X86: uv: xpc_make_first_contact hang due to not accepting ACTIVE state
authorRobin Holt <holt@sgi.com>
Wed, 16 Dec 2009 00:47:58 +0000 (16:47 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Wed, 16 Dec 2009 15:20:14 +0000 (07:20 -0800)
Many times while the initial connection is being made, the contacted
partition will send back both the ACTIVATING and the ACTIVE
remote_act_state changes in very close succescion.  The 1/4 second delay
in the make first contact loop is large enough to nearly always miss the
ACTIVATING state change.

Since either state indicates the remote partition has acknowledged our
state change, accept either.

Signed-off-by: Robin Holt <holt@sgi.com>
Cc: Jack Steiner <steiner@sgi.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
drivers/misc/sgi-xp/xpc_uv.c

index 19bd7b0ede91d7282e42f2e7e5255f750a5f334a..241ea5f01aaa731457f0d69101ec9dc6ef5cbe7a 100644 (file)
@@ -1024,7 +1024,8 @@ xpc_make_first_contact_uv(struct xpc_partition *part)
        xpc_send_activate_IRQ_part_uv(part, &msg, sizeof(msg),
                                      XPC_ACTIVATE_MQ_MSG_SYNC_ACT_STATE_UV);
 
-       while (part->sn.uv.remote_act_state != XPC_P_AS_ACTIVATING) {
+       while (!((part->sn.uv.remote_act_state == XPC_P_AS_ACTIVATING) ||
+                (part->sn.uv.remote_act_state == XPC_P_AS_ACTIVE))) {
 
                dev_dbg(xpc_part, "waiting to make first contact with "
                        "partition %d\n", XPC_PARTID(part));