[Info-vax] Beyond Open Source

[seasoned_geek]

On Monday, May 11, 2015 at 7:09:04 AM UTC-5, John Reagan wrote:
> 
> That feels more like a system spinlock problem or somebody decided to "write their own".  gcc doesn't know from spinlocks.  There are system calls to manage them.  Such bugs are hard to track down on any platform.  We've had them on OpenVMS in the past.  It doesn't appear to be wide-spread since a Google search on several terms didn't find anything.

It is actually pretty wide spread. I've been hearing about it with various distros. The searches aren't turning anything up because of all the finger pointing. Everybody is busy blaming every other package. There have also been various stabs at it because everyone sees it as a different problem.

http://www.securitytracker.com/id/1028147

http://www.gossamer-threads.com/lists/linux/kernel/1726074

It has been moving around for years:

http://www.cvedetails.com/bugtraq-bid/45028/Linux-Kernel-posix-cpu-timers.c-Local-Race-Condition-Vulne.html

Adding insult to injury there is also a MIPS flavor.

http://www.spinics.net/lists/stable/msg90183.html


Blaming the gcc compiler does sound very off-shore-reading-from-script though.