linux-kernel - Re: [PATCH 14/16] usb: musb: dsps: add MUSB_DEVCTL

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Mon, 29 Jul 2013 13:41:38 -0500
From:	Bin Liu <binmlist@...il.com>
To:	Sebastian Andrzej Siewior <bigeasy@...utronix.de>
Cc:	linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org,
	balbi@...com, george.cherian@...com
Subject: Re: [PATCH 14/16] usb: musb: dsps: add MUSB_DEVCTL_SESSION back after removal

Sebastian,

On Mon, Jul 29, 2013 at 12:51 PM, Sebastian Andrzej Siewior
<bigeasy@...utronix.de> wrote:
> On 07/29/2013 07:26 PM, Bin Liu wrote:
>> Sebastian,
>
> Hi Bin,
>
>> I did not say AM335xEVM is not able to run OTG mode. The problem is in
>> OTG mode the SESSION bit will be cleared once the device is unplugged,
>> then there is no way the SESSION bit will come back in the current
>> mainline kernel.
>
> Yes, you did.

If I did, that was wrong statement.

>
>>
>> The TI 3.2 kernel works around this OTG issue by toggling the SESSION
>> bit in b_idle handling in otg_timer(). The workaround makes most users
>> happy but it causes VBUS voltage pulsing. That is why I said I don't
>> know an ideal solution without a ID pin interrupt support.
>
> Now I understand more pieces of the puzzle. And the missing ID pin
> support is Am335x-evm specific since other am335x board may wire up
> that pin.

No, it is not board wiring issue. The AM335x EVM DOES have ID pin
wired to the ID pin of the micro-AB receptacle. I said AM335x does not
have IP pin INTERRUPT support, by which the kernel does not know when
a device is plugged in without setting the SESSION bit.

> But you say there is no way that the phy or anything else could help to
> avoid that pulsing.
>
>> I think you looked at a wrong file, maybe a wrong branch. Please check
>> [1], which defines
>>
>> 2644          * mode[0:3] = USB0PORT's mode
>> 2645          * mode[4:7] = USB1PORT's mode
>> 2646          * AM335X beta EVM has USB0 in OTG mode and USB1 in host mode.
>> 2647          */
>> 2648         .mode           = (MUSB_HOST << 4) | MUSB_OTG,
>
> Oh it is a long way. So looked into the wrong branch.
>
>>> I agree that without that try_idle part things keep working but it
>>> seems, that it is also called in the other tree.
>>>
>>
>> Then because AM335xEVM USB1 port does not call *try_idle() in this 3.2
>> kernel, its SESSION bit is always set.
>>
>> But I don't have a clear idea how to solve this in the mainline kernel
>> since is_otg_enabled() has been cleaned.
>
> So now I think this is a board problem. What about switching port 1
> from OTG to HOST only and avoiding kicking the timer in
> dsps_musb_try_idle() in such a case? It seems we only need to do this
> in OTG mode.

I have not read the MUSB driver in the mainline kernel yet. If somehow
port 1 can could be set to HOST mode then the SESSION bit would stay,
then that should solve the issue.

>
>> Regards,
>> -Bin.
>
> Sebastian
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/