From xemacs-m  Sat Mar  1 18:19:12 1997
Received: from altair.xemacs.org (steve@xemacs.miranova.com [206.190.83.19])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id SAA22521
	for <xemacs-beta@xemacs.org>; Sat, 1 Mar 1997 18:18:57 -0600 (CST)
Received: (from steve@localhost)
	by altair.xemacs.org (8.8.5/8.8.5) id QAA14041;
	Sat, 1 Mar 1997 16:30:30 -0800
Mail-Copies-To: never
To: xemacs-beta@xemacs.org
Subject: Re: 20.1-b2 with dynamic PURESIZE success, mostly
References: <199703010620.WAA04094@infodock.com>
X-Url: http://www.miranova.com/%7Esteve/
X-Face: #!T9!#9s-3o8)*uHlX{Ug[xW7E7Wr!*L46-OxqMu\xz23v|R9q}lH?cRS{rCNe^'[`^sr5"
 f8*@r4ipO6Jl!:Ccq<xoV[Qz2u8<8-+Vwf2gzJ44lf_/y9OaQ`@#Q65{U4/TC)i2`~/M&QI$X>p:9I
 OSS'2{-)-4wBnVeg0S\O4Al@)uC[pD|+
X-Attribution: sb
From: Steven L Baur <steve@miranova.com>
In-Reply-To: Bob Weiner's message of Fri, 28 Feb 1997 22:20:50 -0800
Mime-Version: 1.0 (generated by tm-edit 7.105)
Content-Type: text/plain; charset=US-ASCII
Date: 01 Mar 1997 16:30:28 -0800
Message-ID: <m2lo87i197.fsf@altair.xemacs.org>
Lines: 49
X-Mailer: Gnus v5.4.16/XEmacs 20.1

Bob Weiner writes:

>>>>>> "SLB" == Steven L Baur <steve@miranova.com> writes:
SLB> Weird stuff seems to happen with too small PURESIZEs.  The only time I
SLB> got an actual coredump in testing was when I had it set to 30k.

The last fixes I posted seem to have done the trick.  I've done
multiple tests now starting from 5 bytes of PURESIZE and 1 byte (:-))
of PURESIZE and it all runs smoothly to completion starting with just
one typed `make'.  While composing this message I just did the 1 byte
test on 19.15 and ended up with:

$ make
 ...
Finding pointers to doc strings...
Finding pointers to doc strings...done
Dumping under names xemacs and emacs-19.15-xemacs-beta96
Purespace usage: 624020 of 624036 (100%).
 ...
if [ -f SATISFIED ]; then               make xemacs;    fi
make[3]: Leaving directory `/usr/src/xemacs-19.15/src'
make[2]: Leaving directory `/usr/src/xemacs-19.15/src'
make[1]: Leaving directory `/usr/src/xemacs-19.15/src'

$ 

Looks good to me.

> I agree.  In the distant past, I've seen highly excess garbage
> collection occur for some unknown reason when not enough open
> PURESPACE was left.

That should be fixed.  XEmacs is in a weird state depending upon where
it runs out of pure space, so now garbage collection gets turned off
completely if purespace runs out.

> I like it at around 93% and never have problems then.

The initial betas will not give you a choice.  I'd prefer people
testing right up at the limit.  I haven't had any problems[1], and I
think metrics like 93% purespace used, or 10k of purespace slop will
become a thing of the past.

[1]
I don't recommend dumping icomplete and rsz-minibuf into the same
XEmacs.  You will hit brain-dead abort()s in redisplay. :-(
-- 
steve@miranova.com baur
Unsolicited commercial e-mail will be billed at $250/message.

