Re: [PATCH] arch/i386/kernel/cpu/transmeta.c, kernel 2.6.17.8

[Michael Tokarev <mjt@tls.msk.ru>]

Forrest Voight wrote:
> Corrects warning:
> 
>  CC      arch/i386/kernel/cpu/centaur.o
>  CC      arch/i386/kernel/cpu/transmeta.o
> arch/i386/kernel/cpu/transmeta.c: In function 'init_transmeta':
> arch/i386/kernel/cpu/transmeta.c:12: warning: 'cpu_freq' may be used
> uninitialized in this function
>  CC      arch/i386/kernel/cpu/intel.o
> 

This is a false alarm.
Here's the code (details omitted):

 if ( max >= 0x80860001 ) {
   cpuid(0x80860001, &dummy, &cpu_rev, &cpu_freq, &cpu_flags);
                                       ^^^^^^^^^
 }
 if ( max >= 0x80860002 ) {
    printk(KERN_INFO "CPU: Processor %u MHz\n", cpu_freq);
 }

Note the two conditions: if second is true, the first is
true too, so both branches are executed, so first cpu_freq
is initialized (by cpuid() call) and next it's used in printk.

The same thing will be done by the following code:

 if ( max >= 0x80860001 ) {
   cpuid(0x80860001, &dummy, &cpu_rev, &cpu_freq, &cpu_flags);
   if ( max >= 0x80860002 ) {
      printk(KERN_INFO "CPU: Processor %u MHz\n", cpu_freq);
   }
 }

and in this case gcc will not (hopefully) issue the warning.

BTW, cpu_rev gets initialized to 0 here as well - looks like
it's done also just to prevent warning message.

/mjt

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/