linux-kernel - Re: [PATCH V3] net/dt: Add support for overriding phy configuration from device tree

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:	Fri, 7 Feb 2014 14:43:15 -0800
From:	Florian Fainelli <f.fainelli@...il.com>
To:	David Laight <David.Laight@...lab.com>
Cc:	Matthew Garrett <matthew.garrett@...ula.com>,
	netdev <netdev@...r.kernel.org>,
	"devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Kishon Vijay Abraham I <kishon@...com>
Subject: Re: [PATCH V3] net/dt: Add support for overriding phy configuration
 from device tree

2014-02-05 David Laight <David.Laight@...lab.com>:
> From: Florian Fainelli
>> It would be good to explain exactly how your hardware is broken
>> exactly. I really do not think that such a fine-grained setting where
>> you could disable, e.g: 100BaseT_Full, but allow 100BaseT_Half to
>> remain usable makes that much sense. In general, Gigabit might be
>> badly broken, but 100 and 10Mbits/sec should work fine. How about the
>> MASTER-SLAVE bit, is overriding it really required?
>
> There are plenty of systems out there where you'd want to disable
> either HDX or FDX modes.
> The MAC unit has to know whether the PHY is in HDX or FDX in order
> to work properly. Many do not need to know the speed - since the
> PHY is responsible for the tx/rx fifo clock.
> Getting the negotiated speed out of the PHY can be difficult, while
> the ANAR can easily be set.
> Unfortunately it is usually impossible to disable the 'fall-back'
> 10M HDX.

The problem that I have with that approach in general is that:

- it bloats the code for a set of properties that are going to be used
by hopefully a few percentage of the actual Device Trees out there
- it puts no limits on what is acceptable/best-practice to be put in
terms of configuration in the Device Tree, how about the 16x16 other
register values out there which are standardized?
- a PHY fixup should be registered based on the top-level compatible
property for a given board where the specific PHY on a specific board
is known to be broken
- make things incredibly harder to debug than they are today

I do acknowledge the need to have a solution to these problems, but
this seems to duplicate existing mechanisms available (e.g: PHY
fixups) without leveraging information that should be properly flagged
in the Device Tree (board model, root-node compatible string etc...)
to allow software to take corrective measures.
-- 
Florian
--
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/