#1017: "rx FIFO overrun" prevents traffic from flowing
------------------------------------+---------------------------------------
Reporter: [EMAIL PROTECTED] | Owner:
Type: defect | Status: new
Priority: critical | Milestone:
Component: madwifi: driver | Version: trunk
Resolution: | Keywords:
Patch_attached: 0 |
------------------------------------+---------------------------------------
Comment (by foodoc):
Replying to [comment:161 mdtaylor]:
> foodoc, can you run "iwpriv athX debug_scanbufs 1" command for me when
it's happening and tell me if any rxbuffers are leaked? If none leaked,
then this is just you punishing the box and you can try increasing the rx
buffer count or decreasing other CPU load. Otherwise, I'd like to know
the line numbers where the buffers were taken that had leaked.
Allright! I said I'll report back... After a full day of iperf on a
faster/less loaded machine I didn't see any leaks or RX FIFO overruns
anymore.
Next stop: without the debug-code, because it seems to _eat_ nearly
10mbits (out of 28mbits).
Thanks,
foodoc
--
Ticket URL: <http://madwifi.org/ticket/1017#comment:166>
madwifi.org <http://madwifi.org/>
Multiband Atheros Driver for Wireless Fidelity
-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
Madwifi-tickets mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/madwifi-tickets
