Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | Cleanup 3dnow builtin handling. Most of them were already handled by LLVM ↵ | Craig Topper | 2012-01-30 | 1 | -1/+0 | |
| | | | | | | connecting intrinsics and builtins in IntrinsicsX86.td. llvm-svn: 149233 | |||||
* | Re-enable test that was broken by r148919 | Craig Topper | 2012-01-25 | 1 | -6/+2 | |
| | | | | llvm-svn: 148932 | |||||
* | disable this test for now. | Chris Lattner | 2012-01-25 | 1 | -2/+3 | |
| | | | | llvm-svn: 148928 | |||||
* | Add AVX2 intrinsics for pavg, pblend, and pcmp instructions. Also remove ↵ | Craig Topper | 2011-12-20 | 1 | -6/+0 | |
| | | | | | | unneeded builtins for SSE pcmp. Change SSE pcmpeqq and pcmpgtq to not use builtins and just use vector == and >. llvm-svn: 146969 | |||||
* | Fix _mm256_round_pd, _mm256_round_ps, _mm_permute_pd and _mm256_permute_pd AVX | Chad Rosier | 2011-12-17 | 1 | -2/+2 | |
| | | | | | | | | intrinsics to use "I" (ICE) markings. Fix avxintrin.h to take them into account. Part of rdar://10595450 llvm-svn: 146791 | |||||
* | Remove the 'unaligned load' builtins now that they're no longer used in the ↵ | Bill Wendling | 2011-05-13 | 1 | -3/+0 | |
| | | | | | | *mmintrin.h files. llvm-svn: 131300 | |||||
* | LLVM doesn't always optimize away the four loads from this: | Bill Wendling | 2011-05-12 | 1 | -0/+2 | |
| | | | | | | | | | | (__m128){ p[0], p[1], p[2], p[3] } which produces really bad code. This could be done in instcombine, but it's probably better to do it in the front-end instead. <rdar://problem/9424836> llvm-svn: 131237 | |||||
* | Add 3DNow! Intrinsics. | Michael J. Spencer | 2011-04-15 | 1 | -0/+29 | |
| | | | | llvm-svn: 129570 | |||||
* | Removing the unaligned load tests from builtins-x86.c since they're ↵ | Bill Wendling | 2011-04-13 | 1 | -2/+0 | |
| | | | | | | generated by a regular 'load' now. llvm-svn: 129464 | |||||
* | Implement __builtin_ia32_vec_ext_v2si function (required by Qt). | Argyrios Kyrtzidis | 2010-10-10 | 1 | -0/+1 | |
| | | | | llvm-svn: 116162 | |||||
* | Remove rsqrtps_nr256 and sqrtps_nr256 builtins, at least until we need them | Bruno Cardoso Lopes | 2010-08-11 | 1 | -2/+0 | |
| | | | | llvm-svn: 110844 | |||||
* | Remove 256-bit cast built-ins and make the AVX intrinsic call llvm ↵ | Bruno Cardoso Lopes | 2010-08-11 | 1 | -6/+0 | |
| | | | | | | __builtin_shufflevector with the appropriate arguments llvm-svn: 110771 | |||||
* | Remove 256-bit unpack built-ins and make the AVX intrinsic call llvm ↵ | Bruno Cardoso Lopes | 2010-08-11 | 1 | -4/+0 | |
| | | | | | | __builtin_shufflevector with the appropriate arguments llvm-svn: 110768 | |||||
* | Remove 256-bit shuffle built-ins and make the AVX intrinsic call llvm ↵ | Bruno Cardoso Lopes | 2010-08-11 | 1 | -2/+0 | |
| | | | | | | __builtin_shufflevector with the appropriate arguments llvm-svn: 110766 | |||||
* | Make replicate intrinsics use shufflevector instead of dup builtins, also ↵ | Bruno Cardoso Lopes | 2010-08-10 | 1 | -3/+0 | |
| | | | | | | remove the dup builtins llvm-svn: 110646 | |||||
* | We don't want to support built-ins which aren't needed by the intrinsics. ↵ | Bruno Cardoso Lopes | 2010-08-05 | 1 | -16/+0 | |
| | | | | | | Remove them llvm-svn: 110399 | |||||
* | Add more AVX 256-bit intrinsics and test cases for them | Bruno Cardoso Lopes | 2010-08-04 | 1 | -0/+36 | |
| | | | | llvm-svn: 110178 | |||||
* | Support x86 AVX 256-bit instructions built-ins. Right now support all of ↵ | Bruno Cardoso Lopes | 2010-08-03 | 1 | -2/+112 | |
| | | | | | | | | | | them, but as soon as we properly codegen the simple vector operations, remove the unnecessary built-ins/intrinsics from clang and llvm. Also add tests for the new built-ins llvm-svn: 110096 | |||||
* | Fix __builtin_ia32_roundss and __builtin_ia32_roundsd definitions. | Eric Christopher | 2010-03-04 | 1 | -2/+2 | |
| | | | | | | Re-enable test. llvm-svn: 97707 | |||||
* | Update tests to use %clang_cc1 instead of 'clang-cc' or 'clang -cc1'. | Daniel Dunbar | 2009-12-15 | 1 | -2/+2 | |
| | | | | | | | | | - This is designed to make it obvious that %clang_cc1 is a "test variable" which is substituted. It is '%clang_cc1' instead of '%clang -cc1' because it can be useful to redefine what gets run as 'clang -cc1' (for example, to set a default target). llvm-svn: 91446 | |||||
* | Eliminate &&s in tests. | Daniel Dunbar | 2009-11-08 | 1 | -1/+1 | |
| | | | | | | - 'for i in $(find . -type f); do sed -e 's#\(RUN:.*[^ ]\) *&& *$#\1#g' $i | FileUpdate $i; done', for the curious. llvm-svn: 86430 | |||||
* | Remove a few more vector builtins. | Eli Friedman | 2009-06-07 | 1 | -5/+0 | |
| | | | | llvm-svn: 73022 | |||||
* | Now that LLVM CodeGen can handle the generic variations a bit better, | Eli Friedman | 2009-06-07 | 1 | -4/+0 | |
| | | | | | | get rid of a few more clang vector builtins. llvm-svn: 73015 | |||||
* | Test changes to account for removed builtins. | Eli Friedman | 2009-06-06 | 1 | -143/+0 | |
| | | | | llvm-svn: 73004 | |||||
* | Remove an unused builtin. | Anders Carlsson | 2009-05-18 | 1 | -1/+0 | |
| | | | | llvm-svn: 72033 | |||||
* | Add 'cmp' SSE builtins and get rid of a bunch of other builtins. | Anders Carlsson | 2009-05-18 | 1 | -44/+32 | |
| | | | | llvm-svn: 72032 | |||||
* | Rename clang to clang-cc. | Daniel Dunbar | 2009-03-24 | 1 | -2/+2 | |
| | | | | | | Tests and drivers updated, still need to shuffle dirs. llvm-svn: 67602 | |||||
* | Fix definition of __builtin_ia32_vec_set_v2di and de-XFAIL | Daniel Dunbar | 2009-01-26 | 1 | -5/+4 | |
| | | | | | | builtins-x86.c. llvm-svn: 63069 | |||||
* | Added vec_set intrinsics | Mon P Wang | 2008-10-18 | 1 | -0/+4 | |
| | | | | llvm-svn: 57749 | |||||
* | Add X86 builtin code generation test case. | Daniel Dunbar | 2008-10-05 | 1 | -0/+532 | |
llvm-svn: 57104 |