Re: [PATCH 1/2] dma-mapping: introduce dma_get_seg_boundary_nr_pages()

[Christophe Leroy <christophe.leroy@xxxxxxxxxx>]

Le 03/09/2020 à 18:12, Christoph Hellwig a écrit :
> On Thu, Sep 03, 2020 at 01:57:39PM +0300, Andy Shevchenko wrote:
> > > +{
> > > +       if (!dev)
> > > +               return (U32_MAX >> page_shift) + 1;
> > > +       return (dma_get_seg_boundary(dev) >> page_shift) + 1;
> >
> > Can it be better to do something like
> >    unsigned long boundary = dev ? dma_get_seg_boundary(dev) : U32_MAX;
> >
> >    return (boundary >> page_shift) + 1;
> >
> > ?
>
> I don't really see what that would buy us.

I guess it would avoid the duplication of    >> page_shift) + 1

Christophe