f771 assertion failure

[Jeffrey A Law]

  In message < 199709292148.OAA02957@dot.cygnus.com >you write:
  >   -- The specmarks are good at verifying the results, thus we know
  >      that we are getting valid code out of the compiler.  This was
  >      my primary concern, esp considering that g77 has not supported
  >      Alpha at all until recently.
Yes, absolutely!

  >   -- I have some scheduling patches pending to take better advantage
  >      of haifa's superscalar scheduling, and the change in runtime
  >      will still be meaningful.
Yes absolutely.

Just because the existing loop optimizer doesn't do a good job doesn't
mean running spec is useless -- it makes a good testsuite and a good
way to measure the effects of other optimizations (like haifa).

jeff