linux-kernel - Re: mc13xxx-core: kernel hangs after 'regmap

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:	Thu, 24 May 2012 21:22:29 +1000
From:	Marc Reilly <marc@...esign.com.au>
To:	Mark Brown <broonie@...nsource.wolfsonmicro.com>
Cc:	Shawn Guo <shawn.guo@...escale.com>,
	"Uwe Kleine-König" 
	<u.kleine-koenig@...gutronix.de>,
	Fabio Estevam <festevam@...il.com>,
	Samuel Ortiz <sameo@...ux.intel.com>,
	Sascha Hauer <kernel@...gutronix.de>,
	Philippe Rétornaz <philippe.retornaz@...l.ch>,
	"linux-kernel" <linux-kernel@...r.kernel.org>
Subject: Re: mc13xxx-core: kernel hangs after 'regmap_read'

Hi,

On Thursday, May 24, 2012 08:37:49 PM Mark Brown wrote:
> On Thu, May 24, 2012 at 07:08:37PM +1000, Marc Reilly wrote:
> > I'm wondering about regmap_init where the buf_size is set up. I think it
> > will
> > 
> > only end up being 3 bytes. I think line 249 should be something like:
> > 	map->format.buf_size = (config->reg_bits
> > 	
> >                           + config->val_bits
> > 					         
> > 					         + (config->pad_bits % 8)) / 8;
> > 
> > or perhaps better:
> > 	map->reg_shift = config->pad_bits % 8;
> > 	map->format.buf_size = (config->reg_bits
> > 	
> >                           + config->val_bits
> > 					         
> > 					         + map->reg_shift) / 8;
> 
> Yes, that's been missed in the addition of padding.  We should also be
> using DIV_ROUND_UP() which we aren't at the minute.  

That would break, in _regmap_read_raw():

	ret = map->bus->read(map->bus_context, map->work_buf,
			     map->format.reg_bytes + map->format.pad_bytes,
			     val, val_len);

If pad_bytes was 1 here, then the register size would end up being 2 bytes.

The way I understood the pad_bytes field was it was the number of _complete_ 
padding bytes (ie, full 8 bits) between the register address and value. Any 
remainder of padding bits is incorporated into the register and shifted by 
shift bits.

> We could also do
> reg_bytes + pad_bytes + val_bytes which should cover everything I think?

If we want to cover everything, we could do reg_bytes + pad_bytes + val_bytes 
+ 3 ... :) (I'm not seriously suggesting that).

Cheers,
Marc
--
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/