Re: 答复: [PATCH 3/4]new driver initial patch 3

Michal Suchanek hramrach at
Wed Aug 1 03:40:37 PDT 2012


On 1 August 2012 11:41, Aaron.Chen  陈俊杰 <aaron.chen at> wrote:
> Hi,
> From 7b54308ec7f2d7f2515298aa2989ea21a89161ed Mon Sep 17 00:00:00 2001
> From: Aaron Chen <aaron.chen at>
> Date: Wed, 1 Aug 2012 16:58:35 +0800
> Subject: [PATCH 3/4] new driver initial patch 3

This is again totally useless summary.

> new driver added to the source tree and deleted the old useless files.

This is not true, it was added previously.

If I were to nitpick these files are not useless, the current driver
is probably used by some people. Also the old files are not removed.
They are updated and nearly 100% rewritten in the process but even
when whitespace is considered some code remains unmodified. I suspect
more would be common if diff patience algorithm was used without
regard to whitespace.

> These files works for all of the smi graphics chips.

The old files support a wide range of chips or so it seems by the chip
list enums.

I am sure you did not work on useless driver for years so this might
be one of the right places to list the features added and removed by
the new driver as compared to the current driver.

You replace copyright notices on files you modify. This is NOT how
copyright is managed. When you modify a file that you did not write
you add you name among the author names, ad DO NOT DELETE the names of
authors of previous revisions of the  file.

I also noticed that all copyright notices in all files are dated (c) 2012.

For some reason it is a practice to record history of files in the
copyright notices - eg.

(c) 2007, 2009-2011 John Doe

which means that the file was written in 2007 and modified in 2009,
2010 and 2011.

I would be surprised if the driver files you submit in the previous
patch all just came into existence this year. From this history
standpoint the copyright notice is inaccurate although you  can
probably claim you touched all of the files.

Maybe you should discuss with your legal department if this practice
is relevant to you.



More information about the xorg-devel mailing list