code elimination in the new FS backend. 1154- i965: Add initial folding of constants into operand immediate slots. 1155- i965: In disasm, gen6 fb writes don't put msg reg # in 1156 destreg_conditionalmod. 1157- i965: Add support for gen6 FB writes to the new FS. 1158- i965: Enable the constant propagation code. 1159- i965: Also do constant propagation for the second operand of CMP. 1160- i965: Add back gen6 headerless FB writes to the new FS backend. 1161- i965: Gen6 no longer has the IFF instruction; always use IF. 1162- i965: Fix up IF/ELSE/ENDIF for gen6. 1163- i965: Fix botch in the header_present case in the new FS. 1164- i965: Add some clarification of the WECtrl field. 1165- i965: Don't do 1/w multiplication in new FS for gen6 1166- i965: Gen6's sampler messages are the same as Ironlake. 1167- i965: Refactor gl_FrontFacing setup out of general variable setup. 1168- i965: Add support for gl_FrontFacing on gen6. 1169- i965: Don't assume that WPOS is always provided on gen6 in the new 1170 FS. 1171- i965: Fix gen6 pointsize handling to match pre-gen6. 1172- i965: Disable emitting if () statements on gen6 until we really fix 1173 them. 1174- i965: Normalize cubemap coordinates like is done in the Mesa IR path. 1175- mesa: Simplify a bit of \_mesa_add_state_reference using memcmp. 1176- i965: Drop the check for duplicate \_mesa_add_state_reference. 1177- i965: Drop the check for YUV constants in the param list. 1178- i965: Handle swizzles in the addition of YUV texture constants. 1179- i965: Fix gen6 WM push constants updates. 1180- i965: Fix new FS gen6 interpolation for sparsely-populated arrays. 1181- i965: Enable attribute swizzling (repositioning) in the gen6 SF. 1182- i965: Add register coalescing to the new FS backend. 1183- i965: Split FS_OPCODE_DISCARD into two steps. 1184- i965: Reduce register interference checks for changed 1185 FS_OPCODE_DISCARD. 1186- i965: Move FS backend structures to a header. 1187- i965: Give the math opcodes information on base mrf/mrf len. 1188- i965: Give the FB write and texture opcodes the info on base MRF, 1189 like math. 1190- i965: Compute to MRF in the new FS backend. 1191- i965: Don't consider gen6 math instructions to write to MRFs. 1192- i965: Add a couple of checks for gen6 math instruction limits. 1193- i965: Don't compute-to-MRF in gen6 math instructions. 1194- i965: Expand uniform args to gen6 math to full registers to get 1195 hstride == 1. 1196- i965: Don't compute-to-MRF in gen6 VS math. 1197- i965: Fix gen6 pixel_[xy] setup to avoid mixing int and float src 1198 operands. 1199- i965: Always use the new FS backend on gen6. 1200- i965: Fix missing "break;" in i2b/f2b, and missing AND of CMP result. 1201- intel: Allow CopyTexSubImage to InternalFormat 3/4 textures, like 1202 RGB/RGBA. 1203- i965: Don't rebase the index buffer to min 0 if any arrays are in 1204 VBOs. 1205- i965: Add support for rescaling GL_TEXTURE_RECTANGLE coords to new 1206 FS. 1207- i965: Set class_sizes[] for the aligned reg pair class. 1208- i965: Update the live interval when coalescing regs. 1209- i965: Add a pass to the FS to split virtual GRFs to float channels. 1210- i965: Add a function for handling the move of boolean values to flag 1211 regs. 1212- i965: Add peepholing of conditional mod generation from expressions. 1213- i965: Enable the new FS backend on pre-gen6 as well. 1214- i965: Fix texturing on pre-gen5. 1215- i965: Set the type of the null register to fix gen6 FS comparisons. 1216- i965: Disable the debug printf I added for FS disasm. 1217- i965: Fix a weirdness in NOT handling. 1218- i965: Fix assertion failure on gen6 BufferSubData to busy BO. 1219- i965: Assert out on gen6 VS constant buffer reads that hang the GPU 1220 for now. 1221- i965: Fix scissor-offscreen on gen6 like we did pre-gen6. 1222- i965: Avoid blits in BufferCopySubdata on gen6. 1223- i965: Tell the shader compiler when we expect depth writes for gen6. 1224- i965: Remove the gen6 emit_mi_flushes I sprinkled around the driver. 1225- i965: Disable thread dispatch when the FS doesn't do any work. 1226- i965: Add EU emit support for gen6's new IF instruction with 1227 comparison. 1228- i965: Set the source operand types for gen6 if/else/endif to integer. 1229- i965: Use the new style of IF statement with embedded comparison on 1230 gen6. 1231- i965: Split register allocation out of the ever-growing brw_fs.cpp. 1232- i965: Fix gl_FrontFacing emit on pre-gen6. 1233- i965: Add support for register spilling. 1234- i965: Don't emit register spill offsets directly into g0. 1235- i965: Correct scratch space allocation. 1236- i965: Be more aggressive in tracking live/dead intervals within 1237 loops. 1238- i965: Move the FS disasm/annotation printout to codegen time. 1239- i965: Add support for pull constants to the new FS backend. 1240- i965: Add EU code for dword scattered reads (constant buffer array 1241 indexing). 1242- i965: Clarify an XXX comment in FB writes with real info. 1243- i965: Use SENDC on the first render target write on gen6. 1244- i965: Clear some undefined fields of g0 when using them for gen6 FB 1245 writes. 1246- i965: Add disasm for the flag register. 1247- i965: Add support for discard instructions on gen6. 1248- i965: Handle new ir_unop_round_even in channel expression splitting. 1249- i965: Fix typo in comment about state flags. 1250- i965: Set up the constant buffer on gen6 when it's needed. 1251- i965: Add support for constant buffer loads on gen6. 1252- i965: Drop the eot argument to read messages, which can never be set. 1253- i965: Fix VS URB entry sizing. 1254- i965: Disable register spilling on gen6 until it's fixed. 1255- i965: Make FS uniforms be the actual type of the uniform at upload 1256 time. 1257- i965: Add user clip planes support to gen6. 1258- i965: Update gen6 SF state when point state (sprite or attenuation) 1259 changes. 1260- i965: Upload required gen6 VS push constants even when using pull 1261 constants. 1262- i965: Update the gen6 stencil ref state when stencil state changes. 1263- mesa: Make metaops use program refcounts instead of names. 1264- mesa: Don't compute an unused texture completeness debug string. 1265- intel: For batch, use GTT mapping instead of writing to a malloc and 1266 copying. 1267- intel: Annotate debug printout checks with unlikely(). 1268- intel: Remove the magic unaligned memcpy code. 1269- i965: Remove dead intel_structs.h file. 1270- intel: Avoid taking logbase2 of several things that we max. 1271- intel: Remove duplicated teximage miptree to object miptree 1272 promotion. 1273- intel: Remove leftover dri1 locking fields in the context. 1274- mesa: Fix delayed state flagging for EXT_sso-related program changes. 1275- intel: Fix the client-side swapbuffers throttling. 1276- Revert "intel: Fix the client-side swapbuffers throttling." 1277- i965: Allow OPCODE_SWZ to put immediates in the first arg. 1278- i965: Add support for math on constants in gen6 brw_wm_glsl.c path. 1279- i965: Work around strangeness in swizzling/masking of gen6 math. 1280- i965: re-enable gen6 IF statements in the fragment shader. 1281- glsl: Free the loop state context when we free the loop state. 1282- i965: Fix gl_FragCoord inversion when drawing to an FBO. 1283- i965: Shut up spurious gcc warning about GLSL_TYPE enums. 1284- mesa: Don't spam the console in a debug build unless some spam is 1285 requested. 1286- i965: Add state dumping for sampler state. 1287- i965: Add dumping of the sampler default color. 1288- i965: Fail on loops on gen6 for now until we write the EU emit code 1289 for it. 1290- i965: Eliminate dead code more aggressively. 1291- mesa: Include C++ files in the makedepend of DRI drivers. 1292- i965: Fix compute_to_mrf to not move a MRF write up into another live 1293 range. 1294- i965: Just use memset() to clear most members in FS constructors. 1295- i965: Remove extra n at the end of every instruction in 1296 INTEL_DEBUG=wm. 1297- i965: Fold constants into the second arg of BRW_SEL as well. 1298- glsl: Add a helper function for determining if an rvalue could be a 1299 saturate. 1300- i965: Recognize saturates and turn them into a saturated mov. 1301- ir_to_mesa: Detect and emit MOV_SATs for saturate constructs. 1302- i965: Improve compute-to-mrf. 1303- i965: Remove duplicate MRF writes in the FS backend. 1304- i965: Move gen4 blend constant color to the gen4 blending file. 1305- i965: Don't upload polygon stipple unless required. 1306- i965: Don't upload line stipple pattern unless we're stippling. 1307- i965: Don't upload line smooth params unless we're line smoothing. 1308- i965: Use the new embedded compare in SEL on gen6 for VS MIN and MAX 1309 opcodes. 1310- i965: Fix type of gl_FragData[] dereference for FB write. 1311- glsl: Make the symbol table's add_function just use the function's 1312 name. 1313- glsl: Make the symbol table's add_variable just use the variable's 1314 name. 1315- glsl: Add a helper constructor for expressions that works out result 1316 type. 1317- glsl: Fix structure and array comparisions. 1318- glsl: Quiet unreachable no-return-from-function warning. 1319- i965: Dump the WHILE jump distance on gen6. 1320- i965: Add support for gen6 DO/WHILE ISA emit. 1321- i965: Add support for gen6 BREAK ISA emit. 1322- i965: Add support for gen6 CONTINUE instruction emit. 1323- i965: Enable IF statements in the VS. 1324- i965: Add support for loops in the VS. 1325- glsl: Mark the array access for whole-array comparisons. 1326- glsl: Fix flipped return of has_value() for array constants. 1327- mesa: Add getters for the rest of the supported draw buffers. 1328- mesa: Add getters for ARB_copy_buffer's attachment points. 1329- intel: Add an env var override to execute for a different GPU 1330 revision. 1331- i965: Update gen6 WM state on compiled program change, not just FP 1332 change. 1333- i965: Update gen6 SF state on fragment program change too. 1334- i965: Fix compile warning about missing opcodes. 1335- i965: Move payload reg setup to compile, not lookup time. 1336- i965: Provide delta_xy reg to gen6 non-GLSL path PINTERP. 1337- i965: Fix up 16-wide gen6 FB writes after various refactoring. 1338- i965: Don't smash a group of coordinates doing gen6 16-wide sampler 1339 headers. 1340- i965: Fix gen6 interpolation setup for 16-wide. 1341- i965: Fix up gen6 samplers for their usage by brw_wm_emit.c 1342- i965: Make the sampler's implied move on gen6 be a raw move. 1343- i965: Align gen6 push constant size to dispatch width. 1344- i965: Add support for the instruction compression bits on gen6. 1345- i965: Nuke brw_wm_glsl.c. 1346- i965: Remove INTEL_DEBUG=glsl_force now that there's no brw_wm_glsl.c 1347- i965: Fix comment about gen6_wm_constants. 1348- i965: Handle saturates on gen6 math instructions. 1349- i965: Always hand the absolute value to RSQ. 1350- i965: Add disabled debug code for dumping out the WM constant 1351 payload. 1352- i965: Work around gen6 ignoring source modifiers on math 1353 instructions. 1354- i965: Fix flipped value of the not-embedded-in-if on gen6. 1355- i965: Don't try to store gen6 (float) blend constant color in bytes. 1356- i965: Set up the color masking for the first drawbuffer on gen6. 1357- i965: Set up the per-render-target blend state on gen6. 1358- i965: Set the render target index in gen6 fixed-function/ARB_fp path. 1359- i965: Use the new pixel mask location for gen6 ARB_fp KIL 1360 instructions. 1361- i965: Drop KIL_NV from the ff/ARB_fp path since it was only used for 1362 GLSL. 1363- i965: Drop push-mode reladdr constant loading and always use 1364 constant_map. 1365- i965: Fix VS constants regression pre-gen6. 1366- i965: Clean up VS constant buffer location setup. 1367- i965: Set up the correct texture border color state struct for 1368 Ironlake. 1369- i965: Set render_cache_read_write surface state bit on gen6 constant 1370 surfs. 1371- i965: remove unused variable since brw_wm_glsl.c removal. 1372- intel: Use plain R8 and RG8 for COMPRESSED_RED and COMPRESSED_RG. 1373- intel: Set the swizzling for depth textures using the GL_RED depth 1374 mode. 1375- glsl: Correct the marking of InputsRead/OutputsWritten on in/out 1376 matrices. 1377- i965: Correct the dp_read message descriptor setup on g4x. 1378- intel: Include stdbool so we can stop using GLboolean when we want 1379 to. 1380- i965: Fix ARL to work on gen6. 1381- i956: Fix the old FP path fragment position setup on gen6. 1382- i965: Fix gl_FragCoord.z setup on gen6. 1383- i965: Add support for using the BLT ring on gen6. 1384- intel: Update renderbuffers before looking up CopyTexImage's read 1385 buffer. 1386- intel: Drop commented intel_flush from copy_teximage. 1387- intel: Try to sanely check that formats match for CopyTexImage. 1388- intel: Support glCopyTexImage() from XRGB8888 to ARGB8888. 1389- i965: Avoid using float type for raw moves, to work around SNB issue. 1390- i965: Set the alternative floating point mode on gen6 VS and WM. 1391- i965: Add support for gen6 constant-index constant loading. 1392- i965: Add support for gen6 reladdr VS constant loading. 1393- i965: Improve the hacks for ARB_fp scalar^scalar POW on gen6. 1394- i965: Factor out the ir comparision to BRW_CONDITIONAL\_\* code. 1395- i965: Fix regression in FS comparisons on original gen4 due to gen6 1396 changes. 1397- i965: Do lowering of array indexing of a vector in the FS. 1398- intel: Only do frame throttling at glFlush time when using 1399 frontbuffer. 1400- intel: Handle forced swrast clears before other clear bits. 1401- intel: Use tri clears when we don't know how to blit clear the 1402 format. 1403- intel: Add spans code for the ARB_texture_rg support. 1404- intel: Add a couple of helper functions to reduce rb code 1405 duplication. 1406- intel: Fix segfaults from trying to use \_ColorDrawBuffers in FBO 1407 validation. 1408- intel: When validating an FBO's combined depth/stencil, use the given 1409 FBO. 1410 1411Fabian Bieler (2): 1412 1413- r600g: set address of pop instructions to next instruction 1414- glsl: fix lowering conditional returns in subroutines 1415 1416Francisco Jerez (51): 1417 1418- dri/nv04: Fix PGRAPH_ERRORs when running OA. 1419- dri/nv04: Mipmapping fixes. 1420- dri/nv04: Align SIFM transfer dimensions. 1421- dri/nv04: Fix up color mask. 1422- dri/nv04: Fix maximum texture size. 1423- dri/nv04: Fix provoking vertex. 1424- dri/nouveau: Update nouveau_class.h. 1425- dri/nouveau: Add some more extensions. 1426- dri/nouveau: Fix glRenderbufferStorage with DEPTH_COMPONENT as 1427 internal format. 1428- dri/nouveau: Don't request a fake front unnecessarily. 1429- dri/nouveau: Don't reemit the BO state in nouveau_state_emit(). 1430- dri/nouveau: Cleanup references to the old FBOs on glMakeCurrent(). 1431- meta: Don't bind the created texture object in init_temp_texture(). 1432- dri/nv10: Fix the CLAMP texture wrap mode. 1433- dri/nv04: Use nvgl_wrap_mode(). 1434- dri/nouveau: Remove unnecessary assertion. 1435- dri/nouveau: Cleanup more references to old FBOs and VBOs. 1436- dri/nv10-nv20: Fix texturing in some cases after a base level change. 1437- dri/nouveau: Fix software mipmap generation on 1x1 textures. 1438- dri/nouveau: Have a smaller amount of larger scratch buffers. 1439- dri/nouveau: Remove unnecessary flush. 1440- dri/nv10: Use fast Z clears. 1441- dri/nouveau: Minor cleanup. 1442- dri/nv10: Fake fast Z clears for pre-nv17 cards. 1443- dri/nouveau: Initialize tile_flags when allocating a render target. 1444- nouveau: Get larger push buffers. 1445- dri/nouveau: Force a "slow" Z clear if we're getting a new depth 1446 buffer. 1447- dri/nv20: Clear with the 3D engine. 1448- dri/nouveau: Don't assert(0) on compressed internal formats. 1449- dri/nv25: Bind a hierarchical depth buffer. 1450- dri/nouveau: Call \_mesa_update_state() after framebuffer 1451 invalidation. 1452- dri/nouveau: Honor the access flags in nouveau_bufferobj_map_range. 1453- dri/nouveau: Tell the vbo module we want real hardware BOs. 1454- dri/nouveau: Split out the scratch helpers to a separate file. 1455- dri/nouveau: Avoid recursion in nouveau_bo_context_reset(). 1456- dri/nouveau: Use a macro to iterate over the bound vertex attributes. 1457- dri/nouveau: Split out array handling to its own file. 1458- dri/nouveau: Optimize VBO binding re-emission. 1459- dri/nouveau: Keep small DYNAMIC_DRAW vertex buffers in system ram. 1460- dri/nouveau: Pipeline glTexSubImage texture transfers. 1461- dri/nouveau: Fix type promotion issue on 32bit platforms. 1462- dri/nouveau: Validate the framebuffer state on read buffer changes. 1463- dri/nouveau: Re-emit the BO state when coming back from a software 1464 fallback. 1465- meta: Don't leak alpha function/reference value changes. 1466- meta: Fix incorrect rendering of the bitmap alpha component. 1467- vbo: Avoid unnecessary copy to/from current in vertex format upgrade. 1468- meta: Don't try to disable cube maps if the driver doesn't expose the 1469 extension. 1470- meta: Handle bitmaps with alpha test enabled. 1471- dri/nouveau: Split hardware/software TNL instantiation more cleanly. 1472- dri/nouveau: Fix typo. 1473- dri/nouveau: Kill a bunch of ternary operators. 1474 1475Fredrik Höglund (2): 1476 1477- r600g: Fix texture sampling with swizzled coords 1478- r600g: fix pow(0, 0) evaluating to NaN 1479 1480Guillermo S. Romero (1): 1481 1482- r300g: Do not use buf param before checking for NULL. 1483 1484Henri Verbeet (19): 1485 1486- r600g: Flush upload buffers before draws instead of before flushes. 1487- r600g: Check for other references before checking for existing 1488 mappings in radeon_bo_pb_map_internal(). 1489- r600g: Remove a redundant flush in r600_texture_transfer_map(). 1490- r600g: Buffer object maps imply a wait. 1491- r600g: Respect PB_USAGE_UNSYNCHRONIZED in 1492 radeon_bo_pb_map_internal(). 1493- Revert "r600g: Flush upload buffers before draws instead of before 1494 flushes." 1495- r600g: fix exports_ps to export a number not a mask. 1496- r600g: Mention AMD in the renderer string. 1497- r600g: Cleanup the fenced_bo list in r600_context_fini(). 1498- r600g: Evergreen has two extra frac_bits for the sampler LOD state. 1499- r600: Evergreen has two extra frac_bits for the sampler LOD state. 1500- r600g: Add PIPE_FORMAT_L8A8_UNORM for Evergreen as well. 1501- r600g: Swizzle vertex data only once. 1502- r600g: Synchronize supported color formats between Evergreen and 1503 r600/r700. 1504- r600g: Fix the PIPE_FORMAT_L8A8_UNORM color swaps. 1505- r600g: Fix the PIPE_FORMAT_A8_UNORM color swap for Evergreen as well. 1506- r600g: Cleanup block bo references in r600_context_fini(). 1507- r600g: Cleanup fetch shader resources in r600_pipe_shader_destroy(). 1508- st/mesa: Handle wrapped depth buffers in st_copy_texsubimage(). 1509 1510Hui Qi Tay (10): 1511 1512- llvmpipe: minor changes in llvm coefficient calcs 1513- draw: cliptest and viewport done in a single loop in vertex shader 1514- draw: added viewport and cliptest flags 1515- draw: sanitize llvm variant key 1516- draw: corrections for w coordinate 1517- draw: corrections to allow for different cliptest cases 1518- llvmpipe: Moved draw pipeline twoside function to llvm setup code 1519- llvmpipe: added llvm offset setup code 1520- llvmpipe: clean up polygon offset function in lp setup code 1521- llvmpipe: fix such that offset/twoside function only does in-place 1522 modification 1523 1524Ian Romanick (102): 1525 1526- glsl2: Refactor testing for whether a deref is of a matrix or array 1527- glsl2: Add flags to enable variable index lowering 1528- glsl: Add doxygen comments 1529- EGL DRI2: Silence piles of 'unused variable' warnings 1530- EGL DRI2: Silence 'missing initializer' warnings 1531- egl_glx: Silence piles of 'unused variable' warnings 1532- egl: Fix several 'comparison between signed and unsigned integer' 1533 warnings 1534- dri: Ensure that DRI driver cpp files are in tarballs 1535- mesa: Force GL_ARB_copy_buffer to always be enabled 1536- mesa: Force GL_SGIS_generate_mipmap to always be enabled 1537- Remove GL_MESA_packed_depth_stencil 1538- Remove GL_EXT_cull_vertex 1539- Regenerate files changed by previous commit 1540- Remove unnescessary initializations of UpdateTexturePalette 1541- ARB_texture_rg: Add GLX protocol support 1542- ARB_texture_rg: Correct some errors in RED / RG internal format 1543 handling 1544- ARB_texture_rg: Add GL_TEXTURE_{RED,GREEN}_SIZE query support 1545- ARB_texture_rg: Add GL_RED as a valid GL_DEPTH_TEXTURE_MODE 1546- ARB_texture_rg: Handle RED and RG the same as RGB for tex env 1547- ARB_texture_rg: Add R8, R16, RG88, and RG1616 internal formats 1548- ARB_texture_rg: Allow RED and RG textures as FBO color buffer 1549 attachments 1550- mesa: Enable GL_ARB_texture_rg in software paths 1551- i965: Enable GL_ARB_texture_rg 1552- mesa: Add ARB_texture_compression_rgtc as an alias for 1553 EXT_texture_compression_rgtc 1554- ARB_texture_rg: Add GL_COMPRESSED_{RED,RG} cases in 1555 \_mesa_is_color_format 1556- mesa: Fix misplaced #endif 1557- mesa: Trivial correction to comment 1558- rgtc: Detect RGTC formats as color formats and as compressed formats 1559- docs: Add list of bugs fixed in 7.9 1560- docs: Import 7.9 release notes from 7.9 branch. 1561- docs: Import 7.8.x release notes from 7.8 branch. 1562- docs: download.html does not need to be updated for each release 1563- docs: Update mailing lines from sf.net to freedesktop.org 1564- docs: Import news updates from 7.9 branch 1565- docs: added news item for 7.9 release 1566- glsl: Fail linking if assign_attribute_locations fails 1567- glsl: Refactor 'layout' grammar to match GLSL 1.60 spec grammar 1568- glsl: Slight refactor of error / warning checking for ARB_fcc layout 1569- glsl: Clear type_qualifier using memset 1570- glsl: Wrap ast_type_qualifier contents in a struct in a union 1571- glsl: Regenerate files modified by previous commits 1572- glcpp: Add the define for ARB_explicit_attrib_location when present 1573- glcpp: Regenerate files changes by previous commit 1574- glsl: Add parser support for GL_ARB_explicit_attrib_location layouts 1575- glsl: Regenerate files changes by previous commit 1576- glsl: Track explicit location in AST to IR translation 1577- glsl: Add linker support for explicit attribute locations 1578- main: Enable GL_ARB_explicit_attrib_location for swrast 1579- intel: Enable GL_ARB_explicit_attrib_location 1580- glsl: Remove const decoration from inlined function parameters 1581- docs: skeleton for 7.10 release notes 1582- docs: Update status of GL 3.x related extensions 1583- mesa: Validate assembly shaders when GLSL shaders are used 1584- glsl: Fix incorrect assertion 1585- linker: Reject shaders that have unresolved function calls 1586- mesa: Silence unused variable warning 1587- mesa: Refactor validation of shader targets 1588- mesa: Clean up two 'comparison between signed and unsigned' warnings 1589- mesa: Clean up various 'unused parameter' warnings in shaderapi 1590- glsl: Slightly change the semantic of \_LinkedShaders 1591- linker: Trivial indention fix 1592- i965: Fix indentation after commit 3322fbaf 1593- linker: Improve handling of unread/unwritten shader inputs/outputs 1594- glapi: Add GL_EXT_separate_shader_objects 1595- glapi: Commit files changed by previous commit 1596- mesa: Add infrastructure to track GL_EXT_separate_shader_objects 1597- mesa: Skeletal support for GL_EXT_separate_shader_objects 1598- mesa: Add display list support for GL_EXT_separate_shader_objects 1599 functions 1600- mesa: Track an ActiveProgram distinct from CurrentProgram 1601- Track separate programs for each stage 1602- swrast: Enable GL_EXT_separate_shader_objects in software paths 1603- intel: Enable GL_EXT_separate_shader_objects in Intel drivers 1604- docs: add GL_EXT_separate_shader_objects to release notes 1605- glsl: Fix incorrect gl_type of sampler2DArray and 1606 sampler1DArrayShadow 1607- ir_to_mesa: Refactor code for emitting DP instructions 1608- mesa: Allow query of MAX_SAMPLES with EXT_framebuffer_multisample 1609- glsl: Refactor is_vec_{zero,one} to be methods of ir_constant 1610- glsl: Simplify generation of swizzle for vector constructors 1611- glsl: Make is_zero and is_one virtual methods of ir_rvalue 1612- ir_to_mesa: Generate smarter code for some conditional moves 1613- glsl: Add ir_unop_sin_reduced and ir_unop_cos_reduced 1614- glsl: Eliminate assumptions about size of ir_expression::operands 1615- glsl: Add ir_rvalue::is_negative_one predicate 1616- glsl: Add unary ir_expression constructor 1617- glsl: Add ir_quadop_vector expression 1618- glsl: Fix matrix constructors with vector parameters 1619- i915: Disallow alpha, red, RG, and sRGB as render targets 1620- glsl: Use M_LOG2E constant instead of calling log2 1621- glsl: Lower ir_binop_pow to a sequence of EXP2 and LOG2 1622- i915: Request that POW instructions be lowered 1623- i915: Correctly generate unconditional KIL instructions 1624- glsl: Ensure that equality comparisons don't return a NULL IR tree 1625- i965: Correctly emit constants for aggregate types (array, matrix, 1626 struct) 1627- glsl: Inherrit type of declared variable from initializer 1628- linker: Ensure that unsized arrays have a size after linking 1629- linker: Fix regressions caused by previous commit 1630- glsl: Inherrit type of declared variable from initializer after 1631 processing assignment 1632- linker: Allow built-in arrays to have different sizes between shader 1633 stages 1634- ir_to_mesa: Don't generate swizzles for record derefs of 1635 non-scalar/vectors 1636- Refresh autogenerated file builtin_function.cpp. 1637- glsl: Allow less restrictive uses of sampler array indexing in GLSL 1638 <= 1.20 1639- docs: Import 7.9.1 release notes from 7.9 branch 1640 1641Jakob Bornecrantz (27): 1642 1643- rbug: Cast opcode to corrent int size 1644- rbug: Add function to get opcode name string 1645- scons: Link against talloc in the Gallium DRI drivers 1646- i915g: Link with wrapper sw winsys with scons 1647- tgsi: Actually care what check_soa_dependencies says 1648- tgsi: Fix missing test before check 1649- llvmpipe: Move makefile include to before targets 1650- wrapper: Fix spelling 1651- wrapper: Add a way to dewrap a pipe screen without destroying it 1652- egl: Remove unnecessary headers 1653- target-helpers: Remove per target software wrapper check 1654- graw: Tidy graw xlib scons file a bit 1655- scons: Remove old pipebuffer SConscript 1656- scons: Detabify 1657- scons: Check for pkg-config before trying to use it 1658- scons: Check for libdrm_[intel|radeon] as well 1659- scons: Move dependancy checks to the main gallium scons file 1660- scons: Unify state tracker SConscripts 1661- galahad: Correct the name of the scons library 1662- graw: Use inline sw helper instead of roll your own loader 1663- libgl-xlib: Use sw helper instead of roll your own 1664- libgl-xlib: Use inline debug helper instead of non-inline version 1665- graw: Use inline debug helper instead of non-inline version 1666- gallium: Remove redundant sw and debug target helpers 1667- i915g: Improve debug printing for textures 1668- i915g: Make sure that new vbo gets updated 1669- st/mesa: Unbind all constant buffers 1670 1671Jerome Glisse (75): 1672 1673- r600g: alternative command stream building from context 1674- r600g: move chip class to radeon common structure 1675- r600g: use pipe context for flushing inside map 1676- r600g: add back reference check when mapping buffer 1677- r600g: directly allocate bo for user buffer 1678- r600g: fix multi buffer rendering 1679- r600g: occlusion query for new design 1680- r600g: flush color buffer after draw command 1681- r600g: disable shader rebuild optimization & account cb flush packet 1682- r600g: fix multiple occlusion query on same id 1683- r600g: initial evergreen support in new path 1684- r600g: fix typo in evergreen define (resource are in x range) 1685- r600g: move use_mem_constants flags for new designs structure 1686 alignment 1687- r600g: evergreen fix for new design 1688- r600g: fix compilation after change to evergreend.h 1689- r600g: fixup some evergreen register definitions 1690- r600g: fix evergreen new path 1691- r600g: fix reg definition 1692- r600g: fix evergreen new path 1693- r600g: bring over fix from old path to new path 1694- r600g: fix vertex resource & polygon offset 1695- r600g: disable early cull optimization when occlusion query running 1696- r600g: move around variables to share depth uncompression code 1697- r600g: use depth decompression in new path 1698- r600g: fix index buffer drawing 1699- r600g: build packet header once 1700- r600g: fix pointsprite & resource unbinding 1701- r600g: fix routing btw vertex & pixel shader 1702- r600g: fix occlusion query after change to block structure 1703- r600g: use ptr for blit depth uncompress function 1704- r600g: fix remaining piglit issue in new design 1705- r600g: switch to new design 1706- r600g: suspend/resume occlusion query around clear/copy 1707- r600g: avoid rebuilding the vertex shader if no change to input 1708 format 1709- r600g: use a hash table instead of group 1710- r600g: delete old path 1711- r600g: cleanup 1712- r600g: more cleanup 1713- r600g: use constant buffer instead of register for constant 1714- r600g: fix constant & literal src splitting, also fix mplayer gl2 1715 shader 1716- evergreeng: avoid overlapping border color btw VS & PS 1717- r600g: indentation fixes 1718- r600g: rename radeon_ws_bo to r600_bo 1719- r600g: allow r600_bo to be a sub allocation of a big bo 1720- r600g: use r600_bo for relocation argument, simplify code 1721- r600g: rename radeon_ws_bo to r600_bo 1722- r600g: remove dead label & fix indentation 1723- r600g: store reloc information in bo structure 1724- r600g: improve bo flushing 1725- r600g: simplify block relocation 1726- r600g: userspace fence to avoid kernel call for testing bo busy 1727 status 1728- r600g: avoid segfault due to unintialized list pointer 1729- r600g: fix dirty state handling 1730- r600g: allow driver to work without submitting cmd to GPU 1731- gallium/noop: no operation gallium driver 1732- r600g: code cleanup (indent, trailing space, empty line ...) 1733- r600g: fix occlusion query on evergreen (avoid lockup) 1734- r600g: add fetch shader capabilities 1735- r600g: dump raw shader output for debugging 1736- r600g: update polygon offset only when rasterizer or zbuffer change 1737- r600g: indentation fix 1738- r600g: more indentation fix + warning silencing + dead code removal 1739- r600g: build fetch shader from vertex elements 1740- r600g: avoid useless shader rebuild at draw call 1741- r600g: remove useless flush map 1742- r600g: remove dead code 1743- r600g: fix userspace fence against lastest kernel 1744- r600g: avoid using pb\* helper we are loosing previous cpu cycle with 1745 it 1746- r600g: specialized upload manager 1747- r600g: indentation cleanup 1748- r600g: fix bo size when creating bo from handle 1749- r600g: fix segfault when translating vertex buffer 1750- r600g: need to reference upload buffer as the might still live 1751 accross flush 1752- r600g: properly unset vertex buffer 1753- r600g: avoid segfault 1754 1755Joakim Sindholt (3): 1756 1757- util/u_blitter: fix leak 1758- radeong: fix leaks 1759- r300g: silence guard band cap errors 1760 1761Johann Rudloff (3): 1762 1763- radeon: Implement EGL_MESA_no_surface_extension 1764- radeon: Implement \__DRI_IMAGE and EGL_MESA_image_drm 1765- radeon: Implement GL_OES_EGL_image 1766 1767John Doe (3): 1768 1769- r600g: misc cleanup 1770- r600g: don't double count dirty block 1771- r600g: keep a mapping around for each bo 1772 1773Jon TURNEY (1): 1774 1775- Ensure -L$(TOP)/$(LIB_DIR) appears in link line before any -L in 1776 $LDFLAGS 1777 1778José Fonseca (128): 1779 1780- gallivm: Fix address register swizzle. 1781- gallivm: Start collecting bitwise arithmetic helpers in a new module. 1782- gallivm: Clamp indirect register indices to file_max. 1783- util: linearized sRGB values don't fit into 8bits 1784- llvmpipe: Default to no threading on single processor systems. 1785- tgsi: Don't ignore indirect registers in tgsi_check_soa_dependencies 1786- llvmpipe: Describe how to profile llvmpipe. 1787- llvmpipe: When failing free fs shader too. 1788- util: Flush stdout on util_format. 1789- gallivm: Add unorm support to lp_build_lerp() 1790- llvmpipe: Special case complementary and identify blend factors in 1791 SoA. 1792- llvmpipe: Make rgb/alpha bland func/factors match, when there is no 1793 alpha. 1794- draw: Prevent clipped vertices overflow. 1795- draw: Fulfil the new min_lod/max_lod/lod_bias/border_color dynamic 1796 state 1797- gallivm: Fetch the lod from the dynamic state when min_lod == 1798 max_lod. 1799- gallivm: Remove dead experimental code. 1800- llvmpipe: Decouple sampler view and sampler state updates. 1801- scons: New build= option, with support for checked builds. 1802- scons: New build= option, with support for checked builds. 1803- trace: Fix set_index_buffer and draw_vbo tracing. 1804- python/retrace: Handle set_index_buffer and draw_vbo. 1805- gallivm: Use SSE4.1's ROUNDSS/ROUNDSD for scalar rounding. 1806- gallivm: More comprehensive border usage logic. 1807- retrace: Handle clear_render_target and clear_depth_stencil. 1808- llvmpipe: Dump a few missing shader key flags. 1809- llvmpipe: Fix perspective interpolation for point sprites. 1810- llvmpipe: Fix sprite coord perspective interpolation of Q. 1811- gallivm: Take the type signedness in consideration in 1812 round/ceil/floor. 1813- gallivm: Use a faster (and less accurate) log2 in lod computation. 1814- gallivm: Fast implementation of iround(log2(x)) 1815- gallivm: Combined ifloor & fract helper. 1816- gallivm: Only apply min/max_lod when necessary. 1817- gallivm: Compute lod as integer whenever possible. 1818- util: Cleanup util_pack_z_stencil and friends. 1819- llvmpipe: Cleanup depth-stencil clears. 1820- gallivm: Vectorize the rho computation. 1821- gallivm: Do not do mipfiltering when magnifying. 1822- gallivm: Simplify lp_build_mipmap_level_sizes' interface. 1823- gallivm: Don't compute the second mipmap level when frac(lod) == 0 1824- gallivm: Use lp_build_ifloor_fract for lod computation. 1825- gallivm: Clamp mipmap level and zero mip weight simultaneously. 1826- gallivm: Fix copy'n'paste typo in previous commit. 1827- gallivm: Implement brilinear filtering. 1828- gallivm: Use the wrappers for SSE pack intrinsics. 1829- gallivm: Avoid control flow for two-sided stencil test. 1830- gallivm: Warn when doing inefficient integer comparisons. 1831- gallivm: Move into the as much of the second level code as possible. 1832- llvmpipe: First minify the texture size, then broadcast. 1833- gallivm: Help for combined extraction and broadcasting. 1834- gallivm: Do size computations simultanously for all dimensions (AoS). 1835- llvmpipe: Prevent z > 1.0 1836- llvmpipe: Fix MSVC build. Enable the new SSE2 code on non SSE3 1837 systems. 1838- gallivm: Handle code have ret correctly. 1839- util: Defined M_SQRT2 when not available. 1840- gallivm: Less code duplication in log computation. 1841- gallivm: Special bri-linear computation path for unmodified rho. 1842- gallivm: Don't generate Phis for execution mask. 1843- gallivm: Use varilables instead of Phis for cubemap selection. 1844- gallivm: Remove support for Phi generation. 1845- gallivm: Factor out the SI->FP texture size conversion for SoA path 1846 too 1847- gallivm: Simplify if/then/else implementation. 1848- gallivm: Cleanup the rest of the flow module. 1849- gallivm: Fix a long standing bug with nested if-then-else emission. 1850- gallivm: Allow to disable bri-linear filtering with 1851 GALLIVM_DEBUG=no_brilinear runtime option 1852- gallivm: Use variables instead of Phis in loops. 1853- gallivm: Pass texture coords derivates as scalars. 1854- llvmpipe: Remove outdated comment about stencil testing. 1855- gallivm: Eliminate unsigned integer arithmetic from texture 1856 coordinates. 1857- gallium: Define C99 restrict keyword where absent. 1858- tgsi: Export some names for some tgsi enums. 1859- gallivm: More detailed analysis of tgsi shaders. 1860- llvmpipe: Use lp_tgsi_info. 1861- llvmpipe: Do not dispose the execution engine. 1862- llvmpipe: Fix MSVC build. 1863- llmvpipe: improve mm_mullo_epi32 1864- gallivm: Name anonymous union. 1865- llvmpipe: Unbreak Z32_FLOAT. 1866- gallivm: More accurate float -> 24bit & 32bit unorm conversion. 1867- llvmpipe: Generalize the x8z24 fast path to all depth formats. 1868- llvmpipe: Fix depth-stencil regression. 1869- llvmpipe: Ensure z_shift and z_width is initialized. 1870- gallivm: Fix SoA cubemap derivative computation. 1871- llvmpipe: Fix bad refactoring. 1872- llvmpipe: Initialize bld ctx via lp_build_context_init instead of 1873 ad-hoc and broken code. 1874- gallivm: Comment lp_build_insert_new_block(). 1875- gallivm: Add a note about SSE4.1's nearest mode rounding. 1876- llvmpipe: Don't test rounding of x.5 numbers. 1877- gallium: Avoid using \__doc_\_ in python scripts. 1878- gallivm: always enable LLVMAddInstructionCombiningPass() 1879- gallivm: Remove the EMMS opcodes. 1880- mesa: Fix windows build (uint -> GLuint). 1881- scons: Revamp how to specify targets to build. 1882- scons: Fix MinGW cross-compilation. 1883- scons: Some pipe drivers are not portable for MSVC 1884- scons: Restore x11 tool behavior for backwards compatability. 1885- scons: Disable python state tracker when swig is not present. 1886- r600g: List recently added files in SConscript. 1887- scons: Add aliases for several pipe drivers. 1888- scons: i915 can't build on MSVC either. 1889- scons: Propagate installation targets. 1890- xorg/vmwgfx: Add missing source file to SConscript. 1891- st/xorg: Add missing n to error message. 1892- st/xorg: Detect libkms with scons too. 1893- xorg/vmwgfx: Link libkms when available. 1894- r600g: Swap the util_blitter_destroy call order. 1895- gallivm: Allocate TEMP/OUT arrays only once. 1896- libgl-gdi: Allow to pick softpipe/llvmpipe on runtime. 1897- scons: Use inline wrap helpers more consistently. 1898- svga: Use consistent hexadecimal representation on debug output. 1899- scons: Alias for svga 1900- wgl: Stub WGL_ARB_pbuffer support. 1901- wgl: More complete WGL_ARB_pbuffer support. 1902- svga: Silence debug printf. 1903- scons: Move MSVS_VERSION option to common module. 1904- vega: Remove extraneous ; 1905- retrace: Some fixes. 1906- util: C++ safe. 1907- wgl: Fix double free. Remove dead code. 1908- util: Plug leaks in util_destroy_gen_mipmap. 1909- util: \__builtin_frame_address() doesn't work on mingw. 1910- util: Don't try to use imagehlp on mingw. 1911- wgl: Unreference the current framebuffer after the make_current call. 1912- WIN32_THREADS -> WIN32 1913- mapi: Hack to avoid vgCreateFont being generated as vgCreateFontA. 1914- wgl: Fix visual's buffer_mask configuration. 1915- mesa: Temporary hack to prevent stack overflow on windows 1916- mesa: Bump the number of bits in the register index. 1917- llvmpipe: Plug fence leaks. 1918 1919Julien Cristau (1): 1920 1921- Makefile: don't include the same files twice in the tarball 1922 1923Keith Whitwell (89): 1924 1925- llvmpipe: brackets around macro arg 1926- llvmpipe: remove duplicate code 1927- llvmpipe: return zero from floor_pot(zero) 1928- gallivm: make lp_build_sample_nop public 1929- llvmpipe: add LP_PERF flag to disable various aspects of 1930 rasterization 1931- llvmpipe: add DEBUG_FS to dump variant information 1932- llvmpipe: use llvm for attribute interpolant calculation 1933- graw: add frag-face shader 1934- llvmpipe: fix flatshading in new line code 1935- draw: don't apply flatshading to clipped tris with <3 verts 1936- llvmpipe: handle FACING interpolants in line and point setup 1937- llvmpipe: handle up to 8 planes in triangle binner 1938- llvmpipe: make debug_fs_variant respect variant->nr_samplers 1939- gallivm: don't apply zero lod_bias 1940- llvmpipe: fail gracefully on oom in scene creation 1941- llvmpipe: avoid overflow in triangle culling 1942- gallivm: special case conversion 4x4f to 1x16ub 1943- gallivm: round rather than truncate in new 4x4f->1x16ub conversion 1944 path 1945- llvmpipe: clean up setup_tri a little 1946- llvmpipe: add rast_tri_4_16 for small lines and points 1947- llvmpipe: fix off-by-one in tri_16 1948- llvmpipe: defer attribute interpolation until after mask and ztest 1949- llvmpipe: use alloca for fs color outputs 1950- llvmpipe: store zero into all alloca'd values 1951- llvmpipe: dump fragment shader ir and asm when LP_DEBUG=fs 1952- gallivm: specialized x8z24 depthtest path 1953- gallivm: prefer blendvb for integer arguments 1954- gallivm: simpler uint8->float conversions 1955- llvmpipe: try to be sensible about whether to branch after mask 1956 updates 1957- llvmpipe: clean up shader pre/postamble, try to catch more early-z 1958- llvmpipe: simplified SSE2 swz/unswz routines 1959- llvmpipe: try to do more of rast_tri_3_16 with intrinsics 1960- llvmpipe: add debug helpers for epi32 etc 1961- llvmpipe: try to keep plane c values small 1962- llvmpipe: fix typo in last commit 1963- gallium: move sse intrinsics debug helpers to u_sse.h 1964- r600g: add missing file to sconscript 1965- gallivm: don't branch on KILLs near end of shader 1966- Revert "llvmpipe: try to keep plane c values small" 1967- llvmpipe: make sure intrinsics code is guarded with PIPE_ARCH_SSE 1968- llvmpipe: don't try to emit non-existent color outputs 1969- r600/drm: fix segfaults in winsys create failure path 1970- r600g: emit hardware linewidth 1971- r600g: handle absolute modifier in shader translator 1972- llvmpipe: reintroduce SET_STATE binner command 1973- llvmpipe: don't pass frontfacing as a float 1974- llvmpipe: slightly shrink the size of a binned triangle 1975- llvmpipe: don't store plane.ei value in binned data 1976- gallium: move some intrinsics helpers to u_sse.h 1977- llvmpipe: do plane calculations with intrinsics 1978- llvmpipe: use aligned loads/stores for plane values 1979- llvmpipe: fix non-sse build after recent changes 1980- llvmpipe: check shader outputs are non-null before using 1981- llvmpipe: validate color outputs against key->nr_cbufs 1982- llvmpipe: clean up fields in draw_llvm_variant_key 1983- llvmpipe: remove setup fallback path 1984- llvmpipe: fail cleanly on malloc failure in lp_setup_alloc_triangle 1985- Merge remote branch 'origin/master' into lp-setup-llvm 1986- llvmpipe: remove unused file 1987- llvmpipe: remove unused arg from jit_setup_tri function 1988- Merge branch 'llvm-cliptest-viewport' 1989- draw: make sure viewport gets updated in draw llvm shader 1990- llvmpipe: turn off draw offset/twoside when we can handle it 1991- llvmpipe: avoid generating tri_16 for tris which extend past tile 1992 bounds 1993- llvmpipe: guard against NULL task->query pointer 1994- st/mesa: unbind constant buffer when not in use 1995- r600g: propagate usage flags in texture transfers 1996- r600g: propogate resource usage flags to winsys, use to choose bo 1997 domains 1998- r600g: use a buffer in GTT as intermediate on texture up and 1999 downloads 2000- r600g: remove unused flink, domain fields from r600_resource 2001- r600g: set hardware pixel centers according to gl_rasterization_rules 2002- evergreeng: protect against null constant buffers 2003- r600g: don't call debug_get_bool_option for tiling more than once 2004- evergreeng: respect linewidth state, use integer widths only 2005- evergreeng: set hardware pixelcenters according to 2006 gl_rasterization_rules 2007- r600g: avoid recursion with staged uploads 2008- r600g: attempt to turn on DXTn formats 2009- r600g: translate ARR instruction 2010- r600: fix my pessimism about PIPE_TRANSFER_x flags 2011- ws/r600: match bo_busy shared/fence logic in bo_wait 2012- r600g: guard experimental s3tc code with R600_ENABLE_S3TC 2013- r600g: do not try to use staging resource for depth textures 2014- r600g: enforce minimum stride on render target texture images 2015- llvmpipe: fix up twoside after recent changes 2016- llvmpipe: twoside for specular color also 2017- Merge branch 'lp-offset-twoside' 2018- llvmpipe: raise dirty flag on transfers to bound constbuf 2019- llvmpipe: remove misleading debug string 2020- llvmpipe: shortcircuit some calls to set_scene_state 2021 2022Kenneth Graunke (94): 2023 2024- glsl: Change from has_builtin_signature to has_user_signature. 2025- glsl: Don't print blank (function ...) headers for built-ins. 2026- glsl: Properly handle nested structure types. 2027- glsl/builtins: Fix equal and notEqual builtins. 2028- glsl/builtins: Switch comparison functions to just return an 2029 expression. 2030- glsl: Add comments to clarify the types of comparison binops. 2031- glsl: Fix broken handling of ir_binop_equal and ir_binop_nequal. 2032- glsl: "Copyright", not "Constantright" 2033- i965: Fix incorrect batchbuffer size in gen6 clip state command. 2034- i965: Use logical-not when emitting ir_unop_ceil. 2035- glsl: Add front-end support for the "trunc" built-in. 2036- glsl: Refresh autogenerated file builtin_function.cpp. 2037- i965: Use RNDZ for ir_unop_trunc in the new FS. 2038- i965: Correctly emit the RNDZ instruction. 2039- i965: Clean up a warning in the old fragment backend. 2040- glsl: Add a new ir_unop_round_even opcode for GLSL 1.30's roundEven. 2041- glsl: Add front-end support for GLSL 1.30's roundEven built-in. 2042- i965: Add support for ir_unop_round_even via the RNDE instruction. 2043- glsl: Add support for the 1.30 round() built-in. 2044- glsl: Refresh autogenerated file builtin_function.cpp. 2045- glsl: Don't return NULL IR for erroneous bit-shift operators. 2046- i965: Add missing "break" statement. 2047- glsl: Fix copy and paste error in ast_bit_and node creation. 2048- glsl: Regenerate parser files. 2049- i965: Remove unused variable. 2050- glsl: Remove useless ir_shader enumeration value. 2051- mesa: Remove FEATURE_ARB_shading_language_120 macro. 2052- glcpp: Return NEWLINE token for newlines inside multi-line comments. 2053- glcpp: Refresh autogenerated lexer file. 2054- glsl: Add support for GLSL 1.30's modf built-in. 2055- glsl: Refresh autogenerated file builtin_function.cpp. 2056- generate_builtins.py: Output large strings as arrays of characters. 2057- Refresh autogenerated file builtin_function.cpp. 2058- glsl: Fix constant component count in vector constructor emitting. 2059- Fix build on systems where "python" is python 3. 2060- i965: Add bit operation support to the fragment shader backend. 2061- glsl: Remove unused ARRAY_SIZE macro. 2062- glsl/builtins: Rename 'x' to 'y_over_x' in atan(float) 2063 implementation. 2064- glsl/builtins: Clean up some ugly autogenerated code in atan. 2065- Refresh autogenerated file builtin_function.cpp. 2066- glsl: Don't print a useless space at the end of an S-Expression list. 2067- ir_reader: Return a specific ir_dereference variant. 2068- ir_reader: Remove useless error check. 2069- ir_reader: Fix some potential NULL pointer dereferences. 2070- ir_dead_functions: Actually free dead functions and signatures. 2071- glsl: Remove unnecessary "unused variable" warning suppression. 2072- glsl: Remove GLSL_TYPE_FUNCTION define. 2073- glsl: Convert glsl_type::base_type from #define'd constants to an 2074 enum. 2075- glsl: Rework reserved word/keyword handling in the lexer. 2076- glsl: Add new keywords and reserved words for GLSL 1.30. 2077- glsl: Add support for the 'u' and 'U' unsigned integer suffixes. 2078- glsl: Refresh autogenerated lexer and parser files. 2079- generate_builtins.py: Fix inconsistent use of tabs and spaces 2080 warning. 2081- glsl: Implement the asinh, acosh, and atanh built-in functions. 2082- glsl: Refresh autogenerated file builtin_function.cpp. 2083- glsl: Add constant expression handling for asinh, acosh, and atanh. 2084- glsl: Remove unused and out of date Makefile.am. 2085- glsl: Rename various ir\_\* files to lower\_\* and opt_*. 2086- glcpp: Define GL_FRAGMENT_PRECISION_HIGH if GLSL version >= 1.30. 2087- Refresh autogenerated glcpp parser. 2088- glsl: Fix constant expression handling for <, >, <=, >= on vectors. 2089- glsl: Unconditionally define GL_FRAGMENT_PRECISION_HIGH in ES2 2090 shaders. 2091- Regenerate glcpp parser. 2092- glsl: Reimplement the "cross" built-in without ir_binop_cross. 2093- Refresh autogenerated file builtin_function.cpp. 2094- glsl: Remove the ir_binop_cross opcode. 2095- glsl: Refactor get_num_operands. 2096- glsl: Simplify a type check by using type->is_integer(). 2097- glsl: Combine many instruction lowering passes into one. 2098- mesa: Fix glGet of ES2's GL_MAX_*_VECTORS properties. 2099- glsl: Don't inline function prototypes. 2100- glsl: Use do_common_optimization in the standalone compiler. 2101- glsl: Add a virtual as_discard() method. 2102- glsl: Refactor out cloning of function prototypes. 2103- glsl: Lazily import built-in function prototypes. 2104- glsl: Remove anti-built-in hacks from the print visitor. 2105- glsl/linker: Free any IR discarded by optimization passes. 2106- glsl: Add an optimization pass to simplify discards. 2107- glsl: Add a lowering pass to move discards out of if-statements. 2108- glsl: Remove "discard" support from lower_jumps. 2109- glsl: Add comments to lower_jumps (from the commit message). 2110- ir_print_visitor: Print out constant structure values. 2111- glsl: Factor out code which emits a new function into the IR stream. 2112- symbol_table: Add support for adding a symbol at top-level/global 2113 scope. 2114- glsl: Properly add functions during lazy built-in prototype 2115 importing. 2116- glcpp: Don't emit SPACE tokens in conditional_tokens production. 2117- Refresh autogenerated glcpp parser. 2118- glsl: Clean up code by adding a new is_break() function. 2119- glsl: Consider the "else" branch when looking for loop breaks. 2120- Remove OES_compressed_paletted_texture from the ES2 extension list. 2121- glsl/builtins: Compute the correct value for smoothstep(vec, vec, 2122 vec). 2123- glsl: Support if-flattening beyond a given maximum nesting depth. 2124- i965: Flatten if-statements beyond depth 16 on pre-gen6. 2125- i965: Internally enable GL_NV_blend_square on ES2. 2126 2127Kristian Høgsberg (16): 2128 2129- glx: Hold on to drawables if we're just switching to another context 2130- intel: Fix GL_ARB_shading_language_120 commit 2131- dri2: Make createImageFromName() take a \__DRIscreen instead of 2132 \__DRIcontext 2133- glx: Invalidate buffers after binding a drawable 2134- dri: Pass the \__DRIscreen and the \__DRIscreen private back to image 2135 lookup 2136- glx: Only remove drawables from the hash when we actually delete them 2137- gles2: Add GL_EXT_texture_format_BGRA8888 support 2138- Get rid of GL/internal/glcore.h 2139- gl: Remove unused GLcontextModes fields 2140- Rename GLvisual and \__GLcontextModes to struct gl_config 2141- Drop GLframebuffer typedef and just use struct gl_framebuffer 2142- Drop GLcontext typedef and use struct gl_context instead 2143- Drop the "neutral" tnl module 2144- Only install vtxfmt tables for OpenGL 2145- i965: Don't write mrf assignment for pointsize output 2146- docs: Fix MESA_drm_image typo 2147 2148Krzysztof Smiechowicz (1): 2149 2150- nvfx: Pair os_malloc_aligned() with os_free_aligned(). 2151 2152Luca Barbieri (84): 2153 2154- auxiliary: fix unintended fallthrough 2155- glsl: add pass to lower variable array indexing to conditional 2156 assignments 2157- auxiliary: fix depth-only and stencil-only clears 2158- gallium: avoid the C++ keyword "template" in sw_winsys.h 2159- softpipe: make z/s test always pass if no zsbuf, instead of crashing 2160- tgsi: add switch/case opcodes to tgsi_opcode_tmp.h 2161- softpipe: fix whitespace 2162- d3d1x: add new Direct3D 10/11 COM state tracker for Gallium 2163- d3d1x: add blob and signature extraction APIs 2164- d3d1x: fix compilation with recent Wine versions installed 2165- d3d1x: add missing file 2166- d3d1x: actually enable and fix blob apis 2167- d3d1x: fix build with compilers other than GCC 4.5 2168- d3d1x: add template parameters to base class ctor calls for GCC 4.4 2169- d3d1x: fix GCC 4.1/4.2 build 2170- d3d1x: ignore errors while building docs 2171- d3d1x: attempt to fix/workaround bug #30322 2172- nvfx: remove gl_PointCoord hack 2173- glx: decouple dri2.c and GLX, fixing Gallium EGL and d3d1x build 2174- winsys: automatically build sw winsys needed by EGL and d3d1x 2175- d3d1x: don't build progs automatically 2176- d3d1x: add missing memory barrier 2177- d3d1x: link with CXXFLAGS 2178- d3d1x: fix cf analysis 2179- d3d1x: fix warning 2180- d3d1x: fix segfault when hashing 2181- d3d1x: destroy native_display on adapter destruction 2182- d3d1x: fix GUID declarations 2183- d3d1x: redesign the HWND resolver interface 2184- d3d1x: fix API name 2185- d3d1x: define GUIDs in the normal way 2186- d3d1x: add Wine dlls (tri, tex working, but no other testing) 2187- d3d1x: properly reference count the backend 2188- d3d1x: fix deadlocks on non-recursive mutex 2189- d3d1x: bind NULL CSOs before destroying default CSOs on context dtor 2190- d3d1x: initialize the mutex 2191- d3d1x: autogenerate shader enums and text from def files 2192- d3d1x: s/tpf/sm4/g 2193- d3d1x: normalize whitespace 2194- d3d1x: remove specstrings 2195- d3d1x: minifix 2196- d3d1x: rename context params 2197- d3d11: rename screen params 2198- d3d1x: rename params in misc and objects 2199- d3d1x: rename parameters in dxgi 2200- d3d11: obliterate IDL parameter names 2201- d3d1x: remove specstrings.h include 2202- d3d1x: flush the pipe context when presenting 2203- d3d1x: remove another include specstrings.h 2204- d3d1x: flush properly 2205- d3d1x: add missing guid.cpp 2206- d3d1x: fix build without system EGL/egl.h 2207- d3d1x: add autogenerated files as prerequisites, so make builds them 2208- d3d1x: obliterate IDL parameter names from d3d10.idl from Wine too 2209- d3d1x: add shader dumping 2210- d3d1x: add untested support for geometry shader translation 2211- d3d1x: don't assert on unsupported resource types 2212- d3d1x: fix CheckMultisampleQualityLevels 2213- d3d1x: draw to the correct buffer 2214- d3d1x: fix linking of dxbc2tgsi 2215- nvfx: allow setting NULL constant buffers 2216- nvfx: add RGB framebuffer format support in addition to BGR 2217- d3d1x: don't crash on drivers not supporting vertex or geometry 2218 sampling 2219- d3d1x: assert if X visual is not among enumerated visuals 2220- d3d1x: stop using GLX in demos, just use the default visual 2221- d3d1x: CRLF -> LF in progs 2222- mesa: make makedepend an hard requirement 2223- gallium: add $(PROGS_DEPS) as dependencies for $(PROGS) 2224- d3d1x: fix parallel build 2225- d3d1x: add private gitignore file 2226- d3d1x: fix progs linking if not all EGL platforms are enabled 2227- d3d1x: link progs with CXXFLAGS 2228- d3d11: advertise IDXGIDevice1, not just IDXGIDevice 2229- d3d11: ignore StructureByteStride 2230- d3d1x: link to libdrm for X11 platform too 2231- ureg: support centroid interpolation 2232- d3d1x: support centroid interpolation 2233- d3d1x: properly support specifying MipLevels as 0 2234- d3d1x: put proper calling convention in headers, fixes 64-bit builds 2235- d3d1x: rework DXGI for occlusion testing and default width/height 2236- d3d1x: fix Map 2237- d3d11: fix reference counting so devices get freed 2238- d3d1x: work around crash in widl 2239- glsl: Unroll loops with conditional breaks anywhere (not just the 2240 end) 2241 2242Lucas Stach (1): 2243 2244- nvfx: fill PIPE_CAP_PRIMITIVE_RESTART and 2245 PIPE_CAP_SHADER_STENCIL_EXPORT 2246 2247Marek Olšák (100): 2248 2249- r300g: prevent creating multiple winsys BOs for the same handle 2250- r300g/swtcl: fix CS overrun 2251- st/mesa: fix assertion failure in GetTexImage for cubemaps 2252- util: make calling remove_from_list multiple times in a row safe 2253- r300g: fixup long-lived BO maps being incorrectly unmapped when 2254 flushing 2255- r300g: make accessing map_list and buffer_handles thread-safe 2256- r300g: fix a copy-paste typo for logging 2257- r300g: fix the border color for every format other than 2258 PIPE_FORMAT_B8G8R8A8 2259- Build r300g by default 2260- util: fix util_pack_color for B4G4R4A4 2261- r300g: fix macrotiling on R350 2262- r300g: code cleanups 2263- r300/compiler: fix projective mapping of 2D NPOT textures 2264- r300/compiler: do not use copy propagation if SaturateMode is used 2265- r300/compiler: fix shadow sampling with swizzled coords 2266- r300g: add support for 3D NPOT textures without mipmapping 2267- r300g: fix swizzling of texture border color 2268- configure.ac: look for libdrm_radeon before building 2269 gallium/r300,r600 2270- configure.ac: do not build xorg-r300g by default 2271- Makefile: ensure Gallium's Makefile.xorg and SConscript.dri are in 2272 the tarball 2273- r300g: add support for formats beginning with X, like X8R8G8B8 2274- r300g: fix conditional rendering in non-wait path 2275- r300g: add support for R8G8 colorbuffers 2276- r300g: add support for L8A8 colorbuffers 2277- update release notes for Gallium 2278- r300g: fix microtiling for 16-bits-per-channel formats 2279- r300g: do not print get_param errors in non-debug build 2280- r300g: say no to PIPE_CAP_STREAM_OUTPUT and 2281 PIPE_CAP_PRIMITIVE_RESTART 2282- mesa: allow FBO attachments of formats LUMINANCE, LUMINANCE_ALPHA, 2283 and INTENSITY 2284- r300g: fix texture border for 16-bits-per-channel formats 2285- st/mesa: support RGBA16 and use it for RGBA12 as well 2286- r300g: add a default channel ordering of texture border for unhandled 2287 formats 2288- r300g: mention ATI in the renderer string 2289- r300g: rename has_hyperz -> can_hyperz 2290- r300g: turn magic numbers into names in the hyperz code 2291- gallium: add CAPs for indirect addressing and lower it in st/mesa 2292 when needed 2293- tgsi: fill out CAPs for indirect addressing 2294- i915g: fill out CAPs for indirect addressing 2295- i965g: fill out CAPs for indirect addressing 2296- nv50: fill out CAPs for indirect addressing 2297- nvfx: fill out CAPs for indirect addressing 2298- r300g: fill out CAPs for indirect addressing 2299- r600g: fill out CAPs for indirect addressing 2300- svga: fill out CAPs for indirect addressing 2301- r300g: fix texture border color for all texture formats 2302- r300g: clean up redundancy in draw functions 2303- r300g: return shader caps from Draw for SWTCL vertex shaders 2304- r300g: remove the hack with OPCODE_RET 2305- r300g: print FS inputs uninitialized due to hardware limits to stderr 2306- r300g: fix rendering with no vertex elements 2307- st/mesa: enable ARB_explicit_attrib_location and 2308 EXT_separate_shader_objects 2309- docs: add GL 4.1 status 2310- gallium: add PIPE_SHADER_CAP_SUBROUTINES 2311- st/mesa: set MaxUniformComponents 2312- u_blitter: use PIPE_TRANSFER_DISCARD to prevent cpu/gpu stall 2313- r300/compiler: fix rc_rewrite_depth_out for it to work with any 2314 instruction 2315- r300/compiler: remove duplicate function rc_mask_to_swz 2316- r300/compiler: add a function for swizzling a mask 2317- r300/compiler: move util functions to radeon_compiler_util 2318- u_blitter: interpolate clear color using a GENERIC varying instead of 2319 COLOR 2320- st/mesa: fix texture border color for RED and RG base formats 2321- util: rename u_mempool -> u_slab 2322- r300g: fix texture border color once again 2323- r300/compiler: implement and lower OPCODE_CLAMP 2324- ir_to_mesa: Add support for conditional discards. 2325- r300g: fix texture swizzling with compressed textures on r400-r500 2326- r300g: disable ARB_texture_swizzle if S3TC is enabled on r3xx-only 2327- r300g: fix up cubemap texture offset computation 2328- r300/compiler: disable the swizzle lowering pass in vertex shaders 2329- r300g: fix build 2330- r300g: use internal BO handle for add_buffer and write_reloc 2331- r300g: implement simple transfer_inline_write for buffers 2332- mesa, st/mesa: fix gl_FragCoord with FBOs in Gallium 2333- r300g: fix pointer arithmetic with void\* in transfer_inline_write 2334- r300g: do not remove unused constants if we are not near the limit 2335- r300g: add capability bit index_bias_supported 2336- r300g: one more r500_index_bias_supported leftover 2337- r300g: do not use the index parameter in set_constant_buffer 2338- r300g: cleanup winsys 2339- r300g: optimize looping over atoms 2340- st/mesa: initialize key in st_vp_varient 2341- u_blitter: use util_is_format_compatible in the assert 2342- r300g: cache packet dwords of 3D_LOAD_VBPNTR in a command buffer if 2343 possible 2344- r300g: validate buffers only if any of bound buffers is changed 2345- r300g: also revalidate the SWTCL vertex buffer after its reallocation 2346- r300/compiler: don't terminate regalloc if we surpass max temps limit 2347- r300/compiler: add a function to query program stats (alu, tex, 2348 temps..) 2349- r300/compiler: cleanup rc_run_compiler 2350- r300/compiler: do not print pair/tex/presub program stats for vertex 2351 shaders 2352- r300/compiler: handle DPH and XPD in rc_compute_sources_for_writemask 2353- r300/compiler: make lowering passes possibly use up to two less temps 2354- r300/compiler: remove at least unused immediates if externals cannot 2355 be removed 2356- r300/compiler: fix LIT in VS 2357- r300/compiler: fix swizzle lowering with a presubtract source operand 2358- r300g: fix rendering with a vertex attrib having a zero stride 2359- r300g: finally fix the texture corruption on r3xx-r4xx 2360- r300g/swtcl: re-enable LLVM 2361- r300g: mark vertex arrays as dirty after a buffer_offset change 2362- mesa: fix texel store functions for some float formats 2363- r300/compiler: disable the rename_regs pass for loops 2364 2365Mario Kleiner (1): 2366 2367- mesa/r300classic: Fix dri2Invalidate/radeon_prepare_render for page 2368 flipping. 2369 2370Mathias Fröhlich (3): 2371 2372- r300g: Avoid returning values in a static array, fixing a potential 2373 race 2374- r600g: Only compare active vertex elements 2375- st/mesa: Set PIPE_TRANSFER_DISCARD for 2376 GL_MAP_INVALIDATE_RANGE/BUFFFER_BIT 2377 2378Michal Krol (10): 2379 2380- svga: Fix relative addressing translation for pixel shaders. 2381- svga: Integer constant register file has a separate namespace. 2382- tgsi/exec: Cleanup the remaining arithmetic instructions. 2383- tgsi/exec: Get rid of obsolete condition codes. 2384- tgsi/build: Reduce interface clutter. 2385- graw/gdi: Initial commit. 2386- scons: Hook-up graw-gdi target. 2387- graw/gdi: Fix window dimensions. 2388- os: Open file streams in binary mode. 2389- graw: Export graw_save_surface_to_file(). 2390 2391Nicolas Kaiser (26): 2392 2393- swrast: remove duplicated include 2394- egl: remove duplicated include 2395- gallium/rtasm: remove duplicated include 2396- gallium/util: remove duplicated include 2397- gallium/i915: remove duplicated include 2398- gallium/llvmpipe: remove duplicated include 2399- gallium/softpipe: remove duplicated include 2400- gallium/st: remove duplicated includes 2401- gallium/winsys: remove duplicated include 2402- glx: remove duplicated include 2403- dri/common: remove duplicated include 2404- dri/i810: remove duplicated include 2405- dri/i915: remove duplicated include 2406- dri/i965: remove duplicated include 2407- dri/intel: remove duplicated include 2408- dri/mga: remove duplicated include 2409- dri/r128: remove duplicated include 2410- dri/r300: remove duplicated include 2411- dri/r600: remove duplicated include 2412- dri/radeon: remove duplicated includes 2413- dri/savage: remove duplicated include 2414- main: remove duplicated includes 2415- math: remove duplicated includes 2416- st: remove duplicated include 2417- i965g: use Elements macro instead of manual sizeofs 2418- nv50: fix always true conditional in shader optimization 2419 2420Orion Poplawski (1): 2421 2422- osmesa: link against libtalloc 2423 2424Owen W. Taylor (1): 2425 2426- r600g: Fix location for clip plane registers 2427 2428Peter Clifton (3): 2429 2430- intel: Fix emit_linear_blit to use DWORD aligned width blits 2431- intel: Add assert check for blitting alignment. 2432- meta: Mask Stencil.Clear against stencilMax in \_mesa_meta_Clear 2433 2434Robert Hooker (2): 2435 2436- intel: Add a new B43 pci id. 2437- egl_dri2: Add missing intel chip ids. 2438 2439Roland Scheidegger (16): 2440 2441- gallivm: fix copy&paste bug 2442- gallivm: don't use URem/UDiv when calculating offsets for blocks 2443- gallivm: optimize yuv decoding 2444- gallivm: fix trunc/itrunc comment 2445- gallivm: faster iround implementation for sse2 2446- gallivm: replace sub/floor/ifloor combo with ifloor_fract 2447- gallivm: optimize some tex wrap mode calculations a bit 2448- gallivm: more linear tex wrap mode calculation simplification 2449- gallivm: avoid unnecessary URem in linear wrap repeat case 2450- gallivm: optimize soa linear clamp to edge wrap mode a bit 2451- gallivm: make use of new iround code in lp_bld_conv. 2452- gallivm: fix different handling of [non]normalized coords in linear 2453 soa path 2454- gallivm: only use lp_build_conv 4x4f -> 1x16 ub fastpath with sse2 2455- r200: fix r200 large points 2456- mesa: remove unneeded DD_POINT_SIZE and DD_LINE_WIDTH tricaps 2457- gallium: support for array textures and related changes 2458 2459Shuang He (1): 2460 2461- mesa: allow GLfixed arrays for OpenGL ES 2.0 2462 2463Stephan Schmid (1): 2464 2465- r600g: fix relative addressing when splitting constant accesses 2466 2467Thomas Hellstrom (21): 2468 2469- st/xorg: Don't try to use option values before processing options 2470- xorg/vmwgfx: Make vmwarectrl work also on 64-bit servers 2471- st/xorg: Add a customizer option to get rid of annoying cursor update 2472 flicker 2473- xorg/vmwgfx: Don't hide HW cursors when updating them 2474- st/xorg: Don't try to remove invalid fbs 2475- st/xorg: Fix typo 2476- st/xorg, xorg/vmwgfx: Be a bit more frendly towards cross-compiling 2477 environments 2478- st/xorg: Fix compilation errors for Xservers compiled without 2479 Composite 2480- st/xorg: Don't use deprecated x*alloc / xfree functions 2481- xorg/vmwgfx: Don't use deprecated x*alloc / xfree functions 2482- st/xorg: Fix compilation for Xservers >= 1.10 2483- mesa: Make sure we have the talloc cflags when using the talloc 2484 headers 2485- egl: Add an include for size_t 2486- mesa: Add talloc includes for gles 2487- st/egl: Fix build for include files in nonstandard places 2488- svga/drm: Optionally resolve calls to powf during link-time 2489- gallium/targets: Trivial crosscompiling fix 2490- st/xorg: Add a function to flush pending rendering and damage 2491- gallium/targets/xorg-vmwgfx: Xv fixes 2492- xorg/vmwgfx: Flush even if we don't autopaint the color key 2493- xorg/vmwgfx: Don't clip video to viewport 2494 2495Tilman Sauerbeck (35): 2496 2497- r600g: Fixed a bo leak in r600_blit_state_ps_shader(). 2498- r600g: Use clamped math for RCP and RSQ. 2499- r600g: Formatting fixes. 2500- r600g: Added DB_SHADER_CONTROL defines. 2501- r600g: Only set PA_SC_EDGERULE on rv770 and greater. 2502- r600g: Enable PIPE_SHADER_CAP_TGSI_CONT_SUPPORTED. 2503- r600g: Fixed the shift in S_02880C_KILL_ENABLE. 2504- glsl2: Empty functions can be inlined. 2505- glsl2: Fixed cloning of ir_call error instructions. 2506- r600g: Added support for TGSI_SEMANTIC_FACE. 2507- gallium/docs: Fixed a typo in the SCS opcode description. 2508- r600g: Honour destination operand's writemask in the SCS 2509 implementation. 2510- r600g: Implemented the Z and W component write for the SCS opcode. 2511- python/tests: Fixed tri.py for API and TGSI syntax changes. 2512- r600g: Removed debug code. 2513- gallium/docs: The RET opcode may appear anywhere in a subroutine. 2514- r600g: Destroy the blitter. 2515- r600g: Fixed two texture surface leaks in 2516 r600_blit_uncompress_depth(). 2517- r600g: Cleaned up index buffer reference handling in the draw module. 2518- r600g: Fixed r600_vertex_element leak. 2519- r600g: Added r600_pipe_shader_destroy(). 2520- r600g: Also clear bc data when we're destroying a shader. 2521- r600g: In radeon_bo(), call LIST_INITHEAD early. 2522- r600g: Destroy the blitter. 2523- r600g: Removed unused 'ptr' argument from radeon_bo(). 2524- r600g: Made radeon_bo_pb_map_internal() actually call 2525 radeon_bo_map(). 2526- r600g: Fixed unmap condition in radeon_bo_pb_destroy(). 2527- r600g: Made radeon_bo::map_count signed. 2528- r600g: We don't support PIPE_CAP_PRIMITIVE_RESTART. 2529- r600g: Delete custom_dsa_flush on shutdown. 2530- r600g: Fixed two memory leaks in winsys. 2531- r600g: Destroy the winsys in r600_destroy_screen(). 2532- st/mesa: Reset the index buffer before destroying the pipe context. 2533- st/mesa: Reset the constant buffers before destroying the pipe 2534 context. 2535- r600g: Removed duplicated call to tgsi_split_literal_constant(). 2536 2537Timo Wiren (1): 2538 2539- Fix typos in comments and debug output strings. 2540 2541Tom Fogal (3): 2542 2543- Implement x86_64 atomics for compilers w/o intrinsics. 2544- Prefer intrinsics to handrolled atomic ops. 2545- Revert "Prefer intrinsics to handrolled atomic ops." 2546 2547Tom Stellard (32): 2548 2549- r300/compiler: Refactor the pair instruction data structures 2550- r300g: Always try to build libr300compiler.a 2551- r300/compiler: Fix two mistakes in the presubtract optimization pass. 2552- r300/compiler: Add more helper functions for iterating through 2553 sources 2554- r300/compiler: Print immediate values after "dead constants" pass 2555- r300/compiler: radeon_remove_constants.c: fix indentation 2556- r300/compiler: Use rc_for_all_reads_src() in "dead constants" pass 2557- r300/compiler: Fix segfault in error path 2558- r300/compiler: Don't use rc_error() unless the error is unrecoverable 2559- r300/compiler: Don't merge instructions that write output regs and 2560 ALU result 2561- r300/compiler: Create a helper function for merging presubtract 2562 sources 2563- r300/compiler: Fix incorrect assumption 2564- r300/compiler: Clear empty registers after constant folding 2565- r300/compiler: Add a new function for more efficient dataflow 2566 analysis 2567- r300g: Add new debug option for logging vertex/fragment program stats 2568- r300/compiler: Use rc_get_readers_normal() for presubtract 2569 optimizations 2570- r300/compiler: Don't clobber presubtract sources during optimizations 2571- r300/compiler: Don't track readers into an IF block. 2572- r300/compiler: Make sure presubtract sources use supported swizzles 2573- r300/compiler: Fix register allocator's handling of loops 2574- r300/compiler: Fix instruction scheduling within IF blocks 2575- r300/compiler: Use zero as the register index for unused sources 2576- r300/compiler: Ignore alpha dest register when replicating the result 2577- r300/compiler: Add rc_get_readers() 2578- r300/compiler: Handle BREAK and CONTINUE in rc_get_readers() 2579- r300/compiler: Track readers through branches in rc_get_readers() 2580- r300/compiler: Convert RGB to alpha in the scheduler 2581- r300/compiler: Use presubtract operations as much as possible 2582- r300/compiler: Enable rename_reg pass for r500 cards 2583- r300/compiler: Add a more efficient version of 2584 rc_find_free_temporary() 2585- r300/compiler: Don't allow presubtract sources to be remapped twice 2586- r300/compiler: Fix black terrain in Civ4 2587 2588Victor Tseng (1): 2589 2590- egl/i965: include inline_wrapper_sw_helper.h 2591 2592Viktor Novotný (6): 2593 2594- dri/nouveau: Import headers from rules-ng-ng 2595- dri/nouveau: nv04: Use rules-ng-ng headers 2596- dri/nouveau: nv10: Use rules-ng-ng headers 2597- dri/nouveau nv20: Use rules-ng-ng headers 2598- dri/nouveau: Remove nouveau_class.h, finishing switch to rules-ng-ng 2599 headers 2600- dri/nouveau: Clean up magic numbers in get_rt_format 2601 2602Vinson Lee (214): 2603 2604- llvmpipe: Remove unnecessary header. 2605- r600g: Remove unnecessary headers. 2606- mesa: Include missing header in program.h. 2607- glsl: Fix 'format not a string literal and no format arguments' 2608 warning. 2609- r600g: Silence uninitialized variable warning. 2610- r600g: Silence uninitialized variable warning. 2611- nvfx: Silence uninitialized variable warnings. 2612- r600g: Silence uninitialized variable warning. 2613- r600g: Silence uninitialized variable warning. 2614- r600g: Silence uninitialized variable warning. 2615- r600g: Silence unused variable warning. 2616- nv50: Update files in SConscript to match Makefile. 2617- nv50: Remove unnecessary headers. 2618- nv50: Silence uninitialized variable warning. 2619- nv50: Silence uninitialized variable warning. 2620- nv50: Silence uninitialized variable warning. 2621- gallivm: Remove unnecessary headers. 2622- draw: Remove unnecessary header. 2623- nv50: Silence uninitialized variable warnings. 2624- nv50: Fix 'control reaches end of non-void function' warning. 2625- mesa/st: Silence uninitialized variable warning. 2626- gallivm: Remove unnecessary header. 2627- r600g: Remove unnecessary header. 2628- r600g: Remove unnecessary headers. 2629- r600g: Fix implicit declaration warning. 2630- r600g: Fix memory leak on error path. 2631- r600g: Silence uninitialized variable warning. 2632- r600g: Silence unused variable warnings. 2633- mesa: bump version to 7.10 2634- ir_to_mesa: Remove unused member array_indexed from struct 2635 statevar_element. 2636- mesa: Silence "'valid_texture_object' defined but not used" warning. 2637- x86: Silence unused variable warning on Mac OS X. 2638- glsl: Fix 'control reaches end of non-void function' warning. 2639- nvfx: Remove const qualifer from nvfx_vertprog_translate. 2640- nvfx: Silence uninitialized variable warnings. 2641- r600g: Remove unused variable. 2642- nv50: Silence missing initializer warning. 2643- nv50: Remove dead initialization. 2644- nv50: Remove dead initialization. 2645- tgsi: Remove duplicate case value. 2646- glut: Define markWindowHidden for non-Windows only. 2647- glut: Define eventParser for non-Windows only. 2648- r300g: Silence uninitialized variable warning. 2649- intel: Fix implicit declaration of function '_mesa_meta_Bitmap' 2650 warning. 2651- mesa: Remove unnecessary headers. 2652- r600g: Remove unnecessary header. 2653- unichrome: Remove unnecessary header. 2654- intel: Remove unnecessary headers. 2655- r600g: Remove unused variable. 2656- r600g: Disable unused variables. 2657- r600g: Remove unused variable. 2658- r600g: Silence 'control reaches end of non-void function' warning. 2659- r600g: Remove unused variable. 2660- r600g: Remove unused variable. 2661- r600g: Disable unused variables. 2662- intel: Remove unnecessary header. 2663- st/dri: Remove unnecessary header. 2664- r600g: Remove unused variable. 2665- r300g: Remove unused variable. 2666- r600g: Don't return a value in function returning void. 2667- r600g: Remove unused variables. 2668- r600g: Include p_compiler.h instead of malloc.h. 2669- r600g: Silence uninitialized variable warnings. 2670- scons: Add MinGW-w64 prefixes for MinGW build. 2671- dri: Add GET_PROGRAM_NAME definition for Mac OS X. 2672- scons: Add program/sampler.cpp to SCons build. 2673- mesa: Fix printf format warning. 2674- mesa: Fix printf format warning. 2675- mesa: Fix printf format warning. 2676- r300/compiler: Move declaration before code. 2677- r300/compiler: Move declaration before code. 2678- r300/compiler: Move declaration before code. 2679- r300/compiler: Move declaration before code. 2680- r600g: Update SConscript. 2681- r300/compiler: Move declaration before code. 2682- r600g: Update SConscript. 2683- r300/compiler: Move declaration before code. 2684- r600g: Update SConscript. 2685- r300/compiler: Move declaration before code. 2686- r600g: Fix SCons build. 2687- r300/compiler: Move declaration before code. 2688- r300/compiler: Move declaration before code. 2689- r300/compiler: Move declaration before code. 2690- r300/compiler: Remove declaration before code. 2691- r300/compiler: Move declaration before code. 2692- r300/compiler: Move declaration before code. 2693- r300/compiler: Move declaration before code. 2694- r300/compiler: Move declaration before code. 2695- glsl: Remove unnecessary header. 2696- savage: Remove unnecessary header. 2697- r600g: Remove unused variable. 2698- r600g: Remove unnecessary headers. 2699- r600g: Fix SCons build. 2700- r600g: Remove unnecessary header. 2701- gallivm: Remove unnecessary header. 2702- r600g: Silence uninitialized variable warning. 2703- r600g: Silence uninitialized variable warning. 2704- r600g: Silence uninitialized variable warning. 2705- i915: Silence unused variable warning in non-debug builds. 2706- i915: Silence unused variable warning in non-debug builds. 2707- i965: Silence unused variable warning on non-debug builds. 2708- i965: Silence unused variable warning on non-debug builds. 2709- i965: Initialize member variables. 2710- r300: Silence uninitialized variable warning. 2711- tdfx: Silence unused variable warning on non-debug builds. 2712- gallivm: Remove unnecessary header. 2713- glsl: Initialize variable in 2714 ir_derefence_array::constant_expression_value 2715- mesa: Add missing header to shaderobj.h. 2716- llvmpipe: Return non-zero exit code for lp_test_round failures. 2717- r300/compiler: Remove unused variable. 2718- st/xorg: Fix memory leak on error path. 2719- llvmpipe: Initialize state variable in debug_bin function. 2720- llvmpipe: Initialize variable. 2721- draw: Move loop variable declaration outside for loop. 2722- r600g: Ensure r600_src is initialized in tgsi_exp function. 2723- glsl: Add assert for unhandled ir_shader case. 2724- swrast: Print out format on unexpected failure in 2725 \_swrast_DrawPixels. 2726- llvmpipe: Remove unnecessary header. 2727- draw: Remove unnecessary header. 2728- gallivm: Silence uninitialized variable warnings. 2729- gallivm: Silence uninitialized variable warnings. 2730- gallivm: Silence uninitialized variable warning. 2731- r300g: Silence uninitialized variable warning. 2732- mesa: Remove unnecessary headers. 2733- r600g: Silence uninitialized variable warnings. 2734- st/mesa: Remove unnecessary header. 2735- mesa: Remove unnecessary header. 2736- egl: Remove unnecessary headers. 2737- swrast: Print out format on unexpected failure in 2738 \_swrast_ReadPixels. 2739- st/mesa: Silence uninitialized variable warning. 2740- savage: Remove unnecessary header. 2741- st/vega: Remove unnecessary headers. 2742- dri/nouveau: Silence uninitialized variable warning. 2743- r300/compiler: Move declaration before code. 2744- r300/compiler: Move declaration before code. 2745- i965: Silence uninitialized variable warning. 2746- i965: Silence uninitialized variable warning. 2747- mesa: Clean up header file inclusion in accum.h. 2748- mesa: Clean up header file inclusion in version.h. 2749- mesa: Clean up header file inclusion in api_loopback.h. 2750- mesa: Clean up header file inclusion in api_validate.h. 2751- mesa: Include mfeatures.h in api_loopback for FEATURE_beginend. 2752- mesa: Include mfeatures.h in api_validate.c for FEATURE\_\* symbols. 2753- mesa: Clean up header file inclusion in arrayobj.h. 2754- mesa: Clean up header file inclusion in atifragshader.h. 2755- mesa: Clean up header file inclusion in attrib.h. 2756- mesa: Clean up header file inclusion in blend.h. 2757- mesa: Clean up header file inclusion in buffers.h. 2758- mesa: Clean up header file inclusion in colortab.h. 2759- mesa: Clean up header file inclusion in convolve.h. 2760- mesa: Clean up header file inclusion in debug.h. 2761- mesa: Clean up header file inclusion in depth.h. 2762- mesa: Clean up header file inclusion in depthstencil.h. 2763- mesa: Clean up header file inclusion in drawpix.h. 2764- mesa: Clean up header file inclusion in drawtex.h. 2765- mesa: Clean up header file inclusion in enable.h. 2766- mesa: Clean up header file inclusion in extensions.h. 2767- graw: Add struct pipe_surface forward declaration. 2768- mesa: Clean up header file inclusion in fbobject.h. 2769- mesa: Clean up header file inclusion in ffvertex_prog.h. 2770- mesa: Clean up header file inclusion in fog.h. 2771- mesa: Clean up header file inclusion in framebuffer.h. 2772- mesa: Clean up header file inclusion in hint.h. 2773- mesa: Clean up header file inclusion in histogram.h. 2774- mesa: Clean up header file inclusion in image.h. 2775- mesa: Add missing header and forward declarations in dd.h. 2776- mesa: Clean up header file inclusion in light.h. 2777- mesa: Clean up header file inclusion in lines.h. 2778- mesa: Clean up header file inclusion in matrix.h. 2779- mesa: Clean up header file inclusion in multisample.h. 2780- mesa: Clean up header file inclusion in nvprogram.h. 2781- winsys/xlib: Add cygwin to SConscript. 2782- mesa: Clean up header file inclusion in pixel.h. 2783- mesa: Clean up header file inclusion in pixelstore.h. 2784- mesa: Fix printf format warnings. 2785- mesa: Clean up header file inclusion in points.h. 2786- i965: Silence uninitialized variable warning. 2787- glsl: Add ir_constant_expression.cpp to SConscript. 2788- mesa: Add definitions for inverse hyperbolic function on MSVC. 2789- glsl: Fix 'control reaches end of non-void function' warning. 2790- glsl: Add lower_vector.cpp to SConscript. 2791- glsl: Fix type of label 'default' in switch statement. 2792- st/mesa: Remove unnecessary headers. 2793- swrast: Remove unnecessary header. 2794- r600: Remove unnecesary header. 2795- intel: Remove unnecessary header. 2796- mesa: Clean up header file inclusion in polygon.h. 2797- mesa: Clean up header file inclusion in rastpos.h. 2798- mesa: Clean up header file inclusion in readpix.h. 2799- mesa: Clean up header file inclusion in renderbuffer.h. 2800- mesa: Clean up header file inclusion in scissor.h. 2801- mesa: Clean up header file inclusion in shaderapi.h. 2802- mesa: Clean up header file inclusion in shared.h. 2803- mesa: Clean up header file inclusion in stencil.h. 2804- r600: Remove unnecessary header. 2805- llvmpipe: Remove unnecessary headers. 2806- mesa: Clean up header file inclusion in syncobj.h. 2807- r300/compiler: Move declaration before code. 2808- r300/compiler: Move declaration before code. 2809- mesa: Clean up header file inclusion in texcompress.h. 2810- st/vega: Silence uninitialized variable warning. 2811- mesa: Clean up header file inclusion in texcompress_s3tc.h. 2812- mesa: Clean up header file inclusion in texenvprogram.h. 2813- mesa: Clean up header file inclusion in texformat.h. 2814- mesa: Clean up header file inclusion in texgetimage.h. 2815- mesa: Clean up header file inclusion in texobj.h. 2816- gallium/noop: Add prototype for noop_init_state_functions. 2817- mesa: Clean up header file inclusion in texrender.h. 2818- mesa: Clean up header file inclusion in transformfeedback.h. 2819- mesa: Clean up header file inclusion in varray.h. 2820- mesa: Clean up header file inclusion in viewport.h. 2821- r200: Silence uninitialized variable warning. 2822- r600g: Fix SCons build. 2823- i965: Silence uninitialized variable warning. 2824 2825Xavier Chantry (8): 2826 2827- nv50: fix size of outputs_written array 2828- nv50: apply layout_mask to tile_flags 2829- nvfx: only expose one rt on nv30 2830- nvfx: fb->nr_cbufs <= 1 on nv30 2831- nvfx: reset nvfx->hw_zeta 2832- nvfx: fixes after array textures merge 2833- init ps->context with util_surfaces_get and do_get 2834- gallium/trace: check bind_vertex_sampler_states and 2835 set_vertex_sampler_views 2836 2837Xiang, Haihao (10): 2838 2839- mesa: fix regression from b4bb6680200b5a898583392f4c831c02f41e63f7 2840- i965: add support for polygon mode on Sandybridge. 2841- i965: fix for flat shading on Sandybridge 2842- i965: set minimum/maximum Point Width on Sandybridge 2843- meta: allow nested meta operations 2844- i965: support for two-sided lighting on Sandybridge 2845- i965: fix register region description 2846- i965: use align1 access mode for instructions with execSize=1 in VS 2847- i965: don't spawn GS thread for LINELOOP on Sandybridge 2848- i965: use BLT to clear buffer if possible on Sandybridge 2849 2850Zack Rusin (8): 2851 2852- rbug: fix rbug when contexts are being destroyed 2853- llvmpipe: fix rasterization of vertical lines on pixel boundaries 2854- scons: build the xorg state trackers only when env includes drm 2855- gallivm: implement indirect addressing of the output registers 2856- gallivm: implement indirect addressing over inputs 2857- gallivm: fix storing of the addr register 2858- scons: add alias for identity 2859- gallium/util: add states relevant to geometry shaders 2860 2861Zhenyu Wang (40): 2862 2863- i965: disasm quarter and write enable instruction control on 2864 sandybridge 2865- i965: new state dump for sandybridge 2866- i965: enable accumulator update in PS kernel too on sandybridge 2867- i965: Fix color interpolation on sandybridge 2868- i965: force zero in clipper to ignore RTAIndex on sandybridge 2869- i965: fix point size setting in header on sandybridge 2870- i965: ff sync message change for sandybridge 2871- i965: ignore quads for GS kernel on sandybridge 2872- i965: add sandybridge viewport state bo into validation list 2873- i965: VS use SPF mode on sandybridge for now 2874- i965: fix jump count on sandybridge 2875- i965: Fix sampler on sandybridge 2876- i965: fix const register count for sandybridge 2877- i965: Add all device ids for sandybridge 2878- i965: sandybridge pipe control workaround before write cache flush 2879- i965: only allow SIMD8 kernel on sandybridge now 2880- i965: don't do calculation for delta_xy on sandybridge 2881- i965: fix pixel w interpolation on sandybridge 2882- i965: enable polygon offset on sandybridge 2883- i965: fix scissor state on sandybridge 2884- i965: fix point sprite on sandybridge 2885- i965: fix occlusion query on sandybridge 2886- i965: fallback bitmap operation on sandybridge 2887- i965: Always set tiling for depth buffer on sandybridge 2888- i965: fallback lineloop on sandybridge for now 2889- Revert "i965: Always set tiling for depth buffer on sandybridge" 2890- i965: always set tiling for fbo depth buffer on sandybridge 2891- i965: Fix GS hang on Sandybridge 2892- Revert "i965: fallback lineloop on sandybridge for now" 2893- i965: refresh wm push constant also for BRW_NEW_FRAMENT_PROGRAM on 2894 gen6 2895- i965: fix dest type of 'endif' on sandybridge 2896- Revert "i965: VS use SPF mode on sandybridge for now" 2897- i965: also using align1 mode for math2 on sandybridge 2898- i965: Fix GS state uploading on Sandybridge 2899- i965: upload WM state for \_NEW_POLYGON on sandybridge 2900- i965: Use MI_FLUSH_DW for blt ring flush on sandybridge 2901- i965: explicit tell header present for fb write on sandybridge 2902- i965: Fix occlusion query on sandybridge 2903- i965: Use last vertex convention for quad provoking vertex on 2904 sandybridge 2905- i965: Fix provoking vertex select in clip state for sandybridge 2906 2907Zou Nan hai (1): 2908 2909- i965: skip too small size mipmap 2910 2911delphi (2): 2912 2913- draw: added userclip planes and updated variant_key 2914- draw: some changes to allow for runtime changes to userclip planes 2915 2916nobled (3): 2917 2918- r300g: Abort if atom allocations fail 2919- r300g: Abort if draw_create() fails 2920- r300g: Drop unnecessary cast 2921 2922pontus lidman (1): 2923 2924- mesa: check for posix_memalign() errors 2925 2926richard (2): 2927 2928- evergreen : fix z format setting, enable stencil. 2929- r600c : inline vertex format is not updated in an app, switch to use 2930 vfetch constants. For the 7.9 and 7.10 branches as well. 2931