Re: [PATCH v4 2/2] xen/arm: Enlarge identity map space to 10TB

[Leo Yan <leo.yan@xxxxxxxxxx>]

Hi Julien,

On Mon, Nov 06, 2023 at 09:39:24AM +0000, Julien Grall wrote:

[...]

> > I would like to check if here is anything specific I should follow up
> > on. Based on the discussion in this thread, I've come to the following
> > conclusions:
> > 
> > - Remove the fixes tags;
> > - Add a description in commit log, something like:
> >    "Since commit 1c78d76b67e1 ('xen/arm64: mm: Introduce helpers to
> >     prepare/enable/disable the identity mapping'), Xen will fail to boot
> >     up if it's loaded in memory above 2TB. This commit fixes the
> >     regression introduced by that commit."
> > - Add tages:
> >    A review tag from Michal Orzel
> >    A review tag from Bertrand Marquis
> >    A test tag from Henry Wang
> > 
> > Should I repin a new patch set to address the items mentioned above?
> 
> You will also want to update the documentation after
> "docs/arm: Document where Xen should be loaded in
> memory"

Will do.

> > Another question is for the 'Release-acked-by' tag.  Henry gave this
> > tag, but I don't know how to handle it if I need to respin this patch.
> > Seems to me this is a special tag only for release process, so I don't
> > need to include it in the new patch, right?
> 
> The release-acked-by tag is only necessary during freeze period if the patch
> will land in the next release (i.e. 4.18). In this case, your patch will be
> part of the 4.19, so you can remove the release-acked-by.

Okay, I will _not_ include release-acked-by tag in the respin.

> If your patch was targeting 4.19, then it is usually fine to keep the
> release-acked-by even for the respin. It means that the release manager is
> happy for the patch to go assuming the patch has all the necessary reviews.

Thanks for explaination.

Leo