This is the mail archive of the
mailing list for the GCC project.
Re: [tree-profiling-branch PATCH IPCP extensions + Function cloning
- From: Steven Bosscher <stevenb at suse dot de>
- To: Razya Ladelsky <RAZYA at il dot ibm dot com>
- Cc: Jan Hubicka <jh at suse dot cz>, Ayal Zaks <ZAKS at il dot ibm dot com>, gcc-patches at gcc dot gnu dot org, hubicka at ucw dot cz, Mircea Namolaru <NAMOLARU at il dot ibm dot com>
- Date: Sun, 16 Jan 2005 15:18:43 +0100
- Subject: Re: [tree-profiling-branch PATCH IPCP extensions + Function cloning
- Organization: SUSE Labs
- References: <OF51ED2B04.D7661169-ONC2256F8B.004D4C5E-C2256F8B.004D5BEC@il.ibm.com>
On Sunday 16 January 2005 15:15, Razya Ladelsky wrote:
> > Also it would be really cool to share at least partly this with the
> > copy_decl_for_inlining
> Sharing the code means that we would have to add another flag, saying
> whether we are
> versioning or inlining. This will require to add another argument to all
> to copy_decl_for_inlining(), and we wanted to minimize changes in existing
I would prefer the proper design over minimizing the number of changes.
You can do most of the changes incrementally. For example you could
already sumbit relatively independent changes piece-by-piece and merge
then into the branch. Then the larger, "new-stuff" patches don't seem
so intrusive anymore.
Duplicating code is *bad*.