Pass -offload-lto instead of -lto for cuda/hip kernels and enable fai…

…led tests
intel · Jan 13, 2025 · 6c7fbf7 · 6c7fbf7
1 parent 1a7e2f0
commit 6c7fbf7
Show file tree

Hide file tree

Showing 5 changed files with 7 additions and 9 deletions.
diff --git a/clang/test/Driver/linker-wrapper.c b/clang/test/Driver/linker-wrapper.c
@@ -21,7 +21,7 @@ __attribute__((visibility("protected"), used)) int x;
 // RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run \
 // RUN:   --linker-path=/usr/bin/ld %t.o -o a.out 2>&1 | FileCheck %s --check-prefix=NVPTX-LINK
 
-// NVPTX-LINK: clang{{.*}} -o {{.*}}.img --target=nvptx64-nvidia-cuda -march=sm_70 -O2 -flto {{.*}}.o {{.*}}.o
+// NVPTX-LINK: clang{{.*}} -o {{.*}}.img --target=nvptx64-nvidia-cuda -march=sm_70 -O2 -foffload-lto {{.*}}.o {{.*}}.o
 
 // RUN: clang-offload-packager -o %t.out \
 // RUN:   --image=file=%t.elf.o,kind=openmp,triple=nvptx64-nvidia-cuda,arch=sm_70 \

diff --git a/clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp b/clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp
@@ -1533,7 +1533,6 @@ Expected<StringRef> clang(ArrayRef<StringRef> InputFiles, const ArgList &Args,
   };
 
   // Forward all of the `--offload-opt` and similar options to the device.
-  CmdArgs.push_back("-flto");
   for (auto &Arg : Args.filtered(OPT_offload_opt_eq_minus, OPT_mllvm))
     CmdArgs.append(
         {"-Xlinker",
@@ -1542,6 +1541,12 @@ Expected<StringRef> clang(ArrayRef<StringRef> InputFiles, const ArgList &Args,
   if (!Triple.isNVPTX() && !Triple.isSPIRV())
     CmdArgs.push_back("-Wl,--no-undefined");
 
+  if (Triple.isNVPTX() || Triple.isAMDGPU()) {
+    CmdArgs.push_back("-foffload-lto");
+  } else {
+    CmdArgs.push_back("-flto");
+  }
+
   if (IsSYCLKind && Triple.isNVPTX())
     CmdArgs.push_back("-S");
   for (StringRef InputFile : InputFiles)

diff --git a/sycl/test-e2e/NewOffloadDriver/multisource.cpp b/sycl/test-e2e/NewOffloadDriver/multisource.cpp
@@ -5,8 +5,6 @@
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
-// XFAIL: cuda
-// XFAIL-TRACKER: https://github.com/intel/llvm/issues/16413
 // Separate kernel sources and host code sources
 // Test with `--offload-new-driver`
 // RUN: %{build} --offload-new-driver -c -o %t.kernel.o -DINIT_KERNEL -DCALC_KERNEL

diff --git a/sycl/test-e2e/NewOffloadDriver/split-per-source-main.cpp b/sycl/test-e2e/NewOffloadDriver/split-per-source-main.cpp
@@ -1,9 +1,6 @@
 // RUN: %{build} -Wno-error=unused-command-line-argument -fsycl-device-code-split=per_source -I %S/Inputs -o %t.out %S/Inputs/split-per-source-second-file.cpp \
 // RUN: --offload-new-driver -fsycl-dead-args-optimization
 // RUN: %{run} %t.out
-//
-// XFAIL: cuda
-// XFAIL-TRACKER: https://github.com/intel/llvm/issues/16413
 
 #include "Inputs/split-per-source.h"
 

diff --git a/sycl/test-e2e/NewOffloadDriver/sycl-external-with-optional-features.cpp b/sycl/test-e2e/NewOffloadDriver/sycl-external-with-optional-features.cpp
@@ -3,8 +3,6 @@
 // RUN: %{build} -DSOURCE2 --offload-new-driver -c -o %t2.o
 // RUN: %clangxx -Wno-error=unused-command-line-argument -fsycl -fsycl-targets=%{sycl_triple} --offload-new-driver %t1.o %t2.o -o %t.exe
 // RUN: %{run} %t.exe
-// XFAIL: cuda
-// XFAIL-TRACKER: https://github.com/intel/llvm/issues/16413
 
 #ifdef SOURCE1
 #include <iostream>