First steps towards implementing execution par_unseq #3063
Conversation
hpx/parallel/util/loop.hpp
Outdated
| #elif defined(HPX_INTEL_VERSION) || defined(HPX_CLANG_VERSION) | ||
| #pragma ivdep | ||
| #pragma omp simd | ||
| #endif |
There was a problem hiding this comment.
I'd like to leave omp out of the picture here since we'd require users to compile with -fopenmp then.
There was a problem hiding this comment.
What about other compilers, should we issue a warning?
|
Would be nice to have some confidence if it really has the expected effects (better code gen) before pushing it further. |
I agree, but how can we get confidence without actually trying out things? |
|
> Would be nice to have some confidence if it really has the expected effects (better code gen) before pushing it further.
I agree, but how can we get confidence without actually trying out things?
Trying it out doesn't mean it has be merged ;)
|
hkaiser
force-pushed
the
par_unseq
branch
2 times, most recently
from
85ae248
to
5409978
Compare
SummaryYou can't use OpenMP +#if defined(HPX_HAVE_OPENMP_SIMD)
+#pragma omp simd
+#endif
+ for (/**/; it != end; ++it)
+ {
+ f(it);
+ }OpenMP 5 is expected to fix this. I don't know the precise status of that proposal but can look if you want. I recall that Intel 18 compilers do not object to I would not expect a compiler to attempt to vectorize container accesses if a random access iterator is not supported... DetailsThe relevant portions of the OpenMP 4.5 specification are:
|
|
@jeffhammond thanks for this information. This branch is really meant to be used for experimentation towards implementing |
| find_package(OpenMP QUIET) | ||
| if(OPENMP_FOUND) | ||
| if("${CMAKE_CXX_COMPILER_ID}" STREQUAL "Intel") | ||
| hpx_add_compile_flag_if_available(-openmp-simd) |
There was a problem hiding this comment.
-openmp-simd is deprecated (https://software.intel.com/en-us/node/693432). It was replaced by -qopenmp-simd in version 16 or 17.
I recommend that CMake test for -fopenmp-simd, then -qopenmp-simd then -openmp-simd. ICC supports many of the GCC flags for compatibility...
| @@ -253,9 +253,9 @@ namespace hpx { namespace parallel { namespace util | |||
| HPX_FORCEINLINE void prefetch_addresses(T const& ... ts) | |||
| { | |||
| int const sequencer[] = { | |||
| (_mm_prefetch( | |||
| 0, (_mm_prefetch( | |||
There was a problem hiding this comment.
For the non-x86 code path, you may want to use __builtin_prefetch, which both GCC and Clang support.
There was a problem hiding this comment.
We can do that, MSVC however supports _mm_prefetch only.
|
@hkaiser These are important experiments, in part because the OpenMP standard working group recognizes that support for C++ in OpenMP is lacking. While 5.0 is pretty close to finished, your experience in implementing PSTL will be useful in identifying gaps that should be addressed in the next iteration. I don't know what code you allow yourself to look at (due to licensing), but both RAJA and Intel PSTL may be useful. In particular, RAJA supports a bunch of different pragmas for persuading compilers to vectorize inner loops. |
|
Closing this since it's not actively being worked on, but if someone feels like picking this up again, feel free to do so! |
This is related to #2271