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++] Avoid C++ FE creating !public weak variables

From: Jason Merrill <jason at redhat dot com>
To: gcc-patches at gcc dot gnu dot org
Cc: Mark Mitchell <mark at codesourcery dot com>, gcc-patches at gcc dot gnu dot org, dnovillo at redhat dot com
Date: Thu, 12 Nov 2009 15:39:24 -0500
Subject: Re: [c++] Avoid C++ FE creating !public weak variables
References: <20091112051031.GC14355@kam.mff.cuni.cz> <4AFB9FC7.9070600@codesourcery.com> <20091112193843.GO24874@kam.mff.cuni.cz>

On 11/12/2009 02:38 PM, Jan Hubicka wrote:

With some further sanity checking I noticed that with -fno-weak we also tend to
create DECL_COMDAT !TREE_PUBLIC functions.  This is caused by comdat_linkage function
that simply drops to this path when no weak support is there.
Is this expected behaviour or shall we also clear COMDAT flag here?
Some of cgraph code assumes that COMDAT decls are public, I can fix it on that side too.

I'm not sure what the best answer is for vague-linkage functions that we can't actually share between translation units because of target limitations. There is some code in the front end that expects DECL_COMDAT to be set on these decls, though it could be changed to use a different test.

Jason

References:
- [c++] Avoid C++ FE creating !public weak variables
  - From: Jan Hubicka
- Re: [c++] Avoid C++ FE creating !public weak variables
  - From: Mark Mitchell
- Re: [c++] Avoid C++ FE creating !public weak variables
  - From: Jan Hubicka

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