AArch64: add initial NEON support
Patch by Ana Pazos.
- Completed implementation of instruction formats:
AdvSIMD three same
AdvSIMD modified immediate
AdvSIMD scalar pairwise
- Completed implementation of instruction classes
(some of the instructions in these classes
belong to yet unfinished instruction formats):
Vector Arithmetic
Vector Immediate
Vector Pairwise Arithmetic
- Initial implementation of instruction formats:
AdvSIMD scalar two-reg misc
AdvSIMD scalar three same
- Intial implementation of instruction class:
Scalar Arithmetic
- Initial clang changes to support arm v8 intrinsics.
Note: no clang changes for scalar intrinsics function name mangling yet.
- Comprehensive test cases for added instructions
To verify auto codegen, encoding, decoding, diagnosis, intrinsics.
llvm-svn: 187567
diff --git a/llvm/test/CodeGen/AArch64/complex-copy-noneon.ll b/llvm/test/CodeGen/AArch64/complex-copy-noneon.ll
new file mode 100644
index 0000000..4ae5478
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/complex-copy-noneon.ll
@@ -0,0 +1,21 @@
+; RUN: llc -mtriple=aarch64-none-linux-gnu -mattr=-neon < %s
+
+; The DAG combiner decided to use a vector load/store for this struct copy
+; previously. This probably shouldn't happen without NEON, but the most
+; important thing is that it compiles.
+
+define void @store_combine() nounwind {
+ %src = alloca { double, double }, align 8
+ %dst = alloca { double, double }, align 8
+
+ %src.realp = getelementptr inbounds { double, double }* %src, i32 0, i32 0
+ %src.real = load double* %src.realp
+ %src.imagp = getelementptr inbounds { double, double }* %src, i32 0, i32 1
+ %src.imag = load double* %src.imagp
+
+ %dst.realp = getelementptr inbounds { double, double }* %dst, i32 0, i32 0
+ %dst.imagp = getelementptr inbounds { double, double }* %dst, i32 0, i32 1
+ store double %src.real, double* %dst.realp
+ store double %src.imag, double* %dst.imagp
+ ret void
+}