summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2006-08-24fix fragment.position initialization bugBrian Paul
2006-08-24Make sure that _GNU_SOURCE is enabled in the linux configs that setIan Romanick
HAVE_POSIX_MEMALIGN. This eliminates a compiler warning. Also, clean up some crufty linux-sparc* configs.
2006-08-24add point-position.cKeith Whitwell
2006-08-24Test for fragment.position in points primitives. Currently rendersKeith Whitwell
incorrectly on software mesa.
2006-08-24added glut/mini/ sources to glut tarballBrian Paul
2006-08-24Check if mesa_vp->Base.Parameters is null before dereferencing (bug reportBrian Paul
from Chris Rankin). Also, fix some bad casts.
2006-08-24Fix some problems with the generation of the size tables. EnableIan Romanick
generation of protocol tables for Render.
2006-08-23silence warningBrian Paul
2006-08-23Added _mesa_lookup_program() and _mesa_lookup_bufferobj() functions to avoidBrian Paul
a lot of casting elsewhere. Use _mesa_lookup_texture() in tdfx driver, use _mesa_lookup_bufferobj() in r300 driver.
2006-08-23New script and API description file to enable generate of GLX protocolIan Romanick
decode tables in the server.
2006-08-23update glut fbdev sources listBrian Paul
2006-08-23VT switching now uses correct keys.Sean D'Epagnier
exiting when virtual screen size is larger than regular size doesn't corrupt screen. colormap is reset when using stdin input on exit. return is not reported as ctrl-m but '\r' as it should be backspace keycode was incorrect and not working, fixed
2006-08-22remove duplicate pieces of code.Aapo Tahkola
2006-08-22extend some DrawRangeElements cases to support rendering more than 65535 ↵Aapo Tahkola
verts in one blast. ut2k4 cases in other words ...
2006-08-22fix broken write-only depth setting. use early z where possible.Aapo Tahkola
2006-08-22Correct the name of the GL_NV_texture_expand_normal extension.Ian Romanick
2006-08-22The TLS dispatch functions for x86-64 are fixed-size, just like onIan Romanick
x86. This means that the position of each dispatch function can be calculated as (dispatch_offset * size_of_function) + glNewList. This allows us to not store the function pointers in glprocs_table_t. This same optimization has been done for quite some time on x86. We could probably also do this for some cases of the SPARC dispatch as well.
2006-08-22Add new attribute called static_dispatch to the <function> element. ThisIan Romanick
boolean attribute, which defaults to true, determines whether or not a static dispatch function is available in libGL for applications to link against. Ideally, any new functions that are not part of the ABI should not have directly accessable dispatch functions. This forces applications to use glXGetProcAddress to access these functions. By doing this we can gracefully remove functions from libGL without breaking the linkage of applications. Note that the static dispatch functions are still generated. However, they are given names like gl_dispatch_stub_820 and are marked with the "hidden" linker attribute. All extension functions added since the previous Mesa release (6.5) have been marked as 'static_dispatch="false"'.
2006-08-21move ALL_SOURCES to sources fileBrian Paul
2006-08-21remove FBDEV_DRIVER_OBJECTS assignment (in sources file)Brian Paul
2006-08-21various updatesBrian Paul
2006-08-21More GLSL preprocessor code:Michal Krol
Rename slang_version.syn to slang_pp_version.syn. The #version directive understands version 120. Cosmetic changes in version preprocessor. Checks if requested version is supported by the compiler.
2006-08-21GLSL 1.20 constructors and operators.Michal Krol
2006-08-19remove the now confusing option to manually enable software ARB_vp if drm is ↵Roland Scheidegger
not new enough on r200
2006-08-19Removed ifeq from makefile, and corrected fbdev driver so it compilesSean D'Epagnier
2006-08-18updates for 6.5.1Brian Paul
2006-08-18Fix-ups for linux-fbdev config. No longer build Xlib driver sources.Brian Paul
2006-08-18The driver now compiles correctly without any x headers or libraries installedSean D'Epagnier
The bitmap and stroke code can't be shared with glx anymore because of this. The model for the mini teapot is restored and I have tested it to work with linux-fbdev and linux-solo The driver recognizes 32bpp where there is no alpha (my radeon 7500) It also sets the correct number of cmap entrees (instead of 256 which can be an error)
2006-08-18Fix writemasks on texture arb fp instructions.Alan Hourihane
Cleanup invarient state emission.
2006-08-17sort the FEATURE_ definesBrian Paul
2006-08-17added FEATURE_EXT_timer_query and testsBrian Paul
2006-08-17change #ifdef to #ifBrian Paul
2006-08-16change tex memory allocation calls (debian bug 383209)Brian Paul
2006-08-16When available, which is most of the time, use posix_memalign to implementIan Romanick
_mesa_align_malloc and friends.
2006-08-15Enable support for GL_EXT_gpu_program_parameters for R300.Ian Romanick
2006-08-15Add support for GL_EXT_gpu_program_parameters. Any driver that enablesIan Romanick
either GL_ARB_vertex_program or GL_ARB_fragment_program should enable this extension as well.
2006-08-15Add test for GL_EXT_gpu_program_parameters.Ian Romanick
2006-08-15Remove extraneous trailing semicolon.Ian Romanick
2006-08-15s/GLvoid/void/ to fix gcc 4.2 issue (bug 7875)Brian Paul
2006-08-14Bug #7169: Attempt to make libGL symbols visible to drivers.Michel Dänzer
Some applications end up dlopening libGL without RTLD_GLOBAL, so the libGL symbols referenced by the driver can't be unresolved when libGL dlopens it. This attempts to make the libGL symbols visible to the driver by dlopening libGL (again) with RTLD_GLOBAL before dlopening the driver and dlclosing the obtained handle afterwards.
2006-08-12gcc-4.1.1 compile fixAapo Tahkola
2006-08-12Add comments explaining a couple "mystery" bits.Ian Romanick
2006-08-12Fix white-space problem in the prototype for glBlitFramebufferEXT thatIan Romanick
prevented gl_mangle.h from picking it up.
2006-08-11Fix long standing bug that prevented newer drivers from working withIan Romanick
older libGL. The data in extension_helper.h erroneous instructed drivers to expect all functions with establised offsets, even those not in the ABI, to have a specific offset allocated. This is just wrong. A function is either in the ABI and has a set offset or a function is not in the ABI and has a re-map offset.
2006-08-11Convert comment header for __GLcontextModesRec to Doxygen style.Ian Romanick
2006-08-11support shorts as vertex dataAapo Tahkola
2006-08-10Add padding to match the current xf86-video-intel DDX->DRI interface.Eric Anholt
2006-08-10mkdir -> mkdir -p, since /usr/include might not exist in the installroot yetAdam Jackson
2006-08-10linux-indirect target fixes.Adam Jackson
2006-08-106.5.1 relnotesBrian Paul
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-13 10:37:49 +0000