This patch addresses an ABI compatibility issue with empty aggregate

parameters.  Examples of these are:

  struct { } a;
  union { } b[256];
  int a[0];

An empty aggregate has an address, although dereferencing that address is
pointless.  When passed as a parameter, an empty aggregate does not consume
a protocol register, nor does it consume a doubleword in the parameter save
area.  Passing an empty aggregate by reference passes an address just as
for any other aggregate.  Returning an empty aggregate uses GPR3 as a hidden
address of the return value location, just as for any other aggregate.

The patch modifies PPCTargetLowering::LowerFormalArguments_64SVR4 and
PPCTargetLowering::LowerCall_64SVR4 to properly skip empty aggregate
parameters passed by value.  The handling of return values and by-reference
parameters was already correct.

Built on powerpc64-unknown-linux-gnu and tested with no new regressions.
A test case is included to test proper handling of empty aggregate
parameters on both sides of the function call protocol.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@167090 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 51 insertions, 0 deletions

diff --git a/test/CodeGen/PowerPC/emptystruct.ll b/test/CodeGen/PowerPC/emptystruct.ll
new file mode 100644
index 0000000..36b4abd
--- /dev/null
+++ b/test/CodeGen/PowerPC/emptystruct.ll
@@ -0,0 +1,51 @@
+; RUN: llc -mcpu=pwr7 -O0 < %s | FileCheck %s
+
+; This tests correct handling of empty aggregate parameters and return values.
+; An empty parameter passed by value does not consume a protocol register or
+; a parameter save area doubleword.  An empty parameter passed by reference
+; is treated as any other pointer parameter.  An empty aggregate return value 
+; is treated as any other aggregate return value, passed via address as a 
+; hidden parameter in GPR3.  In this example, GPR3 contains the return value
+; address, GPR4 contains the address of e2, and e1 and e3 are not passed or
+; received.
+
+target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64"
+target triple = "powerpc64-unknown-linux-gnu"
+
+%struct.empty = type {}
+
+define void @callee(%struct.empty* noalias sret %agg.result, %struct.empty* byval %a1, %struct.empty* %a2, %struct.empty* byval %a3) nounwind {
+entry:
+  %a2.addr = alloca %struct.empty*, align 8
+  store %struct.empty* %a2, %struct.empty** %a2.addr, align 8
+  %0 = load %struct.empty** %a2.addr, align 8
+  %1 = bitcast %struct.empty* %agg.result to i8*
+  %2 = bitcast %struct.empty* %0 to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* %2, i64 0, i32 1, i1 false)
+  ret void
+}
+
+; CHECK: callee:
+; CHECK: std 4,
+; CHECK: std 3,
+; CHECK-NOT: std 5,
+; CHECK-NOT: std 6,
+; CHECK: blr
+
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
+
+define void @caller(%struct.empty* noalias sret %agg.result) nounwind {
+entry:
+  %e1 = alloca %struct.empty, align 1
+  %e2 = alloca %struct.empty, align 1
+  %e3 = alloca %struct.empty, align 1
+  call void @callee(%struct.empty* sret %agg.result, %struct.empty* byval %e1, %struct.empty* %e2, %struct.empty* byval %e3)
+  ret void
+}
+
+; CHECK: caller:
+; CHECK: addi 4,
+; CHECK: std 3,
+; CHECK-NOT: std 5,
+; CHECK-NOT: std 6,
+; CHECK: bl callee