linux-kernel - Re: sysfs file creation result nightmare (WAS radeonfb: Fix sysfs_create_bin

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]

Date:	Sat, 9 Dec 2006 12:38:17 -0800
From:	Andrew Morton <akpm@...l.org>
To:	Benjamin Herrenschmidt <benh@...nel.crashing.org>
Cc:	Jean Delvare <khali@...ux-fr.org>,
	Paul Mackerras <paulus@...ba.org>,
	Linux Kernel list <linux-kernel@...r.kernel.org>
Subject: Re: sysfs file creation result nightmare (WAS radeonfb: Fix
 sysfs_create_bin_file warnings)

On Sun, 10 Dec 2006 06:59:10 +1100
Benjamin Herrenschmidt <benh@...nel.crashing.org> wrote:

> On Sat, 2006-12-09 at 16:56 +0100, Jean Delvare wrote:
> 
> > Check for error on radeonfb device sysfs files creation. This fixes the
> > following warnings:
> 
> (Moving to LKML as I think that's a generic issue)
> 
> As usual with most of that crap about return values from
> sysfs_create_file, I disagree. strongly.

Actually, wrongly.

> Why would I prevent the framebuffer from initializing (and thus a
> console to be displayed at all on many machines) just because for some
> reason, I couldn't create a pair of EDID files in sysfs that are not
> even very useful anymore ?

Because there's a bug in your kernel.  We don't hide and work around bugs.

> I have _plenty_ of cases where the failure to create sysfs files, while
> annoying and maybe deserving a warning, certainly doesn't imply
> completely preventing the driver from initializing.

Why does it matter?  Just fix the bugs and the issue won't arise.  If you
hide them and work around them in this manner, they won't get fixed.

> However, all the
> patches I've seen so far to fix the new warnings do just that (make the
> driver fail)
> 
> I'd really like to have some kind of macro or attribute or whatever I
> can put on a function call to say that I'm purposefully ignoring the
> error. Is there some gcc magic that can do that ?

#define HIDE_AND_WORK_AROUND_A_BUG(expr) (void)(expr)

but it might meet some resistance.


Just fix the bugs, for heck's sake.

-
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/