On Tue, Feb 06, 2007 at 10:57:42PM +0100, Carl-Daniel Hailfinger wrote:
Signed-off-by: Adam Kaufman email@example.com Signed-off-by: Stefan Reinauer firstname.lastname@example.org Acked-by: Stefan Reinauer email@example.com Acked-by: Adam Kaufman firstname.lastname@example.org
It seems there is still some confusion about Signed-off-by and Acked-by. The meaning of Acked-by is a real subset of Signed-off-by. So acking a patch you already signed off for is totally meaningless and should be avoided. The Linux kernel guys never ack and sign off the same patch.
I was going to reply to this, too, but for another reason. The last 'Acked-by: Adam Kaufman email@example.com' was not posted publically by Adam anywhere as far as I can see, so it should not appear in the commit message.
As for the meaning of Acked-by, I'm not so sure. We currently use Signed-off-by as a tagging of "I wrote (parts of) this code" (i.e. is has legal importance), and the Acked-by merely as "ok, looks fine" indicator (but you didn't write the code).
Not sure how exactly the Linux kernel folks use these...
Sample scenario: Developer A posts a patch with a Signed-off-by, you (developer B) modify/improve it and add your own Signed-off-by.
Then someone else (developer C) improves the patch even further, and adds his Signed-off-by. Now you (B) post an Acked-by for this last revision of the patch, i.e.
Signed-off-by: A Signed-off-by: B Signed-off-by: C Acked-by: B
This makes sense to me. What would probably _not_ make any sense is
Signed-off-by: A Signed-off-by: B Acked-by: B
because you ack your own patch here, which defeats our purpose of the Acked-by tag (enforce review by at least one developer).
Also, I don't think that a Signed-off-by necessarily means that you ack the patch. It's possible that you just posted an improved version of the patch (and added your Signed-off-by), but you still think that it's not good enough to be committed, i.e. you don't want to ack it at the moment...