Re: [PATCH v2 10/19] spi: dw: Use DMA max burst to set the request thresholds

[Andy Shevchenko <andy.shevchenko@xxxxxxxxx>]

On Mon, May 18, 2020 at 04:43:06PM +0300, Serge Semin wrote:
> On Mon, May 18, 2020 at 04:25:20PM +0300, Andy Shevchenko wrote:
> > On Mon, May 18, 2020 at 3:53 PM Serge Semin
> > <Sergey.Semin@xxxxxxxxxxxxxxxxxxxx> wrote:
> > > On Mon, May 18, 2020 at 02:03:43PM +0300, Andy Shevchenko wrote:
> > > > On Sat, May 16, 2020 at 11:01:33PM +0300, Serge Semin wrote:
> > > > > On Fri, May 15, 2020 at 05:38:42PM +0300, Andy Shevchenko wrote:
> > > > > > On Fri, May 15, 2020 at 01:47:49PM +0300, Serge Semin wrote:

...

> > > > > It's not like anyone cared about padding in this structure in the first place)
> > > >
> > > > I think I have been caring (to some extend).
> > >
> > > Well, If you have then instead of asking to rearrange just two members (which
> > > by the way finely grouped by the Tx-Rx affiliation) why not sending a
> > > patch, which would refactor the whole structure so to be optimal for the x64
> > > platforms? I don't really see why this gets very important for you seeing
> > > Mark is Ok with this. My current commit follows the common driver design
> > > including the DW SSI data members grouping. On the second thought I'll leave
> > > it as is then.
> > 
> > Again same issue here. What is really easy to do for you here, will
> > become a burden and additional churn to anybody else.
> > So, why not to minimize it in the first place? Same with comma in
> > another patch. Sorry, I really don't get it.
> 
> If comma is more or less understandable (though adding it is absolutely
> redundant there and doesn't worth even a bit of time spending for the
> discussion), here you consider the patch from padding point of view.
> The driver developer didn't care about it, but did care about grouping the
> members in a corresponding way. The padding burden will be there anyway and
> should be fixed for the whole structure in an additional patch. Until then
> the way of grouping should be preserved.

Like you said, we spent already much more time than that simple change can be
satisfied. And like you said, "deleloper ... did care about groupping members
in a corresponding way". So, if we look at this in the original code, my
suggestion, besides padding benefit, is consistent with existing pattern in
that data structure.

Note, I agree on extern keyword change can be postponed (it was in the original
code), but here you introduce a new code...

-- 
With Best Regards,
Andy Shevchenko