linux-kernel - Re: [PATCH 01/45] kstrtox: converting strings to integers done (hopefully) right

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]

Message-ID: <20101208085145.GA5320@core2.telecom.by>
Date:	Wed, 8 Dec 2010 10:51:45 +0200
From:	Alexey Dobriyan <adobriyan@...il.com>
To:	Arnd Bergmann <arnd@...db.de>
Cc:	Jesper Juhl <jj@...osbits.net>, akpm@...ux-foundation.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 01/45] kstrtox: converting strings to integers done
 (hopefully) right

On Tue, Dec 07, 2010 at 10:45:33AM +0100, Arnd Bergmann wrote:
> On Tuesday 07 December 2010 10:32:29 Alexey Dobriyan wrote:
> > On Tue, Dec 07, 2010 at 10:04:58AM +0100, Arnd Bergmann wrote:
> > > On Monday 06 December 2010 16:16:36 Alexey Dobriyan wrote:
> > >
> > > > It is used by kstrtol() if "long" and "long long" aren't the same type.
> > > > I can't ifdef it because gcc doesn't allow "#if sizeof(long long) == ".
> > > > "Do not use" hints "use kstrtol()".
> > > 
> > > Can't you use #ifdef CONFIG_64BIT to see if long is 64 bit wide?
> > 
> > I could, but there are no defines for alignment.
> 
> Not sure what you mean here.

One can't write:

	#if sizeof(long) == sizeof(long long) && alignof(long) == alignof(long long)

and there are no _ALIGNOF_LONG or something like that.

> All architectures always naturally align 'long' variables --
> we rely on that elsewhere already. 'long long' is misaligned on x86-32 and
> possibly some other 32 bit architectures, but since that should not cause
> any problems.

The point is that code as written doesn't care and correct wrt alignment.

> Another option would be to export just the strto{u,s}{8,16,32,64}
> functions and define the inline wrappers conditionally on CONFIG_64BIT.

This means more ifdefs. Maybe someone could ifdeflessly fix this later.
--
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/