- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
My group is currently using Inspector to test our code before moving to profiling, and we consistently receive a data race on several MPI calls, such as MPI_GATHERV, MPI_INIT_THREAD and MPI_ISEND. The constant is that the write side of this read/write race is rxm_conn_eq_read from librxm-fi.so. Has anyone else had this experience? If so, was there a resolution other than suppressing the issue in Inspector?
- Tags:
- CC++
- Debugging
- Development Tools
- Fortran
- Intel® Inspector
- Optimization
- Parallel Computing
- Vectorization
Link Copied
0 Replies
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page