summaryrefslogtreecommitdiff
path: root/src/gallium/auxiliary/tgsi
AgeCommit message (Collapse)Author
2008-10-28gallium: ppc: don't replicate/smear immediate values, use vspltw instruction ↵Brian Paul
as with constants
2008-10-28gallium: use some PPC vec registers to store TGSI tempsBrian Paul
This could be a lot better, but already makes for better code.
2008-10-27gallium: ppc: emit fewer 'li' instructions prior to vector loads/storesBrian Paul
2008-10-27gallium: ppc: use a src register cache to avoid redundant loadsBrian Paul
2008-10-23scons: ppc support.Michel Dänzer
2008-10-22gallium: PPC: clamp y to [-128,128] for LITBrian Paul
2008-10-22gallium: remove ppc_vload_float(), rename ppc_vecmove() -> ppc_vmove().Brian Paul
2008-10-22gallium: new PPC built-in constants arrayBrian Paul
It's hard to form PPC vector immediates so load them from an array.
2008-10-22gallium: GALLIUM_NOPPC debug var to disable PPC codegenBrian Paul
2008-10-22gallium: PPC LIT instruction (not quite complete yet)Brian Paul
2008-10-22gallium: var renaming in tgsi_ppc.cBrian Paul
2008-10-22gallium: remove SSE remnants from tgsi_ppc.cBrian Paul
2008-10-22cell: TGSI->PPC for RSQ, RCP and src register sign modesBrian Paul
2008-10-22gallium: TGSI->PPC inequality operatorsBrian Paul
2008-10-22gallium: fix broken TGSI_FILE_CONSTANT case, use ppc_reserver_register()Brian Paul
2008-10-22gallium: TGSI to PPC code generationBrian Paul
Based on the TGSIto SSE2 code generator. Incomplete and lots of SSE stuff still hanging around but the basic dozen or so TGSI opcodes are functioning.
2008-10-07gallium: Introduce PIPE_ARCH_SSE define for SSE support.José Fonseca
Besides meaning x86 and x86-64 architecture, it also depends on SSE2 support enabled on gcc. This fixes the linux-debug build.
2008-10-01tgsi: Include p_config.h.José Fonseca
2008-09-30cell: Moved X86 checks to wrap #include section so that Cell targets will ↵Jonathan White
compile again.
2008-09-30tgsi: SSE2 optimized exp2, log2 and pow implementations.José Fonseca
Special care must be taken when calling compiler generated SSE2 functions from the runtime generated SSE2: saving the xmm registers, and notify gcc the stack is not 16byte aligned. It would be more efficient to keep the stack pointer 16byte aligned, but too hairy, and not consistent in all x86 architectures. This has been tested in linux x86 and windows x86 userspace. Not tested on x86-64 because it is broken for other reasons (even without this change).
2008-09-18Merge commit 'origin/gallium-0.1' into gallium-0.2Keith Whitwell
Conflicts: src/mesa/shader/slang/slang_link.c
2008-09-18tgsi: Build tgsi_text with makeJakob Bornecrantz
2008-09-18tgsi: Make tgsi dumps look more like mesa shader dumps.Jakob Bornecrantz
2008-09-17gallium: fix tgsi sanity checker with respect to END.Brian Paul
Subroutine code may be found after the END instruction so it's not always the last instruction. At least check for presence of exactly one END instruction though.
2008-09-16gallium: fix info entries for KIL, KILPBrian Paul
KIL takes 1 src register. KILP uses no registers (uses cond codes).
2008-09-16tgsi: Make tgsi_sanity.c compile with makeJakob Bornecrantz
2008-09-12gallium: use copy_token() function to avoid type punning/aliasing problemsBrian Paul
This fixes parsing errors seen with optimized builds on PPC (which led to crashes). The memcpy() is heavy-handed, but works. A lighter uint assignment could be used on x86...
2008-09-12gallium: use new compare32() function to fix warnings about type punning and ↵Brian Paul
aliasing
2008-09-12gallium: silence warningBrian Paul
2008-09-10tgsi: Fix newline pos.José Fonseca
2008-09-10tgsi: Verify constants are set before attempting to read them.José Fonseca
2008-09-08tgsi: Cleanup code.Michal Krol
2008-09-05tgsi: Refactor tgsi_dump to avoid using string buffers when dumping.José Fonseca
This fixes a stack overflow when dumping shaders. It ended up being pretty much as the original code Michal had before, before I went on a cleanup rampage on it and took things that ended up needing...
2008-09-02gallium: increase string buffer size to 16000 to avoid truncated output of ↵Brian Paul
long shaders
2008-08-24gallium: refactor/replace p_util.h with util/u_memory.h and util/u_math.hBrian Paul
Also, rename p_tile.[ch] to u_tile.[ch]
2008-08-22gallium: replace LOG2() macro with util_fast_log2() inline funcBrian Paul
2008-08-22gallium: use new util_fast_exp2(), _log2(), pow() functionsBrian Paul
New code surrounded with #if FAST_MATH to allow comparing against original code if we need to debug.
2008-08-19tgsi: Implement LOG opcode for SSE2 codegen.Michal Krol
2008-08-19tgsi: Implement EXP opcode for SSE2.Michal Krol
2008-08-19tgsi: Fix ARL opcode in SSE2 codegen.Michal Krol
2008-08-19tgsi: Workaround debug output buffer size limitations.Michal Krol
2008-08-19tgsi: Use NUM_CHANNELS.Michal Krol
2008-08-18tgsi: Update condition code vector when storing dest register.Michal Krol
2008-08-18tgsi: Saturate modifier obeys ExecMask.Michal Krol
Implement NVIDIA [-1;+1] saturate mode.
2008-08-18tgsi: Respect condition codes when storing destination register.Michal Krol
2008-08-18tgsi: Cosmetic changes.Michal Krol
2008-08-18tgsi: Add condition code (CC) register.Michal Krol
2008-08-13gallium: fix exec_kilp(), fix Exec/FuncMask test for TGSI_OPCODE_RETBrian Paul
Fixes a few glean glsl regressions.
2008-08-13tgsi: Initial code for KILP, needs CCs working to be complete.Michal Krol
2008-08-13tgsi: Use a homebrew version of toupper().Michal Krol
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-15 10:34:54 +0000