rx: Clarify error checks for busy channel check

Commit a84c6b0ece1fdee4f462c6ce27fa78c2e0d419f4 changed this so we
don't just discard an incoming request if the call already had an
error. But if the call already has an error, rxi_WaitforTQBusy is a
no-op, so checking if the error has "changed" is unnecessary and can
be confusing. Just bypass this whole block if the call already has an
error.

Discussed during the 5 Dec 2012 release-team meeting.

Change-Id: Id57d65736f3228d4e7595f56800f42c52e83ef39
Reviewed-on: http://gerrit.openafs.org/8748
Tested-by: BuildBot <buildbot@rampaginggeek.com>
Reviewed-by: Benjamin Kaduk <kaduk@mit.edu>
Reviewed-by: Derrick Brashear <shadow@your-file-system.com>

diff --git a/src/rx/rx.c b/src/rx/rx.c
index 6d7e78d..0c05763 100644 (file)
--- a/src/rx/rx.c
+++ b/src/rx/rx.c
@@ -3379,14 +3379,13 @@ rxi_ReceiveServerCall(osi_socket socket, struct rx_packet *np,
      * flag is cleared.
      */
 #ifdef RX_ENABLE_LOCKS
-    if (call->state == RX_STATE_ACTIVE) {
-       int old_error = call->error;
+    if (call->state == RX_STATE_ACTIVE && !call->error) {
        rxi_WaitforTQBusy(call);
         /* If we entered error state while waiting,
          * must call rxi_CallError to permit rxi_ResetCall
          * to processed when the tqWaiter count hits zero.
          */
-        if (call->error && call->error != old_error) {
+        if (call->error) {
            rxi_CallError(call, call->error);
            MUTEX_EXIT(&call->lock);
             return NULL;