[Openmp-dev] OpenMP and C++

via Openmp-dev openmp-dev at lists.llvm.org
Thu Aug 11 20:20:58 PDT 2016


Where's the face palm emoji when I need it.. That's what I get for early morning reply

  Original Message  
From: Hal Finkel
Sent: Friday, August 12, 2016 10:19
To: C Bergström
Cc: Terry L Wilmarth; openmp-dev
Subject: Re: OpenMP and C++

----- Original Message -----
> From: "C Bergström" <cbergstrom at pathscale.com>
> To: "Hal Finkel" <hfinkel at anl.gov>
> Cc: "Terry L Wilmarth" <terry.l.wilmarth at intel.com>, "openmp-dev" <openmp-dev at lists.llvm.org>
> Sent: Thursday, August 11, 2016 10:13:43 PM
> Subject: OpenMP and C++
> 
> If you plan to introduce C++ officially, would you mind have a

Not that such a discussion is a bad thing, but I'll point out that roughly 25% of the runtime source files are already C++ files, so this is not really an introduction.

> discussion on which parts of the C++ language are ok and which are
> not
> ok. I'd be strongly in favor of not using RTTI and EH. I think both
> of
> those should be disabled in the build and never used. (Performance
> reasons)

Indeed; these things are already a part of LLVM's policy.

-Hal

> 
> There are also a few c++11 things which are popular, but may not
> always make sense. Stuff like that can be handled on a case-by-case
> basis.
> 
> On Fri, Aug 12, 2016 at 8:21 AM, Hal Finkel via Openmp-dev
> <openmp-dev at lists.llvm.org> wrote:
> > ----- Original Message -----
> >> From: "Terry L via Openmp-dev Wilmarth"
> >> <openmp-dev at lists.llvm.org>
> >> To: "Chris Lattner" <clattner at apple.com>,
> >> openmp-dev at lists.llvm.org
> >> Sent: Thursday, August 11, 2016 1:06:09 PM
> >> Subject: Re: [Openmp-dev] LLVM coding conventions an the OpenMP
> >> runtime
> >>
> >> A second thing we are considering (and apologies for leaving this
> >> out
> >> of the original message) is the renaming of .c files to .cpp. As
> >> the runtime evolves, C++ has been used more and more.
> >
> > Yes, please. That would be great.
> >
> >>
> >> So the file renaming is something else to consider, either
> >> separately
> >> or in conjunction with the code formatting changes.
> >
> > Please do this in a separate commit. On the off-chance that this
> > introduces a behavioral change, I'd like to be able to bisect to
> > just this change.
> >
> > -Hal
> >
> >>
> >> -Terry
> >>
> >> -----Original Message-----
> >> From: Openmp-dev [mailto:openmp-dev-bounces at lists.llvm.org] On
> >> Behalf
> >> Of Chris Lattner via Openmp-dev
> >> Sent: Thursday, August 11, 2016 12:17 AM
> >> To: C Bergström <cbergstrom at pathscale.com>
> >> Cc: LLVM-OpenMP (openmp-dev at lists.llvm.org)
> >> <openmp-dev at lists.llvm.org>
> >> Subject: Re: [Openmp-dev] LLVM coding conventions an the OpenMP
> >> runtime
> >>
> >> On Aug 9, 2016, at 1:23 AM, C Bergström via Openmp-dev
> >> <openmp-dev at lists.llvm.org> wrote:
> >> > Pragmatically and just my view - If the research is open and
> >> > there's a
> >> > plan to integrate it back I'm empathetic. If it's closed and
> >> > just
> >> > an
> >> > outside fork, I don't care what you do and it shouldn't block an
> >> > open
> >> > source project.
> >>
> >> +1, this is consistent with the overall LLVM approach.
> >>
> >> The open source project can’t be concerned with every private fork
> >> that may be maintained, and worrying about that often cuts against
> >> the bests interests of the project (which is to always move
> >> forward).
> >>
> >> Additionally, it is arguably *better* for the open source project
> >> to
> >> have a bit of churn in the code, because this makes it more
> >> painful
> >> to maintain out of tree branches. This is one way to encourage
> >> these
> >> private branches to be contributed back to mainline, instead of
> >> being kept proprietary.
> >>
> >> -Chris
> >> _______________________________________________
> >> Openmp-dev mailing list
> >> Openmp-dev at lists.llvm.org
> >> http://lists.llvm.org/cgi-bin/mailman/listinfo/openmp-dev
> >> _______________________________________________
> >> Openmp-dev mailing list
> >> Openmp-dev at lists.llvm.org
> >> http://lists.llvm.org/cgi-bin/mailman/listinfo/openmp-dev
> >>
> >
> > --
> > Hal Finkel
> > Assistant Computational Scientist
> > Leadership Computing Facility
> > Argonne National Laboratory
> > _______________________________________________
> > Openmp-dev mailing list
> > Openmp-dev at lists.llvm.org
> > http://lists.llvm.org/cgi-bin/mailman/listinfo/openmp-dev
> 

-- 
Hal Finkel
Assistant Computational Scientist
Leadership Computing Facility
Argonne National Laboratory


More information about the Openmp-dev mailing list