30edf1c449
We still can try to vectorize the bundle of the instructions, even if the repeated number of instruction is non-power-of-2. In this case need to adjust the cost (calculate the cost only for unique scalar instructions) and cost of the extracts. Also, when scheduling the bundle need to schedule only unique scalars to avoid compiler crash because of the multiple dependencies. Can be safely applied only if all scalars's users are also vectorized and do not require memory accesses (this one is a temporarily requirement, can be relaxed later). --------- Co-authored-by: Alexey Bataev <a.bataev@outlook.com>
25 lines
1009 B
LLVM
25 lines
1009 B
LLVM
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
|
|
; RUN: opt -mtriple=x86_64-- -passes=slp-vectorizer,instcombine -S < %s | FileCheck %s
|
|
|
|
; These conversions should be vectorized by reviews.llvm.org/D57059
|
|
|
|
define dso_local <4 x float> @foo(<4 x i32> %0) {
|
|
; CHECK-LABEL: @foo(
|
|
; CHECK-NEXT: [[TMP2:%.*]] = sitofp <4 x i32> [[TMP0:%.*]] to <4 x float>
|
|
; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <4 x float> [[TMP2]], <4 x float> poison, <4 x i32> <i32 1, i32 1, i32 2, i32 3>
|
|
; CHECK-NEXT: ret <4 x float> [[TMP3]]
|
|
;
|
|
%2 = extractelement <4 x i32> %0, i32 1
|
|
%3 = sitofp i32 %2 to float
|
|
%4 = insertelement <4 x float> undef, float %3, i32 0
|
|
%5 = insertelement <4 x float> %4, float %3, i32 1
|
|
%6 = extractelement <4 x i32> %0, i32 2
|
|
%7 = sitofp i32 %6 to float
|
|
%8 = insertelement <4 x float> %5, float %7, i32 2
|
|
%9 = extractelement <4 x i32> %0, i32 3
|
|
%10 = sitofp i32 %9 to float
|
|
%11 = insertelement <4 x float> %8, float %10, i32 3
|
|
ret <4 x float> %11
|
|
}
|
|
|