On Sun, Feb 15, 2009 at 07:01:41PM +0100, Borislav Petkov wrote:
> Hi,
>
> > > Signed-off-by: Borislav Petkov <petkovbb@xxxxxxxxx>
> > > ---
> > > include/linux/ide.h | 166 +++++++++++++++++++++++++++++++++++++++++++++++++++
> > > 1 files changed, 166 insertions(+), 0 deletions(-)
> > >
> > > diff --git a/include/linux/ide.h b/include/linux/ide.h
> > > index c75631c..f133062 100644
> > > --- a/include/linux/ide.h
> > > +++ b/include/linux/ide.h
> > > @@ -497,6 +497,82 @@ enum {
> > > IDE_AFLAG_NO_AUTOCLOSE = (1 << 24),
> > > };
> > >
> > > +#define ide_drv_drq_int(drive) \
> > > + ((drive)->atapi_flags & IDE_AFLAG_DRQ_INTERRUPT)
> >
> > Why not use a static inline here so we get proper typecheck.
> > And then convert the return result to a bool (0/1) so you
> > do not have to do this at the call site.
> >
> > I counted at least three places in ide-cd that does a local
> > transformation to a bool and I saw nowhere the actual bit value
> > was used.
>
> I'm assuming you're talking about those places (and similar):
>
> drive->dma = !!(drive->dev_flags & IDE_DFLAG_USING_DMA);
And in other places we do:
ide_drv_drq_int(drive) ? 1 : 0
> Well, actually we almost never use the 0/1 bool value and this one
> case is more of an exception. If you take a closer look, we don't have
> setters/getters ...,
I scanned all your patches and I did not find a single place where
the macros was not used as a bool value.
Mostly in if (ide_drv_drq_int(drive)) .. and ide_drv_drq_int(drive) &&
statements.
> So the macros as such are
> simply to save some stack and improve readability and since the whole
> thing keeps changing we might just as well turn them into static inlines
> one fine day :).
gcc will optmize the static inline functions so there is no drawbacks only
better type checking.
Sam
--
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
