This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: Inlining function into itself (via transitivity

From: Josef Zlomek <zlomj9am at artax dot karlin dot mff dot cuni dot cz>
To: Geoff Keating <geoffk at geoffk dot org>
Cc: gcc at gcc dot gnu dot org
Date: Wed, 21 Jan 2004 19:10:17 +0100
Subject: Re: Inlining function into itself (via transitivity
References: <20040121141142.GD13065@artax.karlin.mff.cuni.cz> <jmoesxnuok.fsf@desire.geoffk.org>

> > is it correct to inline function into itself (via transitivity), i.e.
> > 
> > a <-- b <-- c <-- a
> > 
> > where x <-- y means that y is inlined into x.
> > 
> > Thanks for answer.
> 
> Yes, this can be done without changing the meaning of the program.  It
> is often pointless, though.

I have a testcase which GCC (mainline) gets ICE (because of jump to undefined
label) on.
First it inlines a <-- b, then b <-- a.
I have probably found the problem, I'm going to make a patch.

Josef

References:
- Inlining function into itself (via transitivity
  - From: Josef Zlomek
- Re: Inlining function into itself (via transitivity
  - From: Geoff Keating

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]