This is the mail archive of the gcc-patches@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: [C++ patch] Fix thunk emitting code

From: Jan Hubicka <jh at suse dot cz>
To: Mark Mitchell <mark at codesourcery dot com>
Cc: Jan Hubicka <jh at suse dot cz>, gcc-patches at gcc dot gnu dot org
Date: Wed, 28 Jan 2004 19:54:00 +0100
Subject: Re: [C++ patch] Fix thunk emitting code
References: <20040128141026.GR8094@kam.mff.cuni.cz> <4017FA0C.2060104@codesourcery.com>

> Jan Hubicka wrote:
> 
> >Hi,
> >While trying to save memory, tree-optimize release DECL_ARGUMENTS list.  
> >This
> >breaks C++ frontend that still needs argument list to produce thunks.  This
> >happens to worky only because we don't release the memory when inlining 
> >due to
> >bug, but passing -fno-inline causes several testsuite faiulres.
> >
> >I am going to fix the tree-optimize bug next, so this will reproduce
> >more easilly then.
> >
> >OK for mainline/branch?
> > 
> >
> Yes.
> 
> I'd meant to look at that for another reason -- the thunks should come 
> before the function so that they are more likely to be on the same page 
> as the function to which control is being transferred.

This looks like good reason too.  Once unit-at-a-time code is told more
about sane functions orderings, it would be nice to have them being
"usual" functions visible to cgraph rather than something emit on the
side.

Honza
> 
> -- 
> Mark Mitchell
> CodeSourcery, LLC
> (916) 791-8304
> mark@codesourcery.com
>

References:
- [C++ patch] Fix thunk emitting code
  - From: Jan Hubicka
- Re: [C++ patch] Fix thunk emitting code
  - From: Mark Mitchell

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