netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [XFRM] Fix aevent timer
@ 2006-04-13 12:55 jamal
  2006-04-13 13:46 ` Herbert Xu
  2006-04-14 22:03 ` David S. Miller
  0 siblings, 2 replies; 3+ messages in thread
From: jamal @ 2006-04-13 12:55 UTC (permalink / raw)
  To: David S. Miller, netdev; +Cc: Herbert Xu, KOVACS Krisztian

[-- Attachment #1: Type: text/plain, Size: 193 bytes --]


Finally got an opportunity to test this. I have incorporated all the
feedback and it passes all the scenarios i know.

Dave, please apply to Linus tree since this is a bug fix.

cheers,
jamal

[-- Attachment #2: aevent-fix-timer-5 --]
[-- Type: text/plain, Size: 2761 bytes --]


Send aevent immediately if we have sent nothing since last timer and
this is the first packet.
Fixes a corner case when packet threshold is very high, the timer low
and a very low packet rate input which is bursty.

Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
---

 include/net/xfrm.h    |    8 ++++++++
 net/xfrm/xfrm_state.c |   25 +++++++++++++++++++------
 2 files changed, 27 insertions(+), 6 deletions(-)

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 0d5529c..b77d333 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -143,6 +143,11 @@ struct xfrm_state
 	/* Replay detection state at the time we sent the last notification */
 	struct xfrm_replay_state preplay;
 
+	/* internal flag that only holds state for delayed aevent at the 
+	 * moment
+	*/
+	u32			xflags;
+
 	/* Replay detection notification settings */
 	u32			replay_maxage;
 	u32			replay_maxdiff;
@@ -167,6 +172,9 @@ struct xfrm_state
 	 * interpreted by xfrm_type methods. */
 	void			*data;
 };
+
+/* xflags - make enum if more show up */
+#define XFRM_TIME_DEFER	1
 
 enum {
 	XFRM_STATE_VOID,
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index a8e14dc..3dc3e1f 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -805,16 +805,22 @@ void xfrm_replay_notify(struct xfrm_stat
 	case XFRM_REPLAY_UPDATE:
 		if (x->replay_maxdiff &&
 		    (x->replay.seq - x->preplay.seq < x->replay_maxdiff) &&
-		    (x->replay.oseq - x->preplay.oseq < x->replay_maxdiff))
-			return;
+		    (x->replay.oseq - x->preplay.oseq < x->replay_maxdiff)) {
+			if (x->xflags & XFRM_TIME_DEFER)
+				event = XFRM_REPLAY_TIMEOUT;
+			else
+				return;
+		}
 
 		break;
 
 	case XFRM_REPLAY_TIMEOUT:
 		if ((x->replay.seq == x->preplay.seq) &&
 		    (x->replay.bitmap == x->preplay.bitmap) &&
-		    (x->replay.oseq == x->preplay.oseq))
+		    (x->replay.oseq == x->preplay.oseq)) {
+			x->xflags |= XFRM_TIME_DEFER;
 			return;
+		}
 
 		break;
 	}
@@ -825,8 +831,10 @@ void xfrm_replay_notify(struct xfrm_stat
 	km_state_notify(x, &c);
 
 	if (x->replay_maxage &&
-	    !mod_timer(&x->rtimer, jiffies + x->replay_maxage))
+	    !mod_timer(&x->rtimer, jiffies + x->replay_maxage)) {
 		xfrm_state_hold(x);
+		x->xflags &= ~XFRM_TIME_DEFER;
+	}
 }
 EXPORT_SYMBOL(xfrm_replay_notify);
 
@@ -836,10 +844,15 @@ static void xfrm_replay_timer_handler(un
 
 	spin_lock(&x->lock);
 
-	if (xfrm_aevent_is_on() && x->km.state == XFRM_STATE_VALID)
-		xfrm_replay_notify(x, XFRM_REPLAY_TIMEOUT);
+	if (x->km.state == XFRM_STATE_VALID) {
+		if (xfrm_aevent_is_on())
+			xfrm_replay_notify(x, XFRM_REPLAY_TIMEOUT);
+		else
+			x->xflags |= XFRM_TIME_DEFER;
+	}
 
 	spin_unlock(&x->lock);
+	xfrm_state_put(x);
 }
 
 int xfrm_replay_check(struct xfrm_state *x, u32 seq)

^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [XFRM] Fix aevent timer
  2006-04-13 12:55 [XFRM] Fix aevent timer jamal
@ 2006-04-13 13:46 ` Herbert Xu
  2006-04-14 22:03 ` David S. Miller
  1 sibling, 0 replies; 3+ messages in thread
From: Herbert Xu @ 2006-04-13 13:46 UTC (permalink / raw)
  To: jamal; +Cc: David S. Miller, netdev, KOVACS Krisztian

On Thu, Apr 13, 2006 at 08:55:58AM -0400, jamal wrote:
> 
> Finally got an opportunity to test this. I have incorporated all the
> feedback and it passes all the scenarios i know.

Looks good to me.  Thanks Jamal.
-- 
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [XFRM] Fix aevent timer
  2006-04-13 12:55 [XFRM] Fix aevent timer jamal
  2006-04-13 13:46 ` Herbert Xu
@ 2006-04-14 22:03 ` David S. Miller
  1 sibling, 0 replies; 3+ messages in thread
From: David S. Miller @ 2006-04-14 22:03 UTC (permalink / raw)
  To: hadi; +Cc: netdev, herbert, hidden

From: jamal <hadi@cyberus.ca>
Date: Thu, 13 Apr 2006 08:55:58 -0400

> Finally got an opportunity to test this. I have incorporated all the
> feedback and it passes all the scenarios i know.
> 
> Dave, please apply to Linus tree since this is a bug fix.

Applied, thanks.

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2006-04-14 22:03 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-04-13 12:55 [XFRM] Fix aevent timer jamal
2006-04-13 13:46 ` Herbert Xu
2006-04-14 22:03 ` David S. Miller

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).