[PATCH] LLVM OpenMP CMake Overhaul

C Bergström cbergstrom at pathscale.com
Tue Jun 23 10:53:39 PDT 2015

On Wed, Jun 24, 2015 at 12:04 AM, Peyton, Jonathan L
<jonathan.l.peyton at intel.com> wrote:
> Chris,
>> I am strongly apposed to this change. This was designed specifically to ensure that each set of compilers can easily set their own specific flags. Do not remove this feature.
>>Why - who cares? fwiw - most cmake files are named CMakeLists.txt
>> .cmake is typically used for cmake files which are meant to be reused as "modules".
>> I am a bit annoyed with the history of the cmake build system. My original system worked and was removed without much justification "just because". Now it's being refactored again and it's unclear to me the driving motivation.
> Maybe I can help answer these questions with the aid of the email message below.  To get the LLVM OpenMP Runtime set as the default OpenMP runtime library linked when using the -fopenmp flag, we have to get our CMake build to use LLVM conventions/standards.  This meant putting everything in its own pseudo namespace Libomp, libomp_ , or LIBOMP_ and creating a real configuration step which checks for compiler flags, features, etc.  I've followed what libcxx and compiler-rt do by having a config-ix.cmake file which does all this.  Although the compiler flags approach I had worked for gcc,clang,icc,and msvc, this method of checking the compiler flags is more generic and allows any compiler that supports the flag to use it.  Also, there are at least three other methods of adding custom flags (-DLIBOMP_${LANG}FLAGS=' ... ' , -DCMAKE_${LANG}_FLAGS=' ... ' , or good ole CFLAGS=' ... ' envirable).  The ultimate goal here is to get the LLVM OpenMP Runtime set as the default, so I have to do whatever it takes to achieve that.  If I get the ok to keep the old compiler flag method, I will.

We may have conflicting goals or requirements then. I think it's a
great thing to advance llvm, but I don't personally care what llvm
defines as a default for OMP runtime. Anyone building libomp which
isn't using it as part of llvm probably will feel the same way.

My original cmake, which was removed, exposed variables which allowed
the user to define the flags. Am I mistaken that this is now being
required again?

I think you made a mistake - this goes beyond removing unused stuff

I believe this removed a check for FreeBSD and thus setting it to be true.
I don't think this is correct - there are other assemblers on Win
which will handle this.

libomp_append(flags_local -std=c++0x

When will this flag die? That's legacy shit from gcc from a long time
ago.. it should be c++11, which the minimum version of gcc required to
build llvm supports.

looks like a giant mess. This is exactly the reason it was the way it
was before. Please work with others on the llvm side to get permission
to for some middle-ground approach.
Since you're spending all this time.. wouldn't those nasty perl
scripts get in the way as well? (When will those die)

            OUTPUT  ${filename}
${LIBOMP_TOOLS_DIR}/expand-vars.pl --strict ${LIBOMP_EVFLAGS}
                ${libomp_extra_evflags} ${file_dir}/${filename}.var ${filename}
            DEPENDS ${file_dir}/${filename}.var kmp_version.c

More information about the llvm-commits mailing list