Wenduo Wang
Wenduo Wang
https://github.com/open-mpi/ompi/commit/ae98e0453252cbf026dbd784f618d320ef1066fc has been reverted in https://github.com/open-mpi/ompi/pull/12157. Note that this change was only on main. v5.0.x does not have this issue.
Checked MTT history. The test has been passing.
@hzhou @raffenet Could you kindly review the change?
Converted change to draft for testing.
@hzhou Please take a look at the update. I followed your suggestion to deprecate `MPIDI_OFI_MAX_NICS`. As a result I had to change a few static variables and data structures to...
> @wenduwan The PR looks ok to me. It keeps the usage that we had intended too. Please run the tests to make sure the testsuite passes. @tarudoodi Thank you...
@hzhou Thank you for your insights. I think your points are valid. In this case I'm concerned with the implementation of `MPIDI_OFI_addr_exchange_all_ctx` around https://github.com/pmodels/mpich/blob/ab0b77d38b8fe083537d9e2dc5096422e808385e/src/mpid/ch4/netmod/ofi/init_addrxchg.c#L221-L230 IIUC it tries to exchange the...
@bosilca Thanks for your comments. I'm looking into this PR again.
After a fresh look at the change, it appears to have a much larger blast radius than I expected. The original revision might started with the wrong place - it...
@acgoldma Sorry for the 2-year wait. Does this problem still happen with latest romio and ompi 5.0.3?