3.5 Release Date (was Re: A naive question about the speedoptimization of anthony)

Hi all.  According to the 3.5 plan, the original "First Fridays" final release
date for 3.5 was today, Friday the 4th.

I posted in "WeakArray bug" thread that I don't think we need to hold up the
3.5 release for a fix for that particular bug, since it's an old bug, we
haven't agreed on a fix, and even an agreed-upon fix would need some time for
testing.

However, there's also the issue of possibly adding the closure-compatible VM
additions.  I actually don't have a big problem with adding these in 3.5, if
the various VM folks agree that it's a good idea, since they can't really
affect anything else.  I am assuming there are no licensing issues with the VM
additions... they'd just be Squeak-L.  Let's hear some opinions from the
VM-interested folks on this.

The other issue is that we originally planned to have a 3.6 Release Plan ready
as part of the 3.5 final release.  We still have some work to do on that, so
in any case I think it might be a good idea to postpone the 3.5 release to
next Friday, the 11th.  This will give us some time to hash out a rough 3.6
plan (it doesn't have to be that detailed), and also time to decide whether to
include the closure VM additions.  I like the idea of still releasing it on a
Friday, too. :-)

I wouldn't normally be this cavalier about postponing a release date, but in
this case I think there isn't that much demand to see 3.5 right away, since it
doesn't have much new stuff and since 3.4 was released quite recently.  3.6
will probably be a different story.

Does this sound reasonable?

- Doug


-------- Original Message --------
Subject: Re: A naive question about the speed optimization of anthony
Date: Fri, 4 Apr 2003 00:58:08 -0500
From: Doug Way <[email protected]>
Reply-To: The general-purpose Squeak developers
list<[email protected]>
To: The general-purpose Squeak developers
list<[email protected]>


I understand that they're harmless changes also.  I would tend to just 
include them in 3.6alpha and not mess with 3.5gamma at this stage.  
There's not that much disadvantage to adding them in 3.6alpha... they 
could be added right away (since the 3.6alpha update stream is open), 
and the actual closure (non-VM) code wouldn't be added until later in 
3.6alpha anyway, if it is added.

Although if all of the VM maintainers think it is worth adding in 3.5, 
we could consider that.

- Doug Way


On Thursday, April 3, 2003, at 03:48 PM, Daniel Vainsencher wrote:

> Yeah, I see what you mean about it being quite harmless, so I don't 
> feel
> very strongly about it. I do want to make to advance this thing, but I
> don't want to push for something that adds (slight) risk to a release
> that I advocated be short only because it is relatively
> complication-free (ha, theories - see the weakness weaknes thread).
>
> So, I'll accept whatever other people decide, and it's really mostly up
> to Doug and the VM maintainers.
>
> Daniel
>
> Andreas Raab <[email protected]> wrote:
>> Daniel,
>>
>>> Well, making the VMs "closure compatible" soon is a good thing.
>>>
>>> Changing gamma versions isn't.
>>
>> Well, it's your choice. I was just pointing out that the changes are
>> "harmless" in terms of actually affecting anything and therefore 
>> could be
>> considered for inclusion.
>>
>> Cheers,
>>   - Andreas
>>
>>>
>>> I withdraw from this discussion - I don't have anything more to add.
>>>
>>> Doug's call.
>>>
>>> Hmm, just thought about an alternative that might be better -
>>> add it to
>>> 3.6, and use 3.6 to make the 3.5 VMs. If that's the only
>>> difference (and
>>> 3.5 didn't contain a lot), it should be compatible anyway. But, 
>>> again,
>>> whatever you guys decide.
>>>
>>> Daniel
>>>
>>> Andreas Raab <[email protected]> wrote:
>>>>> It'll be 3.6a (3.5 is now gamma),
>>>>
>>>> So what. We're effectively talking about five methods with
>>> no implication
>>>> whatsoever on any part of the VM or the image. Even if they
>>> were entirely
>>>> broken they'd affect nothing. If you want to move towards
>>> that direction
>>>> it's definitely worthwhile considering.
>>>>
>>>> Cheers,
>>>>   - Andreas
>>>
From [email protected] Sat Apr 05 00:53:08 2003
Return-Path: <[email protected]>
Delivered-To: [email protected]
Received: (qmail 1712 invoked from network); 5 Apr 2003 00:53:08 -0000
Received: from mail.gmx.net (213.165.65.60)
  by mail.theinternetone.net with SMTP; 5 Apr 2003 00:53:08 -0000
Received: (qmail 1362 invoked by uid 65534); 5 Apr 2003 00:53:07 -0000
Received: from pD9533734.dip.t-dialin.net (EHLO atlantis) (217.83.55.52)
  by mail.gmx.net (mp006-rz3) with SMTP; 05 Apr 2003 02:53:07 +0200
From: "Andreas Raab" <[email protected]>
To: "'Discussing the Squeak Foundation'"
	<[email protected]>
Subject: RE: [Squeakfoundation]WeakArray bug (was Re: [UPDATES] 3.5gamma)
Date: Sat, 5 Apr 2003 02:52:54 +0200
Message-ID: <000401c2fb0d$b21c9740$8ef7fea9@atlantis>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook, Build 10.0.4510
In-Reply-To: <[email protected]>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106
X-BeenThere: [email protected]
X-Mailman-Version: 2.1
Precedence: list
Reply-To: Discussing the Squeak Foundation
	<[email protected]>
List-Id: Discussing the Squeak Foundation
 <squeakfoundation.lists.squeakfoundation.org>
List-Unsubscribe: <http://lists.squeakfoundation.org/listinfo/squeakfoundation>,
	<mailto:[email protected]?subject=unsubscribe>
List-Archive: <http://lnx-12.ams-2.theinternetone.net/pipermail/squeakfoundation>
List-Post: <mailto:[email protected]>
List-Help: <mailto:[email protected]?subject=help>
List-Subscribe: <http://lists.squeakfoundation.org/listinfo/squeakfoundation>,
	<mailto:[email protected]?subject=subscribe>
X-List-Received-Date: Sat, 05 Apr 2003 00:53:09 -0000

> We need more explicit reviews.

Or more explicit goals ;-) The fix fixed exactly the bug which was =
reported,
namely that the meta class hierarchy doesn't get recompiled correctly.
Nowhere the fix claimed that it "fixes class builder" because there is =
no
definition for it unless you identify the bugs. A word about the SUnit =
tests
I wrote - I wanted this to be a start for more elaborate tests and =
therefore
I wrote a bunch of tests which ensure critical invariants of =
ClassBuilder.=20

To be more explicit: Those tests actually validate if the reshaping
machinery maps instance state correctly. It is not (never has been, =
never
was intended to) a test for recompiling the entire class hierarchy. The
original reason for this was that on a (non-fixed) system, failing the =
test
of recompiling the meta class hierarchy would mean your system is =
@!#^ed.
Nothing I would expect from a test - and therefore this test can only be
added once the bug is fixed.

Cheers,
  - Andreas



> -----Original Message-----
> From: [email protected]=20
> [mailto:[email protected]]=20
> On Behalf Of Daniel Vainsencher
> Sent: Saturday, April 05, 2003 3:14 AM
> To: Discussing the Squeak Foundation
> Subject: [Squeakfoundation]WeakArray bug (was Re: [UPDATES] 3.5gamma)
>=20
>=20
> About the 3.5/3.6 decision, sounds reasonable to me.=20
>=20
> About why we got to the state where one of us says a bug=20
> critical enough
> to stop the release from going out, when we're in gamma - I think we
> included the fix with the implicit goal of "fixing the class builder"
> but with the explicit SUnit test of "classes with instance=20
> variables and
> subclasses can be reshaped properly".=20
>=20
> Nobody said "this fix does X, which is enough. It doesn't do Y, but we
> can live with that", where Y would be "allow the whole=20
> hierarchy to get
> rebuilt", for example.
>=20
> We need more explicit reviews.
>=20
> Daniel
>=20
> Doug Way <[email protected]> wrote:
> >=20
> > Daniel Vainsencher wrote:
> > >=20
> > > <shifting to sqf list>
> > >=20
> > > That's an interesting point. Does that mean we actually=20
> delay 3.5, when
> > > the 3.6 update stream has already started? another=20
> question is, how did
> > > we get into gamma, when apparently, one of our two goals=20
> for the release
> > > isn't actually achieved?
> >=20
> > Tim did mention this as a bug on March 14.  There wasn't=20
> any mention of
> > whether it was a serious enough problem to try to fix in=20
> 3.5, so I ignored it
> > for the moment, hoping that someone would come up with a fix.
> >=20
> > It did appear to be possibly related to the ClassBuilder=20
> problem, but that
> > wasn't certain, either.  The problem occured with or=20
> without the ClassBuilder
> > fix.
> >=20
> > Just now I did a quick check of when the bug was=20
> introduced.  I was guessing
> > that maybe Andreas introduced it with the ClassBuilder=20
> refactoring/cleanup in
> > 3.4alpha, which was when the other ClassBuilder bug was=20
> introduced.  However,
> > that was not the case... I tested 3.2 and the bug is there,=20
> too.  Turns out
> > this bug is as old as the hills... the bug exists back in=20
> 2.6!  It does not
> > exist in 2.4, though.  (It's always nice to have old images=20
> lying around. :-)=20
> > I didn't have a 2.5 image handy, though.)
> >=20
> > So, given that the bug is 3+ years old, and we don't yet=20
> have a fix that we
> > agree on, I think it's pretty safe to say that we don't=20
> need to address this
> > in 3.5.  As soon as someone comes up with a good fix, we=20
> can include it in
> > 3.6alpha.  (Maybe Brent's fix is sufficient, I don't know,=20
> but it hasn't
> > gotten any feedback yet.)
> >=20
> > - Doug Way
> >=20
> >=20
> > > Daniel
> > >=20
> > > Tim Rowledge <[email protected]> wrote:
> > > > Daniel Vainsencher <[email protected]> wrote:
> > > >
> > > > > AFAICT, this fix is KCP territory. Unless there is a=20
> very good reason
> > > > > otherwise, I would await their recommendation (and=20
> not delay the
> > > > > release).
> > > > Well since the 3.5 release was purported to be mostly=20
> to include a fix
> > > > for a very similar bug and the bug in question has=20
> pretty similar
> > > > effects - cannot recompile a number of classes - I'd=20
> say that without a
> > > > fix we really shouldn't even consider 3.5 in beta in=20
> any meaningful
> > > > sense.
> > > >
> > > > tim
> > > > --
> > > > Tim Rowledge, [email protected],=20
> http://sumeru.stanford.edu/tim
> > > > Spellchecker not found. =20
> Press -- to continue ...
> > > _______________________________________________
> > > Squeakfoundation mailing list
> > > [email protected]
> > > http://lists.squeakfoundation.org/listinfo/squeakfoundation
> > _______________________________________________
> > Squeakfoundation mailing list
> > [email protected]
> > http://lists.squeakfoundation.org/listinfo/squeakfoundation
> _______________________________________________
> Squeakfoundation mailing list
> [email protected]
> http://lists.squeakfoundation.org/listinfo/squeakfoundation
>=20