integer moves of floating point values

Henry Cejtin henry@sourcelight.com
Wed, 9 Jan 2002 17:07:42 -0600


Isn't  it more efficient, in terms of number of registers used, to use only a
single register?  I.e., instead of loading the high and low parts of the real
into  different  32-bit  registers and then storing them, just load the high,
store the high and then load the low and store the  low.   Does  this  bother
some  kind of typing?  I would think that making all of this invisible except
to the actual code generator would be the way to go, and  then  the  aliasing
wouldn't  be  a  problem, but I guess that then certain possible (very local)
optimizations wouldn't be possible.

As you say, the pipelining muti-issue stuff between the integer units and the
floating  point  units  could  be  the  cause.   Another  reason might be the
duplicated offset computation.