Re: [PATCH 5/6] zsmalloc: remove unnecessary type casting

[Greg Kroah-Hartman]

On Wed, Apr 25, 2012 at 09:35:25AM -0400, Nitin Gupta wrote:
> On 04/25/2012 02:23 AM, Minchan Kim wrote:
> 
> > Let's remove unnecessary type casting of (void *).
> > 
> > Signed-off-by: Minchan Kim <minchan@xxxxxxxxxx>
> > ---
> >  drivers/staging/zsmalloc/zsmalloc-main.c |    3 +--
> >  1 file changed, 1 insertion(+), 2 deletions(-)
> > 
> > diff --git a/drivers/staging/zsmalloc/zsmalloc-main.c b/drivers/staging/zsmalloc/zsmalloc-main.c
> > index b7d31cc..ff089f8 100644
> > --- a/drivers/staging/zsmalloc/zsmalloc-main.c
> > +++ b/drivers/staging/zsmalloc/zsmalloc-main.c
> > @@ -644,8 +644,7 @@ void zs_free(struct zs_pool *pool, void *obj)
> >  	spin_lock(&class->lock);
> >  
> >  	/* Insert this object in containing zspage's freelist */
> > -	link = (struct link_free *)((unsigned char *)kmap_atomic(f_page)
> > -							+ f_offset);
> > +	link = (struct link_free *)(kmap_atomic(f_page)	+ f_offset);
> >  	link->next = first_page->freelist;
> >  	kunmap_atomic(link);
> >  	first_page->freelist = obj;
> 
> 
> 
> Incrementing a void pointer looks weired and should not be allowed by C
> compilers though gcc and clang seem to allow this without any warnings.
> (fortunately C++ forbids incrementing void pointers)

Huh?  A void pointer can safely be incremented by C I thought, do you
have a pointer to where in the reference it says it is "unspecified"?

> So, we should keep this cast to unsigned char pointer to avoid relying
> on a non-standard, compiler specific behavior.

I do agree about this, more people are starting to build the kernel with
other compilers than gcc, so it would be nice to ensure that we get
stuff like this right.

thanks,

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