summaryrefslogtreecommitdiffstats
path: root/src/gallium/auxiliary/rtasm
Commit message (Collapse)AuthorAgeFilesLines
* gallium/rtasm: remove duplicated includeNicolas Kaiser2010-09-301-1/+0
| | | | | | Remove duplicated include. Signed-off-by: Brian Paul <[email protected]>
* rtasm: Include missing header in rtasm_x86sse.h.Vinson Lee2010-08-261-0/+1
| | | | Include p_compiler.h for stdint.h uint*_t symbols.
* translate_sse: try to fix Win64Luca Barbieri2010-08-191-1/+2
| | | | Not sure whether it works now (it is still disabled).
* rtasm: add minimal x86-64 support and new instructions (v5)Luca Barbieri2010-08-163-40/+551
| | | | | | | | | | | | | | | | | | | | | | | | Changes in v5: - Add sse2_movdqa Changes in v4: - Use _WIN64 instead of WIN64 Changes in v3: - Add target and target caps functions, so that they could be different in principle from the current CPU and they don't need #ifs to check Changes in v2: - Win64 support (untested) - Use u_cpu_detect.h constants instead of #ifs This commit adds minimal x86-64 support: only movs between registers are supported for r8-r15, and x64_rexw() must be used to ask for 64-bit operations. It also adds several new instructions for the new translate_sse code. movdqa
* translate: add support for 8/16-bit indicesLuca Barbieri2010-08-162-0/+16
| | | | | Currently, only 32-bit indices are supported, but some use cases translate needs support for all types.
* rtasm: fix void * / func * casting warningsBrian Paul2010-05-132-4/+24
|
* rtasm: Silence gnu_printf format warnings.Vinson Lee2010-03-261-1/+1
|
* rtasm: Fix typos in comments.Vinson Lee2010-03-251-3/+3
|
* gallium: Move p_thread.h and p_atomic.h out of gallium interfaces.José Fonseca2010-02-021-2/+2
| | | | Into os/os_thread.h and util/u_atomic.h respectively.
* Merge branch 'master' into instanced-arraysMichal Krol2010-01-053-28/+30
|\ | | | | | | | | | | Conflicts: src/gallium/auxiliary/tgsi/tgsi_dump.c src/gallium/include/pipe/p_shader_tokens.h
| * Merge branch 'mesa_7_7_branch'Brian Paul2010-01-041-2/+30
| |\ | | | | | | | | | | | | | | | | | | | | | | | | Conflicts: docs/relnotes.html src/gallium/drivers/llvmpipe/lp_tex_sample_c.c src/gallium/drivers/r300/r300_cs.h src/mesa/drivers/dri/i965/brw_wm_surface_state.c src/mesa/main/enums.c
| | * rtasm: Play nicely with Windows's Data Execution Prevention.José Fonseca2010-01-031-2/+30
| | |
| * | gallium: Generate a single library for auxiliaries with Make too.José Fonseca2010-01-011-13/+0
| | |
| * | scons: Aggregate all tiny libraries in a single library.José Fonseca2010-01-011-13/+0
| |/ | | | | | | | | Makes integration of gallium into out of tree components much easier. No pratical change for components in this tree,
* / rtasm: Add x86_div().Michal Krol2009-12-302-0/+8
|/
* rtasm: export sse_movmskpsKeith Whitwell2009-07-162-0/+10
|
* rtasm: Use 32bit constant.José Fonseca2009-06-151-3/+3
| | | | As we're only using 32bit bitmasks.
* rtasm: Silence `static function not used' warning.Michal Krol2009-04-101-0/+2
|
* Add #ifdefs needed to compile Gallium on Solaris with gcc or Sun ccAlan Coopersmith2009-03-251-3/+3
| | | | Signed-off-by: Alan Coopersmith <[email protected]>
* gallium: Add support for BSD operating systems, tested with FreeBSDBenjamin Close2009-02-251-3/+7
| | | | | | | | | | BSD supports pipe in the same way as linux hence options which are safe for linux are also safe for BSD. Define PIPE_OS_BSD in include/pipe/p_config.h and adjust the defines to make use of it. Also define MAP_ANONYMOUS for BSD systems which use MAP_ANON Signed-off-by: Benjamin Close <[email protected]>
* gallium: Improve makefiles for librariesJakob Bornecrantz2009-02-201-3/+0
| | | | | | | | The template makefile that most libraries in gallium included was based on dri and had a bunch unrelevant junk in it. Update it and improve the depending makefiles.
* util: Move p_debug.h into util module.José Fonseca2009-02-184-4/+4
| | | | | The debug functions depend on several util function for os abstractions, and these depend on debug functions, so a seperate module is not possible.
* cell: datatype clean-ups in SPE rtasmBrian Paul2009-01-112-105/+99
|
* gallium: added comment/annotation support to PPC rtasmBrian Paul2009-01-102-62/+187
|
* gallium: s/false/FALSE/Brian Paul2009-01-101-1/+1
|
* rtasm: Remove spurious semi-colons after function bodies.José Fonseca2008-12-301-5/+5
|
* CELL: use variant-length fragment ops programsRobert Ellison2008-11-212-4/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a set of changes that optimizes the memory use of fragment operation programs (by using and transmitting only as much memory as is needed for the fragment ops programs, instead of maximal sizes), as well as eliminate the dependency on hard-coded maximal program sizes. State that is not dependent on fragment facing (i.e. that isn't using two-sided stenciling) will only save and transmit a single fragment operation program, instead of two identical programs. - Added the ability to emit a LNOP (No Operation (Load)) instruction. This is used to pad the generated fragment operations programs to a multiple of 8 bytes, which is necessary for proper operation of the dual instruction pipeline, and also required for proper SPU-side decoding. - Added the ability to allocate and manage a variant-length struct cell_command_fragment_ops. This structure now puts the generated function field at the end, where it can be as large as necessary. - On the PPU side, we now combine the generated front-facing and back-facing code into a single variant-length buffer (and only use one if the two sets of code are identical) for transmission to the SPU. - On the SPU side, we pull the correct sizes out of the buffer, allocate a new code buffer if the one we have isn't large enough, and save the code to that buffer. The buffer is deallocated when the SPU exits. - Commented out the emit_fetch() static function, which was not being used.
* CELL: fix stencil twiddling, stencil invertRobert Ellison2008-11-131-3/+3
| | | | | | | | | | | | | Many stencil tests were failing because of a failure to read the stencil buffer, due to "twiddling" (or "untwiddling") "an unsupported texture format". This is fixed for the case of a stencil/Z S824Z format (which twiddles just like the 32-bit color formats). tests/stencilwrap.c was failing on the GL_INVERT test, because the emitted code for "spe_xori" turned out not to be an actual "xori" instruction, but rather a "stqd" instruction, because of a typo in the rtasm code. This is now fixed, and tests/stencil_wrap now works.
* gallium: add missing prototypesBrian Paul2008-11-121-0/+6
|
* cell: move semicolons to silence warnings w/ other compilersBrian Paul2008-11-121-189/+189
|
* cell: fix typo in EMIT_ macroBrian Paul2008-11-121-1/+1
|
* rtasm: Use INLINE keyword. Compile for all platforms, not only GALLIUM_CELL.Michal Krol2008-11-121-9/+5
|
* rtasm: Compile only for GALLIUM_CELL.Michal Krol2008-11-121-0/+4
|
* CELL: two-sided stencil fixesRobert Ellison2008-11-111-2/+5
| | | | | | | | | | | | | | | | | | | With these changes, the tests/stencil_twoside test now works. - Eliminate blending from the stencil_twoside test, as it produces an unneeded dependency on having blending working - The spe_splat() function will now work if the register being splatted and the destination register are the same - Separate fragment code generated for front-facing and back-facing fragments. Often these are the same; if two-sided stenciling is on, they can be different. This is easier and faster than generating code that does both tests and merges the results. - Fixed a cut/paste bug where if the back Z-pass stencil operation were different from all the other operations, the back Z-fail results were incorrect.
* gallium: grow SPE instruction buffer as neededBrian Paul2008-10-291-16/+41
|
* gallium: no longer pass max_inst to ppc_init_func()Brian Paul2008-10-292-2/+2
|
* gallium: use execmem for PPC code, grow instruction buffer as neededBrian Paul2008-10-292-21/+50
|
* gallium: fix alignment parameter passed to u_mmAllocMem()Brian Paul2008-10-291-2/+2
| | | | | | Was 32, now 5. The param is expressed as a power of two exponent. The net effect is that the alignment was a no-op on X86 but on PPC we always got the same memory address everytime rtasm_exec_malloc() was called.
* gallium: prefix memory manager functions with u_ to differentiate from ↵Brian Paul2008-10-291-4/+4
| | | | functions in mesa/main/mm.c
* gallium: test for PIPE_OS_LINUX instead of __linux__Brian Paul2008-10-291-4/+5
|
* gallium: added ppc_vnmsubfp()Brian Paul2008-10-292-1/+12
|
* scons: ppc support.Michel Dänzer2008-10-231-0/+1
|
* gallium: remove ppc_vload_float(), rename ppc_vecmove() -> ppc_vmove().Brian Paul2008-10-222-23/+2
|
* gallium: added ppc_vzero()Brian Paul2008-10-222-0/+13
|
* gallium: added ppc_vload_float(), for limited casesBrian Paul2008-10-222-0/+22
|
* gallium: fix-up confusing register allocation masks in rtasm_ppc.cBrian Paul2008-10-222-21/+36
| | | | Plus, add ppc_reserve_register() func.
* gallium: added ppc_lvewx()Brian Paul2008-10-222-0/+11
|
* cell: implement many more PPC instructions for code genBrian Paul2008-10-223-41/+704
|
* cell: add emit_RI10s() which does range checking on the 10-bit signed ↵Brian Paul2008-10-102-10/+30
| | | | | | immediate field This type of checking should be expanded to cover more instructions...
* cell: additional 'offset' checking in spe_lqd(), spe_stqd()Brian Paul2008-10-101-4/+14
|