summaryrefslogtreecommitdiff
path: root/src/gallium/auxiliary/gallivm
AgeCommit message (Collapse)Author
2010-06-02gallivm: Avoid fp arithmetic in lp_build_sample_wrap_nearest().José Fonseca
2010-06-02gallivm: Don't use lp_build_sample_wrap_nearest()José Fonseca
Always clamp with scaled coordinates.
2010-05-27gallivm: add assertions to check type of phi variablesBrian Paul
2010-05-26gallivm: Remove unnecessary headers.Vinson Lee
2010-05-25gallivm: Choose an appropriate code generation optimization level.José Fonseca
'Default' unless GALLIVM_DEBUG=nopt option is set.
2010-05-24gallivm: Efficient implementation of sin/cos.Qicheng Christopher Li
Based on Julien Pommier's SSE and SSE2 algorithms. Signed-off-by: José Fonseca <jfonseca@vmware.com>
2010-05-20gallivm: rename a var: s/val/array_size/Brian Paul
2010-05-20gallivm: fix spelling in commentBrian Paul
2010-05-20gallivm: handle pointer types in lp_dump_llvmtype()Brian Paul
2010-05-17gallivm: Tweak ret_mask handling.José Fonseca
2010-05-17gallivm: implement function calls by inliningZack Rusin
with this approach we inline the entire function body in the caller
2010-05-15gallivm: Centralize some of the LLVM debugging options.José Fonseca
2010-05-15gallivm: Temporarily remove function call supportJosé Fonseca
Commits moved to the gallivm-call feature branch for further experimentation and stabilization.
2010-05-15gallivm: use our util_snprintfZack Rusin
2010-05-15gallivm: implement function callsZack Rusin
2010-05-14gallivm: added some vector length sanity checksBrian Paul
2010-05-14gallivm: added lp_sizeof_llvm_type()Brian Paul
2010-05-14gallivm: comments, const qualifiersBrian Paul
2010-05-14tgsi: clean up in emit_fetch()Brian Paul
2010-05-13gallivm: silence uninitialized var warningBrian Paul
2010-05-13gallivm: silence casting warningBrian Paul
2010-05-13gallivm: silence uninitialized var warningBrian Paul
2010-05-13llvmpipe: silence uninitialized var warningsBrian Paul
2010-05-13gallivm: added lp_dump_llvmtype() debug helperBrian Paul
2010-05-13llvmpipe: additional commentsBrian Paul
2010-05-13gallivm: rename a varBrian Paul
2010-05-13Conditionally revert "gallivm: Use a more compact approach for ↵José Fonseca
lp_build_broadcast_scalar()." This reverts commit a09e46c72461183c879d8472b44fe740ecc79b9f.
2010-05-12gallivm: Ensure PIPE_OS_xxx are defined.José Fonseca
2010-05-12gallivm: Use a more compact approach for lp_build_broadcast_scalar().José Fonseca
It produces exactly the same machine code, but it cuts 5% of the number of instructions generated for a typical shader. Also, preserve the scalar when length is 1.
2010-05-12gallivm: additional commentsBrian Paul
2010-05-12gallivm: s/bool/booleanBrian Paul
2010-05-12gallivm: rename vars, update comments, etcBrian Paul
2010-05-12gallivm: comment out LLVMUnionTypeKind caseBrian Paul
2010-05-12gallivm: rename texel result paramBrian Paul
Be clear that this parameter returns four texel channel results.
2010-05-12gallivm: add a few commentsBrian Paul
2010-05-12gallivm/llvmpipe: add const qualifiersBrian Paul
2010-05-12gallivm: added lp_typekind_name() util functionBrian Paul
2010-05-10gallivm/llvmpipe: move an old comment to a better locationBrian Paul
2010-05-10gallivm: Silent warning.José Fonseca
2010-05-10gallivm: Add missing lvalue.José Fonseca
2010-05-10gallivm: Temporarily disable custom LLVMDumpValue on MSVC.José Fonseca
Seeing very weird crashes during std::cout initialization. The fault probably lies in the way I build LLVM on MSVC, but disable for now to allow more time to investigate.
2010-05-10gallivm: cosf/sinf are macros on MSVC.José Fonseca
So taking the function address does not work.
2010-05-08gallivm: Add an alternative to LLVMDumpValue that works with Windows GUI apps.José Fonseca
2010-05-08gallivm: Fix mipfiltering with negative lod bias.José Fonseca
In particular, don't use the clamped lod to compute level + 1, or lod in [-1, 0] range will actually interpolate with level 1. This makes Mipfilter DCT pass 100%.
2010-05-08gallivm: Actually do floor/ceil/trunc for scalars.José Fonseca
Also start axing the code duplication for scalar case. The olution is to treat the scalar case specially in a few innermost functions, and leave outer functions untouched.
2010-05-08gallivm: Import the code to compute the minimax polynomials.José Fonseca
It's quite a pain to remember the details after a while, and it is quite likely we'll want to use this again, either for different polynomial orders or different functions, so commit it here.
2010-05-08gallivm: Use a minimax polynomial for exp2 in range [0,1] instead [-0.5,5].José Fonseca
The advantage of range[-0.5, 0.5] is that it doesn't require floor (for which intrinsics are only available in SSE4.1). But the EXP opcode pretty much forces us to use floor, and there is a good floor approximation around truncation available anyway. This fixes EXP failures in VShader DCT.
2010-05-08gallivm: Don't hardcode number of args twice.José Fonseca
2010-05-08gallivm: The the JIT engine to use our sinf()/cosf() on Windows.José Fonseca
A quick hack to get the right results, as there are many DCT tests which use these opcodes to generate data to test other opcodes.
2010-05-08gallivm: Fix BREAK/CONT translation.José Fonseca
The cont_mask must be restored and exec mask recomputed in order to decide whether to repeat the loop or not. Unlike the continue mask, the break_mask must be preserved across loop iterations. Fixes several VShader DCT cases, and no regressions with glean.
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-13 13:30:24 +0000