linux-kernel - Re: [git patch] free

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: <alpine.LFD.1.10.0804231914270.2779@woody.linux-foundation.org>
Date:	Wed, 23 Apr 2008 19:19:52 -0700 (PDT)
From:	Linus Torvalds <torvalds@...ux-foundation.org>
To:	Jeff Garzik <jeff@...zik.org>
cc:	Rene Herman <rene.herman@...access.nl>,
	Adrian Bunk <bunk@...nel.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	LKML <linux-kernel@...r.kernel.org>, rmk@....linux.org.uk,
	"Eric W. Biederman" <ebiederm@...ssion.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...hat.com>
Subject: Re: [git patch] free_irq() fixes

On Wed, 23 Apr 2008, Jeff Garzik wrote:
> 
> When drivers make assumptions about system irq numbering, particularly on x86,
> IMO the situation is fragile.

And when people make changes to long-standing and stable infrastructure, 
the situation also gets fragile.

The fact is, stability of interfaces is a really worthy goal in itself. 
Making a change for its own sake is not a good thing. This fixes 
*nothing*, and the driver changes I objected to I objected to because they 
were ugly as sin.

And I want to point out that your patches made it *much* uglier.

So "cleanup" it sure as hell wasn't. That irq number may not be worth all 
that much in itself, but it has no subtle implementation problems (we 
_need_ that irq number for registration and irq handler lookup anyway, so 
it is meaningful from a driver perspective, and is well-defined from a irq 
core standpoint as well).

I don't mind cleanups, but this is "churn". Change for its own sake. If it 
doesn't lead to any _improvement_, it's pointless.

If drivers don't need it, let them ignore it. But let them ignore it in 
ways that work across versions, and in ways that don't cause ridiculous 
and ugly work-arounds for when they do want it (even if it's just for a 
printk() or similar).

		Linus
--
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/