linux-kernel - Re: [PATCH] USB: initialize or shutdown PHY when add or remove host controller

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:	Tue, 18 Jun 2013 18:00:50 +0300
From:	Felipe Balbi <balbi@...com>
To:	Alan Stern <stern@...land.harvard.edu>
CC:	Felipe Balbi <balbi@...com>, Roger Quadros <rogerq@...com>,
	Chao Xie <chao.xie@...vell.com>, <gregkh@...uxfoundation.org>,
	<linux-usb@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
	<linux-arm-kernel@...ts.infradead.org>, <xiechao.mail@...il.com>
Subject: Re: [PATCH] USB: initialize or shutdown PHY when add or remove host
 controller

Hi,

On Tue, Jun 18, 2013 at 10:53:26AM -0400, Alan Stern wrote:
> > > If the controllers don't want HCD core to manage the PHY they can just set it
> > > to some error code.
> > 
> > they shouldn't have the choice, otherwise it'll be a bit of a PITA to
> > maintain the code. ehci core tries to grab the PHY, if it's not there,
> > try to continue anyway. Assume it's not needed.
> 
> Felipe, are all these issues straightened out to your satisfaction?  I
> can't tell from the email thread.
> 
> Looking through the EHCI glue source files, there appears to be a 
> variety of different ways of getting the PHY.  It's not at all clear 
> that they can be moved into the ehci-hcd core (or even better, the USB 
> core -- which is what Chao is trying to do).

yeah, Roger brought up a big problem with OMAP's EHCI depending on the
mode so, at least for now, we should keep phy_get and, in case of EHCI
OMAP, phy_init in the glue :-(

Roger has all the details, and they're also in the list archives, but
basically, depending on the mode, PHY *must* be initialized at a
particular point.

> Given that the glue module has to be responsible for getting the PHY,
> it should also be responsible for error checking.  So the code added to
> hcd.c doesn't need to apply an IS_ERR check; it can simply assume that
> if hcd->phy is NULL then either there is no software-controllable PHY
> or else the HCD doesn't want the core to manage it.

makes sense to me, add the requirement to:

if (IS_ERR(hcd->phy))
	hcd->phy = NULL;

-- 
balbi

Download attachment "signature.asc" of type "application/pgp-signature" (837 bytes)