Re: Unwanted delayed execution of _Qxx EC methods
From: Gabriele Mazzotta
Date: Fri Mar 06 2015 - 07:42:29 EST
On Friday 06 March 2015 01:52:48 Zheng, Lv wrote:
> Hi, Gabriele
>
>
> I couldn't find this in my mail box, but saw it in the spinics.net.
>
> For EC query, there is no spec definitions around its behavior.
> Some EC firmware will have events queued (like edge triggering) while the others will keeps on reporting events when a condition is set (like level triggering).
> I think this case is a "level triggering like events", it is very useful for us to learn.
> We should be able to handle both firmware variations using one software architecture.
> Supporting this becomes possible now because we are able to switch EC handling between polling/interrupt modes using enhanced ACPICA GPE APIs.
> So we are able to switch EC into polling mode when SCI_EVT is detected and blocks further events and handles all transactions in polling mode.
> After completing the _Qxx evaluation we then can switch back to the interrupt mode and unblock further events.
>
> Could you please open a kernel Bugzilla entry at: https://bugzilla.kernel.org
> So we can discuss there with more test cases considered.
>
> Thanks and best regards
> -Lv
Hi,
thanks for the explanation.
I opened a bug report as requested, here the address:
https://bugzilla.kernel.org/show_bug.cgi?id=94411
Regards,
Gabriele
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/