Re: Polling and kern.polling.idle_poll

2009-07-03 Thread Barney Cordoba



--- On Thu, 7/2/09, Larry Baird  wrote:

> From: Larry Baird 
> Subject: Re: Polling and kern.polling.idle_poll
> To: "Barney Cordoba" 
> Cc: freebsd-net@freebsd.org, "Paul Thornton" 
> Date: Thursday, July 2, 2009, 11:31 AM
> > Curious as to why you don't
> simply lay out the $40 for a modern ethernet card that
> doesn't need polling to work well?
> These are low end embedded boxs.  No way to
> upgrade.  But in actuallity
> they have modern (fxp) NICS.  From our testing, live
> lock is a real issue
> for gateways receiving lots of network traffic. 
> Polling prevents live
> lock.

There are few applications where dropping buckets of packets is preferable to 
losing the keyboard. If "livelock is a problem", then your systems are too slow 
for the task.

Livelock can easity be tuned with the "work" parameter. fxp drivers are hard 
coded to (I think) 6000 ints per second, so its not rocket science to set your 
tuning parameters to whatever the system can handle before entering a livelock 
scenario. In reality, interrupt moderation is the equivalent of polling 6000 
times per second, in the case of fxp. You wouldn't want to set hz to 6000 as 
you substantially increase system overhead. Moderation is a much more efficient 
way of managing your ethernet loads, because the "polls" are only directed at 
the driver, and if you dont have traffic then you don't  need to get polled.

Barney



___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"


Re: Slow local TCP transfers on -CURRENT

2009-07-03 Thread Lawrence Stewart

Hi Qing,

Qing Li wrote:
Qing (added to CC) is aware of the problem. Not sure how far 
off the fix is.




I am resuming the work on it, hoping to get it verified and
finalized in a day or so. Sorry about the delay.


Did this ever get fixed? I don't think I saw a follow up to this thread 
so just thought I'd check.


Cheers,
Lawrence
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"


Re: ndis and USB wirelless ethernet

2009-07-03 Thread Nikos Vassiliadis

Hello Weongyo & net,

I did some tests with an AP. The bad news is that I am
getting random crashes. At least I cannot relate them
to something... It just stops responding, I cannot break
to the debugger. Would I try another driver? there are
3 versions if I recall correctly on the vendor's site.

Something strange I've noticed, is that if I set the ssid
before setting the interface to "up" state, it does not
associate with the AP.

ifconfig wlan0 down
ifconfig wlan0 ssid barba
ifconfig wlan0 up

> Setting BSSID to 00:1b:2f:be:78:aa
> ndis_newstate: INIT -> SCAN
> ndis0: link state changed to DOWN
> ndis0: link state changed to UP
> ndis_scan_results: 4 results
> scan: bssid 00:1c:a2:b2:4e:e5 chan 2412MHz (1/1152) rssi 25
> scan: bssid 00:13:33:07:90:90 chan 2442MHz (7/1152) rssi 75
> scan: bssid 00:05:59:0a:b7:c1 chan 2437MHz (6/1152) rssi 60
> scan: bssid 00:1a:4f:20:dc:4a chan 2437MHz (6/1152) rssi 6
> wlan0: ieee80211_new_state_locked: pending SCAN -> RUN transition lost
> ndis_newstate: SCAN -> AUTH
> Setting BSSID to 00:1b:2f:be:78:aa
> Setting BSSID to ff:ff:ff:ff:ff:ff
> Setting ESSID to "barba"
> ndis_newstate: AUTH -> ASSOC

If if delete the ssid, and "up" the interface, it
associates:

> Setting BSSID to 00:1b:2f:be:78:aa
> ndis_newstate: INIT -> SCAN
> ndis0: link state changed to DOWN
> ndis_scan_results: 4 results
> scan: bssid 00:13:33:07:90:90 chan 2442MHz (7/1152) rssi 73
> scan: bssid 00:1c:a2:b2:4e:e5 chan 2412MHz (1/1152) rssi 15
> scan: bssid 00:1a:4f:20:dc:4a chan 2437MHz (6/1152) rssi 7
> scan: bssid 00:05:59:0a:b7:c1 chan 2437MHz (6/1152) rssi 54
> ndis_newstate: SCAN -> AUTH
> Setting BSSID to 00:1b:2f:be:78:aa
> Setting BSSID to ff:ff:ff:ff:ff:ff
> Setting ESSID to "barba"
> ndis_newstate: AUTH -> ASSOC
> ndis_newstate: ASSOC -> RUN
> ndis0: link state changed to UP

If you have some idea, I would be happy to test it.

Thanks, Nikos
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"


tcputo(RFC 5482) can work now

2009-07-03 Thread Fang Wang
Hi,

The tcp UTO option described in RFC 5482 now is accomplished. The
branch is //depot/projects/soc2009/tcputo/.

Regards,
Fang Wang
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"


CFR: 802.11s implementation (mesh11s project)

2009-07-03 Thread Rui Paulo

Hi,
If you would like to review the mesh11s code, you can find a diff  
against HEAD here:

http://people.freebsd.org/~rpaulo/mesh_195323.diff

This is the most recent diff, but expect changes. We will present the  
re@ team with a patch so we can discuss the inclusion of this in the  
8.0 release.


Thanks,
--
Rui Paulo

___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"


Re: kern/134557: [netgraph] [hang] 7.2 with mpd5.3 hanging up - ng_pptp problem

2009-07-03 Thread mav
Synopsis: [netgraph] [hang] 7.2 with mpd5.3 hanging up - ng_pptp problem

State-Changed-From-To: open->closed
State-Changed-By: mav
State-Changed-When: Sat Jul 4 06:01:31 UTC 2009
State-Changed-Why: 
Patch committed and merged to 7-SATBLE.

http://www.freebsd.org/cgi/query-pr.cgi?pr=134557
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"


Re: kern/134557: commit references a PR

2009-07-03 Thread dfilter service
The following reply was made to PR kern/134557; it has been noted by GNATS.

From: dfil...@freebsd.org (dfilter service)
To: bug-follo...@freebsd.org
Cc:  
Subject: Re: kern/134557: commit references a PR
Date: Sat,  4 Jul 2009 06:01:15 + (UTC)

 Author: mav
 Date: Sat Jul  4 06:00:57 2009
 New Revision: 195336
 URL: http://svn.freebsd.org/changeset/base/195336
 
 Log:
   MFC rev.195335:
   Fix infinite loop in ng_iface, that happens when packet passes out via
   two different ng interfaces sequentially due to tunnelling.
   
   PR:  kern/134557
   Submitted by:Mikolaj Golub
 
 Modified:
   stable/7/sys/   (props changed)
   stable/7/sys/contrib/pf/   (props changed)
   stable/7/sys/netgraph/ng_iface.c
 
 Modified: stable/7/sys/netgraph/ng_iface.c
 ==
 --- stable/7/sys/netgraph/ng_iface.c   Sat Jul  4 05:07:52 2009
(r195335)
 +++ stable/7/sys/netgraph/ng_iface.c   Sat Jul  4 06:00:57 2009
(r195336)
 @@ -365,7 +365,8 @@ ng_iface_output(struct ifnet *ifp, struc
}
  
/* Protect from deadly infinite recursion. */
 -  while ((mtag = m_tag_locate(m, MTAG_NGIF, MTAG_NGIF_CALLED, NULL))) {
 +  mtag = NULL;
 +  while ((mtag = m_tag_locate(m, MTAG_NGIF, MTAG_NGIF_CALLED, mtag))) {
if (*(struct ifnet **)(mtag + 1) == ifp) {
log(LOG_NOTICE, "Loop detected on %s\n", ifp->if_xname);
m_freem(m);
 ___
 svn-src-...@freebsd.org mailing list
 http://lists.freebsd.org/mailman/listinfo/svn-src-all
 To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
 
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"