linux-kernel - Re: [PATCH] regmap: Allow read_reg

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: <542B1074.5000802@ti.com>
Date:	Tue, 30 Sep 2014 15:20:04 -0500
From:	Dan Murphy <dmurphy@...com>
To:	Mark Brown <broonie@...nel.org>
CC:	<gregkh@...uxfoundation.org>, <linux-kernel@...r.kernel.org>,
	<linux-omap@...r.kernel.org>
Subject: Re: [PATCH] regmap: Allow read_reg_mask to be 0

On 09/30/2014 12:25 PM, Mark Brown wrote:
> On Tue, Sep 30, 2014 at 11:07:00AM -0500, Dan Murphy wrote:
>
>> -	if (config->read_flag_mask || config->write_flag_mask) {
>> +	if (config->read_flag_mask == REGMAP_NO_READ_MASK)
>> +		map->read_flag_mask = 0x00;
>> +	else if (config->read_flag_mask)
> This breaks the symmetry in handling of read and write masks which isn't
> great, please make the equivalent update for the write mask too.

Hmmm.  If I make the similar change for the write mask I will be adding dead
code as write_flag_mask is not defaulted by the bus like the read_flag_mask which is
defaulted to 0x80 in the regmap-spi code.  The i2c code does not make this assumption.

The original code did not seem to have symmetry as the only instance
that write_flag_mask is modified is if it is non-zero.

Let me know what you think.  I can add the code to make it more symmetrical but
we may be adding code that is dead.  If I remove the defaulted read_flag_mask value out
of the spi driver then I will need to find every spi device that needs that flag and set it.
IMHO that would be really messy and probably mess us some drivers.

>> +#define REGMAP_NO_READ_MASK		0xff
> An actual out of band value might be preferable here though that'd
> involve changing the type and more checking so perhaps inessential.

Thought about making this -1 with a variable change but that seemed
really drastic where a value of 0xff seems to be a value that no one should use.

Dan

-- 
------------------
Dan Murphy

--
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/