Hui Zhou
Hui Zhou
Shouldn't `MPI_File_open` be a collective and thus okay for a barrier?
``` $ mpirun -n 2 ./ireadtest Expected 512, read 1024 Found 1 errors ```
The size of `flat` is proportional to the number of contiguous blocks. We probably should set a threshold and use some generic fallback for the case such as reported here.
There is typerep api (for both dataloop and yaksa) to get a list noncontig segments in a range as an IOV array. The fallback I have in mind is to...
The datatype in this issue is a *contig* type of `count = 1073741832`. For the purpose of file view, this is equivalent to a single inner type. I think this...
> in 2005 Rob Ross wrote "mpitypes" code (also know as 'dataloop' code) which did a more clever job of maintaining datatype state and information. Instead of flattening the entire...
I am not sure how to fix it. One solution is to allocate a few bits in the match bits and set them to a sequence number of the communicator....
> Maybe we can do the equivalent to an MPI_ANY_TAG/MPI_ANY_SOURCE probe when freeing the communicator and keep it around if something is found in the queue? Debug builds would warn...
test:mpich/ch3/tcp
test:mpich/ch3/most test:mpich/ch4/most