Peter Boyle comments

Results 89 comments of


                                            Peter Boyle

illegal memory access, SyCL branch, CUDA 9.2, gcc 6.3.0

What system is this on? Reason for asking is that IBM Spectrum scale MPI causes a segfault like this unless running under mpirun, even for one node.

illegal memory access, SyCL branch, CUDA 9.2, gcc 6.3.0

Run on Develop tonight on Summit V100 (merged SyCL back to develop), getting poorer performance than expected. But runs Grid : Message : 0.343734 s : ==================================================================================================== Grid : Message...

illegal memory access, SyCL branch, CUDA 9.2, gcc 6.3.0

Different Cuda aware MPI's can definitely behave differently about whether you initialise MPI or Cuda first. Are you sure you've got a CUDA aware MPI ?

How to structure input files?

Hi there, I got fed up with dynamic XML for HMC control. I will also have to deprecate JSON because NVCC doesn't like it in the GPU port, and I'm...

JSON_IO.cc fails to compile on BG/Q with bglclang (fix included)

Status of this?

Failure to compile with GCC 9.3.0 (memory issues)

yeah ... I heard >=9 was being bad. See issue 100 ... it's a bit of a GCC graveyard. You might make a comment there.

IPC handles not copy by value between processes

Hi, intended you to ask Camilo or Patrick for it - they have access. But there's nothing secret as this API is on GitHub, so will attach.

IPC handles not copy by value between processes

https://raw.githubusercontent.com/paboyle/ComputationalCourseQCD/master/reproducer_IPC_bandwidth.cpp

IPC handles not copy by value between processes

Sorry for the run around.... all sorts of things forced me to bounce a file through a GitHub repository.

Grid does not compile on Arm with CUDA

Use CXXFLAGS - there are other neon compilers that are happy with Eigen. Or use the -ccbin flag to nvcc to specify a host compiler that actually works for Grace.