Re: [edk2-devel] [PATCH 3/3] MdeModulePkg: Notify BeforeExitBootServices in CoreExitBootServices

["Ard Biesheuvel" <ardb@kernel.org>]

On Thu, 12 Jan 2023 at 13:24, Laszlo Ersek <lersek@redhat.com> wrote:
>
> On 1/12/23 00:08, Dionna Glaze via groups.io wrote:
> > On Thu, Nov 10, 2022 at 8:55 AM Kinney, Michael D
> > <michael.d.kinney@intel.com> wrote:
> >>
> >> Hi Dionna,
> >>
> >> https://github.com/tianocore/tianocore.github.io/wiki/EDK-II-Development-Process
> >>
> >> I think you are at step 13.  If you have not done so already, please update the
> >> commit messages with all the Reviewed-by and Acked-by tags and make sure your
> >> branch is rebased against the latest edk2/master and update the PR with that
> >> content and verify that all EDK II CI checks pass.
> >>
> >> At that point, it is the responsibility of the EDK II Maintainers to review
> >> the final state of the PR and set the "push" label if everything looks correct.
> >>
> >> At this moment, we are in Soft Freeze and will be in Hard Freeze for the next
> >> 2 weeks.  If this is considered critical for the stable tag release, then
> >> please send a request to Liming with justification for stable tag.  Otherwise,
> >> it will be merged after the stable tag release.
> >>
> >> https://github.com/tianocore/tianocore.github.io/wiki/EDK-II-Release-Planning
> >>
> >> Thanks,
> >>
> >> Mike
> >>
> >
> > Hi Mike, my PR was closed without comment
> > https://github.com/tianocore/edk2/pull/3623 so I rebased and created a
> > new PR after the holidays and hard freeze. I hope this isn't
> > considered bad practice https://github.com/tianocore/edk2/pull/3885
> >
>
> I closed the PR -- similarly to how I manually closed 400+ stale PRs
> then -- because it was not a maintainer-submitted PR with the "push"
> label set. In other words, it was just a personal build, for triggering
> a CI run.
>
> And, in fact regardlessly of whether it was a "push"-labeled maintainer
> PR or just a personal PR, the PR was almost two months old. Clearly
> stale and abandoned -- per edk2 contribution workflow, anyway.
>
> Please excuse me for not explaining this in a comment on the PR, but (as
> I said above) I closed 400+ stale PRs manually within 10-15 minutes on
> the github.com WebUI. The necessary muscle memory training didn't allow
> for individual comments.
>
> (I actually tried scripting the mass-closure at first, with the "gh"
> utility, but something in the github.com data model was broken, and the
> server wouldn't allow me to close those PRs with the utility, even
> though I had authenticated the utility under my account, with a complete
> set of scopes -- see my report at
> <https://github.com/cli/cli/discussions/6814>.)
>
> Regarding your new PR: it's again good for a personal CI run only. If it
> completes fine, please post the patches to the mailing list, using
> git-send-email.
>
> (From browsing recent list traffic, it seems that your colleague Moritz
> Fischer <moritzf@google.com> has posted patches like that to the list;
> please consider consulting with Moritz regarding the git setup (SMTP
> server etc) withing Google. Cc'ing Moritz now.)
>
> When sending the patches like that, please CC the maintainers and
> reviewers that are supposed to review them. Once they are happy with the
> series, one of them will submit a PR with them, and set the "push" label
> on the PR. When the CI run succeeds, the mergify bot will merge *that*
> PR automatically.
>

I think we were already way past this with Dionna's work - numerous
iterations have been posted and discussed, and the merge of the final
version was only deferred because of the soft freeze.