From c7b683c126b849dab5c81e7deecfc1e61f8563a0 Mon Sep 17 00:00:00 2001
From: Michael Liao <michael.hliao@gmail.com>
Date: Tue, 4 Aug 2020 13:56:36 -0400
Subject: [PATCH] [PGO][CUDA][HIP] Skip generating profile on the device stub
 and wrong-side functions.

- Skip generating profile data on `__global__` function in the host
  compilation. It's a host-side stub function only and don't have
  profile instrumentation generated on the real function body. The extra
  profile data results in the malformed instrumentation profile data.
- Skip generating region mapping on functions in the wrong-side, i.e.,
  + For the device compilation, skip host-only functions; and,
  + For the host compilation, skip device-only functions (including
    `__global__` functions.)
- As the device-side profiling is not ready yet, only host-side profile
  code generation is checked.

Differential Revision: https://reviews.llvm.org/D85276
---
 clang/lib/CodeGen/CodeGenPGO.cpp                   | 17 +++++++++++++++++
 clang/test/CodeGenCUDA/profile-coverage-mapping.cu | 20 ++++++++++++++++++++
 2 files changed, 37 insertions(+)
 create mode 100644 clang/test/CodeGenCUDA/profile-coverage-mapping.cu

diff --git a/clang/lib/CodeGen/CodeGenPGO.cpp b/clang/lib/CodeGen/CodeGenPGO.cpp
index e810f60..be3c50b 100644
--- a/clang/lib/CodeGen/CodeGenPGO.cpp
+++ b/clang/lib/CodeGen/CodeGenPGO.cpp
@@ -773,6 +773,11 @@ void CodeGenPGO::assignRegionCounters(GlobalDecl GD, llvm::Function *Fn) {
   if (!D->hasBody())
     return;
 
+  // Skip CUDA/HIP kernel launch stub functions.
+  if (CGM.getLangOpts().CUDA && !CGM.getLangOpts().CUDAIsDevice &&
+      D->hasAttr<CUDAGlobalAttr>())
+    return;
+
   bool InstrumentRegions = CGM.getCodeGenOpts().hasProfileClangInstr();
   llvm::IndexedInstrProfReader *PGOReader = CGM.getPGOReader();
   if (!InstrumentRegions && !PGOReader)
@@ -831,6 +836,18 @@ bool CodeGenPGO::skipRegionMappingForDecl(const Decl *D) {
   if (!D->getBody())
     return true;
 
+  // Skip host-only functions in the CUDA device compilation and device-only
+  // functions in the host compilation. Just roughly filter them out based on
+  // the function attributes. If there are effectively host-only or device-only
+  // ones, their coverage mapping may still be generated.
+  if (CGM.getLangOpts().CUDA &&
+      ((CGM.getLangOpts().CUDAIsDevice && !D->hasAttr<CUDADeviceAttr>() &&
+        !D->hasAttr<CUDAGlobalAttr>()) ||
+       (!CGM.getLangOpts().CUDAIsDevice &&
+        (D->hasAttr<CUDAGlobalAttr>() ||
+         (!D->hasAttr<CUDAHostAttr>() && D->hasAttr<CUDADeviceAttr>())))))
+    return true;
+
   // Don't map the functions in system headers.
   const auto &SM = CGM.getContext().getSourceManager();
   auto Loc = D->getBody()->getBeginLoc();
diff --git a/clang/test/CodeGenCUDA/profile-coverage-mapping.cu b/clang/test/CodeGenCUDA/profile-coverage-mapping.cu
new file mode 100644
index 0000000..5eae6f1
--- /dev/null
+++ b/clang/test/CodeGenCUDA/profile-coverage-mapping.cu
@@ -0,0 +1,20 @@
+// RUN: echo "GPU binary would be here" > %t
+// RUN: %clang_cc1 -fprofile-instrument=clang -triple x86_64-linux-gnu -target-sdk-version=8.0 -fcuda-include-gpubinary %t -emit-llvm -o - %s | FileCheck --check-prefix=PGOGEN %s
+// RUN: %clang_cc1 -fprofile-instrument=clang -fcoverage-mapping -triple x86_64-linux-gnu -target-sdk-version=8.0 -fcuda-include-gpubinary %t -emit-llvm -o - %s | FileCheck --check-prefix=COVMAP %s
+// RUN: %clang_cc1 -fprofile-instrument=clang -fcoverage-mapping -dump-coverage-mapping -triple x86_64-linux-gnu -target-sdk-version=8.0 -fcuda-include-gpubinary %t -emit-llvm-only -o - %s | FileCheck --check-prefix=MAPPING %s
+
+#include "Inputs/cuda.h"
+
+// PGOGEN-NOT: @__profn_{{.*kernel.*}} =
+// COVMAP-COUNT-2: section "__llvm_covfun", comdat
+// COVMAP-NOT: section "__llvm_covfun", comdat
+// MAPPING-NOT: {{.*dfn.*}}:
+// MAPPING-NOT: {{.*kernel.*}}:
+
+__device__ void dfn(int i) {}
+
+__global__ void kernel(int i) { dfn(i); }
+
+void host(void) {
+  kernel<<<1, 1>>>(1);
+}
-- 
2.7.4