llvm-project/llvm/test/CodeGen/X86/avx-trunc.ll

23 lines
655 B
LLVM
Raw Normal View History

2012-02-03 03:00:49 +08:00
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=corei7-avx -mattr=+avx | FileCheck %s
define <4 x i32> @trunc_64_32(<4 x i64> %A) nounwind uwtable readnone ssp{
; CHECK-LABEL: trunc_64_32
[x86] Teach the 128-bit vector shuffle lowering routines to take advantage of the existence of a reasonable blend instruction. The 256-bit vector shuffle lowering has leveraged the general technique of decomposed shuffles and blends for quite some time, but this never made it back into the 128-bit code, and there are a large number of patterns where this is substantially better. For example, this removes almost all domain crossing in vector shuffles that involve some blend and some permutation with SSE4.1 and later. See the massive reduction in 'shufps' for integer test cases in this commit. This isn't perfect yet for a few reasons: 1) The v8i16 shuffle lowering continues to plague me. We don't always form an unpack-based blend when that would be better. But the wins pretty drastically outstrip the losses here. 2) The v16i8 shuffle lowering is just a disaster here. I never went and implemented blend support here for some terrible reason. I'll do that next probably. I've not updated it for now. More variations on this technique are coming as well -- we don't shuffle-into-unpack or shuffle-into-palignr, both of which would also be profitable. Note that some test cases grow significantly in the number of instructions, but I expect to actually be faster. We use pshufd+pshufd+blendw instead of a single shufps, but the pshufd's are very likely to pipeline well (two ports on most modern intel chips) and the blend is a *very* fast instruction. The domain switch penalty will essentially always be more than a blend instruction, which is the only increase in tree height. llvm-svn: 229350
2015-02-16 09:52:02 +08:00
; CHECK: pshufd
; CHECK: pshufd
; CHECK: pblendw
2012-02-03 03:00:49 +08:00
%B = trunc <4 x i64> %A to <4 x i32>
ret <4 x i32>%B
}
define <8 x i16> @trunc_32_16(<8 x i32> %A) nounwind uwtable readnone ssp{
; CHECK-LABEL: trunc_32_16
2012-02-03 03:00:49 +08:00
; CHECK: pshufb
%B = trunc <8 x i32> %A to <8 x i16>
ret <8 x i16>%B
}
define <16 x i8> @trunc_16_8(<16 x i16> %A) nounwind uwtable readnone ssp{
; CHECK-LABEL: trunc_16_8
; CHECK: pshufb
%B = trunc <16 x i16> %A to <16 x i8>
ret <16 x i8> %B
}