[Israel.pm] catching segmentation faults and other crashes

[Yona Shlomo]

On Mon, 12 Nov 2007, Mikhael Goikhman wrote:

> Of course, if Shlomo just wanted to detect abnormal exit and not "core
> dump", then checking for non-zero $? should be portable enough.

Yes. This is my fallback -- but can't I get false positives
this wat?

-- 
Shlomo Yona
yona at cs.technion.ac.il
http://yeda.cs.technion.ac.il/~yona/