From owner-sc22wg5+sc22wg5-dom9=www.open-std.org@open-std.org Fri Mar 29 20:15:47 2024 Return-Path: X-Original-To: sc22wg5-dom9 Delivered-To: sc22wg5-dom9@www.open-std.org Received: by www.open-std.org (Postfix, from userid 521) id EA873356DFB; Fri, 29 Mar 2024 20:15:47 +0100 (CET) Delivered-To: sc22wg5@open-std.org X-Greylist: delayed 15857 seconds by postgrey-1.34 at www5.open-std.org; Fri, 29 Mar 2024 20:15:47 CET Received: from mail-4322.protonmail.ch (mail-4322.protonmail.ch [185.70.43.22]) by www.open-std.org (Postfix) with ESMTP id ADE42356DF5 for ; Fri, 29 Mar 2024 20:15:46 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=protonmail.com; s=protonmail3; t=1711739745; x=1711998945; bh=VhGnnY36BkylZtCGuZjE/ME9VxnutksaHrbPRaSmOIA=; h=Date:To:From:Cc:Subject:Message-ID:In-Reply-To:References: Feedback-ID:From:To:Cc:Date:Subject:Reply-To:Feedback-ID: Message-ID:BIMI-Selector; b=XtxL4UovTCAE5Vv6ipHKQVug9y7GjCiyPu2qGy5+cXIPzixrCtGVNdWfAs4WLdGPN SFvVLxP8dsEE5SJKt3XuzzyXfy/phbcFML3K3oj0vOvzJg4yIImr9OPP1RiyT3Bczr teA1epXUjnxHYHoKQxxEYS5N1EMVR46q60oVJ2Uk8N1fA9SQeYcyZy/Y+q5XNjrtAH q85e5qIjIuUA86kU/WdLXrTBHuMjbCjnggxG+uIE8aL2Irj0uX8mQEdGZx/q4F6UV2 vSIJ9FJ1HTknSGGNKyETmV1ZnvXY5Vp9/RAEAF7+5UioiiIxQNWMlNoyQ71ZRX3eqz Ivl4iWdTT0EmQ== Date: Fri, 29 Mar 2024 19:15:33 +0000 To: General J3 interest list , =?utf-8?Q?Thomas_K=C3=B6nig?= From: Brad Richardson Cc: Jeff Hammond , WG5 Subject: Re: [J3] [SC22WG5.6583] Fortran type sizes - please help with MPI Fortran Message-ID: In-Reply-To: <20240329175417.C02A4356DF4@www.open-std.org> References: <20240329145655.E8CFB356DD2@www.open-std.org> <1DFBC5F5-0300-419B-AD1D-C49E73E066A9@nvidia.com> <20240329175417.C02A4356DF4@www.open-std.org> Feedback-ID: 13174433:user:proton MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Sender: owner-sc22wg5@open-std.org Precedence: bulk Hi Jeff, It is my expectation that there is virtually no existing code using MPI where "INTEGER is INTEGER(kind=3Dc_int)" is not the case, else it wouldn't have worked in the first place. Unless you're telling me that somebody is building a version of an MPI library and the Fortran interface to it where that isn't the case, but if so I have faith they could figure out how to continue doing weird things with the right compiler options. Brad On Fri, 2024-03-29 at 17:54 +0000, Jeff Hammond via J3 wrote: > MPI has three types of Fortran support.=C2=A0 The (deprecated) mpif.h and > MPI.mod support an old-fashioned API, where we only use the primitive > types of Fortran 77.=C2=A0 At least with MPI.mod, the compiler can do > interface checking, although the type-safety is quite weak given most > arguments are INTEGER.=C2=A0 The newer MPI_F08.mod, which Bill Long was > instrumental in creating and which WG5 made possible with the Fortran > 2018 CFI extensions, is a properly typed, modern Fortran API. >=20 > The idea that "it will work" is equivalent to MPI asserting that > INTEGER is INTEGER(kind=3Dc_int) or the MPI ABI doesn=E2=80=99t work, whi= ch > certain people will not accept unless WG5 provides some sort of > evidence that I am not a lunatic for wanting to assert this. >=20 > Jeff >=20 > > On 29. Mar 2024, at 19.33, Thomas K=C3=B6nig wrote: > >=20 > > External email: Use caution opening links or attachments > >=20 > >=20 > > Am 29.03.24 um 16:24 schrieb Jeff Hammond via J3: > > > The issue is, we have 30 years of MPI Fortran code that does > > > things > > > that are correct according to MPI and Fortran, but which are > > > broken if > > > we do certain things that are otherwise very nice from a software > > > design perspective.=C2=A0 Breaking user code is bad for business, as > > > everybody here knows :-) > > >=20 > > > If we could redefine MPI Fortran to use ISO_C_BINDING types, we= =E2=80=99d > > > have > > > zero problems, but ISO_C_BINDING came a decade after MPI was > > > created > > > and Fortran 2003 compilers weren=E2=80=99t widely available until wel= l > > > after 2003. > >=20 > > How does MPI define its interfaces?=C2=A0 Does it use Fortran modules o= r > > the traditional F77-style ABI (with no checking) that f2c etc use? > >=20 > > For the former, C interoperable types could be used; I think all > > major compilers support it by now. > >=20 > > For Linux and at least some BSD variants on AMD64, the latter is > > defined in the x86-64-ABI psABI, which can be found at > > https://gitlab.com/x86-psABIs/x86-64-ABI=C2=A0(section 9.2), > > and major other architectures and operating systems effectively do > > the same. > >=20 > > Maybe defining the MPI interface with respect to an existing ABI > > would be less onerous than just saying "it will work". > >=20 > > Best regards > >=20 > > =C2=A0=C2=A0 Thomas > >=20 >=20