RE: [PATCHv2 2/3] iovmm: add superpages support to fixed da address

["Guzman Lugo, Fernando" <fernando.lugo@ti.com>]

 

> -----Original Message-----
> From: Felipe Contreras [mailto:felipe.contreras@gmail.com] 
> Sent: Tuesday, October 12, 2010 6:09 AM
> To: Guzman Lugo, Fernando
> Cc: Hiroshi.DOYU@nokia.com; david.cohen@nokia.com; 
> felipe.contreras@nokia.com; ameya.palande@nokia.com; 
> linux-kernel@vger.kernel.org; andy.shevchenko@gmail.com; 
> linux-omap@vger.kernel.org
> Subject: Re: [PATCHv2 2/3] iovmm: add superpages support to 
> fixed da address
> 
> On Mon, Oct 11, 2010 at 6:33 PM, Guzman Lugo, Fernando 
> <fernando.lugo@ti.com> wrote:
> >> > @@ -420,7 +431,8 @@ static void sgtable_fill_kmalloc(struct
> >> sg_table
> >> > *sgt, u32 pa, size_t len)
> >> >        for_each_sg(sgt->sgl, sg, sgt->nents, i) {
> >> >                size_t bytes;
> >> >
> >> > -               bytes = iopgsz_max(len);
> >> > +               bytes = max_alignment(da | pa);
> >> > +               bytes = min(bytes, (size_t)iopgsz_max(len));
> >>
> >> Why the size_t casting?
> >
> > To void this warning:
> > arch/arm/plat-omap/iovmm.c:440: warning: comparison of distinct 
> > pointer types lacks a cast
> 
> But how is that possible? iopgsz_max is returning constants, 
> like SZ_1M, so they should not need casts, if anything, the 
> cast should be done in iopgsz_max itself.

The min macro make a "typeof" of the parameter, so the typeof of a 
Constants should be int I think and that's the reason of the warning.

I can use min_t instead to avoid the warning, something like this:

			bytes = min_t(size_t, bytes, iopgsz_max(len));

What do you think?

Regards,
Fernando.

> 
> --
> Felipe Contreras
>