On Wed, 2006-06-07 at 11:34 +0200, Wouter Verhelst wrote: > Really, how is that any relevant? Can you come up with a real-life > scenario (as in, something which actually occurred) where a change to, > say, glibc or something similar made some other application break in > such a way that it would no longer behave as documented?
That has actually happened before with Sun's Java on Linux at least once. Back when glibc switched NPTL, Sun's JVM became very unstable. I'd be surprised if a build of Sun's JVM back then would have passed Sun's proprietary compatibility test suite (i.e. behaved as documented) on a NPTL-ed glibc without some nudging in form of LD_KERNEL_ASSUME etc. See http://bugs.sun.com/bugdatabase/view_bug.do;:YfiG?bug_id=4885046 for a particular instance of the problem. If you search Sun's bug database/the web, you should be able to see more instances. cheers, dalibor topic -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
