[U-Boot] [PATCH 1/2] mmc: tmio: Pass full address to tmio_sd_addr_is_dmaable()

[Masahiro Yamada]

On Wed, Oct 10, 2018 at 1:17 AM Marek Vasut <marek.vasut at gmail.com> wrote:
>
> On 10/09/2018 05:35 PM, Masahiro Yamada wrote:
> > On Tue, Oct 9, 2018 at 11:55 PM Marek Vasut <marek.vasut at gmail.com> wrote:
> >>
> >> On 10/09/2018 02:24 PM, Masahiro Yamada wrote:
> >>> Hi Marek,
> >>
> >> Hi,
> >>
> >>> On Tue, Oct 9, 2018 at 8:26 PM Marek Vasut <marek.vasut at gmail.com> wrote:
> >>>>
> >>>> Pass the entire source data pointer to tmio_sd_addr_is_dmaable()
> >>>
> >>>
> >>> This statement sounds like
> >>> the current code is passing the pointer address only partially.
> >>> Is it right?
> >>
> >> With this change it is.
> >
> >
> > Is anything wrong with my code?
>
> Don't think so.
>
> > How about your patch title
> > "mmc: tmio: Pass full address to tmio_sd_addr_is_dmaable()" ?
> >
> > Does it mean my code is not passing full address?
>
> Could use a rephrasing, yeah
>
> >>>> so we don't have to apply casts throughout the code.
> >>>
> >>> I do not understand this either
> >>> since I see a cast in your code too.
> >>
> >> There is a cast, but it's isolated to this function.
> >>
> >>> In the previous code, the caller casts src->address
> >>> when it passes it to tmio_sd_addr_is_dmaable().
> >>>
> >>> In the new code, 'src' is casted
> >>> in tmio_sd_addr_is_dmaable().
> >>>
> >>> To me, you just moved the location of casting.
> >>> What is the difference (i.e. benefit)?
> >>
> >> I moved the cast from the code into the function, which I think is cleaner.
> >
> > I do not think so.
>
> So would you prefer to see stuff like
>
> function foo(long bar)
> {...}
>
> foo((cast)baz);
>
> ...
>
> foo((cast)quux);
>
> In the code  :)



It is a hypothetical situation.

If there were multiple function calls,
I would agree with you.




-- 
Best Regards
Masahiro Yamada