Em Mon, 3 Aug 2009 08:30:12 +0200
Jean-Francois Moine <moinejf@xxxxxxx> escreveu:
> On Sun, 2 Aug 2009 17:25:29 +0400
> Alexey Klimov <klimov.linux@xxxxxxxxx> wrote:
>
> > > + buffer = kmalloc(JEILINJ_MAX_TRANSFER, GFP_KERNEL |
> > > GFP_DMA);
> > > + if (!buffer) {
> > > + PDEBUG(D_ERR, "Couldn't allocate USB buffer");
> > > + goto quit_stream;
> > > + }
> >
> > This clean up on error path looks bad. On quit_stream you have:
> >
> > > +quit_stream:
> > > + mutex_lock(&gspca_dev->usb_lock);
> > > + if (gspca_dev->present)
> > > + jlj_stop(gspca_dev);
> > > + mutex_unlock(&gspca_dev->usb_lock);
> > > + kfree(buffer);
> >
> > kfree() tries to free null buffer after kmalloc for buffer failed.
> > Please, check if i'm not wrong.
>
> Hi Alexey,
>
> AFAIK, kfree() checks the pointer.
Yeah. Theodore's code is ok. kfree(NULL) is legal.
>
> Cheers.
>
Cheers,
Mauro
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
