Re: glibc-2.3.2 NPTL caveat

From: Tim Cutts <tjrc_at_sanger.ac.uk>
Date: 2005-01-21 22:34:26
On 20 Jan 2005, at 6:07 am, David Mosberger wrote:

> We just discovered a nasty bug in glibc-2.3.2 which can lead to
> "random" memory corruption in multi-threaded applications using NPTL.
> This definitely affects Debian/{testing,unstable}, but it may affect
> other distros as well.  I also believe that the CVS glibc still has
> this bug, so I suspect it's not just glibc-2.3.2 that's affected.
>
> For details, see:
>
>   http://sources.redhat.com/bugzilla/show_bug.cgi?id=685

That explains a lot of problems we had with MySQL crashing almost 
instantly on startup on IA64 machines too.  So what needs to be done?  
Is it as simple a fix as always compiling ld-linux with the larger of 
the two thread struct sizes? (he says, extremely naively, since this is 
the sort of thing in Linux that he has no clue about whatsoever - it 
comes under the heading of "scary stuff you leave to Really Clever 
People to think about"  :-)

Tim

-- 
Dr Tim Cutts
Informatics Systems Group, Wellcome Trust Sanger Institute
GPG: 1024D/E3134233 FE3D 6C73 BBD6 726A A3F5  860B 3CDD 3F56 E313 4233

-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Received on Fri Jan 21 06:33:46 2005

This archive was generated by hypermail 2.1.8 : 2005-08-02 09:20:34 EST