linux-kernel - Re: [PATCH 2/4] mfd: mc13783: When probing, unlock the mc13783 before subsystems initialisation.

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:	Sun, 20 Dec 2009 19:49:42 +0100
From:	Uwe Kleine-König 
	<u.kleine-koenig@...gutronix.de>
To:	Alberto Panizzo <maramaopercheseimorto@...il.com>
Cc:	Samuel Ortiz <sameo@...ux.intel.com>,
	Sascha linux-arm <s.hauer@...gutronix.de>,
	Mark Brown <broonie@...nsource.wolfsonmicro.com>,
	linux-kernel <linux-kernel@...r.kernel.org>,
	linux-arm-kernel-infradead <linux-arm-kernel@...ts.infradead.org>,
	Liam Girdwood <lrg@...mlogic.co.uk>
Subject: Re: [PATCH 2/4] mfd: mc13783: When probing, unlock the mc13783
	before subsystems initialisation.

Hi Alberto,

On Sun, Dec 20, 2009 at 02:48:38PM +0100, Alberto Panizzo wrote:
> > > PATCH 1 & 2 are fixes that can go to .33
> > I don't like patch 1.  I'd prefer that drivers touching
> > MC13783_REG_POWER_MISCELLANEOUS are aware of the bit in question and
> > wouldn't rely on mc13783-core.
> 
> Yes, but MC13783_REG_POWER_MISCELLANEOUS contains bit that control
> different aspect of mc13783 chip.
> GPO are regulator related, but those two bits in question maybe 
> apply to a power management driver, so this problem is a matter
> of mc13783-core.
maybe?
 
> Another possible solution, is to trace the writings to those two bits
> in mc13783_reg_rmw storing the value written an reproducing it in next
> mc13783_reg_rmw calls.
> But the problem for this is that we don't know if the bootloader 
> had initialised those with another non default value.
> Another problem is that if another driver make use of 
> mc13783_reg_write for modifying those bits, the state stored will
> be inconsistent.
The next best thing I would consider acceptable are dedicated functions
for MC13783_REG_POWER_MISCELLANEOUS.  I havn't checked what the register
in question is used for, but I think the bootloader isn't generally a
problem as Linux shouldn't rely on things setup by the bootloader (apart
from the things described in
http://www.arm.linux.org.uk/developer/booting.php of course).  And I
don't see a problem for in-kernel users of mc13783_reg_write to modify
the register.  If the mc13783-API is changed that
MC13783_REG_POWER_MISCELLANEOUS must only be modified by using (say)
mc13783_powermisc_rmw() it's a (probably uncatched) bug to use
mc13783_reg_write.

And patch 1 is definitly *not* material for .33, as there is currently
no user of MC13783_REG_POWER_MISCELLANEOUS in .33-rc1, so there is
nothing to fix.

Best regards
Uwe

-- 
Pengutronix e.K.                              | Uwe Kleine-König            |
Industrial Linux Solutions                    | http://www.pengutronix.de/  |
--
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/