summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* gallium: Add SCons as alternative build system for Gallium.José Fonseca2008-01-319-0/+875
|
* I don't know why using uint64_t for "base" doesn't work. Ugh.Ian Romanick2008-01-302-4/+5
|
* Use SPUs for vertex shader processingIan Romanick2008-01-304-1/+132
|
* Add driver_private field for drivers that hook shader_queue_flush.Ian Romanick2008-01-301-0/+2
|
* Numerous small fixed to PPU-SPU vertex shader protocolIan Romanick2008-01-302-12/+34
|
* cell_array_info should not be 16-byte alignedIan Romanick2008-01-301-1/+1
| | | | | Forcing cell_array_info to be 16-byte aligned makes it more difficult to stuff that state in batch commands.
* Correctly read / write vertex header from / to main memoryIan Romanick2008-01-301-8/+9
|
* Handle CELL_CMD_VS_EXECUTE *only* outside batch commands.Ian Romanick2008-01-302-6/+5
|
* Set machine->ProcessorIan Romanick2008-01-301-1/+2
| | | | | The default value is 0, which is TGSI_PROCESSOR_FRAGMENT...not correct for a vertex shader!
* Elts are always ints, pass vOut pointers in-line in commandIan Romanick2008-01-302-15/+5
|
* Implement micro_pow and micro_sqrtIan Romanick2008-01-301-12/+31
| | | | Unimplemented micro ops get assertions for now.
* Fix size calculation in attribute fetch.Ian Romanick2008-01-301-4/+4
|
* Fetch uniforms from main memory.Ian Romanick2008-01-301-5/+16
|
* Missing amperstand in previous commit. Oops.Ian Romanick2008-01-301-1/+1
|
* Pass ptr to local memory copy instead of main memory to exec_instructionIan Romanick2008-01-301-10/+12
| | | | | | This was essentially a cut-and-paste bug when the instruction fetcher was added. Also, the test for TGSI_PROCESSOR_FRAGMENT was moved outside the loop for exec_declaration.
* Provide mechanism to hook in custom vertex shader cache flush functionIan Romanick2008-01-303-1/+8
|
* Implement vertex fetch / vertex shader output write-backIan Romanick2008-01-303-41/+58
|
* Initial pass at instruction / declaration fetchIan Romanick2008-01-302-0/+23
|
* Initial pass at vertex shader on SPU using TGSI VMIan Romanick2008-01-309-1/+3540
| | | | | | All of the code is wired in on the SPU side, but it is not called from the PPU yet. Instruction / declaration fetch still needs to be implemented in spu_exec.c.
* Cell: prototype SIMD code for z testingBrian2008-01-302-34/+123
|
* Cell: start to SIMD-ize triangle attribute interpolationBrian2008-01-304-62/+79
| | | | | Using the spu_add(), etc intrinsics. About a 15% speed-up with some tests.
* Cell: fold setup_tri() into tri_draw()Brian2008-01-301-43/+19
|
* Cell: make 'setup' a regular var instead of passing around a pointer everywhereBrian2008-01-301-210/+209
| | | | | We'll never have more than one of these objects. Avoiding pointer deref improves performance a bit.
* Cell: minor code refactoring, movementBrian2008-01-301-30/+55
|
* Cell: check tile status before wait_on_mask()Brian2008-01-301-2/+2
|
* Cell: make wait_on_mask() static/inlinedBrian2008-01-302-21/+21
|
* Cell: move CELL_MAX_SPUSBrian2008-01-302-3/+2
|
* gallium: Teach draw_vf about draw vertices.José Fonseca2008-01-303-174/+161
| | | | | This reduces the emit overhead, which is significant since we're emiting one vertex at a time.
* gallium: Remove draw_vertex_fetch::lookup.José Fonseca2008-01-302-7/+0
| | | | It is not being used, and would be dangerous to use given the possibility of constants.
* gallium: Fix build on WinXP.Michal Krol2008-01-306-19/+19
|
* Cell: use _pack_rgba8() from pack_rgba8.h to do float[4]->uint color conversionBrian2008-01-291-11/+4
| | | | texcyl.c is twice as fast now in non-texture mode
* gallium: weaken assert slightlyKeith Whitwell2008-01-291-1/+1
|
* gallium: streamline various unfilled & stippled pathsKeith Whitwell2008-01-291-45/+113
|
* gallium: don't rely on assert(0) for error handling - may be disabledKeith Whitwell2008-01-291-0/+2
|
* gallium: Emit point size as a constant.José Fonseca2008-01-291-4/+81
|
* gallium: Emit constants.José Fonseca2008-01-294-16/+62
|
* gallium: Allow draw_vf usage to be controlled at runtime.José Fonseca2008-01-291-78/+71
|
* gallium: Add extern keyword to global.José Fonseca2008-01-291-1/+2
|
* Cell: add OPT_FLAGS varBrian2008-01-281-2/+4
|
* Cell: move cmd_render() into new spu_render.c fileBrian2008-01-285-203/+283
|
* Cell: add a few null texture testsBrian2008-01-282-4/+12
|
* Cell: compute min index referenced in draw command, use it to reduce size of ↵Brian2008-01-283-6/+29
| | | | vertex data payload
* Cell: minor optimization for flat shadingBrian2008-01-281-9/+28
|
* Cell: basic texture mappingBrian2008-01-2810-28/+183
| | | | | | Texture images are tiled in PPU code. SPUs use a texture cache for getting texels from textures. This is very rough code, but demos/texcyl.c works.
* Cell: initial texture cache/sampling codeBrian2008-01-282-0/+182
|
* Cell: emit state in cell_clear_surface() if dirty.Brian2008-01-281-0/+4
| | | | | Without this a program that does nothing but glClear() doesn't work. We need the framebuffer state.
* Cell: re-enable bounding boxesBrian2008-01-283-12/+24
| | | | | | | | The geometry bounding box is used to restrict rasterization to just those tiles that are relevant. Note another dummy field had to be added to the cell_command_render struct. Apparently, every 4th word in a struct is susceptible to corruption in some circumstances. Might be a compiler bug.
* gallium: Use GALLIUM_ prefix for env vars.José Fonseca2008-01-291-1/+1
|
* gallium: Use CALLOC for pb_buffer to ensure that all fields of pipe_buffer ↵José Fonseca2008-01-293-3/+4
| | | | are initiallized.
* gallium: Remove direct dependencies to mesa internals.José Fonseca2008-01-294-64/+31
| | | _mesa_exec_free is still being called. More invasive refactoring is necessary to clean it out.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-14 16:23:18 +0000