>They're set by the application via the rdma_conn_param on the active >side, and they get set to the device max values by default on the >passive side. There is currently no matching of these by the iwcm or >any of the driver code: its up to the ULP/application to negotiate this >via private data or some other way. > >Is that what you're asking?
In part - I don't see that IRD or ORD are set through the modify QP calls in the rdma_cm. I'm guessing that the values are passed to the driver, and are used by the driver when transitioning to RTS? Basically, I'm trying to determine if the fix to the librdmacm affects iWarp at all (I don't believe that it does), and if a similar fix would also work in the kernel. I'm pretty sure that it will for IB, but I'm not sure what the behavior would be for iWarp. - Sean _______________________________________________ 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
