This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
RE: [plugins-ici-cloning-instrumentation] new GCC plugin developements
- From: "Grigori Fursin" <grigori dot fursin at inria dot fr>
- To: <ctuning-discussions at googlegroups dot com>, "'Grigori Fursin'" <grigori dot fursin at inria dot fr>
- Cc: "'Zbigniew Chamski'" <zbigniew dot chamski at gmail dot com>, "'Richard Guenther'" <richard dot guenther at gmail dot com>, "'Basile STARYNKEVITCH'" <basile at starynkevitch dot net>, "'Ian Lance Taylor'" <iant at google dot com>, "'GCC Mailing List'" <gcc at gcc dot gnu dot org>, "'Albert Cohen'" <Albert dot Cohen at inria dot fr>, "'Yuri Kashnikoff'" <yuri dot kashnikoff at gmail dot com>, "'Yuanjie Huang'" <huangyuanjie at ict dot ac dot cn>, "'Liang Peng'" <pengliang at ict dot ac dot cn>, <dorit at il dot ibm dot com>, "'Mircea Namolaru'" <NAMOLARU at il dot ibm dot com>, "'Diego Novillo'" <dnovillo at google dot com>
- Date: Fri, 6 Nov 2009 19:44:22 +0100
- Subject: RE: [plugins-ici-cloning-instrumentation] new GCC plugin developements
- References: <4AE6E471.4020200@starynkevitch.net> <84fc9c000910270518s147e0cc4t7f1cd53049c5d590@mail.gmail.com> <4AE6E5D0.4010401@starynkevitch.net> <mcrvdi1klsb.fsf@dhcp-172-17-9-151.mtv.corp.google.com> <4AE7014A.1000903@starynkevitch.net> <mcr3a54lwb4.fsf@dhcp-172-17-9-151.mtv.corp.google.com> <4AE70C5E.4050005@starynkevitch.net> <84fc9c000910270839v2d9efe0dw829c8647f361c96f@mail.gmail.com> <4AE7164D.9010200@starynkevitch.net> <84fc9c000910270855w736df367qe511d8db280aaeb4@mail.gmail.com> <2dc303d60910271056h17038110ib63c53cfa374f5c7@mail.gmail.com> <002c01ca5746$9dd43da0$d97cb8e0$@fursin@inria.fr> <20091102074959.p8410ulv28sg0w44-nzlynne@webmail.spamcop.net> <008601ca5e09$93164fb0$b942ef10$@fursin@inria.fr> <20091106121424.ph6anlgbk0848sss-nzlynne@webmail.spamcop.net> <01e701ca5f0d$903386f0$b09a94d0$@fursin@inria.fr> <20091106132953.s84iph9egwso8oo8-nzlynne@webmail.spamcop.net>
Hi Joern,
> > In parallel Zbigniew prepared a patch that synchronized ICI with the
> > plugin branch
> > and if I am correct, he removed all the duplicating functionality so
> > the changes became
> > quite minimal.
>
> Where is that patch?
After looking at all my email I think we only had internal copies of the patch
so it has never been sent out. It seems like marriage can break memory completely -
so I am really sorry for those memory blackouts :( ...
This means that we are getting back to the same task of syncing the ICI2.0 from adapt
SVN to the current GCC. Actually, could you please check what is duplicate functionality
between current GCC 4.5 in the ICI and if we can remove it from ICI ...
It should naturally relate mainly to the plugin functions ...
Only when it's done we can see if ICI will be intrusive or not ...
> > BUT we did not use that branch for GSOC projects, so what you see
> > in the gcc-4.4.0-ici2.0-adapt is the ICI with duplicate
> > functionality of the old plugin branch
> > and normally the current mainline GCC 4.5. That's I think you start
> > having conflict names.
>
> The two register_pass functions are quite distinct in their functionality
> (except that logically the passes.c one should call - directly or
> indirectly - the pass_manager.c one). Both are an interface exposed to
> plugins, so we can't simply rename an internal function. One of the
> interfaces has to change.
Well, ICI is not in the mainline so it has to change and adapt to the
current mainline... Can you please check if there are other similar cases apart
from this to decide on our changes in one go since we will need to alter
all our plugins due to that (unless we provide some legacy wrapper but
I think at this stage I would prefer to have a clean ICI in GCC)...
Thanks a lot,
Grigori