87 lines
2.8 KiB
LLVM
87 lines
2.8 KiB
LLVM
|
; RUN: opt -basic-aa -loop-distribute -enable-loop-distribute -verify-loop-info -verify-dom-info -S \
|
||
|
|
; RUN: < %s | FileCheck %s
|
||
|
|
|
||
|
|
target datalayout = "e-m:o-i32:64-f80:128-n8:16:32:64-S128"
|
||
|
|
target triple = "x86_64-apple-macosx10.10.0"
|
||
|
|
|
||
|
|
; NOTE: The tests below use infinite loops to force unknown backedge-taken counts.
|
||
|
|
; Making the exit condition depend on a load would break current loop-distribute,
|
||
|
|
; because it requires all accesses to end up in either of the loops, but not both.
|
||
|
|
|
||
|
|
; TODO
|
||
|
|
; Can distribute with unknown backedge-taken count, because no runtime checks are
|
||
|
|
; required.
|
||
|
|
define void @unknown_btc_distribute_no_checks_needed(i32* noalias %a,
|
||
|
|
i32* noalias %c,
|
||
|
|
i32* noalias %d) {
|
||
|
|
; CHECK-LABEL: @unknown_btc_distribute_no_checks_needed(
|
||
|
|
; CHECK-NEXT: entry:
|
||
|
|
; CHECK-NEXT: br label %for.body
|
||
|
|
;
|
||
|
|
entry:
|
||
|
|
br label %for.body
|
||
|
|
|
||
|
|
for.body: ; preds = %for.body, %entry
|
||
|
|
%ind = phi i32 [ 0, %entry ], [ %add, %for.body ]
|
||
|
|
|
||
|
|
%arrayidxA = getelementptr inbounds i32, i32* %a, i32 %ind
|
||
|
|
%loadA = load i32, i32* %arrayidxA, align 4
|
||
|
|
|
||
|
|
%mulA = mul i32 %loadA, 10
|
||
|
|
|
||
|
|
%add = add nuw nsw i32 %ind, 1
|
||
|
|
%arrayidxA_plus_4 = getelementptr inbounds i32, i32* %a, i32 %add
|
||
|
|
store i32 %mulA, i32* %arrayidxA_plus_4, align 4
|
||
|
|
|
||
|
|
%arrayidxD = getelementptr inbounds i32, i32* %d, i32 %ind
|
||
|
|
%loadD = load i32, i32* %arrayidxD, align 4
|
||
|
|
|
||
|
|
%mulC = mul i32 %loadD, 20
|
||
|
|
|
||
|
|
%arrayidxC = getelementptr inbounds i32, i32* %c, i32 %ind
|
||
|
|
store i32 %mulC, i32* %arrayidxC, align 4
|
||
|
|
|
||
|
|
br i1 false, label %for.end, label %for.body
|
||
|
|
|
||
|
|
for.end: ; preds = %for.body
|
||
|
|
ret void
|
||
|
|
}
|
||
|
|
|
||
|
|
; Cannot distribute with unknown backedge-taken count, because runtime checks for
|
||
|
|
; induction wrapping are required.
|
||
|
|
define void @unknown_btc_do_not_distribute_wrapping_checks(i32* noalias %a,
|
||
|
|
i32* noalias %c,
|
||
|
|
i32* noalias %d) {
|
||
|
|
; CHECK-LABEL: @unknown_btc_do_not_distribute_wrapping_checks(
|
||
|
|
; CHECK-NEXT: entry:
|
||
|
|
; CHECK-NEXT: br label %for.body
|
||
|
|
;
|
||
|
|
entry:
|
||
|
|
br label %for.body
|
||
|
|
|
||
|
|
for.body: ; preds = %for.body, %entry
|
||
|
|
%ind = phi i32 [ 0, %entry ], [ %add, %for.body ]
|
||
|
|
|
||
|
|
%arrayidxA = getelementptr inbounds i32, i32* %a, i32 %ind
|
||
|
|
%loadA = load i32, i32* %arrayidxA, align 4
|
||
|
|
|
||
|
|
%mulA = mul i32 %loadA, 10
|
||
|
|
|
||
|
|
%add = add i32 %ind, 1
|
||
|
|
%arrayidxA_plus_4 = getelementptr inbounds i32, i32* %a, i32 %add
|
||
|
|
store i32 %mulA, i32* %arrayidxA_plus_4, align 4
|
||
|
|
|
||
|
|
%arrayidxD = getelementptr inbounds i32, i32* %d, i32 %ind
|
||
|
|
%loadD = load i32, i32* %arrayidxD, align 4
|
||
|
|
|
||
|
|
%mulC = mul i32 %loadD, 20
|
||
|
|
|
||
|
|
%arrayidxC = getelementptr inbounds i32, i32* %c, i32 %ind
|
||
|
|
store i32 %mulC, i32* %arrayidxC, align 4
|
||
|
|
|
||
|
|
br i1 false, label %for.end, label %for.body
|
||
|
|
|
||
|
|
for.end: ; preds = %for.body
|
||
|
|
ret void
|
||
|
|
}
|