[Cerowrt-devel] Fwd: [Cerowrt - Bug #437] (New) bug in initializing wifi interfaces
dave.taht at gmail.com
Sun Dec 15 18:31:54 EST 2013
On the minus side we still have trouble bringing up all the interfaces
on second boot and later. So I've filed a bug on it, and am
making the firmware available for all and sundry to suffer with
on this crisp/clear sunday.
And I can blow up the driver still with the rrul test
[ 1024.917968] ath: phy0: Failed to stop TX DMA, queues=0x00e!
[ 1028.972656] ath: phy0: Failed to stop TX DMA, queues=0x00e!
[ 1332.210937] ath: phy0: Failed to stop TX DMA, queues=0x00e!
but so far, not with the rrul_unclassified test.
Lastly download performance under rrul has become dismal for some reason
since last I tried it (about august on 2ghz)
On the plus side, sebastian's latest AQM changes made it in under the wire
with (no kidding) 30 seconds to spare, and I can't crash the gui anymore.
I'm going to go see the hobbit. My money's on smaug for roast dwarf.
---------- Forwarded message ----------
From: <cerowrt at lists.bufferbloat.net>
Date: Sun, Dec 15, 2013 at 3:00 PM
Subject: [Cerowrt - Bug #437] (New) bug in initializing wifi interfaces
Issue #437 has been reported by David Taht.
Bug #437: bug in initializing wifi interfaces
Author: David Taht
Since about cerowrt 3.10.18 there has been a bug in initializing the
It looks like a set of race conditions in netfd. On initial boot
(while competing with a jffss init,
it generally does better in bringing up more interfaces) but on the
second boot, it's anybody's guess as to what will come up.
(there is a similar problem in starting dnsmasq)
The problem persists in cerowrt 3.10.24-4 which is built from openwrt svn 39076
You have received this notification because you have either subscribed
to it, or are involved in it.
To change your notification preferences, please click here:
Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html
More information about the Cerowrt-devel