f771 assertion failure

[Richard Henderson]

> Until the non-reduction-of-SUBREG-GIVs problem is solved, there is  
> no point in running benchmark codes compiled by g77/egcs on an Alpha.

True, until that is fixed, the results will pretty much suck.

But my thoughts in running the tests now were these:

  -- 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.

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

> PS: The apsi benchmark from SPEC is, if we may believe the Web page  
> description, a sort of air pollution distribution model, that -  
> qualitate qua - has to include quite a lot of meteorology.  I expect  
> it therefore to be like our own limited area weather forecasting  
> code.

Yep.  I've also got a passle of global climate modeling code from
my mates doing climatology back at A&M that I can do testing on.


r~