linux-kernel - Re: [PATCH v2 4/4] usb: Add APIs to access host registers from Tegra PHY

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: <Pine.LNX.4.44L0.1301171016550.1339-100000@iolanthe.rowland.org>
Date:	Thu, 17 Jan 2013 10:21:53 -0500 (EST)
From:	Alan Stern <stern@...land.harvard.edu>
To:	Felipe Balbi <balbi@...com>
cc:	Venu Byravarasu <vbyravarasu@...dia.com>,
	"gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>,
	"linux-usb@...r.kernel.org" <linux-usb@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"swarren@...dotorg.org" <swarren@...dotorg.org>,
	"linux-tegra@...r.kernel.org" <linux-tegra@...r.kernel.org>
Subject: Re: [PATCH v2 4/4] usb: Add APIs to access host registers from Tegra
 PHY

On Thu, 17 Jan 2013, Felipe Balbi wrote:

> > Bits 31 & 30 from PORTSC register were allocated by our SOC designers
> > to inform the host controller about the PHY type to be used. 
> 
> Wow, that's something you should never do. PORTSC register belongs to
> the EHCI controller and those bits are reserved for future use and they
> *MUST* return zero. I wouldn't be surprised if current EHCI driver
> assumes those bits will be zero and/or makes sure they're set to zero
> when writing to PORTSC register.

In fact, those bits _have_ been assigned an official purpose in the 
EHCI-1.1 addendum.  Presumably the Tegra hardware only supports 
EHCI-1.0.

> What if after configuring those two bits, ehci-hcd.ko clears them by
> accident ? Your platform won't work.

That won't happen; ehci-hcd is careful to preserve the values of PORTSC
register bits it doesn't use.  But this still isn't a good idea.

> > As type of PHY is a property related to PHY DT nodes, PHY driver will get this info.
> > (Will remove phy_type from controller DT node soon, after all patches get merged.)
> > However as PORTSC register is in controller domain, added tegra_ehci_set_pts() API
> >  to serve the purpose.
> > 
> > As per Tegra USB PHY design, need to wait for PHY clock to stabilize once we
> > set/clear PHCD bit. Hence this is being accessed from PHY driver. To serve this
> > purpose added tegra_ehci_set_phcd().
> > 
> > On further analysis, seems tegra_ehci_set_wakeon_events() can be
> > removed.
> > 
> > If you are okay with above explanation, I can send updated patch for
> > review.
> 
> not sure I want to sign-off such a patch. I understand it's how your HW
> was done, but this shouldn't have been done, really.
> 
> Alan, what do you think ? Are you ok with PHY driver overwritting PORTSC
> register ?

It's okay provided it is done correctly -- i.e., in a way that won't 
interfere with the normal EHCI operation.

Alan Stern

--
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/