Re: [PATCH v1 1/1] OvmfPkg/SecMain: Fix stack switching to permanent memory fail

[Leif Lindholm <leif.lindholm@linaro.org>]

On Tue, Sep 05, 2017 at 10:45:57AM -0700, Jordan Justen wrote:
> On 2017-09-04 14:25:48, Laszlo Ersek wrote:
> > On 09/04/17 17:11, Ge Song wrote:
> > > On 09/04/2017 09:34 PM, Jordan Justen wrote:
> > > 
> > >> On 2017-09-03 16:55:36, Laszlo Ersek wrote:
> > >>> On 09/03/17 04:12, Ge Song wrote:
> > >>>> In earlier PEI stage, temporary memory(Cache As Ram) is employed as
> > >>>> stack
> > >>>> and heap. We move them to the new room and do some relocation fixup
> > >>>> when
> > >>>> permanent memory becomes available. TemporaryRamMigration()
> > >>>> is responsible for switching the stack.
> > >>> [...]
> > >>>
> > >>>> Cc: Jordan Justen <jordan.l.justen@intel.com>
> > >>>> Cc: Laszlo Ersek <lersek@redhat.com>
> > >>>> Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
> > >>>> Contributed-under: TianoCore Contribution Agreement 1.1
> > >>>> Signed-off-by: Ge Song <ge.song@hxt-semitech.com>
> > >>>> ---
> > >>>>   OvmfPkg/Sec/SecMain.c | 2 ++
> > >>>>   1 file changed, 2 insertions(+)
> > >>> (No more separate emails, I promise.)
> > >>>
> > >>> Before you send v2, can you please amend the patch like this:
> > >>>
> > >>>    git commit --amend --author='Ge Song <ge.song@hxt-semitech.com>'
> > >>>
> > >>> It's OK if you mail it out from another email address, but IMO the git
> > >>> authorship should match the first Signed-off-by.
> > >> I disagree with this. Unless the individual is well known to the EDK
> > >> II community as the owner of both email addresses, I think the author
> > >> needs to send it from their author email.
> > >>
> > >> I suppose they can send it from another email, and then reply from
> > >> their other email saying they contributed it under the Tianocore
> > >> Contribution Agreement, but that sounds like a stretch...
> > 
> > I'm fine either way, as long as we can establish some kind of agreement
> > between (two of?) the three email addresses (sender, git author, S-o-b).
> 
> We've generally said they need to match. If it is a one-time
> contribution, then I think it is acceptable for the contributor to
> reply to the patch from the author's email saying: "This is me. I
> contributed it under the Tiancore Contribution Agreement"
> 
> I think any change from this policy should be decided by the stewards.
> 
> Regarding Ge Song's contribution, I think things are now even more
> complicated. It sounds like the contribution is being made by
> ge.song@hxt-semitech.com, but due to "IT issues" v2 is now being
> authored/sent from songgebird@gmail.com.
> 
> Does hxt-semitech.com agree to this contribution? I think given that
> we know ge.song@hxt-semitech.com is the intended author, the patch
> should have ge.song@hxt-semitech.com as the author with
> ge.song@hxt-semitech.com in the Signed-off-by.
> 
> Ideally it should be sent from ge.song@hxt-semitech.com, but if that
> is not possible, then ge.song@hxt-semitech.com should reply to the
> patch sent by songgebird@gmail.com re-iterating the
> Signed-off-by/Contributed-under is from ge.song@hxt-semitech.com.
> 
> Ge Song, Can you send a v3 with the author reset to
> ge.song@hxt-semitech.com (see lersek's git commit command above) and
> the Signed-off-by using ge.song@hxt-semitech.com, and then reply to
> the patch posting from ge.song@hxt-semitech.com to re-iterate the
> Signed-off-by and Contributed-under?

While obviously not an ideal situation, it is not the first time I
have come across this sort of issue. (And I doubt it will be the
last.)

Your suggestion sounds like a good workaround.
For a series, a reply to the cover-letter should be sufficient.

> > > OK, I'll try to send it from the same email address next time, the reason
> > > that using two different mail addresses is because the author email is an
> > > exchange account and it's hard to find the mail server preferences for
> > > git-send-email...
> 
> Sorry for the hassle, but maybe if you are able to work with your IT
> department to setup git send-email then you'll be able to contribute
> to other open source project's easier from your
> ge.song@hxt-semitech.com email address.

In my experience, getting IT departments on board with this sort of
thing can be very difficult. Clearly, it should be attempted.

Since hxt-semtech.com is a Linaro member, I will try to raise the
importance of getting this situation resolved internally.

/
    Leif