linux-kernel - Re: [PATCH 3/3] dt: tegra20: Add ehci nodes to Seaboard.

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] [day] [month] [year] [list]

Date:	Tue, 19 Jul 2011 17:14:04 -0700
From:	Olof Johansson <olof@...om.net>
To:	Andrew Chew <AChew@...dia.com>
Cc:	Stephen Warren <swarren@...dia.com>,
	"grant.likely@...retlab.ca" <grant.likely@...retlab.ca>,
	Dan Willemsen <dwillemsen@...dia.com>,
	Rhyland Klein <rklein@...dia.com>,
	Venkat Moganty <vmoganty@...dia.com>,
	"devicetree-discuss@...ts.ozlabs.org" 
	<devicetree-discuss@...ts.ozlabs.org>,
	"linux-tegra@...r.kernel.org" <linux-tegra@...r.kernel.org>,
	"linux-usb@...r.kernel.org" <linux-usb@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 3/3] dt: tegra20: Add ehci nodes to Seaboard.

On Tue, Jul 19, 2011 at 5:07 PM, Andrew Chew <AChew@...dia.com> wrote:
>> > And since there are defaults specified in tegra20.dtsi,
>> does it really make sense to also have default values
>> assigned in ehci-tegra.c (for when a property is not
>> present)?  I worry that the information is now duplicated.
>> If those properties aren't present, then someone's mucked
>> with the tegra20.dtsi ehci properties.
>>
>> Once all platforms are cut over to devicetree-only, that can be the
>> case. Until that happens, there will be need for settings in the C
>> code too. Is Nvidia switching to device trees for android/fastboot?
>
> I'm not talking about the defaults in arch/arm/mach-tegra/usb_phy.c.  I understand that once we cut over to devicetree-only, the defaults there can go away.
>
> The question is, if a property is not present, what should we do?  The default value for that property SHOULD be present somewhere in the dts chain.  But if it's not, should we error out, spit out a warning, or spit out a warning and assign a default?  If it's the latter case, then we now have a set of defaults in dts, as well as C code, regardless of whether we've switched over to devicetree-only or not.

If there are no safe/sane defaults, then there isn't much to be done
there. I'd say fail probe (i.e. error out) to avoid surprises.


-Olof
--
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/