From: Derrick Brashear <shadow@dementia.org>
Date: Mon, 25 Jan 2010 21:02:23 +0000 (-0500)
Subject: rx avoid discarding packets while tq is busy
X-Git-Tag: openafs-devel-1_5_71~21
X-Git-Url: https://git.michaelhowe.org/gitweb/?a=commitdiff_plain;h=f0b1d7ccbbc4056338dde92c8b131ad2070351ad;p=packages%2Fo%2Fopenafs.git

rx avoid discarding packets while tq is busy

we don't check here that no one else is working the tq *after* getting the
lock. do so.

Change-Id: I91d1ac020334b038a6a045734911335299613875
Reviewed-on: http://gerrit.openafs.org/1159
Reviewed-by: Jeffrey Altman <jaltman@openafs.org>
Tested-by: Derrick Brashear <shadow@dementia.org>
Tested-by: Jeffrey Altman <jaltman@openafs.org>
Reviewed-by: Derrick Brashear <shadow@dementia.org>
---

diff --git a/src/rx/rx.c b/src/rx/rx.c
index afb31842f..ec342d8cc 100755
--- a/src/rx/rx.c
+++ b/src/rx/rx.c
@@ -2219,6 +2219,7 @@ rxi_NewCall(struct rx_connection *conn, int channel)
 	CLEAR_CALL_QUEUE_LOCK(call);
 #ifdef	AFS_GLOBAL_RXLOCK_KERNEL
 	/* Now, if TQ wasn't cleared earlier, do it now. */
+	rxi_WaitforTQBusy(call);
 	if (call->flags & RX_CALL_TQ_CLEARME) {
 	    rxi_ClearTransmitQueue(call, 1);
 	    /*queue_Init(&call->tq);*/