Re: [OMPI devel] (no subject)

2018-10-21 Thread Dmitry N. Mikushin
MSYS2 is known for not supplying statfs/statvfs, which are practically meaningless in a Windows-based environment, see e.g. [1] Yet, both calls do not seem to be critical for MPI internals, thus one can potentially develop patches to turn them into an optional feature within OpenMPI. If you are

Re: [OMPI devel] CUDA kernels in OpenMPI

2017-01-27 Thread Dmitry N. Mikushin
It's hard to tell without complete makefile example. Could you please post a minimal reprocase? Note specifically for OpenMPI there is a tricky workaround. You can use nvcc as mpicc compiler by exporting OMPI_CC=nvcc and wrapping out incompatible compiler options. Kind regards, - Dmitry

Re: [OMPI devel] CUDA support not working?

2013-11-23 Thread Dmitry N. Mikushin
VT is getting out of sync with CUDA from time to time, this already happened before. - D. 2013/11/24 Jörg Bornschein : > On 23.11.2013, at 21:42, Jörg Bornschein wrote: > > Sorry, > >> I’m typically compiling with >> >> ./configure —with-cuda > > > I’m

Re: [OMPI devel] Program prefix for OpenMPI binaries?

2012-09-09 Thread Dmitry N. Mikushin
I don't know how hard it would be to implement, but it looks like something > we could do for future releases, perhaps starting in the 1.7 series. I'll > have to look into it and discuss with a few people here and then get back to > you on it. > > > On Sep 9, 2012, at 8:02 AM, Dmitry

Re: [OMPI devel] Program prefix for OpenMPI binaries?

2012-09-09 Thread Dmitry N. Mikushin
ppers, e.g. mpicc -> prefix-mpicc, etc., which correctly link to their local orte-wrapper. Thanks, - D. 2012/9/9 Ralph Castain <r...@open-mpi.org>: > > On Sep 8, 2012, at 1:14 PM, "Dmitry N. Mikushin" <maemar...@gmail.com> wrote: > >> Dear all, >> >&

[OMPI devel] Program prefix for OpenMPI binaries?

2012-09-08 Thread Dmitry N. Mikushin
Dear all, Setting program prefix is disabled for OpenMPI compilation, could you please explain why? Consider we want to make a package containing OpenMPI packed together with our open-source compiler, that is installable on arbitrary Linux system. In this case, it would be very handy to add some

[OMPI devel] Errors during VampirTrace wrappers compilation, when used with CUDA 5

2012-09-04 Thread Dmitry N. Mikushin
Trace folks can comment on > this since the issue seems to be in areas of the code they work in. > > >>-Original Message----- >>From: Dmitry N. Mikushin [mailto:maemar...@gmail.com] >>Sent: Monday, September 03, 2012 1:37 PM >>To: Rolf vandeVaart >>Cc: de..

Re: [OMPI devel] OpenMPI CUDA 5 readiness?

2012-09-03 Thread Dmitry N. Mikushin
CUDA 5 basically changes char* to void* in some functions. Attached is a small patch which changes prototypes, depending on used CUDA version. Tested with CUDA 5 preview and 4.2. - D. 2012/9/2 Dmitry N. Mikushin <maemar...@gmail.com>: > Dear Rolf, > > FYI, looks like with CUDA 5

[OMPI devel] OpenMPI CUDA 5 readiness?

2012-09-02 Thread Dmitry N. Mikushin
Dear Rolf, FYI, looks like with CUDA 5 preview OpenMPI trunk fails to build due to the following errors: $ svn info Path: . URL: http://svn.open-mpi.org/svn/ompi/trunk Repository Root: http://svn.open-mpi.org/svn/ompi Repository UUID: 63e3feb5-37d5-0310-a306-e8a459e722fe Revision: 27216 Node

[OMPI devel] Existing frameworks for remote device memory exclusive read/write

2012-07-20 Thread Dmitry N. Mikushin
Dear OpenMPI developers, My question is not directly related to OpenMPI, but might be related to internal project kitchen and your wide experiences. Say, there is a need to implement a transparent read/write of PCI-Express device internal memory from the host system. It is allowed to use only