Re-apply r110655 with fixes. Epilogue must restore sp from fp if the function stack frame has a var-sized object.
Also added a test case to check for the added benefit of this patch: it's optimizing away the unnecessary restore of sp from fp for some non-leaf functions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@110707 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/test/CodeGen/Thumb/large-stack.ll b/test/CodeGen/Thumb/large-stack.ll
index 02de36a..b05e6bf 100644
--- a/test/CodeGen/Thumb/large-stack.ll
+++ b/test/CodeGen/Thumb/large-stack.ll
@@ -1,20 +1,35 @@
-; RUN: llc < %s -march=thumb | grep {ldr.*LCP} | count 5
+; RUN: llc < %s -mtriple=thumb-apple-darwin | FileCheck %s
define void @test1() {
+; CHECK: test1:
+; CHECK: sub sp, #256
+; CHECK: add sp, #256
%tmp = alloca [ 64 x i32 ] , align 4
ret void
}
define void @test2() {
+; CHECK: test2:
+; CHECK: ldr r0, LCPI
+; CHECK: add sp, r0
+; CHECK: mov sp, r7
+; CHECK: sub sp, #4
%tmp = alloca [ 4168 x i8 ] , align 4
ret void
}
define i32 @test3() {
- %retval = alloca i32, align 4
- %tmp = alloca i32, align 4
- %a = alloca [805306369 x i8], align 16
- store i32 0, i32* %tmp
- %tmp1 = load i32* %tmp
- ret i32 %tmp1
+; CHECK: test3:
+; CHECK: ldr r1, LCPI
+; CHECK: add sp, r1
+; CHECK: ldr r1, LCPI
+; CHECK: add r1, sp
+; CHECK: mov sp, r7
+; CHECK: sub sp, #4
+ %retval = alloca i32, align 4
+ %tmp = alloca i32, align 4
+ %a = alloca [805306369 x i8], align 16
+ store i32 0, i32* %tmp
+ %tmp1 = load i32* %tmp
+ ret i32 %tmp1
}