linux-kernel - Re: [RFC][PATCH] genirq: add IRQF

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] [day] [month] [year] [list]

Date:	Mon, 09 Sep 2013 06:40:58 +0200
From:	Michael Opdenacker <michael.opdenacker@...e-electrons.com>
To:	Josh Triplett <josh@...htriplett.org>
CC:	paul.mckenney@...aro.org, akpm@...ux-foundation.org,
	decot@...glers.com, amirv@...lanox.com,
	linux-kernel@...r.kernel.org
Subject: Re: [RFC][PATCH] genirq: add IRQF_NONE

Hi Josh,

On 09/09/2013 06:02 AM, Josh Triplett wrote:
> On Mon, Sep 09, 2013 at 05:48:39AM +0200, Michael Opdenacker wrote:
>> What about adding an IRQF_NONE flag as in the below patch?
>>
>> I'm currently working on removing the use of the deprecated
>> IRQF_DISABLED flag, and frequently have to replace
>> IRQF_DISABLED by 0, typically in request_irq() arguments.
>>
>> Using IRQF_NONE instead of 0 would make the code more readable,
>> at least for people reading driver code for the first time.
>>
>> Would it worth it?
>>
>> I'm sure this kind of idea has come up many times before...
>>
>> Signed-off-by: Michael Opdenacker <michael.opdenacker@...e-electrons.com>
> I don't think it makes sense, no; it's a flags field, meant to receive a
> set of flags, and 0 is the standard empty set of flags.  I think
> IRQF_NONE would actually reduce readability, especially for readers who
> haven't seen it before, because it isn't immediately obvious that it
> just corresponds to the 0 of "no flags".  My first guess reading it
> would be that it's some non-zero flag with some non-obvious semantic,
> such as "don't actually allocate an IRQ", or something strange like
Thanks for your feedback. It's true that 0 for a flag is clear enough,
and that IRQF_NONE will be more confusing.

I was just thinking the IRQF_NONE would make it clearer that the
corresponding argument is a flag. This way, people reading "0" wouldn't
have to lookup the prototype of request_irq() to know what type of
argument this is (flag, number of resources, boolean....)

So, this may be a little helpful for completely new people, but more
confusing for people with a little more experience, as you explained.

I agree not to sacrifice the latter.

Thanks again,

Michael.

-- 
Michael Opdenacker, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
+33 484 258 098

--
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/