Tue, 8 Apr 2014 13:24:09 +0300 от Tomi Valkeinen <tomi.valkeinen@xxxxxx>:
> On 08/04/14 13:15, Alexander Shiyan wrote:
> > Tue, 8 Apr 2014 12:56:24 +0300 от Tomi Valkeinen <tomi.valkeinen@xxxxxx>:
> >> Hi,
> >>
> >> On 20/03/14 18:24, Alexander Shiyan wrote:
> >>> This series adds a new framebuffer driver for Cirrus Logic CLPS711X
> >>> CPUs. Since all code rewritten from scratch, patch is designed as a
> >>> replacement of the old (not updated for a long time) for a new one.
> >>
> >> Why was it rewritten from scratch? It's almost always better to improve
> >> the old one, than write a new, replacing one.
> > Hello.
> >
> > Most of the code has been rewritten. It will be very difficult to see the changes,
> > if it is presented as a patch for the old driver.
>
> Well, now it's impossible to see the changes, except opening both files
> and going through line by line. You don't even list any new features in
> the description, or give any reason why a new driver is required.
>
> And of course one huge patch changing the old driver to the new form is
> bad also.
>
> The proper way would be to gradually change the old driver with a patch
> series. Then it's possible to review the patches and see what is
> actually changed.
Yes, but in this case, I think it's useless work. Current driver is intended
only for one or two boards. New version supports DT and removes the last
usage of <mach/hardware.h> for this platform. I will show all the new
benefits in v2.
---
��.n��������+%������w��{.n�����{����n�r������&��z�ޗ�zf���h���~����������_��+v���)ߣ�
