In D139287#4031473, @jhuber6 wrote:

In D139287#4031469, @hbae wrote:

Looks like GCC 7.5 cannot build LLVM after this change. Could you please take a look?

In file included from /localdisk/hbae/LLVM/llvm-base/openmp/libomptarget/plugins-nextgen/common/PluginInterface/PluginInterface.cpp:11:0:
/localdisk/hbae/LLVM/llvm-base/openmp/libomptarget/plugins-nextgen/common/PluginInterface/PluginInterface.h: In member function ‘virtual llvm::Expected<std::unique_ptr<llvm::MemoryBuffer> > llvm::omp::target::plugin::GenericDeviceTy::doJITPostProcessing(std::unique_ptr<llvm::MemoryBuffer>) const’:
/localdisk/hbae/LLVM/llvm-base/openmp/libomptarget/plugins-nextgen/common/PluginInterface/PluginInterface.h:389:12: error: could not convert ‘MB’ from ‘std::unique_ptr<llvm::MemoryBuffer>’ to ‘llvm::Expected<std::unique_ptr<llvm::MemoryBuffer> >’
     return MB;
            ^~

Some older GCC's have problem with the implicit move on copy elision AFAIK. I'll add a std::move and let me know if that fixes it.

Looks like GCC 7.5 cannot build LLVM after this change. Could you please take a look?

LGTM

LGTM

Applied suggested changes.

Is it possible for me to access the output of the failed test shown in the log?
/b/1/openmp-clang-x86_64-linux-debian/llvm.build/projects/openmp/runtime/test/ompt/worksharing/for/Output/loop_dispatch.c.tmp.out

Added missing REQUIRES line.

The change looks good in general.
I tested basic usage of the ompd plugin, and it seems to work fine as expected.
I encountered python-related build issues on some of my test systems (e.g., Ubuntu 18.04) which I cannot resolve by installing required python packages, so I wonder if we need to make this (ompd test infrastructure) an optional feature for those who are interested.

LGTM

LGTM

LGTM

LGTM

LGTM

Applied suggested changes.

LGTM

LGTM

Looks good to me except for a couple of typos.

LGTM

LGTM

LGTM

LGTM

LGTM

Undo the change that introduce __kmp_task_alloc_impl since the outcome may vary depending on the build compiler.

LGTM

LGTM

Fixed incorrect parameter type.

Ignore my last action (my mistake).

Added inline after static to avoid unused function warning.
Added a RUN line to the test to catch unused function warning.

Removed omp_is_initial_device() from clang built-ins.
Added a new test under openmp/libomptarget.

Applied suggested changes and added a test.

Added ident_t * parameter.

It is unclear what we should do with ident_t at the moment.
Is it going to be useful if we include the location info in the warning/error message?