| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2011-09-15 | Change all checks regarding the presence of any SSE level to always | Bruno Cardoso Lopes | |
| take into consideration the presence of AVX. This change, together with the SSEDomainFix enabled for AVX, makes AVX codegen to always (hopefully) emit the same code as SSE for 128-bit vector ops. I don't have a testcase for this, but AVX now beats SSE in performance for 128-bit ops in the majority of programas in the llvm testsuite git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139817 91177308-0d34-0410-b5e6-96231b3b80d8 | |||
| 2011-09-12 | Fix PR10845. SUBREG_TO_REG shouldn't be used when the input and | Bruno Cardoso Lopes | |
| destination types are equal! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139553 91177308-0d34-0410-b5e6-96231b3b80d8 | |||
| 2011-08-25 | Add support for 256-bit versions of VSHUFPD and VSHUFPS. | Bruno Cardoso Lopes | |
| git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138546 91177308-0d34-0410-b5e6-96231b3b80d8 | |||
| 2011-08-16 | Instead of always leaving the work to the generic legalizer when | Bruno Cardoso Lopes | |
| there is no support for native 256-bit shuffles, be more smart in some cases, for example, when you can extract specific 128-bit parts and use regular 128-bit shuffles for them. Example: For this shuffle: shufflevector <4 x i64> %a, <4 x i64> %b, <4 x i32> <i32 1, i32 0, i32 7, i32 6> This was expanded to: vextractf128 $1, %ymm1, %xmm2 vpextrq $0, %xmm2, %rax vmovd %rax, %xmm1 vpextrq $1, %xmm2, %rax vmovd %rax, %xmm2 vpunpcklqdq %xmm1, %xmm2, %xmm1 vpextrq $0, %xmm0, %rax vmovd %rax, %xmm2 vpextrq $1, %xmm0, %rax vmovd %rax, %xmm0 vpunpcklqdq %xmm2, %xmm0, %xmm0 vinsertf128 $1, %xmm1, %ymm0, %ymm0 ret Now we get: vshufpd $1, %xmm0, %xmm0, %xmm0 vextractf128 $1, %ymm1, %xmm1 vshufpd $1, %xmm1, %xmm1, %xmm1 vinsertf128 $1, %xmm1, %ymm0, %ymm0 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@137733 91177308-0d34-0410-b5e6-96231b3b80d8 | |||
| 2011-08-11 | Fix PR10492 by teaching MOVHLPS and MOVLPS mask matching to be more strict. | Bruno Cardoso Lopes | |
| git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@137324 91177308-0d34-0410-b5e6-96231b3b80d8 | |||
| 2011-08-09 | Rename and tidy up tests | Bruno Cardoso Lopes | |
| git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@137103 91177308-0d34-0410-b5e6-96231b3b80d8 | |||
