On Wed, 16 Jul 2008 17:31:54 -0700
Andrew Morton <akpm@linux-foundation.org> wrote:
> On Tue, 15 Jul 2008 22:27:51 +0900
> Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp> wrote:
>
> > I update cobalt_lcdfb driver.
> >
> > v3
> > - fix read/write count boundary check.
> > - add <include/uaccess.h>.
> > - fix MODULE_AUTHOR.
> >
> > v2
> > - add dpends MIPS_COBALT in Kconfig.
> > - add handling of signal_pending().
> > - check overflow of read/write count.
> >
> > v1
> > - first release.
>
> This driver has been merged into the subsystem tree for a long time and
> has been reviewed and has been perhaps tested by others. Sending a
> complete new version of the entire thing is really quite an unfriendly
> act. It basically invalidates all the review and test work which
> everyone else has done.
>
> This is why I will almost always turn replacement patches into
> incremental patches - so I can see what changed. But that's only good
> for me. Everyone else who is reading your new version won't bother
> doing that.
>
>
> So I generated the incremental patch and:
>
> - The driver I have already includes linux/uaccess.h, so that change
> was unneeded.
>
> - The driver I have has already fixed the reject in
> drivers/video/Makefile, so I get to fix it again, ho hum.
>
> Here's what I ended up committing:
>
> drivers/video/Kconfig | 2 +-
> drivers/video/cobalt_lcdfb.c | 24 ++++++++++++++++++------
> 2 files changed, 19 insertions(+), 7 deletions(-)
OK.
If I update the driver to v4, I'll send the incremental patch.
Thanks a lot,
Yoichi
|