From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from tretyak2.mcst.ru (tretyak2.mcst.ru [212.5.119.215]) by sourceware.org (Postfix) with ESMTP id 80014394C01A for ; Thu, 29 Apr 2021 15:58:02 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.3.2 sourceware.org 80014394C01A Authentication-Results: sourceware.org; dmarc=none (p=none dis=none) header.from=mcst.ru Authentication-Results: sourceware.org; spf=none smtp.mailfrom=malakhov@mcst.ru Received: from tretyak2.mcst.ru (localhost [127.0.0.1]) by tretyak2.mcst.ru (Postfix) with ESMTP id 3635D20B95 for ; Thu, 29 Apr 2021 18:57:59 +0300 (MSK) Received: from frog.lab.sun.mcst.ru (frog.lab.sun.mcst.ru [172.16.4.50]) by tretyak2.mcst.ru (Postfix) with ESMTP id 10BA720B36 for ; Thu, 29 Apr 2021 18:57:48 +0300 (MSK) Received: from gudgeon (gudgeon [172.16.5.17]) by frog.lab.sun.mcst.ru (8.13.4/8.12.11) with ESMTP id 13TFvkKD012115 for ; Thu, 29 Apr 2021 18:57:46 +0300 From: malakhov@mcst.ru (Ilya Yu. Malakhov) To: libffi-discuss@sourceware.org Subject: Non-matching `dg-output's in testsuite/libffi.complex/cls_align_complex.inc have no effect Date: Thu, 29 Apr 2021 18:57:46 +0300 Message-ID: MIME-Version: 1.0 Content-Type: text/plain X-Anti-Virus: Kaspersky Anti-Virus for Linux Mail Server 5.6.39/RELEASE, bases: 20111107 #2745587, check: 20210429 notchecked X-AV-Checked: ClamAV using ClamSMTP X-Spam-Status: No, score=-12.9 required=5.0 tests=BAYES_00, GIT_PATCH_0, KAM_DMARC_STATUS, KAM_LAZY_DOMAIN_SECURITY, SPF_HELO_NONE, SPF_NONE autolearn=ham autolearn_force=no version=3.4.2 X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on server2.sourceware.org X-BeenThere: libffi-discuss@sourceware.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Libffi-discuss mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 29 Apr 2021 15:58:05 -0000 Because the default precision of `printf ("%f")'s in testsuite/libffi.complex/cls_align_complex.inc is 6, it would be reasonable to fix the related `dg-output's as proposed below if they affected the results of libffi.complex/cls_align_complex_{{,long}double,float}.c tests including this header at all. The reason for why they have no effect is that DejaGnu's `proc dg-get-options' responsible for picking out the dg-xxx options in a testcase just greps unpreprocessed .c files. Therefore, no output pattern checks are actually done in these tests. The latter issue also applies to matching `dg-output's in testsuite/libffi.complex/cls_complex_struct.inc. diff --git a/testsuite/libffi.complex/cls_align_complex.inc b/testsuite/libffi.complex/cls_align_complex.inc index 4a812ed..23d87bf 100644 --- a/testsuite/libffi.complex/cls_align_complex.inc +++ b/testsuite/libffi.complex/cls_align_complex.inc @@ -74,18 +74,18 @@ int main (void) args_c[2] = NULL; ffi_call(&cif, FFI_FN(cls_struct_align_fn), &res_c, args_c); - /* { dg-output "12 4951,7i 127 1 9320,1i 13: 13 14271,8i 140" } */ + /* { dg-output "12 4951.000000,7.000000i 127 1 9320.000000,1.000000i 13: 13 14271.000000,8.000000i 140" } */ printf("res: %d %f,%fi %d\n", res_c.a, T_CONV creal (res_c.b), T_CONV cimag (res_c.b), res_c.c); - /* { dg-output "\nres: 13 14271,8i 140" } */ + /* { dg-output "\nres: 13 14271.000000,8.000000i 140" } */ CHECK(ffi_prep_closure_loc(pcl, &cif, cls_struct_align_gn, NULL, code) == FFI_OK); res_c = ((cls_struct_align(*)(cls_struct_align, cls_struct_align))(code))(g_c, f_c); - /* { dg-output "\n12 4951,7i 127 1 9320,1i 13: 13 14271,8i 140" } */ + /* { dg-output "\n12 4951.000000,7.000000i 127 1 9320.000000,1.000000i 13: 13 14271.000000,8.000000i 140" } */ printf("res: %d %f,%fi %d\n", res_c.a, T_CONV creal (res_c.b), T_CONV cimag (res_c.b), res_c.c); - /* { dg-output "\nres: 13 14271,8i 140" } */ + /* { dg-output "\nres: 13 14271.000000,8.000000i 140" } */ exit(0); }