[clang] [flang] [Flang-new][OpenMP] Add offload related flags for AMDGPU (PR #96742)
Dominik Adamski via cfe-commits
cfe-commits at lists.llvm.org
Mon Jul 1 03:51:51 PDT 2024
https://github.com/DominikAdamski updated https://github.com/llvm/llvm-project/pull/96742
>From 5b487aac3c8414b6f37f6888f361ca7488094048 Mon Sep 17 00:00:00 2001
From: Dominik Adamski <dominik.adamski at amd.com>
Date: Fri, 21 Jun 2024 18:03:53 +0200
Subject: [PATCH 1/2] [Flang-new][OpenMP] Add offload related flags for AMDGPU
Flang-new needs to add mlink-builtin-bitcode objects
to properly support offload code generation for AMD GPU.
fcuda-is-device flag is not used by Flang currently.
In the future it will be needed for Flang equivalent function:
AMDGPUTargetCodeGenInfo::getGlobalVarAddressSpace.
---
clang/include/clang/Driver/Options.td | 4 +-
clang/lib/Driver/ToolChains/Flang.cpp | 3 ++
flang/test/Driver/omp-driver-offload.f90 | 58 +++++++++++++----------
flang/test/Driver/target-cpu-features.f90 | 4 +-
flang/test/Driver/target-gpu-features.f90 | 2 +-
5 files changed, 41 insertions(+), 30 deletions(-)
diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td
index dd55838dcf384..612d5793232ce 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -8016,7 +8016,7 @@ def source_date_epoch : Separate<["-"], "source-date-epoch">,
// CUDA Options
//===----------------------------------------------------------------------===//
-let Visibility = [CC1Option] in {
+let Visibility = [CC1Option, FC1Option] in {
def fcuda_is_device : Flag<["-"], "fcuda-is-device">,
HelpText<"Generate code for CUDA device">,
@@ -8031,7 +8031,7 @@ def fno_cuda_host_device_constexpr : Flag<["-"], "fno-cuda-host-device-constexpr
HelpText<"Don't treat unattributed constexpr functions as __host__ __device__.">,
MarshallingInfoNegativeFlag<LangOpts<"CUDAHostDeviceConstexpr">>;
-} // let Visibility = [CC1Option]
+} // let Visibility = [CC1Option, FC1Option]
//===----------------------------------------------------------------------===//
// OpenMP Options
diff --git a/clang/lib/Driver/ToolChains/Flang.cpp b/clang/lib/Driver/ToolChains/Flang.cpp
index 42b45dba2bd31..2679f284c5016 100644
--- a/clang/lib/Driver/ToolChains/Flang.cpp
+++ b/clang/lib/Driver/ToolChains/Flang.cpp
@@ -333,6 +333,9 @@ void Flang::AddAMDGPUTargetArgs(const ArgList &Args,
StringRef Val = A->getValue();
CmdArgs.push_back(Args.MakeArgString("-mcode-object-version=" + Val));
}
+
+ const ToolChain &TC = getToolChain();
+ TC.addClangTargetOptions(Args, CmdArgs, Action::OffloadKind::OFK_OpenMP);
}
void Flang::addTargetOptions(const ArgList &Args,
diff --git a/flang/test/Driver/omp-driver-offload.f90 b/flang/test/Driver/omp-driver-offload.f90
index 6fb4f4eeeeca1..da81a6ee3ba8f 100644
--- a/flang/test/Driver/omp-driver-offload.f90
+++ b/flang/test/Driver/omp-driver-offload.f90
@@ -14,12 +14,12 @@
! Test regular -fopenmp with offload, and invocation filtering options
! RUN: %flang -S -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a --offload-arch=sm_70 \
-! RUN: --target=aarch64-unknown-linux-gnu \
+! RUN: --target=aarch64-unknown-linux-gnu -nogpulib\
! RUN: | FileCheck %s --check-prefix=OFFLOAD-HOST-AND-DEVICE
! RUN: %flang -S -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a --offload-arch=sm_70 --offload-host-device \
-! RUN: --target=aarch64-unknown-linux-gnu \
+! RUN: --target=aarch64-unknown-linux-gnu -nogpulib\
! RUN: | FileCheck %s --check-prefix=OFFLOAD-HOST-AND-DEVICE
! OFFLOAD-HOST-AND-DEVICE: "{{[^"]*}}flang-new" "-fc1" "-triple" "aarch64-unknown-linux-gnu"
@@ -29,7 +29,7 @@
! RUN: %flang -S -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a --offload-arch=sm_70 --offload-host-only \
-! RUN: --target=aarch64-unknown-linux-gnu \
+! RUN: --target=aarch64-unknown-linux-gnu -nogpulib\
! RUN: | FileCheck %s --check-prefix=OFFLOAD-HOST
! OFFLOAD-HOST: "{{[^"]*}}flang-new" "-fc1" "-triple" "aarch64-unknown-linux-gnu"
@@ -39,7 +39,7 @@
! RUN: %flang -S -### %s 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a --offload-arch=sm_70 --offload-device-only \
-! RUN: --target=aarch64-unknown-linux-gnu \
+! RUN: --target=aarch64-unknown-linux-gnu -nogpulib\
! RUN: | FileCheck %s --check-prefix=OFFLOAD-DEVICE
! OFFLOAD-DEVICE: "{{[^"]*}}flang-new" "-fc1" "-triple" "aarch64-unknown-linux-gnu"
@@ -48,13 +48,13 @@
! OFFLOAD-DEVICE-NOT: "{{[^"]*}}flang-new" "-fc1" "-triple" "aarch64-unknown-linux-gnu"
! Test regular -fopenmp with offload for basic fopenmp-is-target-device flag addition and correct fopenmp
-! RUN: %flang -### -fopenmp --offload-arch=gfx90a -fopenmp-targets=amdgcn-amd-amdhsa %s 2>&1 | FileCheck --check-prefixes=CHECK-OPENMP-IS-TARGET-DEVICE %s
+! RUN: %flang -### -fopenmp --offload-arch=gfx90a -fopenmp-targets=amdgcn-amd-amdhsa -nogpulib %s 2>&1 | FileCheck --check-prefixes=CHECK-OPENMP-IS-TARGET-DEVICE %s
! CHECK-OPENMP-IS-TARGET-DEVICE: "{{[^"]*}}flang-new" "-fc1" {{.*}} "-fopenmp" {{.*}} "-fopenmp-is-target-device" {{.*}}.f90"
! Testing appropriate flags are gnerated and appropriately assigned by the driver when offloading
! RUN: %flang -S -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a \
-! RUN: --target=aarch64-unknown-linux-gnu \
+! RUN: --target=aarch64-unknown-linux-gnu -nogpulib\
! RUN: | FileCheck %s --check-prefix=OPENMP-OFFLOAD-ARGS
! OPENMP-OFFLOAD-ARGS: "{{[^"]*}}flang-new" "-fc1" "-triple" "aarch64-unknown-linux-gnu" {{.*}} "-fopenmp" {{.*}}.f90"
! OPENMP-OFFLOAD-ARGS-NEXT: "{{[^"]*}}flang-new" "-fc1" "-triple" "amdgcn-amd-amdhsa"
@@ -70,23 +70,23 @@
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a \
! RUN: -fopenmp-targets=amdgcn-amd-amdhsa \
-! RUN: -fopenmp-assume-threads-oversubscription \
+! RUN: -fopenmp-assume-threads-oversubscription -nogpulib \
! RUN: | FileCheck %s --check-prefixes=CHECK-THREADS-OVS
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=sm_70 \
! RUN: -fopenmp-targets=nvptx64-nvidia-cuda \
-! RUN: -fopenmp-assume-threads-oversubscription \
+! RUN: -fopenmp-assume-threads-oversubscription -nogpulib \
! RUN: | FileCheck %s --check-prefixes=CHECK-THREADS-OVS
! CHECK-THREADS-OVS: "{{[^"]*}}flang-new" "-fc1" {{.*}} "-fopenmp" {{.*}} "-fopenmp-is-target-device" "-fopenmp-assume-threads-oversubscription" {{.*}}.f90"
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a \
! RUN: -fopenmp-targets=amdgcn-amd-amdhsa \
-! RUN: -fopenmp-assume-teams-oversubscription \
+! RUN: -fopenmp-assume-teams-oversubscription -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-TEAMS-OVS
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=sm_70 \
-! RUN: -fopenmp-targets=nvptx64-nvidia-cuda \
+! RUN: -fopenmp-targets=nvptx64-nvidia-cuda -nogpulib\
! RUN: -fopenmp-assume-teams-oversubscription \
! RUN: | FileCheck %s --check-prefixes=CHECK-TEAMS-OVS
! CHECK-TEAMS-OVS: "{{[^"]*}}flang-new" "-fc1" {{.*}} "-fopenmp" {{.*}} "-fopenmp-is-target-device" "-fopenmp-assume-teams-oversubscription" {{.*}}.f90"
@@ -94,48 +94,48 @@
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a \
! RUN: -fopenmp-targets=amdgcn-amd-amdhsa \
-! RUN: -fopenmp-assume-no-nested-parallelism \
+! RUN: -fopenmp-assume-no-nested-parallelism -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-NEST-PAR
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=sm_70 \
! RUN: -fopenmp-targets=nvptx64-nvidia-cuda \
-! RUN: -fopenmp-assume-no-nested-parallelism \
+! RUN: -fopenmp-assume-no-nested-parallelism -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-NEST-PAR
! CHECK-NEST-PAR: "{{[^"]*}}flang-new" "-fc1" {{.*}} "-fopenmp" {{.*}} "-fopenmp-is-target-device" "-fopenmp-assume-no-nested-parallelism" {{.*}}.f90"
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a \
! RUN: -fopenmp-targets=amdgcn-amd-amdhsa \
-! RUN: -fopenmp-assume-no-thread-state \
+! RUN: -fopenmp-assume-no-thread-state -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-THREAD-STATE
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=sm_70 \
! RUN: -fopenmp-targets=nvptx64-nvidia-cuda \
-! RUN: -fopenmp-assume-no-thread-state \
+! RUN: -fopenmp-assume-no-thread-state -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-THREAD-STATE
! CHECK-THREAD-STATE: "{{[^"]*}}flang-new" "-fc1" {{.*}} "-fopenmp" {{.*}} "-fopenmp-is-target-device" "-fopenmp-assume-no-thread-state" {{.*}}.f90"
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a \
! RUN: -fopenmp-targets=amdgcn-amd-amdhsa \
-! RUN: -fopenmp-target-debug \
+! RUN: -fopenmp-target-debug -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-TARGET-DEBUG
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=sm_70 \
! RUN: -fopenmp-targets=nvptx64-nvidia-cuda \
-! RUN: -fopenmp-target-debug \
+! RUN: -fopenmp-target-debug -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-TARGET-DEBUG
! CHECK-TARGET-DEBUG: "{{[^"]*}}flang-new" "-fc1" {{.*}} "-fopenmp" {{.*}} "-fopenmp-is-target-device" "-fopenmp-target-debug" {{.*}}.f90"
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a \
! RUN: -fopenmp-targets=amdgcn-amd-amdhsa \
-! RUN: -fopenmp-target-debug \
+! RUN: -fopenmp-target-debug -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-TARGET-DEBUG
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=sm_70 \
! RUN: -fopenmp-targets=nvptx64-nvidia-cuda \
-! RUN: -fopenmp-target-debug \
+! RUN: -fopenmp-target-debug -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-TARGET-DEBUG
! CHECK-TARGET-DEBUG-EQ: "{{[^"]*}}flang-new" "-fc1" {{.*}} "-fopenmp" {{.*}} "-fopenmp-is-target-device" "-fopenmp-target-debug=111" {{.*}}.f90"
@@ -144,14 +144,14 @@
! RUN: -fopenmp-targets=amdgcn-amd-amdhsa \
! RUN: -fopenmp-target-debug -fopenmp-assume-threads-oversubscription \
! RUN: -fopenmp-assume-teams-oversubscription -fopenmp-assume-no-nested-parallelism \
-! RUN: -fopenmp-assume-no-thread-state \
+! RUN: -fopenmp-assume-no-thread-state -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-RTL-ALL
! RUN: %flang -S -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=sm_70 \
! RUN: -fopenmp-targets=nvptx64-nvidia-cuda \
! RUN: -fopenmp-target-debug -fopenmp-assume-threads-oversubscription \
! RUN: -fopenmp-assume-teams-oversubscription -fopenmp-assume-no-nested-parallelism \
-! RUN: -fopenmp-assume-no-thread-state \
+! RUN: -fopenmp-assume-no-thread-state -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-RTL-ALL
! CHECK-RTL-ALL: "{{[^"]*}}flang-new" "-fc1" {{.*}} "-fopenmp" {{.*}} "-fopenmp-is-target-device" "-fopenmp-target-debug" "-fopenmp-assume-teams-oversubscription"
! CHECK-RTL-ALL: "-fopenmp-assume-threads-oversubscription" "-fopenmp-assume-no-thread-state" "-fopenmp-assume-no-nested-parallelism"
@@ -160,12 +160,12 @@
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=gfx90a \
! RUN: -fopenmp-targets=amdgcn-amd-amdhsa \
-! RUN: -fopenmp-version=45 \
+! RUN: -fopenmp-version=45 -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-OPENMP-VERSION
! RUN: %flang -### %s -o %t 2>&1 \
! RUN: -fopenmp --offload-arch=sm_70 \
! RUN: -fopenmp-targets=nvptx64-nvidia-cuda \
-! RUN: -fopenmp-version=45 \
+! RUN: -fopenmp-version=45 -nogpulib\
! RUN: | FileCheck %s --check-prefixes=CHECK-OPENMP-VERSION
! CHECK-OPENMP-VERSION: "{{[^"]*}}flang-new" "-fc1" {{.*}} "-fopenmp" "-fopenmp-version=45" {{.*}}.f90"
@@ -190,7 +190,7 @@
! NO-LIBC-GPU-NVPTX-NOT: "-lcgpu-nvptx"
! RUN: %flang -### --target=x86_64-unknown-linux-gnu -fopenmp \
-! RUN: --offload-arch=gfx90a \
+! RUN: --offload-arch=gfx900 --rocm-path=%S/Inputs/rocm \
! RUN: -gpulibc %s 2>&1 \
! RUN: | FileCheck --check-prefix=LIBC-GPU-AMDGPU %s
! LIBC-GPU-AMDGPU-DAG: "-lcgpu-amdgpu"
@@ -198,7 +198,7 @@
! RUN: %flang -### --target=x86_64-unknown-linux-gnu -fopenmp \
! RUN: --offload-arch=gfx90a \
-! RUN: -nogpulibc %s 2>&1 \
+! RUN: -nogpulibc -nogpulib %s 2>&1 \
! RUN: | FileCheck --check-prefix=NO-LIBC-GPU-AMDGPU %s
! NO-LIBC-GPU-AMDGPU-NOT: "-lcgpu-amdgpu"
@@ -220,10 +220,18 @@
! Test -fopenmp-force-usm option with offload
! RUN: %flang -S -### %s -o %t 2>&1 \
! RUN: -fopenmp -fopenmp-force-usm --offload-arch=gfx90a \
-! RUN: --target=aarch64-unknown-linux-gnu \
+! RUN: --target=aarch64-unknown-linux-gnu -nogpulib\
! RUN: | FileCheck %s --check-prefix=FORCE-USM-OFFLOAD
! FORCE-USM-OFFLOAD: "{{[^"]*}}flang-new" "-fc1" "-triple" "aarch64-unknown-linux-gnu"
! FORCE-USM-OFFLOAD-SAME: "-fopenmp" "-fopenmp-force-usm"
! FORCE-USM-OFFLOAD-NEXT: "{{[^"]*}}flang-new" "-fc1" "-triple" "amdgcn-amd-amdhsa"
! FORCE-USM-OFFLOAD-SAME: "-fopenmp" "-fopenmp-force-usm"
+
+! RUN: %flang -### -v --target=x86_64-unknown-linux-gnu -fopenmp \
+! RUN: --offload-arch=gfx900 \
+! RUN: --rocm-path=%S/Inputs/rocm %s 2>&1 \
+! RUN: | FileCheck --check-prefix=MLINK-BUILTIN-BITCODE %s
+! MLINK-BUILTIN-BITCODE: "{{[^"]*}}flang-new" "-fc1" "-triple" "amdgcn-amd-amdhsa"
+! MLINK-BUILTIN-BITCODE-SAME: "-fcuda-is-device"
+! MLINK-BUILTIN-BITCODE-SAME: "-mlink-builtin-bitcode" {{.*Inputs.*rocm.*amdgcn.*bitcode.*}}oclc_isa_version_900.bc
diff --git a/flang/test/Driver/target-cpu-features.f90 b/flang/test/Driver/target-cpu-features.f90
index eea7a0f665b34..215d323ba2038 100644
--- a/flang/test/Driver/target-cpu-features.f90
+++ b/flang/test/Driver/target-cpu-features.f90
@@ -23,10 +23,10 @@
! RUN: %flang --target=riscv64-linux-gnu -c %s -### 2>&1 \
! RUN: | FileCheck %s -check-prefix=CHECK-RV64
-! RUN: %flang --target=amdgcn-amd-amdhsa -mcpu=gfx908 -c %s -### 2>&1 \
+! RUN: %flang --target=amdgcn-amd-amdhsa -mcpu=gfx908 -nogpulib -c %s -### 2>&1 \
! RUN: | FileCheck %s -check-prefix=CHECK-AMDGPU
-! RUN: %flang --target=r600-unknown-unknown -mcpu=cayman -c %s -### 2>&1 \
+! RUN: %flang --target=r600-unknown-unknown -mcpu=cayman -nogpulib -c %s -### 2>&1 \
! RUN: | FileCheck %s -check-prefix=CHECK-AMDGPU-R600
! CHECK-A57: "-fc1" "-triple" "aarch64-unknown-linux-gnu"
diff --git a/flang/test/Driver/target-gpu-features.f90 b/flang/test/Driver/target-gpu-features.f90
index 9cc9ce4baaf4d..b783574370a0f 100644
--- a/flang/test/Driver/target-gpu-features.f90
+++ b/flang/test/Driver/target-gpu-features.f90
@@ -3,7 +3,7 @@
! Test that -mcpu are used and that the -target-cpu and -target-features
! are also added to the fc1 command.
-! RUN: %flang --target=amdgcn-amd-amdhsa -mcpu=gfx902 -c %s -### 2>&1 \
+! RUN: %flang --target=amdgcn-amd-amdhsa -mcpu=gfx902 -nogpulib -c %s -### 2>&1 \
! RUN: | FileCheck %s -check-prefix=CHECK-AMDGCN
! CHECK-AMDGCN: "-fc1" "-triple" "amdgcn-amd-amdhsa"
>From 8252dbee79e47a3dbf9d72c09621e73facd9e5ca Mon Sep 17 00:00:00 2001
From: Dominik Adamski <dominik.adamski at amd.com>
Date: Mon, 1 Jul 2024 05:45:46 -0500
Subject: [PATCH 2/2] Fixes after #96909
---
clang/include/clang/Driver/Options.td | 4 ++--
flang/test/Driver/omp-driver-offload.f90 | 1 -
2 files changed, 2 insertions(+), 3 deletions(-)
diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td
index b0538b3007944..1c2b8cfeef6ce 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -8024,7 +8024,7 @@ def source_date_epoch : Separate<["-"], "source-date-epoch">,
// CUDA Options
//===----------------------------------------------------------------------===//
-let Visibility = [CC1Option, FC1Option] in {
+let Visibility = [CC1Option] in {
def fcuda_is_device : Flag<["-"], "fcuda-is-device">,
HelpText<"Generate code for CUDA device">,
@@ -8039,7 +8039,7 @@ def fno_cuda_host_device_constexpr : Flag<["-"], "fno-cuda-host-device-constexpr
HelpText<"Don't treat unattributed constexpr functions as __host__ __device__.">,
MarshallingInfoNegativeFlag<LangOpts<"CUDAHostDeviceConstexpr">>;
-} // let Visibility = [CC1Option, FC1Option]
+} // let Visibility = [CC1Option]
//===----------------------------------------------------------------------===//
// OpenMP Options
diff --git a/flang/test/Driver/omp-driver-offload.f90 b/flang/test/Driver/omp-driver-offload.f90
index da81a6ee3ba8f..8ce60b53fbcdb 100644
--- a/flang/test/Driver/omp-driver-offload.f90
+++ b/flang/test/Driver/omp-driver-offload.f90
@@ -233,5 +233,4 @@
! RUN: --rocm-path=%S/Inputs/rocm %s 2>&1 \
! RUN: | FileCheck --check-prefix=MLINK-BUILTIN-BITCODE %s
! MLINK-BUILTIN-BITCODE: "{{[^"]*}}flang-new" "-fc1" "-triple" "amdgcn-amd-amdhsa"
-! MLINK-BUILTIN-BITCODE-SAME: "-fcuda-is-device"
! MLINK-BUILTIN-BITCODE-SAME: "-mlink-builtin-bitcode" {{.*Inputs.*rocm.*amdgcn.*bitcode.*}}oclc_isa_version_900.bc
More information about the cfe-commits
mailing list