linux-kernel - Re: [PATCH] pda_power: clean up irq, timer, return usage

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: <46996A53.6000807@garzik.org>
Date:	Sat, 14 Jul 2007 20:29:07 -0400
From:	Jeff Garzik <jeff@...zik.org>
To:	cbou@...l.ru
CC:	Andrew Morton <akpm@...ux-foundation.org>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] pda_power: clean up irq, timer, return usage

Anton Vorontsov wrote:
> On Sat, Jul 14, 2007 at 07:12:04PM -0400, Jeff Garzik wrote:
>> Clean up pda_power interrupt handling:
> 
> Nice, thanks! Just few cosmetic comments.
> 
>> Prior to this patch, the driver would pass information it needed
>> to the interrupt handler dev_id pointer, and then prompt forget it
>> ever did so, recreating that same information after a couple passes
>> through the timer-based state machine.
>>
>> This patch removes the redundant checks by passing the
>> pda_power_supply[] pointer through the state machine.  The current
>> code passed 'irq' through the state machine, as an index to recreate
>> the pointer, when we could more simply pass around the pointer itself.
>>
>> Additionally, bogus "return;" statements were removed.
> 
> My preference is to use "return;" statements in functions returning
> void, even if functions is very small. It's just sugar for my eyes,
> to really see exit points. . Without returns I've feeling that
> something is missing there. Yes, my oddity.
> 
> Plus, so far CodingStyle does not say anything about non obligatory
> return statements.
> 
> These should be "fixed" too, though:
> ~/linux-2.6$ grep -h "return;" -A1 -r drivers/ arch/ | grep "^}$" | \
> wc -l
> 1354
> 
> Obviously, drivers/ata is almost pure (3). ;-)
> 
> Either way, I prefer to leave alone these "return;"s, until CodingStyle
> permits them.

CodingStyle is not the end-all of rulebooks.  See repeated messages by 
Linus, me, and others on the subject.  CodingStyle intentionally does 
not list a rule for every possible C code incarnation.

It's the general style to avoid these return statements in new code.  It 
makes the page longer, meaning less code to review.  It tends to trip up 
people making bombing runs through drivers, and searching for return 
statements in their favorite editor.

If you consider how many void functions there are in the kernel, 1354 is 
the obvious minority, being far smaller than "millions."  Please follow 
what -millions- of void functions already do in the kernel.


>> This patch makes it easier to remove the 'irq' argument in the future,
>> in addition to cleaning up the driver today.
>>
>> P.S.  Where are the MAINTAINERS entries for this driver, and
>> drivers/power in general?
> 
> I'll send patch shortly.
> 
>> +	void *power_supply = (void *) power_supply_ptr;
>                                      ^
> I guess common practice is not to put space here.

incorrect.  a space goes there, as I put it.


>> +	charger_timer.data = (unsigned long) power_supply;
> 
> And here.

ditto.

-
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/