linux-kernel - Re: [PATCH] usb: musb: dsps: handle the otg_state_a_wait_vrise

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]

Message-ID: <87h9jtm37m.fsf@saruman.tx.rr.com>
Date:	Tue, 8 Dec 2015 08:20:45 -0600
From:	Felipe Balbi <balbi@...com>
To:	Gregory CLEMENT <gregory.clement@...e-electrons.com>
CC:	<b-liu@...com>, Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	<linux-usb@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
	<linux-arm-kernel@...ts.infradead.org>, <stable@...r.kernel.org>
Subject: Re: [PATCH] usb: musb: dsps: handle the otg_state_a_wait_vrise_timeout case


Hi,

Gregory CLEMENT <gregory.clement@...e-electrons.com> writes:
>>>> if it is the case then it didn't fix the issue I had.
>>>>
>>>> I activated the following debug line:
>>>>
>>>> [musb_hdrc]musb_interrupt =_ "** IRQ %s usb%04x tx%04x rx%04x\012"
>>>> [musb_dsps]dsps_interrupt =p "usbintr (%x) epintr(%x)\012"
>>>>
>>>> But I didn't get any interrupt while disconnecting the cable without any
>>>> device connected on it (whereas I got an interrupt when I connected it).
>>>>
>>>> Note that I applied this patch instead of the "usb: musb: dsps: handle
>>>> the otg_state_a_wait_vrise_timeout case", is what you had in mind ?
>>
>> yeah, that's what I had in mind. But your patch seems wrong :-)
>>
>> I tried writing a more correct version here and found 2 issues:
>>
>> a) bit 3 doesn't do anything :-p I cannot read IRQs from mentor's
>> registers
>>
>> b) when setting RESET_ISOLATION bit, reads of CTRL register hang. Note
>> that according to TRM, RESET_ISOLATION _must_ be set prior to a soft
>> reset and cleared afterwards. But right after setting RESET_ISOLATION,
>> if I try a read of CTRL, it'll hang forever.
>
> The datasheet seems not very coherent about it,
>
> on one side we have:
> "This bit should be set high prior to setting bit 0 and cleared after bit 0
> is cleared."
>
> and on the other side:
> "Both the soft_reset and soft_reset_isolation bits should be asserted
> simultaneously."
>
> The hang you saw could be explained by the following:
> "Setting only the soft_reset_isolation bit will cause all USB0 output
> signals to go to a known constant value via multiplexers.
> This will
> prevent future access to USB0."  page 2567

good catch. Setting them together makes the hang go away.

I still have the other problem, which is legacy IRQ reporting mode not
really working.

-- 
balbi

Download attachment "signature.asc" of type "application/pgp-signature" (819 bytes)