This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: fix opt/8634
- From: Jan Hubicka <jh at suse dot cz>
- To: Richard Kenner <kenner at vlsi1 dot ultra dot nyu dot edu>
- Cc: rakdver at atrey dot karlin dot mff dot cuni dot cz, gcc-patches at gcc dot gnu dot org
- Date: Tue, 8 Apr 2003 09:37:21 +0200
- Subject: Re: fix opt/8634
- References: <10304080109.AA17330@vlsi1.ultra.nyu.edu>
> can you explain this in greater detail? It is one pass over insns, so it
> should be linear, no?
>
> Each pass, yes, but what I meant is suppose you have situation where a
> particular construct causes a call to purge_addressof. That construct is
> used in some percentage of code. So the number of occurrences is proportional
> to the number of insns. That's where the quadratic behavior comes from.
I don't follow here either. purge_addressof is called via toplev.c
exactly once per function and it does exactly one scan of insns.
Honza