<div dir="ltr"><br><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Jun 23, 2020 at 7:56 PM Philip Reames <<a href="mailto:listmail@philipreames.com">listmail@philipreames.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
  
    
  
  <div>
    <p><br>
    </p>
    <div>On 6/22/20 2:34 AM, Manuel Klimek via
      llvm-dev wrote:<br>
    </div>
    <blockquote type="cite">
      
      <div dir="ltr">
        <div class="gmail_quote">
          <div dir="ltr" class="gmail_attr">On Sat, Jun 20, 2020 at 1:45
            AM Zachary Turner via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>>
            wrote:<br>
          </div>
          <blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
            <div>
              <div dir="auto">There’s also some feature regressions in
                GH vs Phab.</div>
            </div>
            <div dir="auto"><br>
            </div>
            <div dir="auto">You *must* initiate a review via a pull
              request, and pull request by definition compares your
              working copy against master.</div>
            <div dir="auto"><br>
            </div>
            <div dir="auto">This is not very compatible with LLVMs
              approach to incremental development.  For example, if you
              ask someone to break a large patch into 5 smaller patches,
              with Phab this is very easy because you can upload the
              diff between N and N+1, then N+1 and N+2, etc.</div>
            <div dir="auto"><br>
            </div>
            <div dir="auto">But with the GH workflow in order to get a
              review on N+4 you have to include all the changes from all
              the earlier revisions as well.</div>
            <div dir="auto"><br>
            </div>
            <div dir="auto">The way around this is to fork and make 5
              branches in your fork, then base each branch off the
              previous one.  But now what do you do if someone requests
              a change on the first one?</div>
            <div dir="auto"><br>
            </div>
            <div dir="auto">Overall it’s a pretty serious limitation if
              you’re used to Phab, and I would evaluate very carefully
              if you’re thinking of going this route</div>
          </blockquote>
          <div><br>
          </div>
          <div>Are you volunteering to drive Phab maintenance and keep
            it up & running?</div>
        </div>
      </div>
    </blockquote>
    <p>This really seems like a question for the board, rather than any
      individual. </p></div></blockquote><div>I've mentioned that trying to get this funded through official LLVM channels is an option, but from all the information I have, it is not a short-term viable option.</div><div>Having individuals step up and drive it is a significantly less coordination-heavy and thus, overall, cheaper way.</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div><p> If you're resigning and the community values phab,
      having the board weigh in on cost to support the tool seems
      worthwhile.  I suspect that cost will be high enough that we will
      migrate to something free, but we should at least have an informed
      discussion.  <br>
    </p>
    <p>(In case it's not clear, "cost" above is specifically meant in
      both the financial and non-financial sense.  This might be a case
      where a contractor is worth considering instead of relying solely
      on volunteer labor.</p></div></blockquote><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div><blockquote type="cite">
      <div dir="ltr">
        <div class="gmail_quote">
          
          <blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
            <div><br>
              <div class="gmail_quote">
                <div dir="ltr" class="gmail_attr">On Fri, Jun 19, 2020
                  at 4:35 PM Zachary Turner <<a href="mailto:zturner@roblox.com" target="_blank">zturner@roblox.com</a>>
                  wrote:<br>
                </div>
                <blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
                  <div>
                    <div dir="auto">Yes GH has a Squash & Merge
                      option that works well.  It’s what we use.  We use
                      the GH web interface for all of this though, if
                      you’re supporting command line you may need some
                      custom tooling to support this.</div>
                    <div dir="auto"><br>
                    </div>
                    <div dir="auto">The biggest thing is that it
                      requires a lot of mental retraining to get out of
                      the rebasing mindset for daily development </div>
                  </div>
                  <div dir="auto"><br>
                  </div>
                  <div dir="auto">On Fri, Jun 19, 2020 at 4:32 PM David
                    Blaikie <<a href="mailto:dblaikie@gmail.com" target="_blank">dblaikie@gmail.com</a>>
                    wrote:<br>
                  </div>
                  <div>
                    <div class="gmail_quote">
                      <blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On Fri, Jun
                        19, 2020 at 4:23 PM Zachary Turner via llvm-dev<br>
                        <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>>
                        wrote:<br>
                        ><br>
                        > I use GH daily at my current employer and i
                        can tell you that the issues with rebasing are
                        very real.  Unless you only use merge commits
                        you are going to have a very bad time<br>
                        <br>
                        Would it be practical to use merge commits
                        during review (never<br>
                        rebasing) & then rebasing/squashing to
                        commit to the main line?<br>
                        (guessing that might still make looking back at
                        the history of the<br>
                        review difficult?)<br>
                        <br>
                        - Dave<br>
                        <br>
                        ><br>
                        > On Fri, Jun 19, 2020 at 2:23 PM Mehdi AMINI
                        via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>>
                        wrote:<br>
                        >><br>
                        >><br>
                        >><br>
                        >> On Fri, Jun 19, 2020 at 1:15 PM Keith
                        Smiley <<a href="mailto:keithbsmiley@gmail.com" target="_blank">keithbsmiley@gmail.com</a>>
                        wrote:<br>
                        >>><br>
                        >>> FWIW GitHub's code review tools
                        have improved significantly in the past few
                        years. At this point with reviews and manual
                        control over resolving / unresolving comments I
                        think many previous complaints I've seen about
                        GitHub vs Phabricator have been alleviated.<br>
                        >><br>
                        >><br>
                        >> To be clear: this wasn't an outdated
                        comment here, I'm using GitHub very frequently
                        *right now* as I'm reviewing contributions to
                        TensorFlow.<br>
                        >><br>
                        >>><br>
                        >>><br>
                        >>> I also believe there's significant
                        value for newcomers and casual contributors
                        (like myself) in using the same tool as so many
                        other major open source projects.<br>
                        >>><br>
                        >>> On Fri, Jun 19, 2020 at 13:04 Mehdi
                        AMINI via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>>
                        wrote:<br>
                        >>>><br>
                        >>>><br>
                        >>>><br>
                        >>>> On Fri, Jun 19, 2020 at 9:56 AM
                        Hubert Tong via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>>
                        wrote:<br>
                        >>>>><br>
                        >>>>> On Fri, Jun 19, 2020 at
                        12:32 PM Anton Korobeynikov via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>>
                        wrote:<br>
                        >>>>>><br>
                        >>>>>> Just my 2 cents here:
                        we are working on enabling this as a part of<br>
                        >>>>>> bugzilla migration as
                        PRs and issues are very tied inside GitHub. Stay<br>
                        >>>>>> tuned for updates!<br>
                        >>>>><br>
                        >>>>> I am not aware that the
                        previous long thread about usage of GitHub PRs
                        in place of Phabricator reviews got anywhere
                        near the point where the option of Phabricator
                        reviews was being dropped<br>
                        >>>><br>
                        >>>><br>
                        >>>> That's my impression as well, I
                        find GitHub review is frustrating in comparison
                        to phab, in particular the way comments are
                        handled across updates, unless you stick to
                        never rebase and only append commits and merges
                        from master. This is unfortunately not
                        compatible with the LLVM repo history right now.<br>
                        >>>><br>
                        >>>> <a href="https://www.phacility.com" rel="noreferrer" target="_blank">https://www.phacility.com</a>
                        offers hosting for Phabricator, could we look
                        into this instead?<br>
                        >>>><br>
                        >>>> --<br>
                        >>>> Mehdi<br>
                        >>>><br>
                        >>>><br>
                        >>>>><br>
                        >>>>> . The original post on this
                        thread indicated interest in not maintaining
                        Phabricator. How does that affect the
                        availability of Phabricator? Does this mean that
                        the community is going to move to GitHub PRs
                        because the choice of Phabricator is being taken
                        away?<br>
                        >>>>><br>
                        >>>>>><br>
                        >>>>>><br>
                        >>>>>> On Fri, Jun 19, 2020 at
                        3:45 PM Manuel Klimek via llvm-dev<br>
                        >>>>>> <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>>
                        wrote:<br>
                        >>>>>> ><br>
                        >>>>>> > -Chris' outdated
                        email, +Chris' correct email :)<br>
                        >>>>>> ><br>
                        >>>>>> > On Fri, Jun 19,
                        2020 at 2:01 PM Manuel Klimek <<a href="mailto:klimek@google.com" target="_blank">klimek@google.com</a>>
                        wrote:<br>
                        >>>>>> >><br>
                        >>>>>> >> Hi folks,<br>
                        >>>>>> >><br>
                        >>>>>> >> phabricator
                        maintenance is a topic that has been lying
                        dormant for a while now.<br>
                        >>>>>> >><br>
                        >>>>>> >> That
                        subsequently creates a non-optimal user
                        experience.<br>
                        >>>>>> >> For me
                        personally, given that github provides a secure
                        PR infrastructure, the additional effort
                        required to keep Phab going is not an investment
                        I'm personally willing to make. I understand
                        that there are unique selling points for Phab in
                        its UI compared to github PRs, but there are
                        also significant downsides in the effort to
                        integrate with Phab that github PRs make easier.<br>
                        >>>>>> >><br>
                        >>>>>> >> Thus, I see
                        two options:<br>
                        >>>>>> >> 1. somebody
                        volunteers to take on Phabricator maintenance
                        and figures out a way to fund it, either through
                        the LLVM foundation or some other means (I'd
                        love for us at Google to pay for it directly and
                        give folks outside Google access, but that is
                        unfortunately a hard problem for a variety of
                        reasons). I'd be happy to help to provide a DB
                        snapshot for the migration, of course.<br>
                        >>>>>> >><br>
                        >>>>>> >> 2. We switch
                        to github PRs<br>
                        >>>>>> >><br>
                        >>>>>> >> Thoughts?<br>
                        >>>>>> >> /Manuel<br>
                        >>>>>> >><br>
                        >>>>>> >><br>
                        >>>>>> >> On Thu, Jun
                        18, 2020 at 6:42 PM Raphael Isemann <<a href="mailto:teemperor@gmail.com" target="_blank">teemperor@gmail.com</a>>
                        wrote:<br>
                        >>>>>> >>><br>
                        >>>>>> >>> Friendly
                        ping<br>
                        >>>>>> >>><br>
                        >>>>>> >>> Am Do., 9.
                        Apr. 2020 um 16:04 Uhr schrieb Alexandre Ganea<br>
                        >>>>>> >>> <<a href="mailto:alexandre.ganea@ubisoft.com" target="_blank">alexandre.ganea@ubisoft.com</a>>:<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > cc
                        Paul / MyDeveloperDay<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > De :
                        llvm-dev <<a href="mailto:llvm-dev-bounces@lists.llvm.org" target="_blank">llvm-dev-bounces@lists.llvm.org</a>>
                        De la part de David Blaikie via llvm-dev<br>
                        >>>>>> >>> >
                        Envoyé : April 8, 2020 10:21 PM<br>
                        >>>>>> >>> > À :
                        Raphael “Teemperor” Isemann <<a href="mailto:teemperor@gmail.com" target="_blank">teemperor@gmail.com</a>>;
                        Manuel Klimek <<a href="mailto:klimek@google.com" target="_blank">klimek@google.com</a>><br>
                        >>>>>> >>> > Cc :
                        llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>><br>
                        >>>>>> >>> > Objet
                        : Re: [llvm-dev] Outdated Phabricator version on
                        <a href="http://reviews.llvm.org" rel="noreferrer" target="_blank">reviews.llvm.org</a>
                        breaks Google authentication since today<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > hey
                        Manuel - are you/do you know who's likely to be
                        doing any upkeep on Phabricator these days?
                        Might need an update for this...<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > -
                        Dave<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > On
                        Wed, Apr 8, 2020 at 5:57 AM Raphael “Teemperor”
                        Isemann via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>>
                        wrote:<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > Hi
                        all,<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > I’m
                        using my Google account to log into my
                        Phabricator account on <a href="http://reviews.llvm.org" rel="noreferrer" target="_blank">reviews.llvm.org</a> .
                        Since today that no longer works as I don’t seem
                        to get any reply from <a href="http://reviews.llvm.org" rel="noreferrer" target="_blank">reviews.llvm.org</a>
                        when I’m logged into my account. It tried
                        logging out which fixes the issue of <a href="http://reviews.llvm.org" rel="noreferrer" target="_blank">reviews.llvm.org</a>
                        not loading, but when I try to login I just get
                        the following error:<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > >
                        Expected to retrieve an "account" email from
                        Google Plus API call to identify account, but
                        failed.<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > After
                        some searching it seems that this error is due
                        to the Google Plus API being shutdown and the
                        Phabricator folks replaced that logic (including
                        this error message string) a year ago here [1]<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > I
                        assume we haven’t updated <a href="http://reviews.llvm.org" rel="noreferrer" target="_blank">reviews.llvm.org</a> to
                        whatever latest Phabricator release contains
                        that patch.<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > Not
                        sure who’s currently responsible for updating <a href="http://reviews.llvm.org" rel="noreferrer" target="_blank">reviews.llvm.org</a> so
                        I thought I’ll just drop a mail to the list (and
                        maybe save someone else from figuring out why
                        their login is suddenly broken).<br>
                        >>>>>> >>> ><br>
                        >>>>>> >>> > [1] <a href="https://secure.phabricator.com/D20030" rel="noreferrer" target="_blank">https://secure.phabricator.com/D20030</a><br>
                        >>>>>> >>> >
                        _______________________________________________<br>
                        >>>>>> >>> > LLVM
                        Developers mailing list<br>
                        >>>>>> >>> > <a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
                        >>>>>> >>> > <a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
                        >>>>>> ><br>
                        >>>>>> >
                        _______________________________________________<br>
                        >>>>>> > LLVM Developers
                        mailing list<br>
                        >>>>>> > <a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
                        >>>>>> > <a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
                        >>>>>><br>
                        >>>>>><br>
                        >>>>>><br>
                        >>>>>> --<br>
                        >>>>>> With best regards,
                        Anton Korobeynikov<br>
                        >>>>>> Department of
                        Statistical Modelling, Saint Petersburg State
                        University<br>
                        >>>>>>
                        _______________________________________________<br>
                        >>>>>> LLVM Developers mailing
                        list<br>
                        >>>>>> <a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
                        >>>>>> <a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
                        >>>>><br>
                        >>>>>
                        _______________________________________________<br>
                        >>>>> LLVM Developers mailing
                        list<br>
                        >>>>> <a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
                        >>>>> <a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
                        >>>><br>
                        >>>>
                        _______________________________________________<br>
                        >>>> LLVM Developers mailing list<br>
                        >>>> <a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
                        >>>> <a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
                        >>><br>
                        >>> --<br>
                        >>> --<br>
                        >>> Keith Smiley<br>
                        >><br>
                        >>
                        _______________________________________________<br>
                        >> LLVM Developers mailing list<br>
                        >> <a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
                        >> <a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
                        ><br>
                        >
                        _______________________________________________<br>
                        > LLVM Developers mailing list<br>
                        > <a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
                        > <a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
                      </blockquote>
                    </div>
                  </div>
                </blockquote>
              </div>
            </div>
            _______________________________________________<br>
            LLVM Developers mailing list<br>
            <a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
            <a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
          </blockquote>
        </div>
      </div>
      <br>
      <fieldset></fieldset>
      <pre>_______________________________________________
cfe-dev mailing list
<a href="mailto:cfe-dev@lists.llvm.org" target="_blank">cfe-dev@lists.llvm.org</a>
<a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev</a>
</pre>
    </blockquote>
  </div>

</blockquote></div></div>