From c47739b45cad6f075cf4c42420c2c47034ff75d4 Mon Sep 17 00:00:00 2001
From: Momchil Velikov <momchil.velikov@arm.com>
Date: Tue, 25 Oct 2022 18:56:31 +0100
Subject: [PATCH] [FuncSpec] Consider small noinline functions for
 specialisation

Small functions with size under a given threshold are not
considered for specialisaion on the presumption that they
are easy to inline. This does not apply to `noinline`
functions, though.

Reviewed By: ChuanqiXu

Differential Revision: https://reviews.llvm.org/D135862
---
 .../Transforms/IPO/FunctionSpecialization.cpp |  8 ++---
 .../FunctionSpecialization/noinline.ll        | 35 +++++++++++++++++++
 2 files changed, 38 insertions(+), 5 deletions(-)
 create mode 100644 llvm/test/Transforms/FunctionSpecialization/noinline.ll

diff --git a/llvm/lib/Transforms/IPO/FunctionSpecialization.cpp b/llvm/lib/Transforms/IPO/FunctionSpecialization.cpp
index 9884bdfe18cb..8063378f77b1 100644
--- a/llvm/lib/Transforms/IPO/FunctionSpecialization.cpp
+++ b/llvm/lib/Transforms/IPO/FunctionSpecialization.cpp
@@ -552,11 +552,9 @@ private:
     // inlined so that we shouldn't specialize it.
     if (Metrics.notDuplicatable || !Metrics.NumInsts.isValid() ||
         (!ForceFunctionSpecialization &&
-         Metrics.NumInsts < SmallFunctionThreshold)) {
-      InstructionCost C{};
-      C.setInvalid();
-      return C;
-    }
+         !F->hasFnAttribute(Attribute::NoInline) &&
+         Metrics.NumInsts < SmallFunctionThreshold))
+      return InstructionCost::getInvalid();
 
     // Otherwise, set the specialization cost to be the cost of all the
     // instructions in the function and penalty for specializing more functions.
diff --git a/llvm/test/Transforms/FunctionSpecialization/noinline.ll b/llvm/test/Transforms/FunctionSpecialization/noinline.ll
new file mode 100644
index 000000000000..6c566fcb5738
--- /dev/null
+++ b/llvm/test/Transforms/FunctionSpecialization/noinline.ll
@@ -0,0 +1,35 @@
+; RUN: opt -S --passes=function-specialization < %s | FileCheck %s
+define dso_local i32 @p0(i32 noundef %x) {
+entry:
+  %add = add nsw i32 %x, 1
+  ret i32 %add
+}
+
+define dso_local i32 @p1(i32 noundef %x) {
+entry:
+  %sub = add nsw i32 %x, -1
+  ret i32 %sub
+}
+
+define internal fastcc i32 @f(i32 noundef %x, ptr nocapture noundef readonly %p) noinline {
+entry:
+  %call = tail call i32 %p(i32 noundef %x)
+  %add = add nsw i32 %call, %x
+  ret i32 %add
+}
+
+define dso_local i32 @g0(i32 noundef %x) {
+entry:
+  %call = tail call fastcc i32 @f(i32 noundef %x, ptr noundef nonnull @p0)
+  ret i32 %call
+}
+
+define dso_local i32 @g1(i32 noundef %x) {
+entry:
+  %call = tail call fastcc i32 @f(i32 noundef %x, ptr noundef nonnull @p1)
+  ret i32 %call
+}
+
+; Check that a noinline function is specialized, even if it's small.
+; CHECK: @f.1
+; CHECK: @f.2