[J3] [EXTERNAL] Re: [SC22WG5.6371] Kickstarting proposals for F202Y features

Fri May 13 14:00:03 UTC 2022

The generics subgroup has expressed strong interest in adding an analog to ELEMENTAL in F2Y that would work with user-defined containers.   For now I’ve considered that to  be separate from our primary work on generics features.

From: J3 <j3-bounces at mailman.j3-fortran.org> on behalf of j3 <j3 at mailman.j3-fortran.org>
Reply-To: j3 <j3 at mailman.j3-fortran.org>
Date: Friday, May 13, 2022 at 9:42 AM
To: j3 <j3 at mailman.j3-fortran.org>, WG5 List <sc22wg5 at open-std.org>
Cc: Jeff Hammond <jehammond at nvidia.com>
Subject: [EXTERNAL] Re: [J3] [SC22WG5.6371] Kickstarting proposals for F202Y features

My desired features for 2Y are:

1.       fetch-and-op atomics in DO CONCURRENT

2.       asynchronous blocks / tasks (equivalent to OpenACC async and OpenMP non-dependent tasks)

3.       scan / prefix sum (LBNL has a paper on this in preparation)

I am working on the papers for the first two of these for Boulder.

Jeff

From: J3 <j3-bounces at mailman.j3-fortran.org> on behalf of Steve Lionel via J3 <j3 at mailman.j3-fortran.org>
Date: Saturday, 23April 2022 at 11:27 PM
To: WG5 List <sc22wg5 at open-std.org>
Cc: Steve Lionel <steve at stevelionel.com>
Subject: [J3] [SC22WG5.6371] Kickstarting proposals for F202Y features
External email: Use caution opening links or attachments

Back in 2019, Ondrej, Zach and Gary started a "Fortran Proposals" Github repository (https://github.com/j3-fortran/fortran_proposals<https://gcc02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fj3-fortran%2Ffortran_proposals&data=05%7C01%7Cthomas.l.clune%40nasa.gov%7C7298857f24124b1e36bf08da34e65f58%7C7005d45845be48ae8140d43da96dd17b%7C0%7C0%7C637880461300568179%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=%2B7eWniRW8ZVSFhbMXTs%2BVpojshIh3zZcZng8tI6M3L0%3D&reserved=0>), which I envisioned as serving as an "incubator" for features that might be considered in future revisions. A few months ago I asked Ondrej to compile a list, organized by clause, of the entries submitted to date, and this is now available. Some of these are detailed with lots of discussion and some are more abstract concepts. Quite a few of these have been proposed and declined for earlier revisions, but it's always worth taking a second look. Some are already in F2023.

Please look over the list (https://gitlab.com/fortran-lang/github_stats/-/wikis/j3-04232022<https://gcc02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgitlab.com%2Ffortran-lang%2Fgithub_stats%2F-%2Fwikis%2Fj3-04232022&data=05%7C01%7Cthomas.l.clune%40nasa.gov%7C7298857f24124b1e36bf08da34e65f58%7C7005d45845be48ae8140d43da96dd17b%7C0%7C0%7C637880461300568179%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=L0CoagDh32l8hwxkoJsc8gVCjadB8RM0Dfv369ngcnI%3D&reserved=0>) and click through for details on proposals that catch your eye. I'd also encourage additional comments on proposals if you have them.

I'd like to do at the July meeting what we did for F202x (coincidentally, also in Boulder) - have each attendee (including virtual) come up with a list of five features you'd like considered. These can come from the Github list, your own concepts, or anywhere else. Subgroups will then offer opinions on each and rank them.

This is just the start of the process, but we will have the advantage of many of the ideas already being discussed and evaluated. I also need to go through the papers list and make sure we include features that were marked "deferred" or "out-of-order". I expect that other ideas will come in over the next year. Of course, Generics will be a big part of this.

Thanks for all your hard work to get us to the endgame of F2023, despite the setbacks of a global pandemic. I'm proud of what we have accomplished.

Steve
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.j3-fortran.org/pipermail/j3/attachments/20220513/46fd64c8/attachment-0001.htm>