#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 anonymous):
Replying to [comment:175 foodoc]:
> Hmm, this seems to be a problem only when IEEE80211_DEBUG_REFCNT is
disabled...?! either the non-debug code is incomplete, or there's
somewhere a tiny race.
I've noticed that some revisions ago there were comments regarding buffer
changes. I expected to see this bug get solved by these changes, but I
noticed that the overruns came earlier and very frequently. After seeing
the note to build the module with debug support I did it (so to try to
help with debug info) and never got an overrun again. I've let the machine
stay on for almost 3 days and used it every day like I did before and the
connection is always working when I pick the laptop to use.
I'd suggest to the anonymous #176 to rebuild the module with
IEEE80211_DEBUG_REFCNT=1 and check if the overruns occur, if they don't
maybe the devs can track where's the difference between debug code and
normal code that can be triggering this problem.
--
Ticket URL: <http://madwifi.org/ticket/1017#comment:178>
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
