b10ecfa914
Currently SLP vectorizer tries to keep only GEPs as scalar, if they are vectorized but used externally. Same approach can be used for all scalar values. This patch tries to keep original scalars if all its operands remain scalar or externally used, the cost of the original scalar is lower than the cost of the extractelement instruction, or if the number of externally used scalars in the same entry is power of 2. Last criterion allows better revectorization for multiply used scalars. Reviewers: RKSimon Reviewed By: RKSimon Pull Request: https://github.com/llvm/llvm-project/pull/100904
73 lines
3.1 KiB
LLVM
73 lines
3.1 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
|
|
; RUN: opt -mtriple=i386-linux-gnu -passes=slp-vectorizer -S %s | FileCheck %s
|
|
|
|
%struct.a = type { [2 x i64] }
|
|
|
|
@a = external global %struct.a
|
|
@b = external global %struct.a
|
|
@c = external global %struct.a
|
|
|
|
define void @f(i1 %x) #0 {
|
|
; CHECK-LABEL: @f(
|
|
; CHECK-NEXT: entry:
|
|
; CHECK-NEXT: [[A1:%.*]] = load i64, ptr getelementptr inbounds ([[STRUCT_A:%.*]], ptr @a, i32 0, i32 0, i32 1), align 8
|
|
; CHECK-NEXT: [[TMP0:%.*]] = load <2 x i64>, ptr @a, align 8
|
|
; CHECK-NEXT: br i1 [[X:%.*]], label [[WHILE_BODY_LR_PH:%.*]], label [[WHILE_END:%.*]]
|
|
; CHECK: while.body.lr.ph:
|
|
; CHECK-NEXT: [[ICMP_A1:%.*]] = icmp eq i64 [[A1]], 0
|
|
; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, ptr @b, align 8
|
|
; CHECK-NEXT: [[TMP2:%.*]] = insertelement <2 x i1> poison, i1 [[ICMP_A1]], i32 0
|
|
; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <2 x i1> [[TMP2]], <2 x i1> poison, <2 x i32> zeroinitializer
|
|
; CHECK-NEXT: [[TMP4:%.*]] = select <2 x i1> [[TMP3]], <2 x i64> [[TMP1]], <2 x i64> [[TMP0]]
|
|
; CHECK-NEXT: br label [[WHILE_END]]
|
|
; CHECK: while.end:
|
|
; CHECK-NEXT: [[TMP5:%.*]] = phi <2 x i64> [ [[TMP0]], [[ENTRY:%.*]] ], [ [[TMP4]], [[WHILE_BODY_LR_PH]] ]
|
|
; CHECK-NEXT: [[TMP6:%.*]] = load <2 x i64>, ptr @c, align 8
|
|
; CHECK-NEXT: [[TMP7:%.*]] = extractelement <2 x i64> [[TMP5]], i32 0
|
|
; CHECK-NEXT: [[ICMP_D0:%.*]] = icmp eq i64 [[TMP7]], 0
|
|
; CHECK-NEXT: br i1 [[ICMP_D0]], label [[IF_END:%.*]], label [[IF_THEN:%.*]]
|
|
; CHECK: if.then:
|
|
; CHECK-NEXT: [[AND0_TMP:%.*]] = and i64 [[TMP7]], 8
|
|
; CHECK-NEXT: [[TMP8:%.*]] = insertelement <2 x i64> poison, i64 [[AND0_TMP]], i32 0
|
|
; CHECK-NEXT: [[TMP9:%.*]] = shufflevector <2 x i64> [[TMP8]], <2 x i64> [[TMP5]], <2 x i32> <i32 0, i32 3>
|
|
; CHECK-NEXT: [[TMP10:%.*]] = and <2 x i64> [[TMP9]], [[TMP6]]
|
|
; CHECK-NEXT: store <2 x i64> [[TMP10]], ptr @a, align 8
|
|
; CHECK-NEXT: br label [[IF_END]]
|
|
; CHECK: if.end:
|
|
; CHECK-NEXT: ret void
|
|
;
|
|
entry:
|
|
%a0 = load i64, ptr @a, align 8
|
|
%a1 = load i64, ptr getelementptr inbounds (%struct.a, ptr @a, i32 0, i32 0, i32 1), align 8
|
|
br i1 %x, label %while.body.lr.ph, label %while.end
|
|
|
|
while.body.lr.ph:
|
|
%icmp.a1 = icmp eq i64 %a1, 0
|
|
%b0 = load i64, ptr @b, align 8
|
|
%b1 = load i64, ptr getelementptr inbounds (%struct.a, ptr @b, i32 0, i32 0, i32 1), align 8
|
|
%c0 = select i1 %icmp.a1, i64 %b0, i64 %a0
|
|
%c1 = select i1 %icmp.a1, i64 %b1, i64 %a1
|
|
br label %while.end
|
|
|
|
while.end:
|
|
%d0 = phi i64 [ %a0, %entry ], [ %c0, %while.body.lr.ph ]
|
|
%d1 = phi i64 [ %a1, %entry ], [ %c1, %while.body.lr.ph ]
|
|
%e0 = load i64, ptr @c, align 8
|
|
%e1 = load i64, ptr getelementptr inbounds (%struct.a, ptr @c, i32 0, i32 0, i32 1), align 8
|
|
%icmp.d0 = icmp eq i64 %d0, 0
|
|
br i1 %icmp.d0, label %if.end, label %if.then
|
|
|
|
if.then:
|
|
%and0.tmp = and i64 %d0, 8
|
|
%and0 = and i64 %and0.tmp, %e0
|
|
%and1 = and i64 %e1, %d1
|
|
store i64 %and0, ptr @a, align 8
|
|
store i64 %and1, ptr getelementptr inbounds (%struct.a, ptr @a, i32 0, i32 0, i32 1), align 8
|
|
br label %if.end
|
|
|
|
if.end:
|
|
ret void
|
|
}
|
|
|
|
attributes #0 = { "target-features"="+sse2" }
|