[PATCH] i.MX51: Full iomux support

[Nguyen Dinh-R00091]

Hi,

>-----Original Message-----
>From: Arnaud Patard [mailto:arnaud.patard at rtp-net.org]
>Sent: Wednesday, December 15, 2010 9:37 AM
>To: Lothar Waßmann
>Cc: Sascha Hauer; linux-arm-kernel at lists.infradead.org; Richard Zhao; Nguyen Dinh-R00091; Jason Wang
>Subject: Re: [PATCH] i.MX51: Full iomux support
>
>Lothar Waßmann <LW at KARO-electronics.de> writes:
>
>> Hi,
>>
>> Arnaud Patard (Rtp) writes:
>>> Sascha Hauer <s.hauer at pengutronix.de> writes:
>> [...]
>>> > The following series picks up the patch from Lothar Waßmann replacing
>>> > the struct pad_desc with a 64bit variable and adds full i.MX51 iomux
>>> > support based on this patch.
>>> > The iomux configurations are taken from the Freescale pinmux tool, so the
>>> > definitions should be rather complete. Anyway, there are some modes
>>> > not present in the tool.
>>> > I took the padmux settings from the old iomux support where present.
>>>
>>> I'm seeing a lot of changes in the iomux file and I have a patch [1]
>>> setting the SION bit for all gpios configuration so that reading the PSR
>>> value is giving usefull results when the gpio is configured as output. I
>>> wanted to send it this week but it will obviously conflict with
>>> this. Should I test your patchset first and if it's working, wait
>>> for its merge or should I send it anyway ? How do you want to deal with
>>> this (as long as you're fine with setting the SION bit for all gpios) ?
>>>
>> I had done the same, but had some trouble with this.
>> E.g. on our board GPIO1_7 is used as a generic GPIO to enable an
>> external clock oscillator for the USBH1 ULPI PHY. When the SION bit
>> for this pad was set, I got strange errors on the USBH1 port
>> (disconnecting low speed devices behind a hub would stall the
>> bus). When I removed the SION bit for that pin everything worked
>> well.
>
>wow. Would be nice to have feedback from fsl. I was fearing of bad
>effects but I didn't think they could actually break something so
>badly. After all, the manual doesn't warn about possible side effects of
>the SION bit iirc. This would mean this will have to be dealt on a board
>basis. It would be... erm... not really "efficient".
>
>Arnaud

You cannot blindly set the SION bit for all GPIO's. I have to dig through my email history for a more technical explanation.

Thanks,
Dinh