btl/usnic: cap the number of resends per progress iteration

New MCA param: btl_usnic_max_resends_per_iteration.  This is the max
number of resends we'll do in a single pass through usNIC component
progress.  This prevents progress from getting stuck in an endless
loop of retransmissions (i.e., if more retransmissions are triggered
during the sending of retransmissions).  Specifically: we need to
leave the resend loop to allow receives to happen (which may ACK
messages we have sent previously, and therefore cause pending resends
to be moot).

Signed-off-by: Jeff Squyres <[email protected]>

Author: jsquyres

opal/mca/btl/usnic/btl_usnic.h

@@ -190,6 +190,10 @@ typedef struct opal_btl_usnic_component_t {
     /** retrans characteristics */
     int retrans_timeout;
 
+    /** max number of messages re-sent during a single progress
+        iteration */
+    int max_resends_per_iteration;
+
     /** minimum number of times through component progress before
         checking to see if standalone ACKs need to be sent */
     int ack_iteration_delay;

opal/mca/btl/usnic/btl_usnic_mca.c

@@ -249,6 +249,10 @@ int opal_btl_usnic_component_register(void)
                   100000, &mca_btl_usnic_component.retrans_timeout,
                   REGINT_GE_ONE, OPAL_INFO_LVL_5));
 
+    CHECK(reg_int("max_resends_per_iteration", "Maximum number of frames to resend in a single iteration through usNIC component progress",
+                  16, &mca_btl_usnic_component.max_resends_per_iteration,
+                  REGINT_GE_ONE, OPAL_INFO_LVL_5));
+
     CHECK(reg_int("ack_iteration_delay", "Minimum number of times through usNIC \"progress\" function before checking to see if standalone ACKs need to be sent",
                   0, &mca_btl_usnic_component.ack_iteration_delay,
                   REGINT_GE_ZERO, OPAL_INFO_LVL_5));

opal/mca/btl/usnic/btl_usnic_module.c

@@ -954,11 +954,12 @@ usnic_do_resends(
     opal_btl_usnic_send_segment_t *sseg;
     opal_btl_usnic_endpoint_t *endpoint;
     struct opal_btl_usnic_channel_t *data_channel;
-    int ret;
+    int ret, count;
 
     data_channel = &module->mod_channels[USNIC_DATA_CHANNEL];
 
-    while ((get_send_credits(data_channel) > 1) &&
+    count = mca_btl_usnic_component.max_resends_per_iteration;
+    while (count > 0 && (get_send_credits(data_channel) > 1) &&
            !opal_list_is_empty(&module->pending_resend_segs)) {
 
         /*
@@ -999,6 +1000,8 @@ usnic_do_resends(
         if (OPAL_UNLIKELY(OPAL_SUCCESS != ret)) {
             opal_btl_usnic_util_abort("hotel checkin failed\n", __FILE__, __LINE__);
         }
+
+        --count;
     }
 }