[FuncSpec] Compute specialisation gain even when forcing specialisation

When rewriting the call sites to call the new specialised functions, a
single call site can be matched by two different specialisations - a
"less specialised" version of the function and a "more specialised"
version of the function, e.g.  for a function

    void f(int x, int y)

the call like `f(1, 2)` could be matched by either

    void f.1(int x /* int y == 2 */);

or

    void f.2(/* int x == 1, int y == 2 */);

The `FunctionSpecialisation` pass tries to match specialisation in the
order of decreasing gain, so "more specialised" functions are
preferred to "less specialised" functions. This breaks, however, when
using the flag `-force-function-specialization`, in which case the
cost/benefit analysis is not performed and all the specialisations are
equally preferable.

This patch makes the pass calculate specialisation gain and order the
specialisations accordingly even when `-force-function-specialization`
is used, under the assumption that this flag has purely debugging
purpose and it is reasonable to ignore the extra computing effort it
incurs.

Reviewed By: ChuanqiXu, labrinea

Differential Revision: https://reviews.llvm.org/D136180

diff --git a/llvm/lib/Transforms/IPO/FunctionSpecialization.cpp b/llvm/lib/Transforms/IPO/FunctionSpecialization.cpp
index 82110c7..4bf1688 100644 (file)
--- a/llvm/lib/Transforms/IPO/FunctionSpecialization.cpp
+++ b/llvm/lib/Transforms/IPO/FunctionSpecialization.cpp
@@ -446,16 +446,16 @@ private:
         SpecializationInfo &S = I.first->second;
 
         if (I.second)
-          S.Gain = ForceFunctionSpecialization ? 1 : 0 - Cost;
-        if (!ForceFunctionSpecialization)
-          S.Gain += getSpecializationBonus(&FormalArg, ActualArg);
+          S.Gain = 0 - Cost;
+        S.Gain += getSpecializationBonus(&FormalArg, ActualArg);
         S.Args.push_back({&FormalArg, ActualArg});
       }
     }
 
     // Remove unprofitable specializations.
-    Specializations.remove_if(
-        [](const auto &Entry) { return Entry.second.Gain <= 0; });
+    if (!ForceFunctionSpecialization)
+      Specializations.remove_if(
+          [](const auto &Entry) { return Entry.second.Gain <= 0; });
 
     // Clear the MapVector and return the underlying vector.
     WorkList = Specializations.takeVector();

diff --git a/llvm/test/Transforms/FunctionSpecialization/specialization-order.ll b/llvm/test/Transforms/FunctionSpecialization/specialization-order.ll
new file mode 100644 (file)
index 0000000..9dade3f
--- /dev/null
+++ b/llvm/test/Transforms/FunctionSpecialization/specialization-order.ll
@@ -0,0 +1,58 @@
+; RUN: opt -S --passes=function-specialization,deadargelim -force-function-specialization < %s | FileCheck %s
+define dso_local i32 @add(i32 %x, i32 %y) {
+entry:
+  %add = add nsw i32 %y, %x
+  ret i32 %add
+}
+
+define dso_local i32 @sub(i32 %x, i32 %y) {
+entry:
+  %sub = sub nsw i32 %x, %y
+  ret i32 %sub
+}
+
+define internal i32 @f(i32 %x, i32 %y, ptr %u, ptr %v) noinline {
+entry:
+  %call = tail call i32 %u(i32 %x, i32 %y)
+  %call1 = tail call i32 %v(i32 %x, i32 %y)
+  %mul = mul nsw i32 %call1, %call
+  ret i32 %mul
+}
+
+define dso_local i32 @g0(i32 %x, i32 %y) {
+; CHECK-LABEL: @g0
+; CHECK:       call i32 @f.2(i32 [[X:%.*]], i32 [[Y:%.*]])
+entry:
+  %call = tail call i32 @f(i32 %x, i32 %y, ptr @add, ptr @add)
+  ret i32 %call
+}
+
+
+define dso_local i32 @g1(i32 %x, i32 %y) {
+; CHECK-LABEL: @g1(
+; CHECK:       call i32 @f.1(i32 [[X:%.*]], i32 [[Y:%.*]])
+entry:
+  %call = tail call i32 @f(i32 %x, i32 %y, ptr @sub, ptr @add)
+  ret i32 %call
+}
+
+define dso_local i32 @g2(i32 %x, i32 %y, ptr %v) {
+; CHECK-LABEL @g2
+; CHECK       call i32 @f.3(i32 [[X:%.*]], i32 [[Y:%.*]], ptr [[V:%.*]])
+entry:
+  %call = tail call i32 @f(i32 %x, i32 %y, ptr @sub, ptr %v)
+  ret i32 %call
+}
+
+; CHECK-LABEL: define {{.*}} i32 @f.1
+; CHECK:       call i32 @sub(i32 %x, i32 %y)
+; CHECK-NEXT:  call i32 @add(i32 %x, i32 %y)
+
+; CHECK-LABEL: define {{.*}} i32 @f.2
+; CHECK:       call i32 @add(i32 %x, i32 %y)
+; CHECK-NEXT   call i32 @add(i32 %x, i32 %y)
+
+; CHECK-LABEL: define {{.*}} i32 @f.3
+; CHECK:       call i32 @sub(i32 %x, i32 %y)
+; CHECK-NEXT:  call i32 %v(i32 %x, i32 %y)
+