; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -passes=loop-vectorize --force-vector-width=4 --force-vector-interleave=0 -S -o - < %s | FileCheck %s
target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"
%0 = type { i32 }
%1 = type { i64 }
define ptr @foo(ptr %p, ptr %p.last) unnamed_addr #0 {
; CHECK-LABEL: @foo(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[P3:%.*]] = ptrtoint ptr [[P:%.*]] to i64
; CHECK-NEXT: [[P_LAST1:%.*]] = ptrtoint ptr [[P_LAST:%.*]] to i64
; CHECK-NEXT: [[TMP0:%.*]] = add i64 [[P_LAST1]], -1024
; CHECK-NEXT: [[TMP1:%.*]] = sub i64 [[TMP0]], [[P3]]
; CHECK-NEXT: [[TMP2:%.*]] = lshr i64 [[TMP1]], 10
; CHECK-NEXT: [[TMP3:%.*]] = add nuw nsw i64 [[TMP2]], 1
; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP3]], 16
; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK: vector.ph:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[TMP3]], 16
; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP3]], [[N_MOD_VF]]
; CHECK-NEXT: [[TMP4:%.*]] = mul i64 [[N_VEC]], 1024
; CHECK-NEXT: [[IND_END:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP4]]
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[POINTER_PHI:%.*]] = phi ptr [ [[P]], [[VECTOR_PH]] ], [ [[PTR_IND:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP5:%.*]] = getelementptr i8, ptr [[POINTER_PHI]], <4 x i64> <i64 0, i64 1024, i64 2048, i64 3072>
; CHECK-NEXT: [[TMP8:%.*]] = getelementptr i8, ptr [[POINTER_PHI]], <4 x i64> <i64 12288, i64 13312, i64 14336, i64 15360>
; CHECK-NEXT: [[WIDE_MASKED_GATHER6:%.*]] = call <4 x ptr> @llvm.masked.gather.v4p0.v4p0(<4 x ptr> [[TMP8]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x ptr> poison)
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
; CHECK-NEXT: [[PTR_IND]] = getelementptr i8, ptr [[POINTER_PHI]], i64 16384
; CHECK-NEXT: [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[TMP10:%.*]] = extractelement <4 x ptr> [[WIDE_MASKED_GATHER6]], i32 3
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP3]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; CHECK: scalar.ph:
; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi ptr [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[P]], [[ENTRY:%.*]] ]
; CHECK-NEXT: br label [[LOOP:%.*]]
; CHECK: loop:
; CHECK-NEXT: [[P2:%.*]] = phi ptr [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[P_INC:%.*]], [[LOOP]] ]
; CHECK-NEXT: [[P_INC]] = getelementptr inbounds i64, ptr [[P2]], i64 128
; CHECK-NEXT: [[V:%.*]] = load ptr, ptr [[P2]], align 8
; CHECK-NEXT: [[B:%.*]] = icmp eq ptr [[P_INC]], [[P_LAST]]
; CHECK-NEXT: br i1 [[B]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK: exit:
; CHECK-NEXT: [[V_LCSSA:%.*]] = phi ptr [ [[V]], [[LOOP]] ], [ [[TMP10]], [[MIDDLE_BLOCK]] ]
; CHECK-NEXT: ret ptr [[V_LCSSA]]
;
entry:
br label %loop
loop:
%p2 = phi ptr [ %p, %entry ], [ %p.inc, %loop ]
%p.inc = getelementptr inbounds i64, ptr %p2, i64 128
%v = load ptr, ptr %p2, align 8
%b = icmp eq ptr %p.inc, %p.last
br i1 %b, label %exit, label %loop
exit:
ret ptr %v
}
define ptr @bar(ptr %p, ptr %p.last) unnamed_addr #0 {
; CHECK-LABEL: @bar(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[P3:%.*]] = ptrtoint ptr [[P:%.*]] to i64
; CHECK-NEXT: [[P_LAST1:%.*]] = ptrtoint ptr [[P_LAST:%.*]] to i64
; CHECK-NEXT: [[TMP0:%.*]] = add i64 [[P_LAST1]], -1024
; CHECK-NEXT: [[TMP1:%.*]] = sub i64 [[TMP0]], [[P3]]
; CHECK-NEXT: [[TMP2:%.*]] = lshr i64 [[TMP1]], 10
; CHECK-NEXT: [[TMP3:%.*]] = add nuw nsw i64 [[TMP2]], 1
; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP3]], 16
; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK: vector.ph:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[TMP3]], 16
; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP3]], [[N_MOD_VF]]
; CHECK-NEXT: [[TMP4:%.*]] = mul i64 [[N_VEC]], 1024
; CHECK-NEXT: [[IND_END:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP4]]
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[POINTER_PHI:%.*]] = phi ptr [ [[P]], [[VECTOR_PH]] ], [ [[PTR_IND:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP5:%.*]] = getelementptr i8, ptr [[POINTER_PHI]], <4 x i64> <i64 0, i64 1024, i64 2048, i64 3072>
; CHECK-NEXT: [[TMP8:%.*]] = getelementptr i8, ptr [[POINTER_PHI]], <4 x i64> <i64 12288, i64 13312, i64 14336, i64 15360>
; CHECK-NEXT: [[WIDE_MASKED_GATHER6:%.*]] = call <4 x ptr> @llvm.masked.gather.v4p0.v4p0(<4 x ptr> [[TMP8]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x ptr> poison)
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
; CHECK-NEXT: [[PTR_IND]] = getelementptr i8, ptr [[POINTER_PHI]], i64 16384
; CHECK-NEXT: [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[TMP10:%.*]] = extractelement <4 x ptr> [[WIDE_MASKED_GATHER6]], i32 3
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP3]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; CHECK: scalar.ph:
; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi ptr [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[P]], [[ENTRY:%.*]] ]
; CHECK-NEXT: br label [[LOOP:%.*]]
; CHECK: loop:
; CHECK-NEXT: [[P2:%.*]] = phi ptr [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[P_INC:%.*]], [[LOOP]] ]
; CHECK-NEXT: [[P_INC]] = getelementptr inbounds i64, ptr [[P2]], i64 128
; CHECK-NEXT: [[V:%.*]] = load ptr, ptr [[P2]], align 8
; CHECK-NEXT: [[B:%.*]] = icmp eq ptr [[P_INC]], [[P_LAST]]
; CHECK-NEXT: br i1 [[B]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK: exit:
; CHECK-NEXT: [[V_LCSSA:%.*]] = phi ptr [ [[V]], [[LOOP]] ], [ [[TMP10]], [[MIDDLE_BLOCK]] ]
; CHECK-NEXT: ret ptr [[V_LCSSA]]
;
entry:
br label %loop
loop:
%p2 = phi ptr [ %p, %entry ], [ %p.inc, %loop ]
%p.inc = getelementptr inbounds i64, ptr %p2, i64 128
%v = load ptr, ptr %p2, align 8
%b = icmp eq ptr %p.inc, %p.last
br i1 %b, label %exit, label %loop
exit:
ret ptr %v
}
attributes #0 = { "target-cpu"="skylake" }
Codebase Browser
llvm