btl/openib: XRC fix bug that could cause an invalid SRQ# to be used

This commit fixes a bug that occurs when attempting a get or put
operation on an endpoint that is not already connected. In this case
the remote_srqn may be set to an invalid value as the rem_srqs array
on the endpoint is not populated. This commit moves the usage of the
rem_srqs array to the internal put/get functions where it is
guaranteed this array is populated.

Signed-off-by: Nathan Hjelm <[email protected]>

(cherry picked from commit open-mpi/ompi@4dc73d7)

Signed-off-by: Nathan Hjelm <[email protected]>

Author: hjelmn

opal/mca/btl/openib/btl_openib_atomic.c

@@ -1,6 +1,6 @@
 /* -*- Mode: C; c-basic-offset:4 ; indent-tabs-mode:nil -*- */
 /*
- * Copyright (c) 2014      Los Alamos National Security, LLC.  All rights
+ * Copyright (c) 2014-2016 Los Alamos National Security, LLC.  All rights
  *                         reserved.
  * Copyright (c) 2015      Research Organization for Information Science
  *                         and Technology (RIST). All rights reserved.
@@ -73,16 +73,7 @@ static int mca_btl_openib_atomic_internal (struct mca_btl_base_module_t *btl, st
 
     frag->sr_desc.wr.atomic.rkey = rkey;
 
-#if HAVE_XRC
-    if (MCA_BTL_XRC_ENABLED && BTL_OPENIB_QP_TYPE_XRC(qp)) {
-#if OPAL_HAVE_CONNECTX_XRC_DOMAINS
-        frag->sr_desc.qp_type.xrc.remote_srqn = endpoint->rem_info.rem_srqs[qp].rem_srq_num;
-#else
-        frag->sr_desc.xrc_remote_srq_num = endpoint->rem_info.rem_srqs[qp].rem_srq_num;
-#endif
-
-    }
-#endif
+    /* NTH: the SRQ# is set in mca_btl_get_internal */
 
     if (endpoint->endpoint_state != MCA_BTL_IB_CONNECTED) {
         OPAL_THREAD_LOCK(&endpoint->endpoint_lock);

opal/mca/btl/openib/btl_openib_get.c

@@ -12,7 +12,7 @@
  *                         All rights reserved.
  * Copyright (c) 2007-2013 Cisco Systems, Inc.  All rights reserved.
  * Copyright (c) 2006-2009 Mellanox Technologies. All rights reserved.
- * Copyright (c) 2006-2014 Los Alamos National Security, LLC.  All rights
+ * Copyright (c) 2006-2016 Los Alamos National Security, LLC.  All rights
  *                         reserved.
  * Copyright (c) 2006-2007 Voltaire All rights reserved.
  * Copyright (c) 2008-2012 Oracle and/or its affiliates.  All rights reserved.
@@ -92,16 +92,6 @@ int mca_btl_openib_get (mca_btl_base_module_t *btl, struct mca_btl_base_endpoint
         frag->sr_desc.wr.rdma.rkey = remote_handle->rkey;
     }
 
-#if HAVE_XRC
-    if (MCA_BTL_XRC_ENABLED && BTL_OPENIB_QP_TYPE_XRC(qp)) {
-#if OPAL_HAVE_CONNECTX_XRC_DOMAINS
-        frag->sr_desc.qp_type.xrc.remote_srqn = ep->rem_info.rem_srqs[qp].rem_srq_num;
-#else
-        frag->sr_desc.xrc_remote_srq_num = ep->rem_info.rem_srqs[qp].rem_srq_num;
-#endif
-    }
-#endif
-
     if (ep->endpoint_state != MCA_BTL_IB_CONNECTED) {
         OPAL_THREAD_LOCK(&ep->endpoint_lock);
         rc = check_endpoint_state(ep, &to_base_frag(frag)->base, &ep->pending_get_frags);
@@ -138,6 +128,19 @@ int mca_btl_openib_get_internal (mca_btl_base_module_t *btl, struct mca_btl_base
     int qp = to_base_frag(frag)->base.order;
     struct ibv_send_wr *bad_wr;
 
+#if HAVE_XRC
+    if (MCA_BTL_XRC_ENABLED && BTL_OPENIB_QP_TYPE_XRC(qp)) {
+        /* NTH: the remote SRQ number is only available once the endpoint is connected. By
+         * setting the value here instead of mca_btl_openib_get we guarantee the rem_srqs
+         * array is initialized. */
+#if OPAL_HAVE_CONNECTX_XRC_DOMAINS
+        frag->sr_desc.qp_type.xrc.remote_srqn = ep->rem_info.rem_srqs[qp].rem_srq_num;
+#else
+        frag->sr_desc.xrc_remote_srq_num = ep->rem_info.rem_srqs[qp].rem_srq_num;
+#endif
+    }
+#endif
+
     /* check for a send wqe */
     if (qp_get_wqe(ep, qp) < 0) {
         qp_put_wqe(ep, qp);

opal/mca/btl/openib/btl_openib_put.c

@@ -101,19 +101,6 @@ int mca_btl_openib_put (mca_btl_base_module_t *btl, struct mca_btl_base_endpoint
         to_out_frag(frag)->sr_desc.wr.rdma.rkey = remote_handle->rkey;
     }
 
-#if HAVE_XRC
-    if (MCA_BTL_XRC_ENABLED && BTL_OPENIB_QP_TYPE_XRC(qp)) {
-
-#if OPAL_HAVE_CONNECTX_XRC
-        to_out_frag(frag)->sr_desc.xrc_remote_srq_num = ep->rem_info.rem_srqs[qp].rem_srq_num;
-#elif OPAL_HAVE_CONNECTX_XRC_DOMAINS
-        to_out_frag(frag)->sr_desc.qp_type.xrc.remote_srqn = ep->rem_info.rem_srqs[qp].rem_srq_num;
-#else
-#error "that should never happen"
-#endif
-    }
-#endif
-
     if (ep->endpoint_state != MCA_BTL_IB_CONNECTED) {
         OPAL_THREAD_LOCK(&ep->endpoint_lock);
         rc = check_endpoint_state(ep, &to_base_frag(frag)->base, &ep->pending_put_frags);
@@ -153,6 +140,21 @@ int mca_btl_openib_put_internal (mca_btl_base_module_t *btl, struct mca_btl_base
     struct ibv_send_wr *bad_wr;
     int rc;
 
+#if HAVE_XRC
+    if (MCA_BTL_XRC_ENABLED && BTL_OPENIB_QP_TYPE_XRC(qp)) {
+        /* NTH: the remote SRQ number is only available once the endpoint is connected. By
+         * setting the value here instead of mca_btl_openib_put we guarantee the rem_srqs
+         * array is initialized. */
+#if OPAL_HAVE_CONNECTX_XRC
+        to_out_frag(frag)->sr_desc.xrc_remote_srq_num = ep->rem_info.rem_srqs[qp].rem_srq_num;
+#elif OPAL_HAVE_CONNECTX_XRC_DOMAINS
+        to_out_frag(frag)->sr_desc.qp_type.xrc.remote_srqn = ep->rem_info.rem_srqs[qp].rem_srq_num;
+#else
+#error "that should never happen"
+#endif
+    }
+#endif
+
     /* check for a send wqe */
     if (qp_get_wqe(ep, qp) < 0) {
         qp_put_wqe(ep, qp);