I have the following set both on rdma_connect as well as at rdma_accept.<div><br class="webkit-block-placeholder"></div><div><div> conn_param.responder_resources = 4;</div><div> conn_param.initiator_depth = 4;
</div><div><br></div><div>Should initiator_depth be lower for better behavior?</div><div><br class="webkit-block-placeholder"></div><div><br class="webkit-block-placeholder"></div><div><br><div class="gmail_quote">On Jan 9, 2008 10:16 PM, Dotan Barak <
<a href="mailto:dotanb@dev.mellanox.co.il">dotanb@dev.mellanox.co.il</a>> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><div><div></div><div class="Wj3C7c">
Rajouri Jammu wrote:<br>> Occasionally, I'm getting a retry exceeded error on the qp (error 12)<br>> when doing rdma_reads.<br>><br>> Under what conditions would thins kind of problem happen?<br>><br>> I have the retry_count = 5 and 'am using rdma_cm for all the
<br>> connection setup.<br>><br>> OFED version is 1.2.5<br></div></div>Does it happen between different HCAs?<br><br>If this happens during working with the QPs (not in the first message)<br>than check the following thing:
<br><br>If the QP attributes values of max_rd_atomic and max_dest_rd_atomic<br>this may happen.<br><br>The values should be (for sides A and B):<br>A.max_rd_atomic <= B.max_dest_rd_atomic<br>A.max_dest_rd_atomic
>= B.max_rd_atomic<br><br>(which means that RDMA Reads/atomic as initiator shouldn't be larger<br>than the supported value as the destination)<br><br>You can check it by query the used QP and verify those values.<br>
<br><br><br>If it happens at the beginning of the connection, there may be other<br>problem and i need more info ....<br><font color="#888888"><br>Dotan<br></font></blockquote></div><br></div></div>