adulau/aha
1
0
Fork 0
mirror of https://github.com/adulau/aha.git synced 2024-12-27 19:26:25 +00:00
Code Issues Projects Releases Packages Wiki Activity

RDMA/nes: Fix crash in nes_accept()

Browse source 
While running IMP_EXT's window test, we saw a crash in nes_accept().
Here is the sequence of what happened:

(1) In MVAPICH2, connect request is received for port #0.

FIX:  Add a nes_connect() check to make sure local or remote tcp port
      is not 0.

(2) Remote node's (passive) TCP stack sends a reset when it gets a
    connect request because of port = 0.  Active side set the connect
    error to IW_CM_EVENT_STATUS_REJECTED when it received the RST from
    remote node.

FIX: The corect error code is -ECONNRESET.

(3) Wrong error code of IW_CM_EVENT_STATUS_REJECTED causes the core to
    destroy its listener ports.  Here there are connections that may
    have sent an MPA request up and waiting for accept or reject.  But
    the listener and its cm_nodes have been freed already causing the
    crash noticed.

FIX: The cm_node is freed only if its state is not
     NES_CM_STATE_MPAREQ_RCVD.  If cm_node's state is
     NES_CM_STATE_MPAREQ_RCVD then its new state is set to
     NES_CM_STATE_LISTENER_DESTROYED and it is not freed.  When
     nes_accept() or nes_reject() is received, its state is checked
     for NES_CM_STATE_LISTENER_DESTROYED and in this case the cm_node
     is freed and error is returned.

Signed-off-by: Faisal Latif <faisal.latif@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This commit is contained in:
Faisal Latif 2009-12-09 15:54:08 -08:00 committed by Roland Dreier
parent 69524e1aff
commit c5a7d48971
2 changed files with 50 additions and 28 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

77
drivers/infiniband/hw/nes/nes_cm.c
View file

@ -978,6 +978,7 @@ static int mini_cm_dec_refcnt_listen(struct nes_cm_core *cm_core,
reset_entry);
{
struct nes_cm_node *loopback = cm_node->loopbackpartner;
enum nes_cm_node_state old_state;
if (NES_CM_STATE_FIN_WAIT1 <= cm_node->state) {
rem_ref_cm_node(cm_node->cm_core, cm_node);
} else {
@ -989,11 +990,12 @@ static int mini_cm_dec_refcnt_listen(struct nes_cm_core *cm_core,
NES_CM_STATE_CLOSED;
WARN_ON(1);
} else {
cm_node->state =
NES_CM_STATE_CLOSED;
rem_ref_cm_node(
cm_node->cm_core,
cm_node);
old_state = cm_node->state;
cm_node->state = NES_CM_STATE_LISTENER_DESTROYED;
if (old_state != NES_CM_STATE_MPAREQ_RCVD)
rem_ref_cm_node(
cm_node->cm_core,
cm_node);
}
} else {
struct nes_cm_event event;
@ -1009,6 +1011,7 @@ static int mini_cm_dec_refcnt_listen(struct nes_cm_core *cm_core,
loopback->loc_port;
event.cm_info.cm_id = loopback->cm_id;
cm_event_connect_error(&event);
cm_node->state = NES_CM_STATE_LISTENER_DESTROYED;
loopback->state = NES_CM_STATE_CLOSED;
event.cm_node = cm_node;
@ -2131,30 +2134,39 @@ static int mini_cm_reject(struct nes_cm_core *cm_core,
cm_node->state = NES_CM_STATE_CLOSED;
rem_ref_cm_node(cm_core, cm_node);
} else {
ret = send_mpa_reject(cm_node);
if (ret) {
cm_node->state = NES_CM_STATE_CLOSED;
err = send_reset(cm_node, NULL);
if (err)
WARN_ON(1);
} else
cm_id->add_ref(cm_id);
if (cm_node->state == NES_CM_STATE_LISTENER_DESTROYED) {
rem_ref_cm_node(cm_core, cm_node);
} else {
ret = send_mpa_reject(cm_node);
if (ret) {
cm_node->state = NES_CM_STATE_CLOSED;
err = send_reset(cm_node, NULL);
if (err)
WARN_ON(1);
} else
cm_id->add_ref(cm_id);
}
}
} else {
cm_node->cm_id = NULL;
event.cm_node = loopback;
event.cm_info.rem_addr = loopback->rem_addr;
event.cm_info.loc_addr = loopback->loc_addr;
event.cm_info.rem_port = loopback->rem_port;
event.cm_info.loc_port = loopback->loc_port;
event.cm_info.cm_id = loopback->cm_id;
cm_event_mpa_reject(&event);
rem_ref_cm_node(cm_core, cm_node);
loopback->state = NES_CM_STATE_CLOSING;
if (cm_node->state == NES_CM_STATE_LISTENER_DESTROYED) {
rem_ref_cm_node(cm_core, cm_node);
rem_ref_cm_node(cm_core, loopback);
} else {
event.cm_node = loopback;
event.cm_info.rem_addr = loopback->rem_addr;
event.cm_info.loc_addr = loopback->loc_addr;
event.cm_info.rem_port = loopback->rem_port;
event.cm_info.loc_port = loopback->loc_port;
event.cm_info.cm_id = loopback->cm_id;
cm_event_mpa_reject(&event);
rem_ref_cm_node(cm_core, cm_node);
loopback->state = NES_CM_STATE_CLOSING;
cm_id = loopback->cm_id;
rem_ref_cm_node(cm_core, loopback);
cm_id->rem_ref(cm_id);
cm_id = loopback->cm_id;
rem_ref_cm_node(cm_core, loopback);
cm_id->rem_ref(cm_id);
}
}
return ret;
@ -2198,6 +2210,7 @@ static int mini_cm_close(struct nes_cm_core *cm_core, struct nes_cm_node *cm_nod
case NES_CM_STATE_UNKNOWN:
case NES_CM_STATE_INITED:
case NES_CM_STATE_CLOSED:
case NES_CM_STATE_LISTENER_DESTROYED:
ret = rem_ref_cm_node(cm_core, cm_node);
break;
case NES_CM_STATE_TSA:
@ -2716,8 +2729,6 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
struct nes_pd *nespd;
u64 tagged_offset;
ibqp = nes_get_qp(cm_id->device, conn_param->qpn);
if (!ibqp)
return -EINVAL;
@ -2733,6 +2744,13 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
"%s\n", cm_node, nesvnic, nesvnic->netdev,
nesvnic->netdev->name);
if (NES_CM_STATE_LISTENER_DESTROYED == cm_node->state) {
if (cm_node->loopbackpartner)
rem_ref_cm_node(cm_node->cm_core, cm_node->loopbackpartner);
rem_ref_cm_node(cm_node->cm_core, cm_node);
return -EINVAL;
}
/* associate the node with the QP */
nesqp->cm_node = (void *)cm_node;
cm_node->nesqp = nesqp;
@ -3003,6 +3021,9 @@ int nes_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
if (!nesdev)
return -EINVAL;
if (!(cm_id->local_addr.sin_port) || !(cm_id->remote_addr.sin_port))
return -EINVAL;
nes_debug(NES_DBG_CM, "QP%u, current IP = 0x%08X, Destination IP = "
"0x%08X:0x%04X, local = 0x%08X:0x%04X.\n", nesqp->hwqp.qp_id,
ntohl(nesvnic->local_ipaddr),
@ -3375,7 +3396,7 @@ static void cm_event_connect_error(struct nes_cm_event *event)
nesqp->cm_id = NULL;
cm_id->provider_data = NULL;
cm_event.event = IW_CM_EVENT_CONNECT_REPLY;
cm_event.status = IW_CM_EVENT_STATUS_REJECTED;
cm_event.status = -ECONNRESET;
cm_event.provider_data = cm_id->provider_data;
cm_event.local_addr = cm_id->local_addr;
cm_event.remote_addr = cm_id->remote_addr;

1
drivers/infiniband/hw/nes/nes_cm.h
View file

@ -200,6 +200,7 @@ enum nes_cm_node_state {
NES_CM_STATE_TIME_WAIT,
NES_CM_STATE_LAST_ACK,
NES_CM_STATE_CLOSING,
NES_CM_STATE_LISTENER_DESTROYED,
NES_CM_STATE_CLOSED
};