Teach lib/VMCore/ConstantFold.cpp how to set the inbounds keyword and
how to fold notionally-out-of-bounds array getelementptr indices instead
of just doing these in lib/Analysis/ConstantFolding.cpp, because it can
be done in a fairly general way without TargetData, and because not all
constants are visited by lib/Analysis/ConstantFolding.cpp. This enables
more constant folding.
Also, set the "inbounds" flag when the getelementptr indices are
one-past-the-end.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@81483 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/test/Assembler/getelementptr.ll b/test/Assembler/getelementptr.ll
index 10e5011..803d6d3 100644
--- a/test/Assembler/getelementptr.ll
+++ b/test/Assembler/getelementptr.ll
@@ -1,11 +1,21 @@
-; RUN: llvm-as < %s | llvm-dis | llvm-as | llvm-dis
+; RUN: llvm-as < %s | llvm-dis | llvm-as | llvm-dis | FileCheck %s
+
+; Verify that over-indexed getelementptrs are folded.
+@A = external global [2 x [3 x [5 x [7 x i32]]]]
+@B = global i32* getelementptr ([2 x [3 x [5 x [7 x i32]]]]* @A, i64 0, i64 0, i64 2, i64 1, i64 7523)
+; CHECK: @B = global i32* getelementptr ([2 x [3 x [5 x [7 x i32]]]]* @A, i64 36, i64 0, i64 1, i64 0, i64 5) ; <i32**> [#uses=0]
+@C = global i32* getelementptr ([2 x [3 x [5 x [7 x i32]]]]* @A, i64 3, i64 2, i64 0, i64 0, i64 7523)
+; CHECK: @C = global i32* getelementptr ([2 x [3 x [5 x [7 x i32]]]]* @A, i64 39, i64 1, i64 1, i64 4, i64 5) ; <i32**> [#uses=0]
;; Verify that i16 indices work.
@x = external global {i32, i32}
@y = global i32* getelementptr ({i32, i32}* @x, i16 42, i32 0)
+; CHECK: @y = global i32* getelementptr (%0* @x, i16 42, i32 0)
; see if i92 indices work too.
define i32 *@test({i32, i32}* %t, i92 %n) {
+; CHECK: @test
+; CHECK: %B = getelementptr %0* %t, i92 %n, i32 0
%B = getelementptr {i32, i32}* %t, i92 %n, i32 0
ret i32* %B
}
diff --git a/test/Transforms/InstCombine/constant-fold-gep.ll b/test/Transforms/InstCombine/constant-fold-gep.ll
index 953f204..281d3b3 100644
--- a/test/Transforms/InstCombine/constant-fold-gep.ll
+++ b/test/Transforms/InstCombine/constant-fold-gep.ll
@@ -44,7 +44,11 @@
store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 16), align 8
; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 2, i32 1, i64 2), align 8
store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 17), align 8
-; CHECK: store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 1, i64 0, i32 0, i64 0), align 8
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 1, i64 0, i32 0, i64 0), align 8
store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 18), align 8
+; CHECK: store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 2, i64 0, i32 0, i64 0), align 8
+ store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 36), align 8
+; CHECK: store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 1, i64 0, i32 0, i64 1), align 8
+ store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 19), align 8
ret void
}
diff --git a/test/Transforms/InstCombine/getelementptr.ll b/test/Transforms/InstCombine/getelementptr.ll
index 444a0ee..037190d 100644
--- a/test/Transforms/InstCombine/getelementptr.ll
+++ b/test/Transforms/InstCombine/getelementptr.ll
@@ -222,7 +222,7 @@
getelementptr (i32* @B, i64 2)
ret i1 %C
; CHECK: @test22
-; CHECK: icmp ult (i32* getelementptr (i32* @A, i64 1), i32* getelementptr (i32* @B, i64 2))
+; CHECK: icmp ult (i32* getelementptr inbounds (i32* @A, i64 1), i32* getelementptr (i32* @B, i64 2))
}
@@ -463,7 +463,7 @@
@A37 = external constant [1 x i8]
define i1 @test37() nounwind {
; CHECK: @test37
-; CHECK: ret i1 icmp eq (i8* getelementptr ([1 x i8]* @A37, i64 0, i64 1), i8* getelementptr ([1 x i8]* @A37, i64 1, i64 0))
+; CHECK: ret i1 true
%t = icmp eq i8* getelementptr ([1 x i8]* @A37, i64 0, i64 1),
getelementptr ([1 x i8]* @A37, i64 1, i64 0)
ret i1 %t