Aren
/
NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Carefully set IFF_OACTIVE in mvgbe_start(). Even if mvgbe_encap() 

returns other than 0, the TX ring might not full. Check whether
the TX ring has one or more packets. If the ring is empty,
dont' set IFF_OACTIVE because an TX complete interrupt never
occur and IFF_OACTIVE flags is left. The interface's timer
isn't reset, so a device timeout desn't occur.

 Fixes a bug that IFF_OACTIVE flag is left on heavy load.
Part of PR#48568.
This commit is contained in:
msaitoh 2014-02-03 05:56:23 +00:00
parent 55b0c96054
commit 3443d04260
1 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
sys/dev/marvell/if_mvgbe.c
View File

@ -1,4 +1,4 @@
/* $NetBSD: if_mvgbe.c,v 1.35 2013/12/23 02:23:25 kiyohara Exp $ */
/* $NetBSD: if_mvgbe.c,v 1.36 2014/02/03 05:56:23 msaitoh Exp $ */
/*
* Copyright (c) 2007, 2008, 2013 KIYOHARA Takashi
* All rights reserved.
@ -25,7 +25,7 @@
* POSSIBILITY OF SUCH DAMAGE.
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: if_mvgbe.c,v 1.35 2013/12/23 02:23:25 kiyohara Exp $");
__KERNEL_RCSID(0, "$NetBSD: if_mvgbe.c,v 1.36 2014/02/03 05:56:23 msaitoh Exp $");
#include "opt_multiprocessor.h"
@ -1059,7 +1059,8 @@ mvgbe_start(struct ifnet *ifp)
* for the NIC to drain the ring.
*/
if (mvgbe_encap(sc, m_head, &idx)) {
ifp->if_flags |= IFF_OACTIVE;
if (sc->sc_cdata.mvgbe_tx_cnt > 0)
ifp->if_flags |= IFF_OACTIVE;
break;
}