https://github.com/jhuber6 updated
https://github.com/llvm/llvm-project/pull/68264
>From 52d11d0d975f03a3cd2760b42236b5ec8097331b Mon Sep 17 00:00:00 2001
From: Joseph Huber <jhub...@vols.utk.edu>
Date: Wed, 4 Oct 2023 16:50:20 -0500
Subject: [PATCH 1/2] [OpenMP] Prevent AMDGPU from overriding visibility on
DT_nohost variables
Summary:
There's some logic in the AMDGPU target that manually resets the
requested visibility of certain variables. This was triggering when we
set a constant variable in OpenMP. However, we shouldn't do this for
OpenMP when the variable has the `nohost` type. That implies that the
variable is not visible to the host and therefore does not need to be
visible, so we should respect the original value of it.
---
clang/lib/CodeGen/Targets/AMDGPU.cpp | 4 ++++
clang/test/OpenMP/target_visibility.cpp | 6 +++++-
2 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/clang/lib/CodeGen/Targets/AMDGPU.cpp
b/clang/lib/CodeGen/Targets/AMDGPU.cpp
index dc628b7345f59fd..70f906a42e6d9c4 100644
--- a/clang/lib/CodeGen/Targets/AMDGPU.cpp
+++ b/clang/lib/CodeGen/Targets/AMDGPU.cpp
@@ -308,6 +308,10 @@ static bool requiresAMDGPUProtectedVisibility(const Decl
*D,
if (GV->getVisibility() != llvm::GlobalValue::HiddenVisibility)
return false;
+ if (auto *Attr = D->getAttr<OMPDeclareTargetDeclAttr>())
+ if (OMPDeclareTargetDeclAttr::DT_NoHost == Attr->getDevType())
+ return false;
+
return D->hasAttr<OpenCLKernelAttr>() ||
(isa<FunctionDecl>(D) && D->hasAttr<CUDAGlobalAttr>()) ||
(isa<VarDecl>(D) &&
diff --git a/clang/test/OpenMP/target_visibility.cpp
b/clang/test/OpenMP/target_visibility.cpp
index 938d164df89bffb..2ed90d4b8ba9a77 100644
--- a/clang/test/OpenMP/target_visibility.cpp
+++ b/clang/test/OpenMP/target_visibility.cpp
@@ -1,5 +1,5 @@
// RUN: %clang_cc1 -debug-info-kind=limited -verify -fopenmp -x c++ -triple
nvptx64-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm %s
-fopenmp-is-target-device -o - | FileCheck %s
-// RUN: %clang_cc1 -debug-info-kind=limited -verify -fopenmp -x c++ -triple
nvptx-unknown-unknown -fopenmp-targets=nvptx-nvidia-cuda -emit-llvm %s
-fopenmp-is-target-device -o - | FileCheck %s
+// RUN: %clang_cc1 -debug-info-kind=limited -verify -fopenmp -x c++ -triple
amdgcn-amd-amdhsa -fopenmp-targets=amdgcn-amd-amdhsa -emit-llvm %s
-fopenmp-is-target-device -o - | FileCheck %s
// expected-no-diagnostics
@@ -21,6 +21,10 @@ void B::bar() { A a; a.foo(); }
void B::sbar() { A::sfoo(); }
#pragma omp declare target to(B::bar, B::sbar)
+[[gnu::visibility("hidden")]] extern const int x = 0;
+#pragma omp declare target to(x) device_type(nohost)
+
+// CHECK-DAG: @x = hidden{{.*}} constant i32 0
// CHECK-DAG: define hidden void @_ZN1B4sbarEv()
// CHECK-DAG: define linkonce_odr hidden void @_ZN1A4sfooEv()
// CHECK-DAG: define hidden void @_ZN1B3barEv(
>From 0be9c9fc769eb580a602ce075d8904d683c3596b Mon Sep 17 00:00:00 2001
From: Joseph Huber <jhub...@vols.utk.edu>
Date: Thu, 5 Oct 2023 07:41:24 -0500
Subject: [PATCH 2/2] Refine logic
We need to change the visibility to protected for declare target
variables, this is because they need to be readable by the host runtime.
---
clang/lib/CodeGen/Targets/AMDGPU.cpp | 4 ++--
clang/test/OpenMP/target_visibility.cpp | 6 +++++-
2 files changed, 7 insertions(+), 3 deletions(-)
diff --git a/clang/lib/CodeGen/Targets/AMDGPU.cpp
b/clang/lib/CodeGen/Targets/AMDGPU.cpp
index 70f906a42e6d9c4..def8bf45af761c2 100644
--- a/clang/lib/CodeGen/Targets/AMDGPU.cpp
+++ b/clang/lib/CodeGen/Targets/AMDGPU.cpp
@@ -309,8 +309,8 @@ static bool requiresAMDGPUProtectedVisibility(const Decl *D,
return false;
if (auto *Attr = D->getAttr<OMPDeclareTargetDeclAttr>())
- if (OMPDeclareTargetDeclAttr::DT_NoHost == Attr->getDevType())
- return false;
+ return isa<VarDecl>(D) && OMPDeclareTargetDeclAttr::DT_NoHost !=
+ Attr->getDevType();
return D->hasAttr<OpenCLKernelAttr>() ||
(isa<FunctionDecl>(D) && D->hasAttr<CUDAGlobalAttr>()) ||
diff --git a/clang/test/OpenMP/target_visibility.cpp
b/clang/test/OpenMP/target_visibility.cpp
index 2ed90d4b8ba9a77..16c1e5aa6940bec 100644
--- a/clang/test/OpenMP/target_visibility.cpp
+++ b/clang/test/OpenMP/target_visibility.cpp
@@ -1,5 +1,5 @@
// RUN: %clang_cc1 -debug-info-kind=limited -verify -fopenmp -x c++ -triple
nvptx64-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm %s
-fopenmp-is-target-device -o - | FileCheck %s
-// RUN: %clang_cc1 -debug-info-kind=limited -verify -fopenmp -x c++ -triple
amdgcn-amd-amdhsa -fopenmp-targets=amdgcn-amd-amdhsa -emit-llvm %s
-fopenmp-is-target-device -o - | FileCheck %s
+// RUN: %clang_cc1 -debug-info-kind=limited -verify -fopenmp -x c++ -triple
amdgcn-amd-amdhsa -fopenmp-targets=amdgcn-amd-amdhsa -emit-llvm %s
-fopenmp-is-target-device -o - | FileCheck %s --check-prefixes=CHECK,AMDGPU
// expected-no-diagnostics
@@ -24,7 +24,11 @@ void B::sbar() { A::sfoo(); }
[[gnu::visibility("hidden")]] extern const int x = 0;
#pragma omp declare target to(x) device_type(nohost)
+[[gnu::visibility("hidden")]] int y = 0;
+#pragma omp declare target to(y)
+
// CHECK-DAG: @x = hidden{{.*}} constant i32 0
+// AMDGPU-DAG: @y = protected{{.*}} i32 0
// CHECK-DAG: define hidden void @_ZN1B4sbarEv()
// CHECK-DAG: define linkonce_odr hidden void @_ZN1A4sfooEv()
// CHECK-DAG: define hidden void @_ZN1B3barEv(
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
