From owner-sc22wg5+sc22wg5-dom9=www.open-std.org@open-std.org Fri Mar 15 18:29:02 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 ED9FA356D6C; Fri, 15 Mar 2024 18:29:01 +0100 (CET) Delivered-To: sc22wg5@open-std.org Received: from NAM10-DM6-obe.outbound.protection.outlook.com (mail-dm6nam10on2063.outbound.protection.outlook.com [40.107.93.63]) by www.open-std.org (Postfix) with ESMTP id 4BA57356D48 for ; Fri, 15 Mar 2024 18:28:59 +0100 (CET) ARC-Seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=A+7BrFsvV1oqLZPymwJHboCnn2FPpHfGXIP044/d+nzurk7nd6hOqYbJkBTjxhfoH6TbwVADrZ7chfLInImHuAIi09tJGzTy+vc0zuZs4rAhN1MMsAPC7rf4Nj4IOij2i5xdWCzk1blFX4PU+Nwm0ahwveUtw69EjXpKl/Kt5W38ZIHIwA2zvc02hsm0Jqu9vP5Anp8xWlKG1qZXJBqAke21dJlI4+g2NOKeh4PR/CF9/1MXhkLQtggA4aQ2Br2pUeesMbQx6/vRL58iojOWC1Roe1jNGqzD3Ao/f9zYjOz4P1h0Gz7cQ9TpRz7Y22pfMOBbDGI/Ty3dniIjnn2Q4g== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=Kv7mJnTx/xVj3i9H25kSZMVngXB91w31MaUA5qW7j+k=; b=g+y8R9NVlTVI6eQMQUFzDZrKwlNbzPq+aUUm2wJyPRNumpGHeXjeeMbErzSNJ4EhKkA0iVI/5MoVMjy/lfyphndgg6Gb1txxEf6YT5wu+WodLc99BsFwqzh90hLH1+1aoTyb/EN3UW6Q1H5XeSjhZKK81BHDnAKz8MAWsqeFMa396qyQri4gdQcqr33ilBqCY8gJ7Lc+5qjlTPmcmYM8bcOu6mEaFCaj7fTZbAAEtasU4r6uDrI4d1QzK4HRiYKXCncwRFizntuh7PkpBkpdNdnb85pIK6WMpIRVoe1FYA4Lzm5xzMq/Uk/CdmPfQmlQkaLsPGA5T07tvYPFk3g1LQ== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=nvidia.com; dmarc=pass action=none header.from=nvidia.com; dkim=pass header.d=nvidia.com; arc=none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=Nvidia.com; s=selector2; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=Kv7mJnTx/xVj3i9H25kSZMVngXB91w31MaUA5qW7j+k=; b=Vd9QNe3G2oj+/R9S6zjwbTmmxhFqES68vWrpe/NQ3lphLICQJzE4jriKGiB2+FEkP42/0AUXwz7oa5j6/q2ABMPmV3mzMP5AYxa3ouRhsvQKo4OrjgasinE+0ajmgE3JDucD3m4wxoIyGFgf5sjIG1cwIQIhIBtFwZlGEXTqa8A6sgtzmR9INONUaOsP3HzFcBxlhPQvvp9ZAf8LqGnVBxGHGOrVoBya8rpuNWTYrbPkOj7dimMSO0oc/5YJ7LguylgqsngJfl5EsL4IODubgjqeTLbFykWAoutFGZRGAWFzwtu6wlPziqirxIkBjMQAPrEYx8wZ/cznXavbQoSohg== Received: from DM6PR12MB3130.namprd12.prod.outlook.com (2603:10b6:5:11b::16) by PH8PR12MB6962.namprd12.prod.outlook.com (2603:10b6:510:1bd::18) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.7386.21; Fri, 15 Mar 2024 17:28:55 +0000 Received: from DM6PR12MB3130.namprd12.prod.outlook.com ([fe80::3b3e:8b99:dc87:e5b3]) by DM6PR12MB3130.namprd12.prod.outlook.com ([fe80::3b3e:8b99:dc87:e5b3%5]) with mapi id 15.20.7386.017; Fri, 15 Mar 2024 17:28:55 +0000 From: Jeff Hammond To: WG5 Subject: bind(C) external linkage issues on MacOS Thread-Topic: bind(C) external linkage issues on MacOS Thread-Index: AQHadvWkjc0rsKcmVECLDT/CNDzOMA== Date: Fri, 15 Mar 2024 17:28:55 +0000 Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: authentication-results: dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=nvidia.com; x-ms-publictraffictype: Email x-ms-traffictypediagnostic: DM6PR12MB3130:EE_|PH8PR12MB6962:EE_ x-ms-office365-filtering-correlation-id: cb1a9564-26ec-4bbb-bcda-08dc451563f6 x-ms-exchange-senderadcheck: 1 x-ms-exchange-antispam-relay: 0 x-microsoft-antispam: BCL:0; x-microsoft-antispam-message-info: 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 x-forefront-antispam-report: CIP:255.255.255.255;CTRY:;LANG:en;SCL:1;SRV:;IPV:NLI;SFV:NSPM;H:DM6PR12MB3130.namprd12.prod.outlook.com;PTR:;CAT:NONE;SFS:(13230031)(366007)(1800799015)(376005)(38070700009);DIR:OUT;SFP:1101; x-ms-exchange-antispam-messagedata-chunkcount: 1 x-ms-exchange-antispam-messagedata-0: =?Windows-1252?Q?HiTkU7EgSpH9kma1g6ygQgovb91qBIsh8mxlBrCcMZ3uBTqEzXUovbpw?= =?Windows-1252?Q?PSgE7C191arnFzFRZ8yEUyoGmZtqjw2HDX8lPTd1FMfOtnZ6hrhq7Yiy?= =?Windows-1252?Q?0SNfLgyWZU1l5qey0RlhUWuhdHWGtWPwctOd08mpB9qdlPX9zJeeQpJx?= =?Windows-1252?Q?5/GiyNlmqDCqcBfglf7qdXt0QgAODsLaeo0VhS14++PGPznn1v/b/jk6?= =?Windows-1252?Q?Om8HK1ZN8ZeL4ZTQKGqiHL3I5mFVO2vtJ6DONmOaguxkZrcMS630gUvP?= =?Windows-1252?Q?H7Qzd8S+1jsk2wbCOkafsdwH3O3J3AzUb4n3y8WmZSMkt86daYnfixRL?= =?Windows-1252?Q?A0neIFm1Ybk/rzv/fbzsjKSpk0jsJxPMmm9FxCZTRMxorycKQCndwjbQ?= =?Windows-1252?Q?O0KN3vCW8rqapORp6CYu6lfFwjFkOc8IHUb4mjgHpjGYw+Qz9wl1k95M?= =?Windows-1252?Q?nKpWGRe7ryyINN3qvbVLV50hj9uazejiUAIDkrebUHesKz+RZ+EWLAJg?= =?Windows-1252?Q?s4Rf8LsUp2VCaGADBt+Tj4WUgOzDdvCP9ZglID+CiJVEQQ34nicTzLcx?= =?Windows-1252?Q?kCNa8xhBy2IFkO0xvWdEwgjkhgu2Qc/Tl+XjurjXAPcJr9F/eD1MGNyr?= =?Windows-1252?Q?tgTzPPr6f46MJ9sQ1ALrd0m8Jx0Jri0j7lcOGv40lt5m3ZSKXC8GULMh?= =?Windows-1252?Q?zDW6rfd9TR8y7RnCh49n+97ItDXjrTdHXyPyaoOPqp2S1+OWJIdo+0SM?= =?Windows-1252?Q?RDAR6MuiAixyQgcNApSWRYPGFOLw/P9lBphrdaW33170HZt//Omi7u1v?= =?Windows-1252?Q?KdtEgxGEiy6W8aH6G1rFDoyH8fQWFlNA/53WY1RiiqdZ03V6VCkcgamN?= =?Windows-1252?Q?uJiXRkhLgUwZF/elLgzcP0mn+tXCjNZKGCjdskd6nNtnX4veU05sYqh1?= =?Windows-1252?Q?YqL+MrEt60tK1qSUuMpQE9w/K8n4fK6Xivn8ajkLJ8QCvruVTuY8re2t?= =?Windows-1252?Q?p4vVTi4V/a7Q8kbhzIHxz5xV1EJP/iJSWTYI68hT/PDGu2l7NfBgom3q?= =?Windows-1252?Q?ffgK/vLm5DDoEICcYXLxbgeWcSIA13nBmqg4SZh7JQzIbPI4wOzbui10?= =?Windows-1252?Q?xdoQz8QQmSVs0/nnQhQhjD6KYbNQ3D9zxOW1GQqHF5fTj7nQJp7RVwXj?= =?Windows-1252?Q?Dlc2RfAWxuP0AnPmZYL8TCxEF00e7dl9s7L8pDXT7PxeMvwaW0Ck9eIQ?= =?Windows-1252?Q?GnKGE7ol5xfvOPuCFEAI/EWQFu+N5vvKrSIeF2XRlwmDZBLDhsDd0Iob?= =?Windows-1252?Q?xvv4v7LI10SD8he5dJ/z9BrpxvT+36V1nlJPH9tvHU8C3art9AQhXCZc?= =?Windows-1252?Q?ABXL86zPBksnEOisaNJbeJfpuVi3zpJkB8LKqJlI/bf3FHtlWf0DG0Zp?= =?Windows-1252?Q?7Fzbpekf3HEGOBm7HfOrcb6EOGEXC2xZQi9UXxC0KUW4ohHycOEzLmO8?= =?Windows-1252?Q?wrhRBNpdjrAGDUEx71LEjCQhas2sw4gwMSbfVqSMyEMzTapNTiwUZU2k?= =?Windows-1252?Q?EuG7NrJbGj6bFJZmdjnHQBAMK+V8R8Mt76vzhYEsY8OF8zYRsWxO0be2?= =?Windows-1252?Q?N7ETfoxU0msdjgyWS33fUrCbhLwl6qBAQ5CCpLrL7KHvaMWBEBfrbxUW?= =?Windows-1252?Q?/pXQoPj2l3a60fwc6x1I9Vdu2Cj8zvadpMMiEdi0/EPbUZ9ZN81sXw?= =?Windows-1252?Q?=3D=3D?= Content-Type: multipart/alternative; boundary="_000_DM6PR12MB3130DA4A03A92CC08EF2F8F5CB282DM6PR12MB3130namp_" MIME-Version: 1.0 X-OriginatorOrg: Nvidia.com X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-AuthSource: DM6PR12MB3130.namprd12.prod.outlook.com X-MS-Exchange-CrossTenant-Network-Message-Id: cb1a9564-26ec-4bbb-bcda-08dc451563f6 X-MS-Exchange-CrossTenant-originalarrivaltime: 15 Mar 2024 17:28:55.2879 (UTC) X-MS-Exchange-CrossTenant-fromentityheader: Hosted X-MS-Exchange-CrossTenant-id: 43083d15-7273-40c1-b7db-39efd9ccc17a X-MS-Exchange-CrossTenant-mailboxtype: HOSTED X-MS-Exchange-CrossTenant-userprincipalname: ckckmYnyTdswYgdRTSTOnWeyzFYua4S36sEWKgfuMzImtiWiPOAbAoh0WAIkPbrJfJzgBhvCpvRJx8s7TiY8rw== X-MS-Exchange-Transport-CrossTenantHeadersStamped: PH8PR12MB6962 Sender: owner-sc22wg5@open-std.org Precedence: bulk --_000_DM6PR12MB3130DA4A03A92CC08EF2F8F5CB282DM6PR12MB3130namp_ Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable This is both an FYI for those of you who use MacOS but also a possible iden= tification of a problem in how we specify 19.5.1.5 Linkage association. Apple has removed =93-commons use_dylibs=94 in recent versions of MacOS. T= his flag sets a non-default behavior of COMMON symbols (which include linka= ge associated bind(C) symbols that are not using obsolescent features) to b= e =93U=94 (undefined) instead of =93S=94 (uninitialized, small) https://lin= ux.die.net/man/1/nm. This flag is required to get a bind(C) module variable that points to a C g= lobal variable to work properly, as is potentially required by MPI. There = are many reports of failures online with both GCC and Intel compilers (I ca= n only verify the former). It has been noted that our definition of linkage association may not be ade= quate to interact with C shared libraries. I don=92t understand this topic= well enough to have an opinion either way. Has anybody else run into this or thought about it enough to know if there = is a way to enhance our definition of linkage association so as to not requ= ire Fortran-specific linker options and behave more like C extern in shared= libraries? Thanks, Jeff jhammond@Jeff-M2 fortran % gcc -fPIC -shared extern2.c -o libxxx.so && gfor= tran -c extern.F90 && ld extern.o libxxx.so -L/opt/homebrew/Cellar/gcc/13.2= .0/lib/gcc/current/ -lgfortran -commons use_dylibs -o extern && ./extern ;= nm extern | grep MPI MPIR_F08_MPI_IN_PLACE=3D0 &MPIR_F08_MPI_IN_PLACE=3D0x1003a4000 &MPIR_F08_MP= I_IN_PLACE=3D4298784768 LOC(MPI_IN_PLACE)=3D 4298784768 LOC(buf)=3D 6168523360 sendbuf=3D0x1003a4000, sendbuf=3D4298784768 sendbuf is MPI_IN_PLACE? yes recvbuf=3D0x16fac3260, recvbuf=3D6168523360 *count=3D1, *datatype=3D2, *op=3D3, *comm=3D4 911 U _MPIR_F08_MPI_IN_PLACE U _MPI_Allreduce jhammond@Jeff-M2 fortran % gcc -fPIC -shared extern2.c -o libxxx.so && gfor= tran -c extern.F90 && ld extern.o libxxx.so -L/opt/homebrew/Cellar/gcc/13.2= .0/lib/gcc/current/ -lgfortran -o extern && ./extern ; nm extern | grep M= PI MPIR_F08_MPI_IN_PLACE=3D0 &MPIR_F08_MPI_IN_PLACE=3D0x102b1c000 &MPIR_F08_MP= I_IN_PLACE=3D4340170752 LOC(MPI_IN_PLACE)=3D 4339777664 LOC(buf)=3D 6127137376 sendbuf=3D0x102abc080, sendbuf=3D4339777664 sendbuf is MPI_IN_PLACE? no recvbuf=3D0x16d34b260, recvbuf=3D6127137376 *count=3D1, *datatype=3D2, *op=3D3, *comm=3D4 911 0000000100008080 S _MPIR_F08_MPI_IN_PLACE U _MPI_Allreduce =3D=3D> extern2.c <=3D=3D #include #include int MPIR_F08_MPI_IN_PLACE; void p(void) { printf("MPIR_F08_MPI_IN_PLACE=3D%d &MPIR_F08_MPI_IN_PLACE=3D%p &MPIR_F0= 8_MPI_IN_PLACE=3D%zu\n", MPIR_F08_MPI_IN_PLACE, &MPIR_F08_MPI_IN_PLACE, (intptr_t)&M= PIR_F08_MPI_IN_PLACE); } void MPI_Allreduce(void ** sendbuf, void ** recvbuf, int * count, int * datatype, int * op, int * comm, int * ierror) { printf("sendbuf=3D%p, sendbuf=3D%zu\n", sendbuf, (intptr_t)sendbuf); printf("sendbuf is MPI_IN_PLACE? %s\n", (intptr_t)sendbuf=3D=3D(intptr_t)&MPIR_F08_MPI_IN_PLACE ? "yes" = : "no"); printf("recvbuf=3D%p, recvbuf=3D%zu\n", recvbuf, (intptr_t)recvbuf); printf("*count=3D%d, *datatype=3D%d, *op=3D%d, *comm=3D%d\n", *count, *datatype, *op, *comm); *ierror =3D 911; } =3D=3D> extern.F90 <=3D=3D module mpi use iso_c_binding !type(c_ptr), bind(C,name=3D"MPI_F_IN_PLACE") :: MPI_IN_PLACE integer(c_int), bind(C, name=3D"MPIR_F08_MPI_IN_PLACE"), target :: MPI_= IN_PLACE interface subroutine p() bind(C,name=3D"p") end subroutine end interface interface SUBROUTINE MPI_ALLREDUCE(SENDBUF, RECVBUF, COUNT, DATATYPE, OP, COM= M, IERROR) & bind(C,name=3D"MPI_Allreduce") use iso_c_binding import :: MPI_IN_PLACE !DEC$ ATTRIBUTES NO_ARG_CHECK :: sendbuf,recvbuf !GCC$ ATTRIBUTES NO_ARG_CHECK :: sendbuf,recvbuf !$PRAGMA IGNORE_TKR sendbuf,recvbuf !DIR$ IGNORE_TKR sendbuf,recvbuf !IBM* IGNORE_TKR sendbuf,recvbuf INTEGER(kind=3Dc_int) :: SENDBUF(*), RECVBUF(*) INTEGER(kind=3Dc_int) :: COUNT, DATATYPE, OP, COMM, IERROR END SUBROUTINE MPI_ALLREDUCE end interface end module mpi program main use mpi implicit none real :: buf(100) integer :: ierror call p buf =3D 17 print*,'LOC(MPI_IN_PLACE)=3D',LOC(MPI_IN_PLACE) print*,'LOC(buf)=3D',LOC(buf) call MPI_ALLREDUCE(MPI_IN_PLACE,buf,1,2,3,4,ierror) print*,ierror end program main --_000_DM6PR12MB3130DA4A03A92CC08EF2F8F5CB282DM6PR12MB3130namp_ Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable

This= is both an FYI for those of you who use MacOS but also a possible identifi= cation of a problem in how we specify 19.5.1.5 Linkage association.

 

Appl= e has removed =93-commons use_dylibs=94 in recent versions of MacOS.  = This flag sets a non-default behavior of COMMON symbols (which include link= age associated bind(C) symbols that are not using obsolescent features) to be =93U=94 (undefined) instead of =93S=94 (uninit= ialized, small) https://linux.die.net/man/1/nm.

 

This= flag is required to get a bind(C) module variable that points to a C globa= l variable to work properly, as is potentially required by MPI.  There= are many reports of failures online with both GCC and Intel compilers (I can only verify the former).<= /p>

 

It h= as been noted that our definition of linkage association may not be adequat= e to interact with C shared libraries.  I don=92t understand this topi= c well enough to have an opinion either way.

 

Has = anybody else run into this or thought about it enough to know if there is a= way to enhance our definition of linkage association so as to not require = Fortran-specific linker options and behave more like C extern in shared libraries?

 

Than= ks,

Jeff

 

jham= mond@Jeff-M2 fortran % gcc -fPIC -shared extern2.c -o libxxx.so && = gfortran -c extern.F90 && ld extern.o libxxx.so -L/opt/homebrew/Cel= lar/gcc/13.2.0/lib/gcc/current/  -lgfortran -commons use_dylibs -o extern && ./extern ; nm extern | grep MPI

MPIR= _F08_MPI_IN_PLACE=3D0 &MPIR_F08_MPI_IN_PLACE=3D0x1003a4000 &MPIR_F0= 8_MPI_IN_PLACE=3D4298784768

LOC(= MPI_IN_PLACE)=3D          = ; 4298784768

LOC(= buf)=3D           6168523= 360

send= buf=3D0x1003a4000, sendbuf=3D4298784768

send= buf is MPI_IN_PLACE? yes

recv= buf=3D0x16fac3260, recvbuf=3D6168523360

*cou= nt=3D1, *datatype=3D2, *op=3D3, *comm=3D4

&nbs= p;        911

           &nb= sp;     U _MPIR_F08_MPI_IN_PLACE

&nbs= p;            &= nbsp;   U _MPI_Allreduce

 

 

jham= mond@Jeff-M2 fortran % gcc -fPIC -shared extern2.c -o libxxx.so && = gfortran -c extern.F90 && ld extern.o libxxx.so -L/opt/homebrew/Cel= lar/gcc/13.2.0/lib/gcc/current/  -lgfortran  -o extern && ./extern ; nm extern | grep MPI

MPIR= _F08_MPI_IN_PLACE=3D0 &MPIR_F08_MPI_IN_PLACE=3D0x102b1c000 &MPIR_F0= 8_MPI_IN_PLACE=3D4340170752

LOC(= MPI_IN_PLACE)=3D          = ; 4339777664

LOC(= buf)=3D           6127137= 376

send= buf=3D0x102abc080, sendbuf=3D4339777664

send= buf is MPI_IN_PLACE? no

recv= buf=3D0x16d34b260, recvbuf=3D6127137376

*cou= nt=3D1, *datatype=3D2, *op=3D3, *comm=3D4

&nbs= p;        911

0000000100008080 S _MPIR_F08_MPI_IN_PLACE

&nbs= p;            &= nbsp;   U _MPI_Allreduce

 

 

=3D= =3D> extern2.c <=3D=3D

#inc= lude <stdio.h>

#inc= lude <stdint.h>

 

int = MPIR_F08_MPI_IN_PLACE;

 

void= p(void)

{

&nbs= p;   printf("MPIR_F08_MPI_IN_PLACE=3D%d &MPIR_F08_MPI_IN= _PLACE=3D%p &MPIR_F08_MPI_IN_PLACE=3D%zu\n",

&nbs= p;           MPIR_F08_MPI= _IN_PLACE,   &MPIR_F08_MPI_IN_PLACE,   (intptr_t)&a= mp;MPIR_F08_MPI_IN_PLACE);

}

 

void= MPI_Allreduce(void ** sendbuf, void ** recvbuf,

&nbs= p;            &= nbsp;     int * count, int * datatype,

&nbs= p;            &= nbsp;     int * op, int * comm, int * ierror)

{

&nbs= p;   printf("sendbuf=3D%p, sendbuf=3D%zu\n", sendbuf, (= intptr_t)sendbuf);

&nbs= p;   printf("sendbuf is MPI_IN_PLACE? %s\n",=

&nbs= p;          (intptr_t)sendbuf= =3D=3D(intptr_t)&MPIR_F08_MPI_IN_PLACE ? "yes" : "no&quo= t;);

&nbs= p;   printf("recvbuf=3D%p, recvbuf=3D%zu\n", recvbuf, (= intptr_t)recvbuf);

&nbs= p;   printf("*count=3D%d, *datatype=3D%d, *op=3D%d, *comm=3D= %d\n",

&nbs= p;           *count, *dat= atype, *op, *comm);

&nbs= p;   *ierror =3D 911;

}

 

=3D= =3D> extern.F90 <=3D=3D

modu= le mpi

&nbs= p;   use iso_c_binding

&nbs= p;   !type(c_ptr), bind(C,name=3D"MPI_F_IN_PLACE") :: M= PI_IN_PLACE

&nbs= p;   integer(c_int), bind(C, name=3D"MPIR_F08_MPI_IN_PLACE&q= uot;), target :: MPI_IN_PLACE

&nbs= p;   interface

&nbs= p;       subroutine p() bind(C,name=3D"p= ")

&nbs= p;       end subroutine

&nbs= p;   end interface

&nbs= p;   interface

&nbs= p;       SUBROUTINE MPI_ALLREDUCE(SENDBUF, RE= CVBUF, COUNT, DATATYPE, OP, COMM, IERROR) &

&nbs= p;            &= nbsp;     bind(C,name=3D"MPI_Allreduce")=

&nbs= p;           use iso_c_bi= nding

&nbs= p;           import :: MP= I_IN_PLACE

&nbs= p;           !DEC$ ATTRIB= UTES NO_ARG_CHECK :: sendbuf,recvbuf

&nbs= p;           !GCC$ ATTRIB= UTES NO_ARG_CHECK :: sendbuf,recvbuf

&nbs= p;           !$PRAGMA IGN= ORE_TKR sendbuf,recvbuf

&nbs= p;           !DIR$ IGNORE= _TKR sendbuf,recvbuf

&nbs= p;           !IBM* IGNORE= _TKR sendbuf,recvbuf

&nbs= p;           INTEGER(kind= =3Dc_int) :: SENDBUF(*), RECVBUF(*)

&nbs= p;           INTEGER(kind= =3Dc_int) :: COUNT, DATATYPE, OP, COMM, IERROR

&nbs= p;       END SUBROUTINE MPI_ALLREDUCE

&nbs= p;   end interface

end = module mpi

 

prog= ram main

&nbs= p;   use mpi

&nbs= p;   implicit none

&nbs= p;   real :: buf(100)

&nbs= p;   integer :: ierror

&nbs= p;   call p

&nbs= p;   buf =3D 17

&nbs= p;   print*,'LOC(MPI_IN_PLACE)=3D',LOC(MPI_IN_PLACE)

&nbs= p;   print*,'LOC(buf)=3D',LOC(buf)

&nbs= p;   call MPI_ALLREDUCE(MPI_IN_PLACE,buf,1,2,3,4,ierror)

&nbs= p;   print*,ierror

end = program main

--_000_DM6PR12MB3130DA4A03A92CC08EF2F8F5CB282DM6PR12MB3130namp_--