[ath9k-devel] need help debugging/pinning down obscure freezes/hangs using AR9287

[Felix Fietkau <nbd@openwrt.org>]

On 2011-10-12 5:35 PM, Peter Stuge wrote:
>>  >>  would you please stop complaining and provide some help ?
>>  >
>>  >  It took me many months time to get an idea of the ath9k community,
>>  >  and I would have appreciated tremendously if someone had explained
>>  >  to me how things worked and what I could expect, so that I could
>>  >  take independent action accordingly. I believe I help people when I
>>  >  talk about how this community works.
>>
>>  sorry, this community works much better than what you speak about it.
>
> The last thing that happened that comes to mind is that Luis asked an
> open source project developer to not mention a particular Atheros
> software name on public mailing lists. The *name* of the software!
>
> The kind of information flow I expected when subscribing to the list
> is all about register details, meanings of bits, how the on-chip
> firmware works and can be controlled, and every other technical
> detail about Atheros hardware that is relevant for working with
> driver development. You know very well that this discussion does not
> happen on the public mailing list, and thus not within the ath9k
> community.
If I understand the point you're trying to make here correctly, it 
sounds like complete BS to me. Let me get this straight: from Luis 
asking people not to refer to internal codenames of Atheros codebases 
you extrapolate that detailed technical discussion on the public lists 
is unwanted? That's ridiculous!

> I've understood that there's also another community, with developers,
> who do get to talk about developer level details, but the barrier of
> entry to that community is higher than simply subscribing to a
> mailing list, and approaching ath9k naively as I did it makes no
> sense to divide resources like that for a wifi driver.
The barrier of entry is there for a reason. Having detailed information 
about the inner workings of the hardware is useless if you don't have a 
basic understanding of how the driver works. For many of the bugs that 
were fixed, you don't actually need any kind of detailed hardware info 
to fix them.

What ath9k needs most is people that are willing to spend enough time 
reading the driver and playing with the code until they understand the 
software side well enough to be able to move on to looking into the 
hardware. The people that I know that did spend enough time on the 
driver to understand the software side also did get direct help from 
Atheros and/or access to hardware documentation.

Unfortunately with many questions that are posted to these lists, the 
effort that went into formulating the question is just a microscopic 
fraction of the effort required to provide a good answer. Also, with 
many of these questions I get the impression that providing an answer 
would be a complete waste of time, since it does not appear that 
anything useful would come out of it. Such questions are usually easily 
recognized, since they're often one-liners like:
'What does $random_register do?' or 'What's the purpose of 
$random_feature?', or even things like 'How do I enable 
$random_unsupported_feature?'

And even with competent users knowing Linux well, it can be quite hard 
to walk them through the right steps to track down bugs, because often 
it's very hard even for people with access to all hardware docs to even 
identify a likely source or general area of the bug. If such a bug can 
then not be reproduced internally, there is no good way to it, and it 
may take some time for people to come up with good ideas.

So if somebody from the community is showing *serious* effort at fixing 
issues or improving ath9k and needs to know about details of how the 
hardware works, there are enough people (including myself) from inside 
or outside of Atheros that are usually helpful with filling the gaps.

- Felix