#18483: DIR-825 running CC r43485 - WIFI cannot handle IRQ 40
-----------------------+------------------------
Reporter: papaj0e | Owner: developers
Type: defect | Status: new
Priority: normal | Milestone:
Component: packages | Version: Trunk
Resolution: | Keywords:
-----------------------+------------------------
Comment (by papaj0e):
I can confirm that after reverting the patch "ath9k: do not access
hardware on IRQs during reset" and running trunk r43901 for almost two
days I'm not able to trigger this bug anymore.
And this is after heavy wireless traffic load with almost 200 million
interrupts fired:
{{{
root@OpenWrt:/tmp# cat /proc/interrupts
CPU0
3: 0 MIPS ehci_hcd:usb1
4: 34298818 MIPS eth0
5: 192964035 MIPS eth1
7: 16214486 MIPS timer
11: 48 MISC serial
14: 0 MISC ohci_hcd:usb2
40: 192117084 AR71XX PCI ath9k
41: 12490162 AR71XX PCI ath9k
ERR: 0
}}}
The only issue I have is the infamous "Failed to stop TX DMA" one:
{{{
[ 1831.230000] ath: phy0: Failed to stop TX DMA, queues=0x002!
[46139.170000] ath: phy0: Failed to stop TX DMA, queues=0x002!
[46139.280000] ath: phy0: Failed to stop TX DMA, queues=0x102!
[46139.380000] ath: phy0: Failed to stop TX DMA, queues=0x002!
[46139.480000] ath: phy0: Failed to stop TX DMA, queues=0x002!
[46139.580000] ath: phy0: Failed to stop TX DMA, queues=0x002!
[46139.680000] ath: phy0: Failed to stop TX DMA, queues=0x002!
[46139.790000] ath: phy0: Failed to stop TX DMA, queues=0x002!
[46139.890000] ath: phy0: Failed to stop TX DMA, queues=0x002!
[46139.990000] ath: phy0: Failed to stop TX DMA, queues=0x002!
[144410.320000] ath: phy0: Failed to stop TX DMA, queues=0x002!
}}}
So it seems to me that this patch is the culprit and it should be revised
by ath9k developers.
--
Ticket URL: <https://dev.openwrt.org/ticket/18483#comment:30>
OpenWrt <http://openwrt.org>
Opensource Wireless Router Technology
_______________________________________________
openwrt-tickets mailing list
[email protected]
https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-tickets
