Re: [PATCH] ep93xx: clock: Fix off by one in ep93xx_div_recalc_rate()

From: Nikita Shubin
Date: Wed Sep 11 2024 - 12:35:52 EST


Hello Alexander, Arnd,

On Wed, 2024-09-11 at 16:59 +0200, Alexander Sverdlin wrote:
> Hi Arnd, Nikita,
>
> On Wed, 2024-09-11 at 14:54 +0000, Arnd Bergmann wrote:
> > On Wed, Sep 11, 2024, at 08:14, Nikita Shubin wrote:
> > > Hi Dan!
> > >
> > > Reviewed-by: Nikita Shubin <nikita.shubin@xxxxxxxxxxx>
> > >
> > > Alexander, Arnd
> > >
> > > unfortunately, the ep93xx DT conversion series is also affected
> > > by this
> > > bug.
> >
> > Here is what I did now:
> >
> > 1. applied Dan's patch on a new branch
> > 2. applied the DT conversion series on top of that,
> >    removing that file.
> > 3. applied the first patch (with minor context changes)
> >    in drivers/clk/clk-ep93xx.c again, along with
> >    the MODULE_LICENSE fix I did.
> > 4. finally, merged the entire branch into my for-next
> >    branch so it actually makes it into linux-next
> >
> > My plan now is to keep the branch in linux-next for at
> > least a week and send all the other pull requests for
> > the merge window first. If no other problems show up
> > (either with this branch or my other 6.12 contents),
> > I hope to send it all later in the merge window. If
> > something goes wrong, I'll send only the bugfix as part
> > of my first fixes branch for 6.12 and we'll defer the
> > DT conversion once more.
> >
> > I should have merged it earlier, but wasn't sure about
> > interdependencies with the parts that already got merged
> > elsewhere and with the comments about DTC warnings.
> >
> > From what I can tell, the current state is as good as
> > it gets, as we'll always get more comments or conflicts
> > with new reversions of the series. Let's hope we can
> > address any other issues on top of what I've merged
> > now and stop rebasing.
>
> thanks Arnd for resolving this finally and Nikita for
> your relentless efforts!
>
> PS. I've archived Subject patch now in soc patchwork
> (because I think I've messed up the author info, but
> all of this seems to be obsolete now)
>

thank you for your work and support!