linux-kernel - Re: [PATCH] drivers/char/specialix.c: stop inlining largish static functions

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: <200804090018.53053.vda.linux@googlemail.com>
Date:	Wed, 9 Apr 2008 00:18:53 +0200
From:	Denys Vlasenko <vda.linux@...glemail.com>
To:	Jiri Slaby <jirislaby@...il.com>
Cc:	Alan Cox <alan@...rguk.ukuu.org.uk>, linux-kernel@...r.kernel.org,
	linux-serial@...r.kernel.org, R.E.Wolff@...wizard.nl
Subject: Re: [PATCH] drivers/char/specialix.c: stop inlining largish static functions

On Tuesday 08 April 2008 23:38, Jiri Slaby wrote:
> > drivers/char/specialix.c has unusually large number
> > of static inline functions - 22.
> > 
> > I looked through them. The file is positively inline-happy.
> > Inlines with udelay() calls. Inlines with complex loops.
> > Nested inlines. Rarely called inlines (e.g. with request_region
> > inside).
> > 
> > This patch removes "inline" from 15 static functions
> > (regardless of number of callsites - gcc nowadays auto-inlines
> > statics with one callsite).
> > 
> > Size difference for 32bit x86:
> >    text    data     bss     dec     hex filename
> >   21669     204    8780   30653    77bd linux-2.6-ALLYES/drivers/char/specialix.o
> >   18470     204    8780   27454    6b3e linux-2.6.inline-ALLYES/drivers/char/specialix.o
> > 
> > 
> > Signed-off-by: Denys Vlasenko <vda.linux@...glemail.com>
> 
> -	static const char *badmagic =
> +	static const char badmagic[] =
>   		KERN_ERR "sx: Warning: bad specialix port magic number for device %s in %s\n";
> -	static const char *badinfo =
> +	static const char badinfo[] =
>   		KERN_ERR "sx: Warning: null specialix port for device %s in %s\n";
> 
> 
> BTW what's this good for? I mean, why we need this as a variable not directly as 
> a parameter?
> 
>   	if (!port) {
>   		printk(badinfo, name, routine);
>   		return 1;
>   	}
>   	if (port->magic != SPECIALIX_MAGIC) {
>   		printk(badmagic, name, routine);
>   		return 1;
>   	}

I am sure these strings can be used directly in printk,
there should be no size difference (sans gcc adding padding
to string arrays "just because").

I chose to make minimal change which only eliminates the waste
of having a pinter to these strings (char *msg = "xxx"
versus char msg[] = "xxx"), leaving more extensive editing
to someone who wants to attack specialix.c on the wider front.
--
vda
--
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/