Re: [PATCH 2/2] dw_dmac: make usage of dw_dma_slave optional

[Andy Shevchenko <andriy.shevchenko@linux.intel.com>]

On Fri, 2012-11-16 at 09:59 +0530, Viresh Kumar wrote: 
> On 15 November 2012 23:28, Andy Shevchenko <andy.shevchenko@gmail.com> wrote:
> > On Thu, Nov 15, 2012 at 5:38 PM, Viresh Kumar <viresh.kumar@linaro.org> wrote:
> >> On 15 November 2012 20:57, Andy Shevchenko
> >> <andriy.shevchenko@linux.intel.com> wrote:
> >>> Well, the prep_* should assign the value due to changes of check in the
> >>> dwc_descriptor_complete. Otherwise we will potentially skip some
> >>> important piece of code.
> >>
> >> What i meant to say was, set_runtime_config() must have already done this part.
> >
> > On one hand it is true. On the other - *_prep* functions use
> > explicitly passed parameter. I doubt there is a consistency between
> > value in slave config passed via dwc_control and value passed as
> > explicit function parameter.
> 
> I believe it should be consistent.
> 
> @Vinod: Why have we duplicated direction? Once in prep_* and then in
> slave_config?

Documentation and TODO list say that slave_config.direction is kinda
deprecated.

> 
> --
> viresh

-- 
Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Intel Finland Oy