4 files changed, 107 insertions, 0 deletions
diff --git a/llvm/test/CodeGen/ARM/call-tc.ll b/llvm/test/CodeGen/ARM/call-tc.ll
new file mode 100644
index 00000000000..8103fab2092
--- /dev/null
+++ b/llvm/test/CodeGen/ARM/call-tc.ll
@@ -0,0 +1,36 @@
+; RUN: llc < %s -march=arm | FileCheck %s -check-prefix=CHECKV4
+; RUN: llc < %s -march=arm -mattr=+v5t | FileCheck %s -check-prefix=CHECKV5
+; RUN: llc < %s -march=arm -mtriple=arm-linux-gnueabi\
+; RUN:   -relocation-model=pic | FileCheck %s -check-prefix=CHECKELF
+
+@t = weak global i32 ()* null           ; <i32 ()**> [#uses=1]
+
+declare void @g(i32, i32, i32, i32)
+
+define void @f() {
+; CHECKELF: PLT
+        call void @g( i32 1, i32 2, i32 3, i32 4 )
+        ret void
+}
+
+define void @g.upgrd.1() {
+; CHECKV4: bx r0 @ TAILCALL
+; CHECKV5: bx r0 @ TAILCALL
+        %tmp = load i32 ()** @t         ; <i32 ()*> [#uses=1]
+        %tmp.upgrd.2 = tail call i32 %tmp( )            ; <i32> [#uses=0]
+        ret void
+}
+
+define i32* @m_231b(i32, i32, i32*, i32*, i32*) nounwind {
+; CHECKV4: m_231b
+; CHECKV4: bx r{{.*}}
+BB0:
+  %5 = inttoptr i32 %0 to i32*                    ; <i32*> [#uses=1]
+  %t35 = volatile load i32* %5                    ; <i32> [#uses=1]
+  %6 = inttoptr i32 %t35 to i32**                 ; <i32**> [#uses=1]
+  %7 = getelementptr i32** %6, i32 86             ; <i32**> [#uses=1]
+  %8 = load i32** %7                              ; <i32*> [#uses=1]
+  %9 = bitcast i32* %8 to i32* (i32, i32*, i32, i32*, i32*, i32*)* ; <i32* (i32, i32*, i32, i32*, i32*, i32*)*> [#uses=1]
+  %10 = call i32* %9(i32 %0, i32* null, i32 %1, i32* %2, i32* %3, i32* %4) ; <i32*> [#uses=1]
+  ret i32* %10
+}
diff --git a/llvm/test/CodeGen/ARM/ifcvt6-tc.ll b/llvm/test/CodeGen/ARM/ifcvt6-tc.ll
new file mode 100644
index 00000000000..5b28804f380
--- /dev/null
+++ b/llvm/test/CodeGen/ARM/ifcvt6-tc.ll
@@ -0,0 +1,23 @@
+; RUN: llc < %s -march=arm -mtriple=arm-apple-darwin | \
+; RUN:   grep cmpne | count 1
+; RUN: llc < %s -march=arm -mtriple=arm-apple-darwin | \
+; RUN:   grep bhi | count 1
+; Here, tail call wins over eliminating branches.  It is 1 fewer instruction
+; and removes all stack accesses, so seems like a win.
+
+define void @foo(i32 %X, i32 %Y) {
+entry:
+	%tmp1 = icmp ult i32 %X, 4		; <i1> [#uses=1]
+	%tmp4 = icmp eq i32 %Y, 0		; <i1> [#uses=1]
+	%tmp7 = or i1 %tmp4, %tmp1		; <i1> [#uses=1]
+	br i1 %tmp7, label %cond_true, label %UnifiedReturnBlock
+
+cond_true:		; preds = %entry
+	%tmp10 = tail call i32 (...)* @bar( )		; <i32> [#uses=0]
+	ret void
+
+UnifiedReturnBlock:		; preds = %entry
+	ret void
+}
+
+declare i32 @bar(...)
diff --git a/llvm/test/CodeGen/ARM/insn-sched1-tc.ll b/llvm/test/CodeGen/ARM/insn-sched1-tc.ll
new file mode 100644
index 00000000000..c457c8c5a55
--- /dev/null
+++ b/llvm/test/CodeGen/ARM/insn-sched1-tc.ll
@@ -0,0 +1,11 @@
+; RUN: llc < %s -march=arm -mattr=+v6
+; RUN: llc < %s -mtriple=arm-apple-darwin -mattr=+v6 |\
+; RUN:   grep mov | count 2
+
+define i32 @test(i32 %x) {
+        %tmp = trunc i32 %x to i16              ; <i16> [#uses=1]
+        %tmp2 = tail call i32 @f( i32 1, i16 %tmp )             ; <i32> [#uses=1]
+        ret i32 %tmp2
+}
+
+declare i32 @f(i32, i16)
diff --git a/llvm/test/CodeGen/ARM/ldm-tc.ll b/llvm/test/CodeGen/ARM/ldm-tc.ll
new file mode 100644
index 00000000000..3819192429e
--- /dev/null
+++ b/llvm/test/CodeGen/ARM/ldm-tc.ll
@@ -0,0 +1,37 @@
+; RUN: llc < %s -mtriple=arm-apple-darwin | FileCheck %s
+
+@X = external global [0 x i32]          ; <[0 x i32]*> [#uses=5]
+
+define i32 @t1() {
+; CHECK: t1:
+; CHECK: ldmia
+        %tmp = load i32* getelementptr ([0 x i32]* @X, i32 0, i32 0)            ; <i32> [#uses=1]
+        %tmp3 = load i32* getelementptr ([0 x i32]* @X, i32 0, i32 1)           ; <i32> [#uses=1]
+        %tmp4 = tail call i32 @f1( i32 %tmp, i32 %tmp3 )                ; <i32> [#uses=1]
+        ret i32 %tmp4
+}
+
+define i32 @t2() {
+; CHECK: t2:
+; CHECK: ldmia
+        %tmp = load i32* getelementptr ([0 x i32]* @X, i32 0, i32 2)            ; <i32> [#uses=1]
+        %tmp3 = load i32* getelementptr ([0 x i32]* @X, i32 0, i32 3)           ; <i32> [#uses=1]
+        %tmp5 = load i32* getelementptr ([0 x i32]* @X, i32 0, i32 4)           ; <i32> [#uses=1]
+        %tmp6 = tail call i32 @f2( i32 %tmp, i32 %tmp3, i32 %tmp5 )             ; <i32> [#uses=1]
+        ret i32 %tmp6
+}
+
+define i32 @t3() {
+; CHECK: t3:
+; CHECK: ldmib
+; CHECK: b.w _f2 @ TAILCALL
+        %tmp = load i32* getelementptr ([0 x i32]* @X, i32 0, i32 1)            ; <i32> [#uses=1]
+        %tmp3 = load i32* getelementptr ([0 x i32]* @X, i32 0, i32 2)           ; <i32> [#uses=1]
+        %tmp5 = load i32* getelementptr ([0 x i32]* @X, i32 0, i32 3)           ; <i32> [#uses=1]
+        %tmp6 = tail call i32 @f2( i32 %tmp, i32 %tmp3, i32 %tmp5 )             ; <i32> [#uses=1]
+        ret i32 %tmp6
+}
+
+declare i32 @f1(i32, i32)
+
+declare i32 @f2(i32, i32, i32)