linux-kernel - Re: [PATCH v5 0/6] drivers/i2c: Add FSI-attached I2C master algorithm

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: <538605fe-f109-b84c-cc21-f1b9824ad5d0@axentia.se>
Date:   Tue, 15 Aug 2017 19:35:47 +0200
From:   Peter Rosin <peda@...ntia.se>
To:     Christopher Bostic <cbostic@...ux.vnet.ibm.com>,
        Joel Stanley <joel@....id.au>
Cc:     Eddie James <eajames@...ux.vnet.ibm.com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        linux-i2c@...r.kernel.org, devicetree <devicetree@...r.kernel.org>,
        Wolfram Sang <wsa@...-dreams.de>,
        Rob Herring <robh+dt@...nel.org>, Jeremy Kerr <jk@...abs.org>,
        Andrew Jeffery <andrew@...id.au>,
        "Edward A. James" <eajames@...ibm.com>
Subject: Re: [PATCH v5 0/6] drivers/i2c: Add FSI-attached I2C master algorithm

On 2017-08-15 18:28, Christopher Bostic wrote:
> On 8/15/17 3:10 AM, Joel Stanley wrote:
>> On Tue, Aug 15, 2017 at 4:06 PM, Peter Rosin <peda@...ntia.se> wrote:
>>> On 2017-07-26 19:13, Eddie James wrote:
>>>> From: "Edward A. James" <eajames@...ibm.com>
>>>>
>>>> This series adds an algorithm for an I2C master physically located on an FSI
>>>> slave device. The I2C master has multiple ports, each of which may be connected
>>>> to an I2C slave. Access to the I2C master registers is achieved over FSI bus.
>>>>
>>>> Due to the multi-port nature of the I2C master, the driver instantiates a new
>>>> I2C adapter for each port connected to a slave. The connected ports should be
>>>> defined in the device tree under the I2C master device.
>>> Hmmm, AFAIU fsi is a bus, and on this bus you have some "client" device that
>>> happens to be an i2c master, and this is a driver for that "client". Is it
>>> totally inconceivable to have some other client device in the future that is
>>> implementing an i2c master differently, but still using the fsi bus?
>>>
>>> With that in mind, is it wise to pick the driver name from the bus that the
>>> device is connected to, and nothing else without further qualification?
>>>
>>> I don't see any "i2c-usb" driver, but I think there are a couple of i2c master
>>> drivers that communicate via usb.
>> You make a fair point. When I did a prototype of this driver I called
>> it i2c-cfam, as it is part of the CFAM hardware unit inside of the
>> Power8/Power9 processors.
>>
>> The documentation does call it FSI_I2CM, so that's an argument for the
>> current name.
>>
>> I'm not sure how accurate that name is. Chris, Eddie, do you have any
>> other suggestions?
> The I2C engine up to now has been always accessed via the FSI bus so 
> historically I assume that's why its labelled as FSI_I2CM in the p8/p9 
> specs.  There isn't any reason this I2C device couldn't be implemented 
> in some other topology independent of FSI / CFAMs.  In other words there 
> are no FSI details internal to this I2C engine, an argument for removing 
> the 'FSI' tag.

Note that I wasn't primarily concerned with this i2c engine growing some other
non-fsi interface (like many devices have both i2c and spi interface versions).
I was more concerned with some future and totally different i2c engine that
naturally sports a totally different register map but still uses the fsi bus.

But you have a point. If this i2c engine evolves and ends up supporting some
other interface, then that too would be cause to regret the i2c-fsi name.

Cheers,
peda