Files
clang-p2996/llvm/test/Transforms/SLPVectorizer/X86/int-bitcast-minbitwidth.ll
Alexey Bataev 01d9528ef9 [SLP]Improve final minbitwidth analysis attempt.
Added part for demanded bits analysis in the IsPotentiallyTruncated to
improve minbitwidth analysis final attempts.

Metric: size..text

Program                                                                                                                                                size..text
                                                                                                                                                       results     results0    diff
                                                                           test-suite :: MultiSource/Benchmarks/MiBench/telecomm-gsm/telecomm-gsm.test    43069.00    42973.00 -0.2%
                                                                                  test-suite :: MultiSource/Benchmarks/mediabench/gsm/toast/toast.test    43066.00    42970.00 -0.2%

Extra trunc instructions are emitted to operate with <32 x i8> instead
of <32 x i16>, will be removed in the next patches.

Reviewers: RKSimon

Reviewed By: RKSimon

Pull Request: https://github.com/llvm/llvm-project/pull/87786
2024-04-08 15:54:30 -04:00

34 lines
1.4 KiB
LLVM

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
; RUN: opt -S --passes=slp-vectorizer -mtriple=x86_64-unknown-linux-gnu -slp-threshold=-9 < %s | FileCheck %s
define void @t(i64 %v) {
; CHECK-LABEL: define void @t(
; CHECK-SAME: i64 [[V:%.*]]) {
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP0:%.*]] = insertelement <4 x i64> poison, i64 [[V]], i32 0
; CHECK-NEXT: [[TMP1:%.*]] = shufflevector <4 x i64> [[TMP0]], <4 x i64> poison, <4 x i32> zeroinitializer
; CHECK-NEXT: [[TMP2:%.*]] = trunc <4 x i64> [[TMP1]] to <4 x i16>
; CHECK-NEXT: [[TMP3:%.*]] = mul <4 x i16> [[TMP2]], <i16 5, i16 6, i16 3, i16 2>
; CHECK-NEXT: [[TMP4:%.*]] = call i16 @llvm.vector.reduce.or.v4i16(<4 x i16> [[TMP3]])
; CHECK-NEXT: [[TMP5:%.*]] = sext i16 [[TMP4]] to i32
; CHECK-NEXT: [[TMP6:%.*]] = and i32 [[TMP5]], 65535
; CHECK-NEXT: store i32 [[TMP6]], ptr null, align 4
; CHECK-NEXT: ret void
;
entry:
%conv12.1.i = trunc i64 %v to i32
%mul.i.1.i = mul i32 %conv12.1.i, 2
%conv12.i = trunc i64 %v to i32
%mul.i.i = mul i32 %conv12.i, 3
%conv14104.i = or i32 %mul.i.1.i, %mul.i.i
%conv12.1.i.1 = trunc i64 %v to i32
%mul.i.1.i.1 = mul i32 %conv12.1.i.1, 6
%conv12.i.1 = trunc i64 %v to i32
%mul.i.i.1 = mul i32 %conv12.i.1, 5
%conv14104.i.1 = or i32 %mul.i.1.i.1, %mul.i.i.1
%0 = or i32 %conv14104.i, %conv14104.i.1
%1 = and i32 %0, 65535
store i32 %1, ptr null, align 4
ret void
}