a99dd8344e
Many tests were running extra passes after loop-vectorize, but they made no difference to the outcome. If tests don't require these extra passes, then we shouldn't run them because it increases the testing time.
38 lines
1.3 KiB
LLVM
38 lines
1.3 KiB
LLVM
; RUN: opt < %s -passes=loop-vectorize -force-vector-interleave=1 -force-vector-width=4 -enable-if-conversion -S | FileCheck %s
|
|
|
|
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
|
|
|
|
;CHECK-LABEL: @reduction_func(
|
|
;CHECK-NOT: load <4 x i32>
|
|
;CHECK: ret i32
|
|
define i32 @reduction_func(ptr nocapture %A, i32 %n) readonly {
|
|
entry:
|
|
%cmp10 = icmp sgt i32 %n, 0
|
|
br i1 %cmp10, label %for.body, label %for.end
|
|
|
|
for.body:
|
|
%indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
|
|
%sum.011 = phi i32 [ %sum.1, %for.inc ], [ 0, %entry ]
|
|
%arrayidx = getelementptr inbounds i32, ptr %A, i64 %indvars.iv
|
|
%0 = load i32, ptr %arrayidx, align 4
|
|
%cmp1 = icmp sgt i32 %0, 30
|
|
br i1 %cmp1, label %if.then, label %for.inc
|
|
|
|
if.then:
|
|
%add = add i32 %sum.011, 2
|
|
%add4 = add i32 %add, %0
|
|
br label %for.inc
|
|
|
|
for.inc:
|
|
%sum.1 = phi i32 [ %add4, %if.then ], [ %sum.011, %for.body ]
|
|
%indvars.iv.next = add i64 %indvars.iv, 1
|
|
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
|
%exitcond = icmp eq i32 %lftr.wideiv, %n
|
|
br i1 %exitcond, label %for.end, label %for.body
|
|
|
|
for.end:
|
|
%sum.0.lcssa = phi i32 [ 0, %entry ], [ 4, %for.inc ]
|
|
ret i32 %sum.0.lcssa
|
|
}
|
|
|