[5311] in testers
Re: one Linux 9.2 problem solved
daemon@ATHENA.MIT.EDU (andrew m. boardman)
Wed Apr 23 18:53:39 2003
Date: Wed, 23 Apr 2003 18:53:37 -0400
Message-Id: <200304232253.SAA21149@pothole.mit.edu>
From: "andrew m. boardman" <amb@MIT.EDU>
To: alexp@MIT.EDU
CC: testers@MIT.EDU, alexp@MIT.EDU, facdev@MIT.EDU, cfyi@MIT.EDU
In-reply-to: <200304232058.h3NKwiLj012010@astrophel.mit.edu> (message from
Alex T Prengel on Wed, 23 Apr 2003 16:58:44 -0400)
How odd. I thought I'd sent mail about this when the JRE issues came up,
but I don't see it. But anyway...
>Setting LD_ASSUME_KERNEL to 2.4.1 gets around this and lets the
>application run normally (in each case I've run into so far). I don't
>know if this is likely to have any bad consequences but I haven't seen
>any so far.
I can't see how this would be a problem. It causes you to get the old
threads implementation instead of NPTL (Posix Threads), the 9.0 default.
Generally speaking, the redhat beta list suggests that this causes lots
and lots of things to just work again. The 9.0 release notes also
suggest that this might also help with JRE problems, although given your
error messages, it looks like something else entirely. Still, easy
enough to give it a go.
From /afs/dev/system/rhlinux/redhat-9.0/RELEASE-NOTES:
- IBM JRE
If an application does not work properly with NPTL, it can be run
using the old LinuxThreads implementation by setting the following
environment variable:
LD_ASSUME_KERNEL=<kernel-version>
The following versions are available:
- 2.4.1 - Linuxthreads with floating stacks
- 2.2.5 - Linuxthreads without floating stacks
NPTL support for all dynamically-linked applications can be disabled
by using the following boot-time option:
nosysinfo
