From patchwork Tue Sep 25 13:11:42 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Chung-Lin Tang X-Patchwork-Id: 974407 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=pass (mailfrom) smtp.mailfrom=gcc.gnu.org (client-ip=209.132.180.131; helo=sourceware.org; envelope-from=gcc-patches-return-486327-incoming=patchwork.ozlabs.org@gcc.gnu.org; receiver=) Authentication-Results: ozlabs.org; dmarc=none (p=none dis=none) header.from=mentor.com Authentication-Results: ozlabs.org; dkim=pass (1024-bit key; unprotected) header.d=gcc.gnu.org header.i=@gcc.gnu.org header.b="ECJqRXoU"; dkim-atps=neutral Received: from sourceware.org (server1.sourceware.org [209.132.180.131]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ozlabs.org (Postfix) with ESMTPS id 42KM0z2Dfmz9s3C for ; Tue, 25 Sep 2018 23:12:03 +1000 (AEST) DomainKey-Signature: a=rsa-sha1; c=nofws; d=gcc.gnu.org; h=list-id :list-unsubscribe:list-archive:list-post:list-help:sender :reply-to:from:subject:to:message-id:date:mime-version :content-type; q=dns; s=default; b=mOlYcOZKh/d6WtQzTMdI7lIA85eKB f//zAkySUb6Zq2OUnDWb3+Ul1KDNylL1IuT1HG/pJrGWeC7bZMlzumQHAYMkadeO emp+a2nu2qxKgpncabI5d0n6MZjBDBSD0k0RQRn96myF6yuUYYsaUybGM+gt1hDW LxAugw7JjeSxXg= DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=gcc.gnu.org; h=list-id :list-unsubscribe:list-archive:list-post:list-help:sender :reply-to:from:subject:to:message-id:date:mime-version :content-type; s=default; bh=SXQGVf504W5Cc2N8j4GpK0bA2LA=; b=ECJ qRXoUA6tATCnsg57LImBCarB8MjBpXMpWW9m05WG4fb88ddNdrIgfVBfhEGk4Ed4 xs1wZdP3ektb3DmTh1/es6Qk+IVjQMQ/95ybHCoXeP8xLNLjqegixQANFoCGS4IP SW715YJ71ZT4y6g4alyYvxe7VAaWU/05QLnF5erY= Received: (qmail 124256 invoked by alias); 25 Sep 2018 13:11:55 -0000 Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Unsubscribe: List-Archive: List-Post: List-Help: Sender: gcc-patches-owner@gcc.gnu.org Delivered-To: mailing list gcc-patches@gcc.gnu.org Received: (qmail 124186 invoked by uid 89); 25 Sep 2018 13:11:54 -0000 Authentication-Results: sourceware.org; auth=none X-Spam-SWARE-Status: No, score=-24.9 required=5.0 tests=AWL, BAYES_00, GIT_PATCH_0, GIT_PATCH_1, GIT_PATCH_2, GIT_PATCH_3, RCVD_IN_DNSWL_NONE, SPF_PASS autolearn=ham version=3.3.2 spammy=2.0 X-HELO: relay1.mentorg.com Received: from relay1.mentorg.com (HELO relay1.mentorg.com) (192.94.38.131) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with ESMTP; Tue, 25 Sep 2018 13:11:50 +0000 Received: from svr-orw-mbx-02.mgc.mentorg.com ([147.34.90.202]) by relay1.mentorg.com with esmtps (TLSv1.2:ECDHE-RSA-AES256-SHA384:256) id 1g4n80-0004LV-IT from ChungLin_Tang@mentor.com for gcc-patches@gcc.gnu.org; Tue, 25 Sep 2018 06:11:48 -0700 Received: from [0.0.0.0] (147.34.91.1) by svr-orw-mbx-02.mgc.mentorg.com (147.34.90.202) with Microsoft SMTP Server (TLS) id 15.0.1320.4; Tue, 25 Sep 2018 06:11:45 -0700 Reply-To: From: Chung-Lin Tang Subject: [PATCH 5/6, OpenACC, libgomp] Async re-work, C/C++ testsuite changes To: , Thomas Schwinge Message-ID: <8086c63b-f729-891b-3d21-76871d360734@mentor.com> Date: Tue, 25 Sep 2018 21:11:42 +0800 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.13; rv:52.0) Gecko/20100101 Thunderbird/52.9.1 MIME-Version: 1.0 These are the testsuite/libgomp.oacc-c-c++-common/* changes. Thanks, Chung-Lin * testsuite/libgomp.oacc-c-c++-common/asyncwait-2.c: New testcase. * testsuite/libgomp.oacc-c-c++-common/data-2-lib.c: Adjust testcase. * testsuite/libgomp.oacc-c-c++-common/data-2.c: Likewise. * testsuite/libgomp.oacc-c-c++-common/data-3.c: Likewise. * testsuite/libgomp.oacc-c-c++-common/lib-71.c: Likewise. * testsuite/libgomp.oacc-c-c++-common/lib-77.c: Likewise. * testsuite/libgomp.oacc-c-c++-common/lib-79.c: Likewise. * testsuite/libgomp.oacc-c-c++-common/lib-81.c: Likewise. diff --git a/libgomp/testsuite/libgomp.oacc-c-c++-common/asyncwait-2.c b/libgomp/testsuite/libgomp.oacc-c-c++-common/asyncwait-2.c new file mode 100644 index 0000000..9420540 --- /dev/null +++ b/libgomp/testsuite/libgomp.oacc-c-c++-common/asyncwait-2.c @@ -0,0 +1,904 @@ +/* { dg-do run { target openacc_nvidia_accel_selected } } */ +/* { dg-additional-options "-lcuda" } */ + +#include +#include +#include + +#include +#include +#include + +int +main (int argc, char **argv) +{ + CUresult r; + CUstream stream1; + int N = 128; //1024 * 1024; + float *a, *b, *c, *d, *e; + int i; + int nbytes; + + srand (time (NULL)); + int s = rand () % 100; + + acc_init (acc_device_nvidia); + + nbytes = N * sizeof (float); + + a = (float *) malloc (nbytes); + b = (float *) malloc (nbytes); + c = (float *) malloc (nbytes); + d = (float *) malloc (nbytes); + e = (float *) malloc (nbytes); + + for (i = 0; i < N; i++) + { + a[i] = 3.0; + b[i] = 0.0; + } + + acc_set_default_async (s); + +#pragma acc data copy (a[0:N]) copy (b[0:N]) copyin (N) + { + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = a[ii]; + } + +#pragma acc wait + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 3.0) + abort (); + + if (b[i] != 3.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 2.0; + b[i] = 0.0; + } + +#pragma acc data copy (a[0:N]) copy (b[0:N]) copyin (N) + { + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = a[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 2.0) + abort (); + + if (b[i] != 2.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 3.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + } + +#pragma acc data copy (a[0:N]) copy (b[0:N]) copy (c[0:N]) copy (d[0:N]) copyin (N) + { + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + d[ii] = ((a[ii] * a[ii] + a[ii]) / a[ii]) - a[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 3.0) + abort (); + + if (b[i] != 9.0) + abort (); + + if (c[i] != 4.0) + abort (); + + if (d[i] != 1.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 2.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + e[i] = 0.0; + } + +#pragma acc data copy (a[0:N], b[0:N], c[0:N], d[0:N], e[0:N]) copyin (N) + { + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + d[ii] = ((a[ii] * a[ii] + a[ii]) / a[ii]) - a[ii]; + } + +#pragma acc parallel wait (s) async (s) + { + int ii; + + for (ii = 0; ii < N; ii++) + e[ii] = a[ii] + b[ii] + c[ii] + d[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 2.0) + abort (); + + if (b[i] != 4.0) + abort (); + + if (c[i] != 4.0) + abort (); + + if (d[i] != 1.0) + abort (); + + if (e[i] != 11.0) + abort (); + } + + + r = cuStreamCreate (&stream1, CU_STREAM_NON_BLOCKING); + if (r != CUDA_SUCCESS) + { + fprintf (stderr, "cuStreamCreate failed: %d\n", r); + abort (); + } + + acc_set_cuda_stream (1, stream1); + + for (i = 0; i < N; i++) + { + a[i] = 5.0; + b[i] = 0.0; + } + +#pragma acc data copy (a[0:N], b[0:N]) copyin (N) + { + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = a[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 5.0) + abort (); + + if (b[i] != 5.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 7.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + } + +#pragma acc data copy (a[0:N]) copy (b[0:N]) copy (c[0:N]) copy (d[0:N]) copyin (N) + { + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + d[ii] = ((a[ii] * a[ii] + a[ii]) / a[ii]) - a[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 7.0) + abort (); + + if (b[i] != 49.0) + abort (); + + if (c[i] != 4.0) + abort (); + + if (d[i] != 1.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 3.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + e[i] = 0.0; + } + +#pragma acc data copy (a[0:N], b[0:N], c[0:N], d[0:N], e[0:N]) copyin (N) + { + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + d[ii] = ((a[ii] * a[ii] + a[ii]) / a[ii]) - a[ii]; + } + +#pragma acc parallel wait (s) async (s) + { + int ii; + + for (ii = 0; ii < N; ii++) + e[ii] = a[ii] + b[ii] + c[ii] + d[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 3.0) + abort (); + + if (b[i] != 9.0) + abort (); + + if (c[i] != 4.0) + abort (); + + if (d[i] != 1.0) + abort (); + + if (e[i] != 17.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 4.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + e[i] = 0.0; + } + +#pragma acc data copyin (a[0:N], b[0:N], c[0:N]) copyin (N) + { + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + +#pragma acc update host (a[0:N], b[0:N], c[0:N]) wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 4.0) + abort (); + + if (b[i] != 16.0) + abort (); + + if (c[i] != 4.0) + abort (); + } + + + for (i = 0; i < N; i++) + { + a[i] = 5.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + e[i] = 0.0; + } + +#pragma acc data copyin (a[0:N], b[0:N], c[0:N]) copyin (N) + { + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc parallel async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + +#pragma acc update host (a[0:N], b[0:N], c[0:N]) async + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 5.0) + abort (); + + if (b[i] != 25.0) + abort (); + + if (c[i] != 4.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 3.0; + b[i] = 0.0; + } + +#pragma acc data copy (a[0:N]) copy (b[0:N]) copyin (N) + { + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = a[ii]; + } + +#pragma acc wait + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 3.0) + abort (); + + if (b[i] != 3.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 2.0; + b[i] = 0.0; + } + +#pragma acc data copy (a[0:N]) copy (b[0:N]) copyin (N) + { + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = a[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 2.0) + abort (); + + if (b[i] != 2.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 3.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + } + +#pragma acc data copy (a[0:N]) copy (b[0:N]) copy (c[0:N]) copy (d[0:N]) copyin (N) + { + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + d[ii] = ((a[ii] * a[ii] + a[ii]) / a[ii]) - a[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 3.0) + abort (); + + if (b[i] != 9.0) + abort (); + + if (c[i] != 4.0) + abort (); + + if (d[i] != 1.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 2.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + e[i] = 0.0; + } + +#pragma acc data copy (a[0:N], b[0:N], c[0:N], d[0:N], e[0:N]) copyin (N) + { + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + d[ii] = ((a[ii] * a[ii] + a[ii]) / a[ii]) - a[ii]; + } + +#pragma acc kernels wait (s) async (s) + { + int ii; + + for (ii = 0; ii < N; ii++) + e[ii] = a[ii] + b[ii] + c[ii] + d[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 2.0) + abort (); + + if (b[i] != 4.0) + abort (); + + if (c[i] != 4.0) + abort (); + + if (d[i] != 1.0) + abort (); + + if (e[i] != 11.0) + abort (); + } + + + r = cuStreamCreate (&stream1, CU_STREAM_NON_BLOCKING); + if (r != CUDA_SUCCESS) + { + fprintf (stderr, "cuStreamCreate failed: %d\n", r); + abort (); + } + + acc_set_cuda_stream (1, stream1); + + for (i = 0; i < N; i++) + { + a[i] = 5.0; + b[i] = 0.0; + } + +#pragma acc data copy (a[0:N], b[0:N]) copyin (N) + { + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = a[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 5.0) + abort (); + + if (b[i] != 5.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 7.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + } + +#pragma acc data copy (a[0:N]) copy (b[0:N]) copy (c[0:N]) copy (d[0:N]) copyin (N) + { + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + d[ii] = ((a[ii] * a[ii] + a[ii]) / a[ii]) - a[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 7.0) + abort (); + + if (b[i] != 49.0) + abort (); + + if (c[i] != 4.0) + abort (); + + if (d[i] != 1.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 3.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + e[i] = 0.0; + } + +#pragma acc data copy (a[0:N], b[0:N], c[0:N], d[0:N], e[0:N]) copyin (N) + { + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + d[ii] = ((a[ii] * a[ii] + a[ii]) / a[ii]) - a[ii]; + } + +#pragma acc kernels wait (s) async (s) + { + int ii; + + for (ii = 0; ii < N; ii++) + e[ii] = a[ii] + b[ii] + c[ii] + d[ii]; + } + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 3.0) + abort (); + + if (b[i] != 9.0) + abort (); + + if (c[i] != 4.0) + abort (); + + if (d[i] != 1.0) + abort (); + + if (e[i] != 17.0) + abort (); + } + + for (i = 0; i < N; i++) + { + a[i] = 4.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + e[i] = 0.0; + } + +#pragma acc data copyin (a[0:N], b[0:N], c[0:N]) copyin (N) + { + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + +#pragma acc update host (a[0:N], b[0:N], c[0:N]) wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 4.0) + abort (); + + if (b[i] != 16.0) + abort (); + + if (c[i] != 4.0) + abort (); + } + + + for (i = 0; i < N; i++) + { + a[i] = 5.0; + b[i] = 0.0; + c[i] = 0.0; + d[i] = 0.0; + e[i] = 0.0; + } + +#pragma acc data copyin (a[0:N], b[0:N], c[0:N]) copyin (N) + { + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + b[ii] = (a[ii] * a[ii] * a[ii]) / a[ii]; + } + +#pragma acc kernels async + { + int ii; + + for (ii = 0; ii < N; ii++) + c[ii] = (a[ii] + a[ii] + a[ii] + a[ii]) / a[ii]; + } + +#pragma acc update host (a[0:N], b[0:N], c[0:N]) async + +#pragma acc wait (s) + + } + + for (i = 0; i < N; i++) + { + if (a[i] != 5.0) + abort (); + + if (b[i] != 25.0) + abort (); + + if (c[i] != 4.0) + abort (); + } + + acc_shutdown (acc_device_nvidia); + + return 0; +} diff --git a/libgomp/testsuite/libgomp.oacc-c-c++-common/data-2-lib.c b/libgomp/testsuite/libgomp.oacc-c-c++-common/data-2-lib.c index 2ddfa7d..f553d3d 100644 --- a/libgomp/testsuite/libgomp.oacc-c-c++-common/data-2-lib.c +++ b/libgomp/testsuite/libgomp.oacc-c-c++-common/data-2-lib.c @@ -153,7 +153,7 @@ main (int argc, char **argv) d[ii] = ((a[ii] * a[ii] + a[ii]) / a[ii]) - a[ii]; #pragma acc parallel present (a[0:N], b[0:N], c[0:N], d[0:N], e[0:N], N) \ - async (4) + wait (1, 2, 3) async (4) for (int ii = 0; ii < N; ii++) e[ii] = a[ii] + b[ii] + c[ii] + d[ii]; diff --git a/libgomp/testsuite/libgomp.oacc-c-c++-common/data-2.c b/libgomp/testsuite/libgomp.oacc-c-c++-common/data-2.c index 0c6abe6..81d623a 100644 --- a/libgomp/testsuite/libgomp.oacc-c-c++-common/data-2.c +++ b/libgomp/testsuite/libgomp.oacc-c-c++-common/data-2.c @@ -162,7 +162,7 @@ main (int argc, char **argv) d[ii] = ((a[ii] * a[ii] + a[ii]) / a[ii]) - a[ii]; #pragma acc parallel present (a[0:N], b[0:N], c[0:N], d[0:N], e[0:N]) \ - wait (1) async (4) + wait (1, 2, 3) async (4) for (int ii = 0; ii < N; ii++) e[ii] = a[ii] + b[ii] + c[ii] + d[ii]; diff --git a/libgomp/testsuite/libgomp.oacc-c-c++-common/data-3.c b/libgomp/testsuite/libgomp.oacc-c-c++-common/data-3.c index 0bf706a..5ec50b8 100644 --- a/libgomp/testsuite/libgomp.oacc-c-c++-common/data-3.c +++ b/libgomp/testsuite/libgomp.oacc-c-c++-common/data-3.c @@ -138,7 +138,7 @@ main (int argc, char **argv) d[ii] = ((a[ii] * a[ii] + a[ii]) / a[ii]) - a[ii]; #pragma acc parallel present (a[0:N], b[0:N], c[0:N], d[0:N], e[0:N]) \ - wait (1,5) async (4) + wait (1, 2, 3, 5) async (4) for (int ii = 0; ii < N; ii++) e[ii] = a[ii] + b[ii] + c[ii] + d[ii]; diff --git a/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-71.c b/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-71.c index c85e824..6afe2a0 100644 --- a/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-71.c +++ b/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-71.c @@ -92,16 +92,22 @@ main (int argc, char **argv) abort (); } - fprintf (stderr, "CheCKpOInT\n"); - if (acc_async_test (1) != 0) + if (acc_async_test (0) != 0) { fprintf (stderr, "asynchronous operation not running\n"); abort (); } + /* Test unseen async number. */ + if (acc_async_test (1) != 1) + { + fprintf (stderr, "acc_async_test failed on unseen number\n"); + abort (); + } + sleep ((int) (dtime / 1000.0f) + 1); - if (acc_async_test (1) != 1) + if (acc_async_test (0) != 1) { fprintf (stderr, "found asynchronous operation still running\n"); abort (); @@ -116,7 +122,3 @@ main (int argc, char **argv) return 0; } - -/* { dg-output "CheCKpOInT(\n|\r\n|\r).*" } */ -/* { dg-output "unknown async \[0-9\]+" } */ -/* { dg-shouldfail "" } */ diff --git a/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-77.c b/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-77.c index f4f196d..2821f88 100644 --- a/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-77.c +++ b/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-77.c @@ -111,7 +111,7 @@ main (int argc, char **argv) start_timer (0); - acc_wait (1); + acc_wait (0); atime = stop_timer (0); @@ -132,7 +132,3 @@ main (int argc, char **argv) return 0; } - -/* { dg-output "CheCKpOInT(\n|\r\n|\r).*" } */ -/* { dg-output "unknown async \[0-9\]+" } */ -/* { dg-shouldfail "" } */ diff --git a/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-79.c b/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-79.c index ef3df13..b22af26 100644 --- a/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-79.c +++ b/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-79.c @@ -114,6 +114,7 @@ main (int argc, char **argv) for (i = 0; i < N; i++) { + stream = (CUstream) acc_get_cuda_stream (i & 1); r = cuLaunchKernel (delay, 1, 1, 1, 1, 1, 1, 0, stream, kargs, 0); if (r != CUDA_SUCCESS) { @@ -122,11 +123,11 @@ main (int argc, char **argv) } } - acc_wait_async (0, 1); - if (acc_async_test (0) != 0) abort (); + acc_wait_async (0, 1); + if (acc_async_test (1) != 0) abort (); diff --git a/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-81.c b/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-81.c index d5f18f0..30a4b57 100644 --- a/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-81.c +++ b/libgomp/testsuite/libgomp.oacc-c-c++-common/lib-81.c @@ -133,7 +133,7 @@ main (int argc, char **argv) for (i = 0; i <= N; i++) { - if (acc_async_test (i) != 0) + if (acc_async_test (i) == 0) abort (); }