commit cd5b657c39a348bd7652a19c5be5035e9df44a42
author Tobias Grosser <grosser@google.com> Mon Jul 01 15:04:07 2013 -0700
committer Tobias Grosser <grosser@google.com> Mon Jul 01 15:52:52 2013 -0700
tree f6fa9e12b0472890b9cb156f7cf83b44bfe5e91b
parent e7ec64eef990a22d765b624dd9225bf3d625a9a0

ForEach: Set internal linkage for kernel functions

The expanded function is the external visible interface. Setting internal
linkage will influence the cost function of the inliner. Specifically, the
inliner knowns that no external users will use the unexpanded kernel function,
which means it will inline it and remove the original function.

This gives a 22% speedup on the colorcube 3D, 19% on convolve3x3
as well as minor speedups on fisheye.

Change-Id: I9770c92d24dfc7978145e694d1a35829c5432264

lib/Renderscript/RSForEachExpand.cpp

diff --git a/lib/Renderscript/RSForEachExpand.cpp b/lib/Renderscript/RSForEachExpand.cpp
index 59c3725..93b5672 100644
--- a/lib/Renderscript/RSForEachExpand.cpp
+++ b/lib/Renderscript/RSForEachExpand.cpp
@@ -27,6 +27,7 @@
 #include <llvm/Pass.h>
 #include <llvm/Support/raw_ostream.h>
 #include <llvm/IR/DataLayout.h>
+#include <llvm/IR/Function.h>
 #include <llvm/IR/Type.h>
 #include <llvm/Transforms/Utils/BasicBlockUtils.h>
 
@@ -606,11 +607,15 @@
       const char *name = func_iter->first;
       uint32_t signature = func_iter->second;
       llvm::Function *kernel = M.getFunction(name);
-      if (kernel && isKernel(signature)) {
-        Changed |= ExpandKernel(kernel, signature);
-      }
-      else if (kernel && kernel->getReturnType()->isVoidTy()) {
-        Changed |= ExpandFunction(kernel, signature);
+      if (kernel) {
+        if (isKernel(signature))
+          Changed |= ExpandKernel(kernel, signature);
+        else if (kernel->getReturnType()->isVoidTy())
+          Changed |= ExpandFunction(kernel, signature);
+        else
+          llvm_unreachable("Unknown kernel type");
+
+        kernel->setLinkage(llvm::GlobalValue::InternalLinkage);
       }
     }