[PATCH 1/2] drivers/staging/nvec/nvec_kbd.c: Fix checkpatch warnings
Julian Andres Klode
jak at debian.org
Thu Sep 22 09:07:36 PDT 2011
On Thu, Sep 22, 2011 at 08:47:04AM -0700, Greg KH wrote:
> On Thu, Sep 22, 2011 at 04:53:41PM +0300, Leon Romanovsky wrote:
> > On Thu, Sep 22, 2011 at 12:31, Marcos Paulo de Souza
> > <marcos.mage at gmail.com> wrote:
> > > Was fixed some checkpatch warning, like spaces before if statements and others.
> > >
> > Thanks Marcos for your effort, but all nvec related files under heavy rewrite.
> > If we will success in time the code is going to change next merge window.
> > Greg, it is up to you that to do with this patch..
> "heavy rewrite"? Why?
The code was not very well understandable and not really stable
either (as in, it crashed my machine). That's why I decided to
rewrite it, so it can actually be understood and work reliable.
The new code works reliable, is documented, and should finally
be understandable immediately.
> I will not accept a "replace the old version of the driver with the new
> one", so you all hopefully are going to provide me with a series of
> incremental patches, right?
There is one larger patch that replaces the interrupt handler, but
otherwise, yes, nice small patches. It is not really possible to
break most of the interrupt rewrite down into separate chunks due
to the completely changing nature of it.
> Why are you sitting on patches? Please never do that.
We are currently testing them on our 2.6.38-based kernels, they yet
have to be ported to mainline kernels. Mainline kernels still miss
most of the video hardware support, so testing them there is more
complicated. And they are not even a week old, the first patchs
were on Monday, and took a bit of stabilisation until today.
I'm sure that the patches will be ready for you soon (tomorrow?),
they just need some polishing and/or porting.
> I'll take this, and any other cleanup patch that is sent to me, you need
> to be able to handle merges with stuff like that.
This one is OK at least for me, but the next one for nvec_power is
definitely not, as it changes behavior (changing counter to 1 from 0).
Julian Andres Klode - Debian Developer, Ubuntu Member
See http://wiki.debian.org/JulianAndresKlode and http://jak-linux.org/.
More information about the devel