How would you solve it if you would have used TCP/IP sockets? Dotan
On Wed, Sep 3, 2008 at 9:49 PM, AJ Guillon <[EMAIL PROTECTED]> wrote: > Hrrrm. That's really too bad because I would like to use RDMA to steal work > from other nodes along with dependent memory. If I'm loading memory for a > task on one node, and another node steals the task, the node from which the > task was stolen should stop fetching memory required for the now stolen > task. A more complex scheduler might be able to deal with this but maybe not > optimally. > > Suggestions for workarounds? > > AJ > > On Sep 1, 2008, at 2:00 AM, "Dotan Barak" <[EMAIL PROTECTED]> wrote: > >> As much as i know, once you posted a WR, you can not cancel it. >> The only thing that you can do is flush the whole QP by changing the >> QP state to ERROR (which flushes the work Queues and produces >> completion for every WR) or to RESET, which cleans the Queues from the >> WRs. >> >> >> Dotan >> >> On Sun, Aug 31, 2008 at 9:48 PM, Adrien Guillon <[EMAIL PROTECTED]> >> wrote: >>> >>> Hey, >>> >>> How can I interrupt an RDMA read cleanly? In my case, I might decide >>> that I don't need to read some memory anymore (because something else >>> happened), so I want to abort. >>> >>> AJ >>> _______________________________________________ >>> general mailing list >>> [email protected] >>> http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general >>> >>> To unsubscribe, please visit >>> http://openib.org/mailman/listinfo/openib-general >>> > _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
