[Flang][Driver] Fix -foffload-device misspelling#201857
Merged
Merged
Conversation
jhuber6
approved these changes
Jun 5, 2026
🐧 Linux x64 Test Results
✅ The build succeeded and all tests passed. |
🪟 Windows x64 Test Results
✅ The build succeeded and all tests passed. |
Meinersbur
force-pushed
the
flang_driver_foffload-device
branch
from
8c1fd56 to
115eb15
Compare
llvmorg-github-actions
Bot
added
clang:driver
|
@llvm/pr-subscribers-flang-driver Author: Michael Kruse (Meinersbur) Changes#200863 added a new Also fix the condition and add a regression test for the driver. 2 Files Affected:
diff --git a/clang/lib/Driver/ToolChains/Flang.cpp b/clang/lib/Driver/ToolChains/Flang.cpp
index ae9ae8176e281..a7e92254d2768 100644
--- a/clang/lib/Driver/ToolChains/Flang.cpp
+++ b/clang/lib/Driver/ToolChains/Flang.cpp
@@ -692,8 +692,8 @@ void Flang::addOffloadOptions(Compilation &C, const InputInfoList &Inputs,
// Tell the frontend when it is compiling for an offloading device, regardless
// of offloading programming model.
- if (IsHostOffloadingAction)
- CmdArgs.push_back("-offload-device");
+ if (JA.getOffloadingDeviceKind() > Action::OFK_Host)
+ CmdArgs.push_back("-foffload-device");
// Skips the primary input file, which is the input file that the compilation
// proccess will be executed upon (e.g. the host bitcode file) and
diff --git a/flang/test/Driver/offload-device.f90 b/flang/test/Driver/offload-device.f90
new file mode 100644
index 0000000000000..2bab35ff93d5b
--- /dev/null
+++ b/flang/test/Driver/offload-device.f90
@@ -0,0 +1,24 @@
+! -foffload-device tells the frontend we are compiling for the auxiliary target,
+! i.e. not the host device. Test for CUDA and OpenMP offloading modes.
+
+! RUN: %flang -target aarch64-linux-gnu --no-offloadlib --offload-arch=sm_80 --offload-arch=gfx90a %s -fopenmp -### 2>&1 | FileCheck %s --check-prefixes=CHECK,OPENMP
+! RUN: %flang -target aarch64-linux-gnu --no-offloadlib --offload-arch=sm_80 -xcuda %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK,CUDA
+
+! Compiled as CUDA, device-compilation is done first
+! CUDA: flang{{(\.exe)?}}" "-fc1" "-triple" "nvptx64-nvidia-cuda"
+! CUDA-SAME: "-foffload-device"
+
+! Host invocation
+! CHECK: flang{{(\.exe)?}}" "-fc1" "-triple" "aarch64-unknown-linux-gnu"
+! CHECK-NOT: -foffload-device
+
+! Compiled as OpenMP, device-code is compiled after host-code compilation,
+! once for each --offload-arch argument
+! OPENMP: flang{{(\.exe)?}}" "-fc1" "-triple" "amdgcn-amd-amdhsa"
+! OPENMP-SAME: "-foffload-device"
+! OPENMP: flang{{(\.exe)?}}" "-fc1" "-triple" "nvptx64-nvidia-cuda"
+! OPENMP-SAME: "-foffload-device"
+
+
+module offload_device
+end module
|
|
@llvm/pr-subscribers-clang-driver Author: Michael Kruse (Meinersbur) Changes#200863 added a new Also fix the condition and add a regression test for the driver. 2 Files Affected:
diff --git a/clang/lib/Driver/ToolChains/Flang.cpp b/clang/lib/Driver/ToolChains/Flang.cpp
index ae9ae8176e281..a7e92254d2768 100644
--- a/clang/lib/Driver/ToolChains/Flang.cpp
+++ b/clang/lib/Driver/ToolChains/Flang.cpp
@@ -692,8 +692,8 @@ void Flang::addOffloadOptions(Compilation &C, const InputInfoList &Inputs,
// Tell the frontend when it is compiling for an offloading device, regardless
// of offloading programming model.
- if (IsHostOffloadingAction)
- CmdArgs.push_back("-offload-device");
+ if (JA.getOffloadingDeviceKind() > Action::OFK_Host)
+ CmdArgs.push_back("-foffload-device");
// Skips the primary input file, which is the input file that the compilation
// proccess will be executed upon (e.g. the host bitcode file) and
diff --git a/flang/test/Driver/offload-device.f90 b/flang/test/Driver/offload-device.f90
new file mode 100644
index 0000000000000..2bab35ff93d5b
--- /dev/null
+++ b/flang/test/Driver/offload-device.f90
@@ -0,0 +1,24 @@
+! -foffload-device tells the frontend we are compiling for the auxiliary target,
+! i.e. not the host device. Test for CUDA and OpenMP offloading modes.
+
+! RUN: %flang -target aarch64-linux-gnu --no-offloadlib --offload-arch=sm_80 --offload-arch=gfx90a %s -fopenmp -### 2>&1 | FileCheck %s --check-prefixes=CHECK,OPENMP
+! RUN: %flang -target aarch64-linux-gnu --no-offloadlib --offload-arch=sm_80 -xcuda %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK,CUDA
+
+! Compiled as CUDA, device-compilation is done first
+! CUDA: flang{{(\.exe)?}}" "-fc1" "-triple" "nvptx64-nvidia-cuda"
+! CUDA-SAME: "-foffload-device"
+
+! Host invocation
+! CHECK: flang{{(\.exe)?}}" "-fc1" "-triple" "aarch64-unknown-linux-gnu"
+! CHECK-NOT: -foffload-device
+
+! Compiled as OpenMP, device-code is compiled after host-code compilation,
+! once for each --offload-arch argument
+! OPENMP: flang{{(\.exe)?}}" "-fc1" "-triple" "amdgcn-amd-amdhsa"
+! OPENMP-SAME: "-foffload-device"
+! OPENMP: flang{{(\.exe)?}}" "-fc1" "-triple" "nvptx64-nvidia-cuda"
+! OPENMP-SAME: "-foffload-device"
+
+
+module offload_device
+end module
|
Meinersbur
added a commit
to ROCm/llvm-project
that referenced
this pull request
Jun 8, 2026
llvm#200863 added a new `-foffload-device` argument for informing the frontend that it compiling for the device-side (and as a consequence must not overwrite any module files compiled for the host), but the driver was mistakenly adding `-offload-device`. Also fix the condition and add a regression test for the driver.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
#200863 added a new
-foffload-deviceargument for informing the frontend that it compiling for the device-side (and as a consequence must not overwrite any module files compiled for the host), but the driver was mistakenly adding-offload-device.Also fix the condition and add a regression test for the driver.