llvm-6502/test/Analysis/ScalarEvolution/nsw.ll
Benjamin Kramer 19ea37059a SCEV: Make the final add of an inbounds GEP nuw if we know that the index is positive.
We can't do this for the general case as saying a GEP with a negative index
doesn't have unsigned wrap isn't valid for negative indices.
  %gep = getelementptr inbounds i32* %p, i64 -1

But an inbounds GEP cannot run past the end of address space. So we check for
the very common case of a positive index and make GEPs derived from that NUW.
Together with Andy's recent non-unit stride work this lets us analyze loops
like

  void foo3(int *a, int *b) {
    for (; a < b; a++) {}
  }

PR12375, PR12376.

Differential Revision: http://llvm-reviews.chandlerc.com/D2033

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193514 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-28 07:30:06 +00:00

161 lines
5.7 KiB
LLVM

; RUN: opt < %s -analyze -scalar-evolution | FileCheck %s
; The addrecs in this loop are analyzable only by using nsw information.
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64"
; CHECK: Classifying expressions for: @test1
define void @test1(double* %p) nounwind {
entry:
%tmp = load double* %p, align 8 ; <double> [#uses=1]
%tmp1 = fcmp ogt double %tmp, 2.000000e+00 ; <i1> [#uses=1]
br i1 %tmp1, label %bb.nph, label %return
bb.nph: ; preds = %entry
br label %bb
bb: ; preds = %bb1, %bb.nph
%i.01 = phi i32 [ %tmp8, %bb1 ], [ 0, %bb.nph ] ; <i32> [#uses=3]
; CHECK: %i.01
; CHECK-NEXT: --> {0,+,1}<nuw><nsw><%bb>
%tmp2 = sext i32 %i.01 to i64 ; <i64> [#uses=1]
%tmp3 = getelementptr double* %p, i64 %tmp2 ; <double*> [#uses=1]
%tmp4 = load double* %tmp3, align 8 ; <double> [#uses=1]
%tmp5 = fmul double %tmp4, 9.200000e+00 ; <double> [#uses=1]
%tmp6 = sext i32 %i.01 to i64 ; <i64> [#uses=1]
%tmp7 = getelementptr double* %p, i64 %tmp6 ; <double*> [#uses=1]
; CHECK: %tmp7
; CHECK-NEXT: --> {%p,+,8}<%bb>
store double %tmp5, double* %tmp7, align 8
%tmp8 = add nsw i32 %i.01, 1 ; <i32> [#uses=2]
; CHECK: %tmp8
; CHECK-NEXT: --> {1,+,1}<nuw><nsw><%bb>
br label %bb1
bb1: ; preds = %bb
%phitmp = sext i32 %tmp8 to i64 ; <i64> [#uses=1]
; CHECK: %phitmp
; CHECK-NEXT: --> {1,+,1}<nuw><nsw><%bb>
%tmp9 = getelementptr double* %p, i64 %phitmp ; <double*> [#uses=1]
; CHECK: %tmp9
; CHECK-NEXT: --> {(8 + %p),+,8}<%bb>
%tmp10 = load double* %tmp9, align 8 ; <double> [#uses=1]
%tmp11 = fcmp ogt double %tmp10, 2.000000e+00 ; <i1> [#uses=1]
br i1 %tmp11, label %bb, label %bb1.return_crit_edge
bb1.return_crit_edge: ; preds = %bb1
br label %return
return: ; preds = %bb1.return_crit_edge, %entry
ret void
}
; CHECK: Classifying expressions for: @test2
define void @test2(i32* %begin, i32* %end) ssp {
entry:
%cmp1.i.i = icmp eq i32* %begin, %end
br i1 %cmp1.i.i, label %_ZSt4fillIPiiEvT_S1_RKT0_.exit, label %for.body.lr.ph.i.i
for.body.lr.ph.i.i: ; preds = %entry
br label %for.body.i.i
for.body.i.i: ; preds = %for.body.i.i, %for.body.lr.ph.i.i
%__first.addr.02.i.i = phi i32* [ %begin, %for.body.lr.ph.i.i ], [ %ptrincdec.i.i, %for.body.i.i ]
; CHECK: %__first.addr.02.i.i
; CHECK-NEXT: --> {%begin,+,4}<nuw><%for.body.i.i>
store i32 0, i32* %__first.addr.02.i.i, align 4
%ptrincdec.i.i = getelementptr inbounds i32* %__first.addr.02.i.i, i64 1
; CHECK: %ptrincdec.i.i
; CHECK-NEXT: --> {(4 + %begin),+,4}<nuw><%for.body.i.i>
%cmp.i.i = icmp eq i32* %ptrincdec.i.i, %end
br i1 %cmp.i.i, label %for.cond.for.end_crit_edge.i.i, label %for.body.i.i
for.cond.for.end_crit_edge.i.i: ; preds = %for.body.i.i
br label %_ZSt4fillIPiiEvT_S1_RKT0_.exit
_ZSt4fillIPiiEvT_S1_RKT0_.exit: ; preds = %entry, %for.cond.for.end_crit_edge.i.i
ret void
}
; Various checks for inbounds geps.
define void @test3(i32* %begin, i32* %end) nounwind ssp {
entry:
%cmp7.i.i = icmp eq i32* %begin, %end
br i1 %cmp7.i.i, label %_ZSt4fillIPiiEvT_S1_RKT0_.exit, label %for.body.i.i
for.body.i.i: ; preds = %entry, %for.body.i.i
%indvar.i.i = phi i64 [ %tmp, %for.body.i.i ], [ 0, %entry ]
; CHECK: %indvar.i.i
; CHECK: {0,+,1}<nuw><nsw><%for.body.i.i>
%tmp = add nsw i64 %indvar.i.i, 1
; CHECK: %tmp =
; CHECK: {1,+,1}<nuw><nsw><%for.body.i.i>
%ptrincdec.i.i = getelementptr inbounds i32* %begin, i64 %tmp
; CHECK: %ptrincdec.i.i =
; CHECK: {(4 + %begin),+,4}<nsw><%for.body.i.i>
%__first.addr.08.i.i = getelementptr inbounds i32* %begin, i64 %indvar.i.i
; CHECK: %__first.addr.08.i.i
; CHECK: {%begin,+,4}<nsw><%for.body.i.i>
store i32 0, i32* %__first.addr.08.i.i, align 4
%cmp.i.i = icmp eq i32* %ptrincdec.i.i, %end
br i1 %cmp.i.i, label %_ZSt4fillIPiiEvT_S1_RKT0_.exit, label %for.body.i.i
; CHECK: Loop %for.body.i.i: backedge-taken count is ((-4 + (-1 * %begin) + %end) /u 4)
; CHECK: Loop %for.body.i.i: max backedge-taken count is ((-4 + (-1 * %begin) + %end) /u 4)
_ZSt4fillIPiiEvT_S1_RKT0_.exit: ; preds = %for.body.i.i, %entry
ret void
}
; A single AddExpr exists for (%a + %b), which is not always <nsw>.
; CHECK: @addnsw
; CHECK-NOT: --> (%a + %b)<nsw>
define i32 @addnsw(i32 %a, i32 %b) nounwind ssp {
entry:
%tmp = add i32 %a, %b
%cmp = icmp sgt i32 %tmp, 0
br i1 %cmp, label %greater, label %exit
greater:
%tmp2 = add nsw i32 %a, %b
br label %exit
exit:
%result = phi i32 [ %a, %entry ], [ %tmp2, %greater ]
ret i32 %result
}
; TODO: This could fold down to '1'
; CHECK-LABEL: PR12375
; CHECK: --> {(4 + %arg),+,4}<nuw><%bb1> Exits: (4 + (4 * ((-1 + (-1 * %arg) + ((4 + %arg) umax (8 + %arg)<nsw>)) /u 4)) + %arg)
define i32 @PR12375(i32* readnone %arg) {
bb:
%tmp = getelementptr inbounds i32* %arg, i64 2
br label %bb1
bb1: ; preds = %bb1, %bb
%tmp2 = phi i32* [ %arg, %bb ], [ %tmp5, %bb1 ]
%tmp3 = phi i32 [ 0, %bb ], [ %tmp4, %bb1 ]
%tmp4 = add nsw i32 %tmp3, 1
%tmp5 = getelementptr inbounds i32* %tmp2, i64 1
%tmp6 = icmp ult i32* %tmp5, %tmp
br i1 %tmp6, label %bb1, label %bb7
bb7: ; preds = %bb1
ret i32 %tmp4
}
; CHECK-LABEL: PR12376
; CHECK: --> {(4 + %arg),+,4}<nuw><%bb2> Exits: (4 + (4 * ((3 + (-1 * %arg) + (%arg umax %arg1)) /u 4)) + %arg)
define void @PR12376(i32* nocapture %arg, i32* nocapture %arg1) {
bb:
br label %bb2
bb2: ; preds = %bb2, %bb
%tmp = phi i32* [ %arg, %bb ], [ %tmp4, %bb2 ]
%tmp3 = icmp ult i32* %tmp, %arg1
%tmp4 = getelementptr inbounds i32* %tmp, i64 1
br i1 %tmp3, label %bb2, label %bb5
bb5: ; preds = %bb2
ret void
}