|A question about subsumption in register allocators firstname.lastname@example.org (Thomas Johnsson) (1991-04-08)|
|Re: A question about subsumption in register allocators email@example.com (1991-04-12)|
|From:||firstname.lastname@example.org (Preston Briggs)|
|Keywords:||optimize, registers question|
|Organization:||Rice University, Houston|
|Date:||Fri, 12 Apr 91 19:03:07 GMT|
Thomas Johnsson <email@example.com> writes:
>Before the actual coloring, subsumption is normally done, to remove
>unnecessary copy operations, and to make the graph to be coloured
>Now the question is: What do register allocators out there do if there
>is more than one way of doing the subsumption, and they are mutually
One of two things:
Start at the top of the routine and head toward the bottom,
coalescing as you go.
Find the loop structure (need for spill costs anyway)
and examine the most deeply nested loops first.
There's very little cost difference between the two.
I don't know about the payoff.
Return to the
Search the comp.compilers archives again.