Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 28 May 2001 10:54:14 +0930
From:      Greg Lehey <grog@lemis.com>
To:        David O'Brien <obrien@FreeBSD.ORG>
Cc:        "Daniel C. Sobral" <dcs@newsguy.com>, alpha@freebsd.org, FreeBSD Core Team <core@FreeBSD.org>
Subject:   Re: broken loader?
Message-ID:  <20010528105414.S29739@wantadilla.lemis.com>
In-Reply-To: <20010527170810.A91713@hub.freebsd.org>; from obrien@FreeBSD.ORG on Sun, May 27, 2001 at 05:08:10PM -0700
References:  <Pine.LNX.4.21.0105251953130.14962-100000@zeppo.feral.com> <20010526145104.C11876@dragon.nuxi.com> <3B111787.220345BB@newsguy.com> <20010527170810.A91713@hub.freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sunday, 27 May 2001 at 17:08:10 -0700, David O'Brien wrote:
> On Sun, May 27, 2001 at 12:04:39PM -0300, Daniel C. Sobral wrote:
>> My guess is that the newly imported FICL 2.05 + the forth stuff we use
>> at boot got too big for the space allocated for the dictionary on the
>> alpha. This happened to me when using my more comprehensive stuff
>> instead of the standard stuff on the i386. Reason why I made commit to
>> remedy this.
>
> Daniel this is not the first time you've screwed the Alpha users with
> loader commits/upgrades.  Once was forgivable, twice is frankly quite
> unforgivable.  You have once again _wasted_ a significant amount of your
> fellow developer's time.
>
> I want the backing of Core that NO FICL upgrades or major .4th changes
> can be committed until they have been tested (compiled and booted) on all
> platforms.  I really don't think I would be allowed to upgrade the
> toolchain if it had only been tested on Alpha and x86.  I fail to see why
> something as important and fundamental as the loader (and harder to
> replace if you FUBAR it and loader.old) should be any different.

Well, you've waved the C word, so this is your confirmation that we've
received the message, and that you should get some kind of response
within a week.  I've assigned the item to Doug Rabson, since he has
the best overview of the topics.

Having said that, I think this is a more general matter.  I think I'm
correct in saying that *nobody* has access to all platforms we're
currently supporting, and certainly not to all that we want to support
Real Soon Now.  This kind of breakage could occur almost anywhere,
though obviously some areas are more likely than others.  We should
come up with a strategy for minimizing breakage.

Ideas?

Greg
--
Finger grog@lemis.com for PGP public key
See complete headers for address and phone numbers

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-alpha" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010528105414.S29739>