[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH 0/3] x86: mwait_idle improvements ported from Linux

>>> On 29.08.13 at 22:20, Matt Wilson <msw@xxxxxxxxxx> wrote:
> On Thu, Aug 29, 2013 at 02:04:54PM +0100, Jan Beulich wrote:
>> 1: x86/mwait_idle: remove assumption of one C-state per MWAIT flag
>> 2: x86/mwait_idle: export both C1 and C1E
>> 3: x86/mwait_idle: initial C8, C9, C10 support
>> Signed-off-by: Len Brown <len.brown@xxxxxxxxx>
>> Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
> Some procedure questions:
> Should these commits mention the upstream Linux commits from which
> they are derived?

That's of questionable use imo.

> Are we following the [] convention for modifications made when
> applying work?

For brief notes I do this; for more extended notes we're used to
insert text between the individual Signed-off-by's (on the
hypervisor side at least).

> Should Len be the Author on these patches?

He being the first one in the Signed-off-by sequence, he'll end up
being the author with the way my commit script works (yes, I
could have added a From: line, but I generally consider this useful
only when this doesn't match the first Signed-off-by, as then it'll
be the person named by From: who's going to be the author; in
no case do I consider the From: line to be useful in a git commit's
description, as there it's clearly redundant with git's meta data).

Albeit it's always questionable whom to consider being the author
when the original commit was significantly reworked. I'm trying to
set some boundary when doing this: When the changes are too
heavy, I drop the original Signed-off-by and name original
authorship in the description instead. For the first two patches on
this series, this was really on the edge between the two models
(but when in doubt I prefer crediting the original author).


Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.