Date: Mon, 30 Apr 2001 19:43:43 +0900 From: "Akinori MUSHA" <knu@iDaemons.org> To: Kris Kennaway <kris@FreeBSD.org> Cc: ports@FreeBSD.org, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: ports/x11-toolkits/fox/files patch-ad Message-ID: <86bspe4qhs.wl@archon.local.idaemons.org> In-Reply-To: <20010430023347.A70094@xor.obsecurity.org> References: <200104300810.f3U8AGY60114@freefall.freebsd.org> <86elua4wf1.wl@archon.local.idaemons.org> <20010430023347.A70094@xor.obsecurity.org>
next in thread | previous in thread | raw e-mail | index | archive | help
At Mon, 30 Apr 2001 02:33:47 -0700,
Kris Kennaway wrote:
> > Also, I haven't yet tested if this big monster of a toolkit works with
> > optimization flags other than "-O2 -ffast-math -finline-functions
> > -fomit-frame-pointer -fexpensive-optimizations". I wish you had
> > pointed it to me rather than just committing directly.
>
> Well, the policy from all quarters of the project is that everything
> over -O is dangerous and is known to break. I tested compilation of
> all of the patches I've been committing today, but since I've just
Please notice that I'm not talking about the principle. Some pieces
of software actually do NOT run with -O, but run with -O2. -O is not
perfect. In this case, at least -O2 is proven to work already, but -O
is not even tested by you or me. (soon I will ;)
> been making ports conform to accepted rules I didn't see the need to
> send each and every patch through maintainers (that would have at
> least doubled the total time it would take me to get through these).
> For a couple of the ports I came across functional (non-style) changes
> which should be made; those I bounced to the maintainers for approval.
Sometimes circumstances override the rules. You cannot conform to the
rules when one of the rules breaks a port.
Also, the change was not obvious. I am sure I'm one of the most
active ports committers, and I think you could just suggest the point
and leave the modifications and the due tests to me, although I fully
appreciate that you were trying to do the right thing.
> Basically I think the approach we should be taking wrt -O2 and so on
> is that people should need to justify why their port should default to
> add anything of the sort to the user-supplied values. It's (almost)
> never a good idea.
Why don't you yell so loudly on the ports@ list first? We are no
dummies, but all ears. ;)
But anyway, you have shown good examples we can follow. Thanks.
> I take full responsibility for any errors I caused during this series
> of commits. It looks like the /usr/local thing snuck through
> unexpectedly in this case; I've already fixed it and will go back and
> re-check the others.
Well, I'm afraid you would have to... I wish this whole thread will
remind ports staffs of the issue and go fix their ports. :)
Regards,
--
/
/__ __ Akinori.org / MUSHA.org
/ ) ) ) ) / FreeBSD.org / Ruby-lang.org
Akinori MUSHA aka / (_ / ( (__( @ iDaemons.org / and.or.jp
"Freeze this moment a little bit longer, make each impression
a little bit stronger.. Experience slips away -- Time stand still"
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?86bspe4qhs.wl>