linux-kernel - Re: [PATCH 2/3] ASoC: pcm179x: Add I2C interface driver

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, 19 Jan 2016 15:11:44 +0100
From:	Johan Hovold <johan@...nel.org>
To:	Mark Brown <broonie@...nel.org>
Cc:	Johan Hovold <johan@...nel.org>,
	Michael Trimarchi <michael@...rulasolutions.com>,
	Jacob Siverskog <jacob@...nage.engineering>,
	Jaroslav Kysela <perex@...ex.cz>,
	Takashi Iwai <tiwai@...e.com>,
	Liam Girdwood <lgirdwood@...il.com>,
	"alsa-devel@...a-project.org" <alsa-devel@...a-project.org>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/3] ASoC: pcm179x: Add I2C interface driver

On Tue, Jan 19, 2016 at 12:25:24PM +0000, Mark Brown wrote:
> On Mon, Jan 18, 2016 at 05:56:57PM +0100, Johan Hovold wrote:
> > On Mon, Jan 18, 2016 at 05:14:42PM +0100, Michael Trimarchi wrote:
> 
> > > > +       { .compatible = "ti,pcm1792a", },
> > > > +       { }
> > > > +};
> > > > +MODULE_DEVICE_TABLE(of, pcm179x_of_match);
> 
> > > can match go in the common part?
> 
> > For matching purposes it could, but that would prevent the OF-aliases
> > from being defined for the driver modules.
> 
> > There are a couple of codec drivers that have taken this route, and it
> > works as long as they also define a legacy (e.g. i2c_device_id) table
> > that modalias autoloading can fall back to.
> 
> > But having having the common module carry the OF-aliases does not seem
> > like the right thing to do (e.g. as the common module cannot drive
> > anything on its own) even if it avoids a copy of the OF id table.
> 
> This does seem like something that the modules code should be able to
> cope with - we have a reference from the driver struct to the table
> which you'd really expect it to be using.  We already have some issues
> there with userspace not handling multiple modaliases terribly well IIRC.

That reference is only used for matching, while the aliases are
generated from the MODULE_DEVICE_TABLE macro (and elf magic).

Even if this sometimes means duplicating an id-table (and i2c and spi OF
id tables aren't necessarily always identical either), it seems to me
that the MODULE_DEVICE_TABLE should go in the actual driver rather than
any module dependency.

Thanks,
Johan