Is there any hope (format-patch)??

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Is there any hope (format-patch)??

Matti Kaasinen
Hi!

Is there any hope with following procedure:
I took reporitory from linux git:
# git clone
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git

For getting patches to make recent version from v2.6.26-rc3 I executed:
# git format-patch -o patchdir v2.6.26-rc3..origin

This produced 9486 patches, two of which were empty. They both appeared
to have two references in git.kernel.org archive so that newer ones were
empy.

Then I checked out v2.6.26-rc3 to a new branch and patched it with
at91patch/maxim.org.za that was produced against v2.6.26-rc3. That
worked out without complaints.

Then I ran:

#git quiltimport --patches patchdir

I suppose, it was the fourth patch of 9486 that failed. Fail did not
come from any of those parts patched with at91 patch. Is this anything
that should work? at91 patch fails if I try it right-away over the HEAD
branch.

cheers,
Matti Kaasinen

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to [hidden email]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Reply | Threaded
Open this post in threaded view
|

Re: Is there any hope (format-patch)??

Daniel Barkalow
On Fri, 1 Aug 2008, Matti Kaasinen wrote:

> Hi!
>
> Is there any hope with following procedure:
> I took reporitory from linux git:
> # git clone
> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
>
> For getting patches to make recent version from v2.6.26-rc3 I executed:
> # git format-patch -o patchdir v2.6.26-rc3..origin

format-patch isn't going to work too well with non-linear history. When
two people make nearby or overlapping changes which get merged later, and
this gets turned into a linear sequence of changes, there's no
possible patch that will accurately reflect the change which got ordered
second.

> Then I checked out v2.6.26-rc3 to a new branch and patched it with
> at91patch/maxim.org.za that was produced against v2.6.26-rc3. That worked out
> without complaints.

It sounds like you really just want to do "merge origin" now, and skip the
whole patch series thing.

You'll probably get some conflicts (or applying the patch directly to
origin would have worked), but they should be relatively easy to resolve.

        -Daniel
*This .sig left intentionally blank*
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to [hidden email]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Reply | Threaded
Open this post in threaded view
|

Re: Is there any hope (format-patch)??

Matti Kaasinen-2
In reply to this post by Matti Kaasinen
Thanks Daniel,
some comments

> ------------------------------------------------------------------------
> On Fri, 1 Aug 2008, Matti Kaasinen wrote:
>
> >/ Hi!/
> >/ /
> >/ Is there any hope with following procedure:/
> >/ I took reporitory from linux git:/
> >/ # git clone/
> >/ git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git/
> >/ /
> >/ For getting patches to make recent version from v2.6.26-rc3 I executed:/
> >/ # git format-patch -o patchdir v2.6.26-rc3..origin/
>
> format-patch isn't going to work too well with non-linear history. When
> two people make nearby or overlapping changes which get merged later, and
> this gets turned into a linear sequence of changes, there's no
> possible patch that will accurately reflect the change which got ordered
> second.
>
> >/ Then I checked out v2.6.26-rc3 to a new branch and patched it with/
> >/ at91patch/maxim.org.za that was produced against v2.6.26-rc3. That worked out/
> >/ without complaints./
>
> It sounds like you really just want to do "merge origin" now, and skip the
> whole patch series thing.
>  
Ok, I'll try how it works, when I get back to my desk. Well, I'm running
out of time and possibly I need to stay in the current version.
> You'll probably get some conflicts (or applying the patch directly to
> origin would have worked),
There came quite a lot of complaints from patching directly the origin.
Strange that some were complaints of trying to re-create files. So, I
just wonder if that patching really is needed. I suppose I should try to
ask that from maxim.org.za
>  but they should be relatively easy to resolve.
>  
Well, hope so - direct patching to origin gave quite lot of complaints.
-Matti
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to [hidden email]
More majordomo info at  http://vger.kernel.org/majordomo-info.html