Re: schedule_block

[Joern Rennecke <amylaar at cygnus dot co dot uk>]

>   > So what is the proper way to achieve the effect that this piece of
>   > code was trying to do?  A rewrite of the register allocator?
> There is no way to do it, nor should there be a need.
> 
> If the copy out of the arg register is on the critical path, then it should
> issue as early as possible because the copy will have a high priority.


The critical path as preceived by sched does not take register pressure
into account.
Maybe this optimization should be implemented by adding a dependence between
the copy and the birth of the next pseudo with a preferenced register class
that includes this hard register.  Or just the birth of the next pseudo if
you want to keep it simple.

If there is no such insn on a patch that the scheduler would consider,
there is no register pressure problem.