On Sun, 2013-06-30 at 16:30 +0100, Ian Campbell wrote: > Which could be destructive, for all we know, since we don't know what is > on the other end nor what it will do. Well but AFAIU, right now you also just check for gpio_keys, right? And anything else could have added that as well?! Also there is no guarantee that the order of ttyS0 / S1 is always the same...
> > > Even once we know it is a QNAP I don't think we can safely just probe > > > serial ports looking for things. > > Well... I would make the devices configurable and fully put it into the > > user's responsibility to set up a qcontrol.conf... > > Yes, I think on x86 the default qcontrol.conf will be empty and the user > will have to configure it. Or maybe I'll stick in a debconf question. Yeah... sounds ok... > It's all a bit moot until there is working support for at least one x86 > platform though. Well I don't see that coming unfortunately.... I mean the A125 is working... so we could support at least that... but I couldn't get the rest working via serial console (either I make something wrong... or it simply doesn't work anymore)... and I really would have preferred that over a real kernel modules. You don't have any good contact at QNAP, do you? And for the kernel level support... I've had started some discussion at lkml[0], but in the end it came down to: not going to happen if QNAP doesn't implement it properly. Which brings me to some vague contact[1] I've had with QNAP itself (they didn't answer any of my mails)... but all they said was they "will include [my] suggestions for internal discussions" Well,... not sure if the A125 support is enough for you to build it for non-arm... it think though it would be worth it. Cheers, Chris. [0] http://thread.gmane.org/gmane.linux.kernel/1508763 [1] https://plus.google.com/102329811815220905914/posts/6ewT8DmCPxd scroll down in the comments
smime.p7s
Description: S/MIME cryptographic signature
