On Jan 3, 2011, at 5:17 PM, Eugene Loh wrote:
> Shouldn't that inner loop be preceded by initialization of
> ep->btl_max_send_size to some very large value (ironically enough, perhaps
> "-1")?
Won't ep->btl_max_send_size already be set to the max value of all the btl's on
the endpoint?
--
Je
As the endpoint's btl_max_send_size has been initialized to the min of the
max_size of all BTLs in the send (respectively rdma) array, the loop you
pinpointed will have no effect (as it is impossible to find a smaller value
than the minimum already computed). Pre-setting to (size_t)-1 should fix
We discussed on the call today that Rainer and I are waiting for another week
in case any "oops!"es are found in the 1.5.1 release such that we would need to
do an immediate 1.5.2 release.
If all goes well (meaning that no "oops"es are found), starting next Tuesday,
we'll open the floodgates an
Thanks for the sanity check. Fix in r24202.
George Bosilca wrote:
As the endpoint's btl_max_send_size has been initialized to the min of the
max_size of all BTLs in the send (respectively rdma) array, the loop you
pinpointed will have no effect (as it is impossible to find a smaller value
t
