On Thu, Mar 17, 2005 at 09:24:01PM +0100, Frank van der Linden wrote:
> I've looked at these crashes, and I can't really find anything wrong
> with the library code.
> 
> The reason that the fstpl instruction may fault, is because there's a
> NaN value on the top of the FPU stack. But, the input parameter doesn't
> seem to be a value that would cause this.
> 
> All I can think of is that there may be some FPU state save/restore
> trouble here, but then again, such problems have not been reported
> with the amd64 port for a long time.

Hm.  How long of a long time?

I'm running with a fresh kernel, now, and the Audacity crashes *seem*
to have gone away.  Unfortunately, it's hard to be completely sure
of that, as the crash could not be reliably induced...  (Of course,
ipf is out of sync between userland and kernel, now, so I'll have
to upgrade userland.)

Most of the files in /usr/bin date to Sept. 25.  Is a 6 month old kernel
old enough to have been "a long time" ago?


-- 
  "I probably don't know what I'm talking about."  http://www.olib.org/~rkr/