[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Subtle bug in intervals code
From: |
Paul Eggert |
Subject: |
Re: Subtle bug in intervals code |
Date: |
Wed, 18 Jul 2012 12:27:47 -0700 |
User-agent: |
Mozilla/5.0 (X11; Linux i686; rv:14.0) Gecko/20120714 Thunderbird/14.0 |
On 07/18/2012 11:33 AM, Eli Zaretskii wrote:
> Do you have an easy test case that triggers this?
My test involves creating bool vectors of various sizes.
Usually it works, sometimes the intervals.c assert fails.
There doesn't seem to be a pattern; no single test case
always works.
That is, in *scratch* I do this:
(make-bool-vector 1000000 nil)
(make-bool-vector 1000000 nil)
(make-bool-vector 10000 nil)
(make-bool-vector 100000 nil)
(make-bool-vector 10000000 nil)
and evaluate each one in turn. I do some more if the first
few don't dump core. It fairly reliably dumps core eventually.
This is Fedora 15 x86-64, compiled with GCC 4.7.1,
with -g -O0 -DENABLE_CHECKING. I ran emacs -Q
under X.