summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* llvmpipe: also test the new lp_build_assert() functionBrian Paul2010-07-291-0/+4
|
* llvmpipe: don't call LLVMCreateJITCompiler() twiceBrian Paul2010-07-295-45/+14
| | | | | | | | | Fixes a failed assertion with LLVM 2.6: <unnamed>::JITResolver::JITResolver(llvm::JIT&): Assertion `TheJITResolver == 0&& "Multiple JIT resolvers?"' failed. Though, not everyone seems to experience this problem.
* gallivm: added lp_build_assert() function to make assertions in LLVM codeBrian Paul2010-07-294-0/+144
|
* Revert "gallivm: fix lp_build_sample_offset() crash when indexing a 1-D texture"José Fonseca2010-07-291-3/+0
| | | | | | This reverts commit 5f90e76c54bbf4456c977b3cbca450d7a570179e. Bad cherry-pick.
* gallivm: fix lp_build_sample_offset() crash when indexing a 1-D textureBrian Paul2010-07-291-0/+3
| | | | | If y==NULL and y_stride==NULL it means the texture is 1D. Return zero for out_i and the offset instead of garbage.
* util: add uint version of pack_z_stencilKeith Whitwell2010-07-291-0/+47
| | | | Useful for packing mask values.
* scons: Use the current python executable for code generation.José Fonseca2010-07-292-3/+4
| | | | Less susceptible to be broken.
* draw: Also emit EMMS on generated LLVM IR.José Fonseca2010-07-291-0/+11
|
* llvmpipe: Avoid corrupting the FPU stack with MMX instructions on 32bit OSes.José Fonseca2010-07-292-0/+29
| | | | | | | Unfortunately LLVM doesn't emit EMMS itself, and there is no easy/effective way to disable MMX. http://llvm.org/bugs/show_bug.cgi?id=3287
* util: Don't include xmmintrin.h.José Fonseca2010-07-291-1/+0
| | | | Unnecessary.
* mesa: implement RCC opcodeBrian Paul2010-07-291-0/+54
|
* mesa: update table of opcodes used by GLSLBrian Paul2010-07-291-8/+8
|
* r600: since 8744c36e added asserts - use another random register for shader ↵Andre Maasikas2010-07-291-1/+1
| | | | with no output
* st/python: Adapt to interface change.Chia-I Wu2010-07-291-21/+18
| | | | This is only compile tested.
* graw/tests: Use pipe_context::draw_vbo.Chia-I Wu2010-07-297-17/+28
| | | | | | | The other drawing variants such as draw_arrays or draw_elements_instanced were removed. This fixes fdo bug #29287.
* gallium: Avoid void pointer arithmetic.Chia-I Wu2010-07-294-7/+7
| | | | This fixes fdo bug #29286.
* mesa: Reduce arrayobj.h header file inclusion.Vinson Lee2010-07-281-1/+1
| | | | | Directly include mtypes.h instead of including context.h to include mtypes.h.
* mesa: Remove unnecessary header.Vinson Lee2010-07-281-1/+0
|
* gallium/docs: Document draw_vbo and set_index_buffer.Chia-I Wu2010-07-292-58/+27
| | | | | Document the new unified drawing method and remove references to old ones.
* gallium: Keep only pipe_context::draw_vbo.Chia-I Wu2010-07-2925-1188/+6
| | | | | | | That is, remove pipe_context::draw_arrays, pipe_context::draw_elements, pipe_context::draw_arrays_instanced, pipe_context::draw_elements_instanced, pipe_context::draw_range_elements.
* gallium: Use unified pipe_context::draw_vbo.Chia-I Wu2010-07-293-103/+71
| | | | Update u_draw_quad, st/vega, and st/mesa to use pipe_context::draw_vbo.
* gallium: Implement draw_vbo and set_index_buffer for all drivers.Chia-I Wu2010-07-2948-202/+1004
| | | | | | | | | | | | | | | Some drivers define a generic function that is called by all drawing functions. To implement draw_vbo for such drivers, either draw_vbo calls the generic function or the prototype of the generic function is changed to match draw_vbo. Other drivers have no such generic function. draw_vbo is implemented by calling either draw_arrays and draw_elements. For most drivers, set_index_buffer does not mark the state dirty for tracking. Instead, the index buffer state is emitted whenever draw_vbo is called, just like the case with draw_elements. It surely can be improved.
* gallium: Add pipe_context::draw_vbo and pipe_context::set_index_buffer.Chia-I Wu2010-07-292-0/+41
| | | | | | This commit adds a new unified draw_vbo method to pipe_context. Unlike other draw methods, draw_vbo treats the index buffer as a state which is set with set_index_buffer.
* mesa: Remove unnecessary header.Vinson Lee2010-07-281-1/+1
|
* egl: Fix deref before NULL-checkKristian Høgsberg2010-07-281-1/+2
|
* intel: Implement EGL_KHR_surfaceless extensionKristian Høgsberg2010-07-284-8/+28
|
* egl_dri2: Implement EGL_KHR_surfaceless_* extensionsKristian Høgsberg2010-07-281-25/+55
|
* egl: EGL_KHR_surfaceless_* extensionsKristian Høgsberg2010-07-284-36/+67
| | | | | | | | | These extensions allow an application to make a context current by passing EGL_NO_SURFACE for the write and read surface in the call to eglMakeCurrent. The motivation is that applications that only want to render to client API targets (such as OpenGL framebuffer objects) should not need to create a throw-away EGL surface just to get a current context.
* Untangle gallium/egl/glx source sharing mess and make it compile againKristian Høgsberg2010-07-283-114/+58
|
* r600g: state context ptr in sampler_view & add I8/L8 buffer formatJerome Glisse2010-07-282-0/+3
| | | | Signed-off-by: Jerome Glisse <[email protected]>
* r600g: switch btw flat/linear interpolationJerome Glisse2010-07-283-3/+7
| | | | | | | | | I am not sure how to properly handle flat shading regarding non color parameter to fragment shader. It seems we should still interpolate non color using linear interpolation and flat shade only apply to color. Signed-off-by: Jerome Glisse <[email protected]>
* r600g: split pipe state creating/binding from hw state creationJerome Glisse2010-07-289-612/+1077
| | | | | | | | | Split hw vs pipe states creation handling as hw states group doesn't match pipe state group exactly. Right now be dumb about that and rebuild all hw states on each draw call. More optimization on that side coming. Signed-off-by: Jerome Glisse <[email protected]>
* glx: Split indirect and applegl implementations into different filesKristian Høgsberg2010-07-287-547/+624
|
* glx: Move bind and unbind to context vtableKristian Høgsberg2010-07-2812-276/+206
|
* glx: Rename __GLXcontext and __GLXdisplayPrivate to struct types.Kristian Høgsberg2010-07-2830-768/+781
|
* glx: Rename __GLXscreenConfigs to struct glx_screenKristian Høgsberg2010-07-2810-101/+99
| | | | Because double underscores in private type names is painful.
* glx: Use _X_EXPORT instead of our own PUBLIC macroKristian Høgsberg2010-07-287-101/+74
| | | | We're an X client library, so we can use Xfuncproto.h.
* glx: Rename glcontextmodes.[ch] to glxconfig.[ch]Kristian Høgsberg2010-07-287-7/+4
|
* glx: Rename __GLcontextModes to struct glx_configKristian Høgsberg2010-07-2817-289/+258
| | | | | With this rename, we use 'config' consitently to refer to GLX configurations instead of the modes/configs/visual mess before.
* glx: Delete unused glcontextmodes.c functionsKristian Høgsberg2010-07-283-101/+0
|
* glx: Stop using glcore.h and glxint.h in glxKristian Høgsberg2010-07-283-93/+90
| | | | We'd like to get rid of these header files.
* Avoid a compiler warning about a potentially unused variable.Carl Worth2010-07-281-1/+1
| | | | | | | | | There is no assignment to the "ret" variable if X_DRI2SwapBuffers is not defined. In this case, the earlier explicit "return 0" is likely to be used, but the compiler can't be sure of that, (nor can I for that matter). We cover this case by explicitly initializing "ret" to 0.
* Avoid compiler warning for unused glx_dpy variable.Carl Worth2010-07-281-1/+1
| | | | | | | | | | My earlier attempt to eliminate this warning (c0ca2bfb2ad8c) was invalid as it removed the variable declaration. Jerome correctly reverted that (600c85efdb0ff) since the variable is used when X_DRI2SwapBuffers is defined. Here, instead of removing the declaration, we move it to inside the correct #ifdef.
* r600g: cleanup resource buffer/texture messJerome Glisse2010-07-2810-259/+217
| | | | | | Use a common function, fix the mess it was before. Signed-off-by: Jerome Glisse <[email protected]>
* r600g: add lrp instruction supportJerome Glisse2010-07-281-3/+80
| | | | Signed-off-by: Jerome Glisse <[email protected]>
* r600g: actualy fix the literal emissionJerome Glisse2010-07-281-3/+3
| | | | | | | Previous patch added literal emission to wrong place, we want to emit literal before emitting a new alu group. Signed-off-by: Jerome Glisse <[email protected]>
* glapi: Fix VMware spelling.Vinson Lee2010-07-281-1/+1
|
* gallium/docs: Fix VMware spelling.Vinson Lee2010-07-281-2/+2
|
* draw: Fix VMware spelling.Vinson Lee2010-07-283-3/+3
|
* r600g: fix up segfault with variation between views and count.Dave Airlie2010-07-281-4/+18
| | | | | For some reason gallium hands us something with lots of empty views, and we are expected to deal with it, just do what r300g does for this bit.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-26 23:06:25 +0000