linux-kernel - Re: [PATCH] driver core: prevent deferred probe with platform_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:	Mon, 23 Sep 2013 14:28:06 +0200
From:	Johan Hovold <jhovold@...il.com>
To:	Mark Brown <broonie@...nel.org>
Cc:	Johan Hovold <jhovold@...il.com>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	Grant Likely <grant.likely@...aro.org>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] driver core: prevent deferred probe with
 platform_driver_probe

On Mon, Sep 23, 2013 at 12:01:40PM +0100, Mark Brown wrote:
> On Mon, Sep 23, 2013 at 10:48:47AM +0200, Johan Hovold wrote:
> > Prevent drivers relying on platform_driver_probe from requesting
> > deferred probing in order to avoid further futile probe attempts (either
> > the driver has been unregistered or its probe function has been set to
> > platform_drv_probe_fail when probing is retried).
> > 
> > Note that several platform drivers currently return subsystem errors
> > from probe and that these can include -EPROBE_DEFER (e.g. if a gpio
> > request fails).
> 
> This doesn't seem like the right end to address the problem from, it
> seems like it would be better to move these drivers over to being normal
> plaform drivers.  Using module_platform_driver() means relying on init
> ordering which is the sort of thing we're trying to get away from.

I actually started out doing that, but it's getting a bit hard to audit
which drivers could actually request probe deferral since gpio and later
other subsystems started returning -EPROBE_DEFER. I found six by just
grepping for gpio_request, but some of these calls can be made in helper
functions (e.g. mmc_gpio_request_cd even though that one was easy to
find).

Having a warning printed by platform_drv_probe if a platform driver
inadvertently requests probe deferral could be useful to catch any
mistakes even if we start moving probe functions out of __init.

I'll fix up the six drivers I found meanwhile.

Johan
--
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/