blob: 146eb2a7fe4fda3e2366a26ef96de8f3f0f14def [file]
; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --filter "LV\(REG\): (?!At)" --version 6
; REQUIRES: asserts
; RUN: opt -passes=loop-vectorize -mtriple riscv64-linux-gnu \
; RUN: -mattr=+v,+d -debug-only=loop-vectorize,vplan --disable-output \
; RUN: -riscv-v-register-bit-width-lmul=8 \
; RUN: -S < %s 2>&1 | FileCheck %s --check-prefix=CHECK
define void @add(ptr noalias nocapture readonly %src1, ptr noalias nocapture readonly %src2, i32 signext %size, ptr noalias nocapture writeonly %result) {
; CHECK-LABEL: 'add'
; CHECK: LV(REG): Calculating max register usage:
; CHECK: LV(REG): VF = 1
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 3 registers
; CHECK: LV(REG): RegisterClass: RISCV::FPRRC, 2 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): Calculating max register usage:
; CHECK: LV(REG): VF = vscale x 1
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 6 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 2 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 2
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 6 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 2 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 4
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 6 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 4 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 8
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 6 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 8 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 16
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 6 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 16 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
;
entry:
%conv = zext i32 %size to i64
%cmp10.not = icmp eq i32 %size, 0
br i1 %cmp10.not, label %for.cond.cleanup, label %for.body
for.cond.cleanup:
ret void
for.body:
%i.011 = phi i64 [ %add4, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float, ptr %src1, i64 %i.011
%0 = load float, ptr %arrayidx, align 4
%arrayidx2 = getelementptr inbounds float, ptr %src2, i64 %i.011
%1 = load float, ptr %arrayidx2, align 4
%add = fadd float %0, %1
%arrayidx3 = getelementptr inbounds float, ptr %result, i64 %i.011
store float %add, ptr %arrayidx3, align 4
%add4 = add nuw nsw i64 %i.011, 1
%exitcond.not = icmp eq i64 %add4, %conv
br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
}
define void @goo(ptr nocapture noundef %a, i32 noundef signext %n) {
; CHECK-LABEL: 'goo'
; CHECK: LV(REG): Calculating max register usage:
; CHECK: LV(REG): VF = 1
; CHECK: LV(REG): Found max usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 3 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): Calculating max register usage:
; CHECK: LV(REG): VF = vscale x 1
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 1 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 2
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 2 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 4
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 4 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 8
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 8 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
;
entry:
%cmp3 = icmp sgt i32 %n, 0
br i1 %cmp3, label %for.body.preheader, label %for.cond.cleanup
for.body.preheader:
%wide.trip.count = zext i32 %n to i64
br label %for.body
for.cond.cleanup.loopexit:
br label %for.cond.cleanup
for.cond.cleanup:
ret void
for.body:
%indvars.iv = phi i64 [ 0, %for.body.preheader ], [ %indvars.iv.next, %for.body ]
%arrayidx = getelementptr inbounds ptr, ptr %a, i64 %indvars.iv
%0 = load ptr, ptr %arrayidx, align 8
%add.ptr = getelementptr inbounds i32, ptr %0, i64 1
store ptr %add.ptr, ptr %arrayidx, align 8
%indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
%exitcond.not = icmp eq i64 %indvars.iv.next, %wide.trip.count
br i1 %exitcond.not, label %for.cond.cleanup.loopexit, label %for.body
}
define void @red_smax_smin(ptr %base.0, ptr %base.1, ptr %base.2, ptr %base.3, i64 %end) {
; CHECK-LABEL: 'red_smax_smin'
; CHECK: LV(REG): Calculating max register usage:
; CHECK: LV(REG): VF = 1
; CHECK: LV(REG): Found max usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 10 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): Calculating max register usage:
; CHECK: LV(REG): VF = vscale x 1
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 6 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 2
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 12 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 4
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 24 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 8
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 48 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): Cost of 32 from 16 spills of RISCV::VRRC
;
entry:
br label %loop.body
loop.body:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop.body ]
%red.0 = phi i64 [ 0, %entry ], [ %red.0.next, %loop.body ]
%red.1 = phi i64 [ 0, %entry ], [ %red.1.next, %loop.body ]
%red.2 = phi i64 [ 0, %entry ], [ %red.2.next, %loop.body ]
%red.3 = phi i64 [ 0, %entry ], [ %red.3.next, %loop.body ]
%ptr.0 = getelementptr i8, ptr %base.0, i64 %iv
%0 = load i64, ptr %ptr.0, align 8
%ptr.1 = getelementptr i8, ptr %base.1, i64 %iv
%1 = load i64, ptr %ptr.1, align 8
%red.0.next = tail call i64 @llvm.smin.i64(i64 %0, i64 %red.0)
%red.2.next = tail call i64 @llvm.smax.i64(i64 %0, i64 %red.2)
%red.1.next = tail call i64 @llvm.smin.i64(i64 %1, i64 %red.1)
%red.3.next = tail call i64 @llvm.smax.i64(i64 %1, i64 %red.3)
%iv.next = add i64 %iv, 1
%exitcond = icmp eq i64 %iv, %end
br i1 %exitcond, label %loop.exit, label %loop.body
loop.exit:
store i64 %red.0.next, ptr %base.0
store i64 %red.1.next, ptr %base.1
store i64 %red.2.next, ptr %base.2
store i64 %red.3.next, ptr %base.3
ret void
}
define void @red_umax_umin(ptr %base.0, ptr %base.1, ptr %base.2, ptr %base.3, i64 %end) {
; CHECK-LABEL: 'red_umax_umin'
; CHECK: LV(REG): Calculating max register usage:
; CHECK: LV(REG): VF = 1
; CHECK: LV(REG): Found max usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 10 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): Calculating max register usage:
; CHECK: LV(REG): VF = vscale x 1
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 6 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 2
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 12 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 4
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 24 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): VF = vscale x 8
; CHECK: LV(REG): Found max usage: 2 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 5 registers
; CHECK: LV(REG): RegisterClass: RISCV::VRRC, 48 registers
; CHECK: LV(REG): Found invariant usage: 1 item
; CHECK: LV(REG): RegisterClass: RISCV::GPRRC, 1 registers
; CHECK: LV(REG): Cost of 32 from 16 spills of RISCV::VRRC
;
entry:
br label %loop.body
loop.body:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop.body ]
%red.0 = phi i64 [ 0, %entry ], [ %red.0.next, %loop.body ]
%red.1 = phi i64 [ 0, %entry ], [ %red.1.next, %loop.body ]
%red.2 = phi i64 [ 0, %entry ], [ %red.2.next, %loop.body ]
%red.3 = phi i64 [ 0, %entry ], [ %red.3.next, %loop.body ]
%ptr.0 = getelementptr i8, ptr %base.0, i64 %iv
%0 = load i64, ptr %ptr.0, align 8
%ptr.1 = getelementptr i8, ptr %base.1, i64 %iv
%1 = load i64, ptr %ptr.1, align 8
%red.0.next = tail call i64 @llvm.umin.i64(i64 %0, i64 %red.0)
%red.2.next = tail call i64 @llvm.umax.i64(i64 %0, i64 %red.2)
%red.1.next = tail call i64 @llvm.umin.i64(i64 %1, i64 %red.1)
%red.3.next = tail call i64 @llvm.umax.i64(i64 %1, i64 %red.3)
%iv.next = add i64 %iv, 1
%exitcond = icmp eq i64 %iv, %end
br i1 %exitcond, label %loop.exit, label %loop.body
loop.exit:
store i64 %red.0.next, ptr %base.0
store i64 %red.1.next, ptr %base.1
store i64 %red.2.next, ptr %base.2
store i64 %red.3.next, ptr %base.3
ret void
}