On Wed, 2015-12-09 at 04:10 -0600, Bucur Madalin-Cristian-B32716 wrote:

quoted

-----Original Message-----
From: David Miller [mailto:davem@davemloft.net]

From: Scott Wood <redacted>
Date: Tue, 8 Dec 2015 16:44:38 -0600

quoted

On Tue, 2015-12-08 at 14:18 -0600, Andy Fleming wrote:

quoted

On Thu, Dec 3, 2015 at 1:19 AM,  [off-list ref] wrote:

quoted

From: Igal Liberman <redacted>

This patch adds the Ethernet MAC driver supporting the three
different types of MACs: dTSEC, tGEC and mEMAC.

Signed-off-by: Igal Liberman <redacted>

[...]

quoted

+
+MODULE_LICENSE("Dual BSD/GPL");
+
+MODULE_AUTHOR("Emil Medve [off-list ref]");

I imagine this email address doesn't exist anymore, or won't soon.
This is also an issue in the ethernet driver (with my old address).
I'm not sure what the right approach is, but we shouldn't be putting
obsolete email addresses in the kernel.

I don't think a MODULE_AUTHOR tag makes sense for drivers like this that

had a

quoted

lot of people work on them.  git history is better for giving credit in

such

quoted

cases.

Agreed.

I've already removed the MODULE_AUTHOR tag from the dpaa_eth code, this will
be removed as well. We are required to keep the Signed-off by: tags of the
original authors for the patches and also were instructed to keep the
original
Freescale email addresses even if the patch authors have different email
addresses now.

What does it mean to say that you're "keeping" the Signed-off-by when that
signoff was made on a very different patch and/or some smaller patch that got
squashed in?  You only need to include other people's signoff if you're
passing the patch on under section (c) of the DCO.  In this case you can use
section (a), and just write a sentence in the changelog giving any appropriate
credit to others.

-Scott