Hi,
  Ok I have played with the povray binaries that people have made available and
one that neil compiled. So far I have not found indication of a FPE problem.
All the SEGV's I have had from povray have all occurred on normal instructions
and (out of interest) all in colour related functions.

There was reported a SEGV from povray at address 0x0000000c. That is one that I
would like to trace at that one may be FPE related (have seens X do this)
but the SEGVs that the povrays I have tested have all generated when I would
call normaly SEGVs. In several cases the SEGV instructions was a
ldr pc, [rx, #xxxx] so this looks more likes a trashed function pointer.

What Rob, Neil and myself are considering to be the bug in the FPE is a problem
that can occur when delivering signals that occured while in the FPE, in
particular when multiple signals are raised while in the FPE. The effect on the
process appears to be to cause a SEGV but typically at a zero page address.

I am looking into a new idea as to the cause of some of the crashes that we
believe are due to the FPE and I hope to have more new soon.

Any comments, thoughts welcome ;-)

Cheers,
				Mark

-- 
Mark Brinicombe				amb@physig.ph.kcl.ac.uk
Research Associate			http://www.ph.kcl.ac.uk/~amb/
Department of Physics			tel: 0171 873 2894
King's College London			fax: 0171 873 2716