Re: PATCH to remove unnecessary NOTE_INSN_DELETED

[Bernd Schmidt <bernds at cygnus dot co dot uk>]

On Thu, 4 Nov 1999, Joern Rennecke wrote:

> : debug these monsters, we just bite the bullet and eat the extra memory needed
> : when we do not squash uids & regnos.
> 
> That is fine if it works, but in acses of memory corruption or reading
> uninitialized memory it might be impossible to observe the bug once the
> squashing is disabled.

I have to agree here.  This is just asking for problems when debugging the
compiler.
Why don't we just add another uid field to every insn?  We'd have one that
stays constant during the entire compilation (for rtl dumps and debugging),
and one that can get renumbered at any time.  This would cost an additional
word per insn, but we'd save the various uid_luid tables that are scattered
throughout most of the optimization passes.  In some situations it might even
make things simpler, e.g. when inserting a new insn we no longer face the
problem of resizing the luid table.

Bernd