Matthieu Dorier
Matthieu Dorier
I'm not sure if this is a bug or a misunderstanding on my part of what `HG_Get_input_extra_buf` does (the same behavior happens with `HG_Get_ouput_extra_buf`), but I can't seem to be...
**Describe the bug** I have a simple Mercury client and server (code available at the end of this issue). The client sends RPCs to the server indefinitely and the server...
I know busy-spinning is meant for networks like GNI, where blocking is inefficient, but if I try it with na+sm, it simply doesn't work. I do this on the server:...
It would be very useful for NA to allow sending to self (both expected and unexpected messages), at least with the `na+sm` and the `ofi` providers. I'm posting this feature...
Something that could be useful in Mercury for upper libraries like Margo is the possibility to inject user-provided file descriptors in the Mercury polling mechanism. Some use-cases would include injecting...
When SELF_FORWARD is disabled, the address returned by HG_Addr_self cannot be used to forward RPCs. Instead, one has to convert it into a string (HG_Addr_to_string) then do a lookup to...
I'm trying to integrate ROSS simulations into a workflow, so a set of processes should be able to call a ROSS-based simulation (CODES model-net-mpi-replay) multiple time. This means calling tw_init/tw_end...
It would be useful to have an `ABT_eventual_wait_any` function that takes an array of eventuals and blocks until one has been set (kind of the ABT equivalent of MPI_Waitany). A...
It would be great to have an ABT_thread_sleep(double timeout_ms) function that puts the calling ABT_thread to sleep for a given amount of time (or "at least the given amount of...
It would be very useful to provide recursive RW lock. It is not uncommon for a code to read-lock a structure and eventually write-lock it for a small modification. Right...