linux-kernel - Re: [PATCH 3/3] gpiolib: use chip->names for symlinks, always use gpioN for device names

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: <1260508949.12048.227.camel@ben-desktop>
Date:	Fri, 11 Dec 2009 16:22:29 +1100
From:	Ben Nizette <bn@...sdigital.com>
To:	David Brownell <david-b@...bell.net>
Cc:	Greg KH <gregkh@...e.de>,
	Jani Nikula <ext-jani.1.nikula@...ia.com>,
	dbrownell@...rs.sourceforge.net, linux-kernel@...r.kernel.org,
	dsilvers@...tec.co.uk, ben@...tec.co.uk,
	Artem.Bityutskiy@...ia.com, akpm@...ux-foundation.org
Subject: Re: [PATCH 3/3] gpiolib: use chip->names for symlinks, always use
 gpioN for device names

On Thu, 2009-12-10 at 20:13 -0800, David Brownell wrote:
> On Thursday 10 December 2009, Greg KH wrote:
> > 
> > > IMO a "good" solution in this space needs to accept that
> > > those names are not going to be globally unique ... but
> > > that they'll be unique within some context, of necessity.
> > > 
> > > If Greg doesn't want to see those names under classes,
> > > so be it ... but where should they then appear?
> > 
> > As a sysfs file within the device directory called 'name'?  Then just
> > grep through the tree to find the right device, that also handles
> > duplicates just fine, right?
> 
> I want a concrete example.  Those chip->names things don't
> seem helpful to me though...
> 
> If for example I were building a JTAG adapter on Linux, it
> might consist of a spidev node (chardev) plus a handful of
> GPIOs.  So "the device directory" would be the sysfs home
> of that spidev node (or some variant)?  And inside that
> directory would be files named after various signals that
> are used as GPIOs ... maybe SRST, TRST, and DETECT to start
> with?  Holding some cookie that gets mapped to those GPIO's
> sysfs entries?

If you've got a spidev node then you've got a dev to pass to
gpio_export_link, all's right with the world.  I think the best thing
(which I think is what Greg was thinking) would be to
have /sys/class/gpio/gpioN/name which reads null unless the kernel has
registered a name against that gpio.

Now I don't like the chip->name thing as a way to hold these labels, I'd
prefer an IDR indexed by gpio number, but that's an implementation
detail for later.

A concrete use case: I've got a board here, the support patch for which
will hopefully appear in a few days.  On it I've got an 8-bit uC which
does a bunch of I/O and on certain conditions, toggles gpio lines so a
userspace daemon knows what's going on.  At the moment, I
have /sys/class/<board name>/ which exports things like revision
information as well as attributes telling me which gpio numbers are used
for which purpose.  My scripts therefore look something like

echo 1 > /sys/class/gpio/gpio`cat /sys/class/<board-name>/ack-gpio`/value

I'm actually fairly happy with that but I don't pretend it's everyone's
cup of tea.  The point is though, the microprocessor has no in-kernel
incarnation except these few gpio so there's no convenient dev node to
link the names through; gpio_export_link() isn't useful.

This problem could be solved as I have, but it could equally-well be
solved by exporting gpioN/name attributes which I could search for.

	--Ben.

--
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/