Up: SGI apps Frequently Asked Questions (FAQ)
Next: -53- Why don't IRIX 5.3 executables run on IRIX 6.0.1?
Previous: -51- Why does stripping executables corrupt them in IRIX 6.x?
Subject:   -52- Why does my floating point signal processing routine,
                when given certain inputs, run incredibly slowly and
                consume all of the CPU in _system_ or _interrupt_ time ?
Date: 03 Sep 1995 22:28:13 PDT

  You may be experiencing an undesirable "floating point underflow"
  behavior of the floating point unit on R3k's and beyond. You'll need
  to add a workaround to your code and, on some machines, install a
  patch as well. A thorough writeup by Chris Pirazzi of SGI
  <cpirazzi@cp.esd.sgi.com> is at
  ftp://viz.tamu.edu/pub/sgi/software/programming/fp-underflow.

Up: SGI apps Frequently Asked Questions (FAQ)
Next: -53- Why don't IRIX 5.3 executables run on IRIX 6.0.1?
Previous: -51- Why does stripping executables corrupt them in IRIX 6.x?