[Openmp-commits] [openmp] r293094 - [OpenMP] Initial implementation of OpenMP offloading library - libomptarget.

[Hal Finkel via Openmp-commits]

Hi George,



On 01/25/2017 03:27 PM, George Rokos via Openmp-commits wrote:
> Author: grokos
> Date: Wed Jan 25 15:27:24 2017
> New Revision: 293094
>
> URL: http://llvm.org/viewvc/llvm-project?rev=293094&view=rev
> Log:
> [OpenMP] Initial implementation of OpenMP offloading library - libomptarget.
>
> This is the patch upstreaming the device-agnostic part of libomptarget.
>
> Differential Revision: https://reviews.llvm.org/D14031
>
>
> ...
> +  # Use add_lit_testsuite() from LLVM CMake. This also depends on OpenMP
> +  # implementation because it uses omp.h.
> +  add_lit_testsuite(check-libomptarget
> +    "Running libomptarget tests"
> +    ${CMAKE_CURRENT_BINARY_DIR}
> +    ARGS "-j 1"

Adding -j 1 may be okay for the stand-alone testing configuration, but 
we can't add it here like this. The argument applies not only to the 
tests in this subdirectory, but to all regression tests being run. That 
means that running check-all after a complete build might take 20 
minutes instead of 1. Is there a reason we need to be (more) careful 
about this here than we are for the openmp tests? As a note, we 
definitely over-subscribe the system for the openmp tests, but they're 
not performance tests, and the stressing has often proved helpful.

Thanks again,
Hal

> +    DEPENDS omptarget omp
> +  )
> +
> ...
>
> _______________________________________________
> Openmp-commits mailing list
> Openmp-commits at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/openmp-commits

-- 
Hal Finkel
Lead, Compiler Technology and Programming Languages
Leadership Computing Facility
Argonne National Laboratory