You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by "William A. Rowe, Jr." <wr...@covalent.net> on 2001/05/10 06:12:10 UTC
Proposed Code Freeze for 1.3.20
I know Roy's model changes things a bit, and we can always go back
and tag the state of the tree as of midnight tonight, but...
can we please freeze development now excepting fixes to absolutely broken stuff?
Bill
Re: Proposed Code Freeze for 1.3.20
Posted by "Roy T. Fielding" <fi...@ebuilt.com>.
> Of course, last-minute "bug fixes" have also resulted in problems
> that have required re-releases in short order. So yeah, maybe "freeze"
> isn't a good term, but generally what we mean is "we plan on
> tagging and rolling soon and want to reduce the risk of
> problems on the present code tree so we would really like
> it if no new features are added and bug fixes be as risk-free
> as possible". :)
Remember when I started the STATUS file for 1.2bINF? I would start a
countdown on the mailing list and then just barrel my way through all
of the status items and either apply the patches or move them to a
"cannot be applied until..." category. The RM job was more about
cleaning house in preparation for a release than anything else. People
weren't prevented from fixing problems -- they just had to put them
in STATUS first and listen if someone else said the item should be
postponed until after some other item was done.
....Roy
Re: Proposed Code Freeze for 1.3.20
Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
Hmmm? isnan/isfnf, while problematic, is a very appropriate patch since
it closes an opportunity for segfaults (even if they are caused by a third
party module, and not our code.)
I'd tend to agree on the ab patch, but it's here.
I'm preparing an alternate solution till those changes are all cleaned up.
An announce will follow.
Bill
----- Original Message -----
From: "Roy T. Fielding" <fi...@ebuilt.com>
To: <ne...@apache.org>
Sent: Friday, May 11, 2001 2:56 PM
Subject: Re: Proposed Code Freeze for 1.3.20
> If at any time progress is stopped because someone committed a bad
> patch to the tree, the best way to fix it is to revert the patch
> (unless for some reason that patch is necessary for the release).
> I'd say you can revert all of the changes for isnan and ab if you
> are in a hurry to get this release out.
>
> People just shouldn't be committing stuff like that to 1.3.x.
>
> ....Roy
>
>
Re: Proposed Code Freeze for 1.3.20
Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
I'm satisfied that David McCreedy, Brian Havard, Brad Nicholes and I have
isnan/isfnf sorted out on the obscure platforms, and JimJag's patch for
unicies appears to be significatly robust. Of course, testing the tarball
is the most efficient way for everyone to be certain.
I'm proceeding to back out the ab comments, tag prior to the ab improvements,
retag the comments for 1.3.20, and leave that code in it's current state.
Tag and roll coming up here in the next 20 minutes or so.
----- Original Message -----
From: "Roy T. Fielding" <fi...@ebuilt.com>
To: <ne...@apache.org>
Sent: Friday, May 11, 2001 2:56 PM
Subject: Re: Proposed Code Freeze for 1.3.20
> If at any time progress is stopped because someone committed a bad
> patch to the tree, the best way to fix it is to revert the patch
> (unless for some reason that patch is necessary for the release).
> I'd say you can revert all of the changes for isnan and ab if you
> are in a hurry to get this release out.
>
> People just shouldn't be committing stuff like that to 1.3.x.
>
> ....Roy
>
>
Re: Proposed Code Freeze for 1.3.20
Posted by "Roy T. Fielding" <fi...@ebuilt.com>.
If at any time progress is stopped because someone committed a bad
patch to the tree, the best way to fix it is to revert the patch
(unless for some reason that patch is necessary for the release).
I'd say you can revert all of the changes for isnan and ab if you
are in a hurry to get this release out.
People just shouldn't be committing stuff like that to 1.3.x.
....Roy
Re: Proposed Code Freeze for 1.3.20
Posted by "William A. Rowe, Jr." <wr...@covalent.net>.
What he said [erm, I meant] ... no you are always welcome to put bug fixes into the
code regardless of the 'freeze' status. I mean, and meant, to propose we freeze any
new feature patches. I'll absolutely wait for dirk to re-sync ab to his tree.
From: "Roy T. Fielding" <fi...@ebuilt.com>
Sent: Friday, May 11, 2001 1:46 AM
> On Wed, May 09, 2001 at 11:12:10PM -0500, William A. Rowe, Jr. wrote:
> > I know Roy's model changes things a bit, and we can always go back
> > and tag the state of the tree as of midnight tonight, but...
> >
> > can we please freeze development now excepting fixes to absolutely broken stuff?
>
> -1. If you want to freeze the tree, tag it and then do your work on
> the branch created by that tag. It is completely ridiculous to ask people
> to freeze a tree when it doesn't even compile on half a dozen platforms.
> Tag it when it is in sufficient shape for a release (or tag it now and
> move the tags back manually to the last version of a file that doesn't
> suffer from breakage). In any case, if it isn't ready for a release now
> then the last thing you want to do is stop people from working on the tree.
So we want to start vendor branching? Certainly could check out 1.3.19 and
release a 010319101 on win32 with the additional patch.
The ab and ap_sprintf patches broke a number of platforms. Either I'd branch
at 1.3.19 and apply the fix, or roll forward. Since there are other very good
bug fixes in the -dev branch, I'd like to continue to push for a prompt release.
Let's drop the word freeze from our vocabulary, and use the phrase feature-freeze
so we stop having this silly discussion. We aren't agreed on how to fork the code
for maintenance, and until we come to a concensus, we need to continue on the old
way of doing things.
Bill
Re: Proposed Code Freeze for 1.3.20
Posted by "Roy T. Fielding" <fi...@ebuilt.com>.
On Wed, May 09, 2001 at 11:12:10PM -0500, William A. Rowe, Jr. wrote:
> I know Roy's model changes things a bit, and we can always go back
> and tag the state of the tree as of midnight tonight, but...
>
> can we please freeze development now excepting fixes to absolutely broken stuff?
-1. If you want to freeze the tree, tag it and then do your work on
the branch created by that tag. It is completely ridiculous to ask people
to freeze a tree when it doesn't even compile on half a dozen platforms.
Tag it when it is in sufficient shape for a release (or tag it now and
move the tags back manually to the last version of a file that doesn't
suffer from breakage). In any case, if it isn't ready for a release now
then the last thing you want to do is stop people from working on the tree.
....Roy
Re: Proposed Code Freeze for 1.3.20
Posted by Cynic <cy...@mail.cz>.
If you mean the build failure, then it seems to be fixed.
I built a fresh snapshot yesterday, and it works. No real
tests, but at least it runs and serves HTML + PHP pages on
several VHosts.
Oh, and BTW, it's on NT 4.
At 06:25 10.5. 2001, William A. Rowe, Jr. wrote the following:
--------------------------------------------------------------
>From: <rb...@covalent.net>
>Sent: Wednesday, May 09, 2001 11:22 PM
>
>
>> On Wed, 9 May 2001, William A. Rowe, Jr. wrote:
>>
>> > I know Roy's model changes things a bit, and we can always go back
>> > and tag the state of the tree as of midnight tonight, but...
>> >
>> > can we please freeze development now excepting fixes to absolutely broken stuff?
>>
>> Isn't the psprintf stuff still broken?
>
>I believe I just patched it for Win32 several days ago (you are asking about the
>isnan() and isinf() tests?)
>
>Would _everyone_ please validate several platforms, please, and status the list?!?
>
>Bill
------end of quote------
cynic@mail.cz
-------------
And the eyes of them both were opened and they saw that their files
were world readable and writable, so they chmoded 600 their files.
- Book of Installation chapt 3 sec 7
Re: Proposed Code Freeze for 1.3.20
Posted by "William A. Rowe, Jr." <wr...@covalent.net>.
From: <rb...@covalent.net>
Sent: Wednesday, May 09, 2001 11:22 PM
> On Wed, 9 May 2001, William A. Rowe, Jr. wrote:
>
> > I know Roy's model changes things a bit, and we can always go back
> > and tag the state of the tree as of midnight tonight, but...
> >
> > can we please freeze development now excepting fixes to absolutely broken stuff?
>
> Isn't the psprintf stuff still broken?
I believe I just patched it for Win32 several days ago (you are asking about the
isnan() and isinf() tests?)
Would _everyone_ please validate several platforms, please, and status the list?!?
Bill
Re: Proposed Code Freeze for 1.3.20
Posted by rb...@covalent.net.
On Wed, 9 May 2001, William A. Rowe, Jr. wrote:
> I know Roy's model changes things a bit, and we can always go back
> and tag the state of the tree as of midnight tonight, but...
>
> can we please freeze development now excepting fixes to absolutely broken stuff?
Isn't the psprintf stuff still broken?
Ryan
_______________________________________________________________________________
Ryan Bloom rbb@apache.org
406 29th St.
San Francisco, CA 94131
-------------------------------------------------------------------------------