[PowerPC] Fix PR17354: Generate nop after local calls for PIC code.

When generating code for shared libraries, even local calls may be
intercepted, so we need a nop after the call for the linker to fix up the
TOC.  Test case adapted from the one provided in PR17354.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@191440 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/PowerPC/PPCISelLowering.cpp b/lib/Target/PowerPC/PPCISelLowering.cpp
index 5e12664e728ca56ee5309e8611c08ebaa4c01810..8da5f0563c6a94445efa779da948f0b81651e9c2 100644 (file)
--- a/lib/Target/PowerPC/PPCISelLowering.cpp
+++ b/lib/Target/PowerPC/PPCISelLowering.cpp
@@ -3487,7 +3487,9 @@ PPCTargetLowering::FinishCall(CallingConv::ID CallConv, SDLoc dl,
       // from allocating it), resulting in an additional register being
       // allocated and an unnecessary move instruction being generated.
       needsTOCRestore = true;
-    } else if ((CallOpc == PPCISD::CALL) && !isLocalCall(Callee)) {
+    } else if ((CallOpc == PPCISD::CALL) &&
+               (!isLocalCall(Callee) ||
+                DAG.getTarget().getRelocationModel() == Reloc::PIC_)) {
       // Otherwise insert NOP for non-local calls.
       CallOpc = PPCISD::CALL_NOP;
     }

diff --git a/test/CodeGen/PowerPC/pr17354.ll b/test/CodeGen/PowerPC/pr17354.ll
new file mode 100644 (file)
index 0000000..dca81b1
--- /dev/null
+++ b/test/CodeGen/PowerPC/pr17354.ll
@@ -0,0 +1,39 @@
+; RUN: llc -mcpu=pwr7 -relocation-model=pic <%s | FileCheck %s
+
+; Test that PR17354 is fixed.  We must generate a nop following even
+; local calls when generating code for shared libraries, to permit
+; TOC fixup.
+
+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.CS = type { i32 }
+
+@_ZL3glb = internal global [1 x %struct.CS] zeroinitializer, align 4
+@llvm.global_ctors = appending global [1 x { i32, void ()* }] [{ i32, void ()* } { i32 65535, void ()* @_GLOBAL__I_a }]
+
+define internal void @__cxx_global_var_init() section ".text.startup" {
+entry:
+  call void @_Z4funcv(%struct.CS* sret getelementptr inbounds ([1 x %struct.CS]* @_ZL3glb, i64 0, i64 0))
+  ret void
+}
+
+; CHECK-LABEL: __cxx_global_var_init:
+; CHECK: bl _Z4funcv
+; CHECK-NEXT: nop
+
+; Function Attrs: nounwind
+define void @_Z4funcv(%struct.CS* noalias sret %agg.result) #0 {
+entry:
+  %a_ = getelementptr inbounds %struct.CS* %agg.result, i32 0, i32 0
+  store i32 0, i32* %a_, align 4
+  ret void
+}
+
+define internal void @_GLOBAL__I_a() section ".text.startup" {
+entry:
+  call void @__cxx_global_var_init()
+  ret void
+}
+
+attributes #0 = { nounwind "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }