On Tue, Jan 19, 2021 at 10:11:16AM +0100, Borislav Petkov wrote:
> On Tue, Jan 19, 2021 at 01:19:42PM +0800, Feng Tang wrote:
> > Sorry, after testing on more platforms, the following is needed to fix
> > a potential array overflow ((a full patch with fix is also attached)
> > 
> > diff --git a/tools/arch/x86/kcpuid/kcpuid.c b/tools/arch/x86/kcpuid/kcpuid.c
> > index 3ea607b..bf68335 100644
> > --- a/tools/arch/x86/kcpuid/kcpuid.c
> > +++ b/tools/arch/x86/kcpuid/kcpuid.c
> > @@ -329,7 +329,8 @@ static int parse_line(char *line)
> >             range = leafs_basic;
> >  
> >     index &= 0x7FFFFFFF;
> > -   if ((int)index > range->nr)
> > +   /* range->nr equals to max index plus 1 */
> > +   if ((int)index >= range->nr)
> 
> That comment is basically repeating the code. What it should do is
> explain *why* that test is being done or not be there at all.
 
Right! So this is better?
/* Skip line parsing for non-existing indexes */ 

Thanks,
Feng

> -- 
> Regards/Gruss,
>     Boris.
> 
> https://people.kernel.org/tglx/notes-about-netiquette

Reply via email to