Thank you Adrian,

Could you point me towards the reference driver? Also, any idea where in
ath9k is a good place to check for decryption failures?
Instrumenting ath9k_cmn_rx_accept in ath9k/common.c showed some decrypt
errors, but it was about same amount for working and broken states.

Alexis

On Thu, May 14, 2015 at 5:49 PM, Adrian Chadd <[email protected]> wrote:

> Hi,
>
> Try writing something that periodically checks if the key is in there.
> I have a feeling that I've seen this bug in the reference driver, and
> they "fixed" it by noticing decrypt failures for a given destination
> and replumbing the key if it didn't match. :(
>
> (I always wondered if this would ever surface as a real issue..)
>
>
> -adrian
>
_______________________________________________
ath9k-devel mailing list
[email protected]
https://lists.ath9k.org/mailman/listinfo/ath9k-devel

Reply via email to