33 messages in org.freebsd.freebsd-currentRe: dev.bce.X.com_no_buffers increasi...| From | Sent On | Attachments |
|---|---|---|
| Ian FREISLICH | Mar 5, 2010 3:20 am | |
| Pyun YongHyeon | Mar 5, 2010 9:56 am | |
| Ian FREISLICH | Mar 5, 2010 10:16 am | |
| Pyun YongHyeon | Mar 5, 2010 10:40 am | |
| Ian FREISLICH | Mar 5, 2010 12:19 pm | |
| Pyun YongHyeon | Mar 5, 2010 1:04 pm | |
| Ian FREISLICH | Mar 5, 2010 1:16 pm | |
| Pyun YongHyeon | Mar 5, 2010 1:55 pm | |
| Ian FREISLICH | Mar 8, 2010 6:44 am | |
| Pyun YongHyeon | Mar 8, 2010 9:49 am | |
| Ian FREISLICH | Mar 9, 2010 12:26 am | |
| Ian FREISLICH | Mar 9, 2010 5:31 am | |
| Pyun YongHyeon | Mar 9, 2010 12:49 pm | |
| Pyun YongHyeon | Mar 9, 2010 1:21 pm | |
| David Christensen | Mar 9, 2010 1:31 pm | |
| Pyun YongHyeon | Mar 9, 2010 1:39 pm | |
| Ian FREISLICH | Mar 9, 2010 1:55 pm | |
| David Christensen | Mar 9, 2010 2:04 pm | |
| Pyun YongHyeon | Mar 9, 2010 2:12 pm | |
| Ryan Stone | Mar 9, 2010 2:30 pm | |
| Fabien Thomas | Mar 9, 2010 2:55 pm | |
| David Christensen | Mar 9, 2010 3:00 pm | |
| Ryan Stone | Mar 9, 2010 3:07 pm | |
| Ian FREISLICH | Mar 9, 2010 9:47 pm | |
| Ian FREISLICH | Mar 10, 2010 1:04 am | |
| David Christensen | Mar 10, 2010 11:10 am | |
| Pyun YongHyeon | Mar 10, 2010 11:51 am | |
| David Christensen | Mar 10, 2010 2:45 pm | |
| Pyun YongHyeon | Mar 10, 2010 3:01 pm | |
| Ian FREISLICH | Mar 10, 2010 10:45 pm | |
| Ian FREISLICH | Mar 10, 2010 11:05 pm | |
| David Christensen | Mar 12, 2010 3:58 pm | |
| Ian FREISLICH | Mar 13, 2010 9:05 am |
| Subject: | Re: dev.bce.X.com_no_buffers increasing and packet loss | |
|---|---|---|
| From: | Ian FREISLICH (ia...@clue.co.za) | |
| Date: | Mar 10, 2010 1:04:59 am | |
| List: | org.freebsd.freebsd-current | |
"David Christensen" wrote:
Yeah, but the question is why bce(4) has no available RX buffers. The system has a lot of available mbufs so I don't see the=20 root cause here.
What's the traffic look like? Jumbo, standard, short frames? Any=20 good ideas on profiling the code? I haven't figured out how to use the CPU TSC but there is a free running timer on the device that might be usable to calculate where the driver's time is spent.
It looks like the traffic that provoked it was this:
10:18:42.319370 IP X.4569 > X.4569: UDP, length 12 10:18:42.319402 IP X.4569 > X.4569: UDP, length 12 10:18:42.319438 IP X.4569 > X.4569: UDP, length 12 10:18:42.319484 IP X.4569 > X.4569: UDP, length 12 10:18:42.319517 IP X.4569 > X.4569: UDP, length 12
A flurry of UDP tinygrams on an IAX2 trunk. The packet rate isn't spectacular at about 30kpps which on top of the base load of 60kpps still isn't a fantastic packet rate. The interesting thing is that while this storm was inprogress, it almost entirely excluded other traffic on the network.
There have been reports of backplane congestion on the switches we use when UDP packets smaller than 400 bytes arrive within 40us of eachother. But that still doesn't explain the counter increases and high interrupt CPU usage, unless the switch was producing garbage output in response.
Ian
-- Ian Freislich
_______________________________________________ free...@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "free...@freebsd.org"