Re: [External] Re: [PATCH 01/16] Cherry pick a set of patches that enables multifd zero page feature.

[Hao Xiang]

On Fri, Oct 27, 2023 at 6:22 AM Peter Maydell <peter.maydell@linaro.org> wrote:
>
> On Fri, 27 Oct 2023 at 13:32, Fabiano Rosas <farosas@suse.de> wrote:
> >
> > Hao Xiang <hao.xiang@bytedance.com> writes:
> >
> > > Juan Quintela had a patchset enabling zero page checking in multifd
> > > threads.
> > >
> > > https://lore.kernel.org/all/20220802063907.18882-13-quintela@redhat.com/
> >
> > Hmm, risky to base your series on code more than an year old. We should
> > bother Juan so he sends an updated version for review.
>
> Also, if we do need to base another series on some older
> patches, the usual way to do this is to include those
> patches in the series in a suitably split-up manner, not
> to squash them all together into a single patch (with no
> signed-off-by tag).

Thanks for the suggestion. I didn't think that through earlier but I
will make sure I
send the next patchset version with Juan's changes in its original
commits format.

>
> For "assume this as a baseline, please review the other
> patches but this patch is not fit to go in", that should make
> the whole patchset into an RFC patch so wo don't accidentally
> apply this patch.
>

Doing zero page checking in the sender threads improves performance. I
would like to see that change getting in sooner. But if that is not
likely the case, I can either make this an RFC or just rebase the
changes to the current implementation (zero page checking in the main
loop).

> thanks
> -- PMM