On Dec 15, 2015, at 5:18 AM, Justin Cinkelj <justin.cink...@xlab.si> wrote: > > Second question. > With two threads in one process, I got attempt to > opal_shmem_segment_dettach() and munmap() on same mmap-ed address, from both > threads. I 'fixed' that by replacing "ds_buf->seg_cpid = getpid()" with > gettid(), and then each thread munmap-s only address allocated by itself. Is > that correct? Or is it possible, that the second thread might still try to > access data at that address?
I echo Gilles' comments here: we never really planned for MPI processes to be threads -- I'm surprised you haven't run into more problems like this. -- Jeff Squyres jsquy...@cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/
