On Thu, Apr 05, 2007 at 08:48:21PM +0200, svn@openbios.org wrote:
Author: uwe
[..]
Signed-off-by: Corey Osgood corey_osgood@verizon.net Acked-by: Uwe Hermann uwe@hermann-uwe.de
Sorry to bring this up again, but wasn't Uwe supposed to sign off if he commits? And a commiter's sign-off is equivalent to an ack?
I never understood the result of the discussion for this case. (How to sign off/ack when also commiting a patch.)
//Peter
On Thu, Apr 05, 2007 at 09:52:57PM +0200, Peter Stuge wrote:
On Thu, Apr 05, 2007 at 08:48:21PM +0200, svn@openbios.org wrote:
Author: uwe
[..]
Signed-off-by: Corey Osgood corey_osgood@verizon.net Acked-by: Uwe Hermann uwe@hermann-uwe.de
Sorry to bring this up again, but wasn't Uwe supposed to sign off if he commits? And a commiter's sign-off is equivalent to an ack?
I never understood the result of the discussion for this case. (How to sign off/ack when also commiting a patch.)
I don't think we had a clear resolution when we last discussed this.
IMHO this procedure is ok:
- Only add a Signed-off-by if you modified the code. If you're just committing someone else's code unmodified, you don't sign-off (but the patch creator must have signed-off of course; patches without sign-offs must never be committed).
- Any code which gets committed must have at least one Signed-off-by _and_ at least one Acked-by. Thus, if you commit other people's code unmodified you add your Ack (if and only if you think the code looks good, of course).
We should really document the exact details in the wiki, though...
Uwe.
* Uwe Hermann uwe@hermann-uwe.de [070411 22:20]:
On Thu, Apr 05, 2007 at 09:52:57PM +0200, Peter Stuge wrote: I don't think we had a clear resolution when we last discussed this.
IMHO this procedure is ok:
Only add a Signed-off-by if you modified the code. If you're just committing someone else's code unmodified, you don't sign-off (but the patch creator must have signed-off of course; patches without sign-offs must never be committed).
Any code which gets committed must have at least one Signed-off-by _and_ at least one Acked-by. Thus, if you commit other people's code unmodified you add your Ack (if and only if you think the code looks good, of course).
What about the case when a patch is slightly reworked.
I think in this case there should be a Signed-off-by: and an Acked-by: by the committer. Going through the whole review process and finding yet another person person to ack it should not be possible.
This would mean if two people are working on a patch and they both agree it should go in, they can "override" the system. That's no problem for the 4-eyes principle though.
On Wed, Apr 11, 2007 at 10:38:08PM +0200, Stefan Reinauer wrote:
- Uwe Hermann uwe@hermann-uwe.de [070411 22:20]:
On Thu, Apr 05, 2007 at 09:52:57PM +0200, Peter Stuge wrote: I don't think we had a clear resolution when we last discussed this.
IMHO this procedure is ok:
Only add a Signed-off-by if you modified the code. If you're just committing someone else's code unmodified, you don't sign-off (but the patch creator must have signed-off of course; patches without sign-offs must never be committed).
Any code which gets committed must have at least one Signed-off-by _and_ at least one Acked-by. Thus, if you commit other people's code unmodified you add your Ack (if and only if you think the code looks good, of course).
What about the case when a patch is slightly reworked.
I think in this case there should be a Signed-off-by: and an Acked-by: by the committer.
Yes, I agree. For trivial modifications just add your Signed-off-by (and Acked-by) and commit.
For nontrivial stuff, sending a modified patch for review is better/required, of course.
Uwe.
-
Peter Stuge -
Stefan Reinauer -
Uwe Hermann