[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1457392822.25457.4.camel@ellerman.id.au>
Date: Tue, 08 Mar 2016 10:20:22 +1100
From: Michael Ellerman <mpe@...erman.id.au>
To: Jiri Kosina <jikos@...nel.org>
Cc: Petr Mladek <pmladek@...e.com>, linuxppc-dev@...abs.org,
jeyu@...hat.com, linux-kernel@...r.kernel.org, rostedt@...dmis.org,
kamalesh@...ux.vnet.ibm.com, duwe@....de,
live-patching@...r.kernel.org, mbenes@...e.cz,
Torsten Duwe <duwe@...e.de>, vojtech@...e.com,
jpoimboe@...hat.com, sjenning@...hat.com
Subject: Re: How to merge? (was Re: [PATCH][v4] livepatch/ppc: Enable
livepatching on powerpc)
On Mon, 2016-03-07 at 23:52 +0100, Jiri Kosina wrote:
> On Mon, 7 Mar 2016, Michael Ellerman wrote:
>
> > > This aligns with my usual workflow, so that'd be my preferred way of doing
> > > things; i.e. you put all the ftrace changes into a separate topic branch,
> > > and then
> > >
> > > - you pull that branch into powerpc#next
> > > - I pull that branch into livepatching tree
> > > - I apply the ppc livepatching support on top of that
> > > - I send a pull request to Linus only after powerpc#next gets merged to
> > > Linus' tree
> > >
> > > Sounds good?
> >
> > Yep, here it is:
> >
> > https://git.kernel.org/cgit/linux/kernel/git/powerpc/linux.git/log/?h=topic/mprofile-kernel
> >
> > aka:
> >
> > git fetch git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux.git topic/mprofile-kernel
>
> Excellent, thanks.
>
> > I haven't merged it into my next yet, but I will tomorrow unless you tell me
> > there's something wrong with it.
>
> There is one remaining issue which I think would be really nice to
> have(TM), and that's Steven's Ack for the whole thing :)
Yeah. He's been on CC the whole time, but he's probably getting a bit sick of
it all, as we're up to about version 15. So I figure if he really hated it he'd
have said so by now :) - but an Ack would still be good.
> For the livepatching part, I don't think we are quite there yet (so maybe
> it'll miss the upcoming merge window anyway).
>
> My primary worry there is what Torsten pointed out, i.e. functions with
> either varargs or more than 8 args needing special care.
Yeah true. My preference would be to merge it, but mark LIVEPATCH as
experimental on powerpc. I think having it in the tree would help it get more
testing, and probably find other bugs too. But it's up to you guys.
> Also, I'd like to have this positively reviewed by at least one more
> livepatching maintainer (I am currently looking into it myself, but my
> understanding of powerpc arch is rather low, so the more eyes, the
> better).
Sure. I can answer powerpc questions, though Torsten is probably the person who
has the best understanding of (livepatching && powerpc).
cheers
Powered by blists - more mailing lists