From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 67838 invoked by alias); 11 Feb 2016 19:33:34 -0000 Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org Received: (qmail 67823 invoked by uid 89); 11 Feb 2016 19:33:33 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-0.9 required=5.0 tests=BAYES_00,KAM_LAZY_DOMAIN_SECURITY autolearn=no version=3.3.2 spammy= X-HELO: usplmg20.ericsson.net Received: from usplmg20.ericsson.net (HELO usplmg20.ericsson.net) (198.24.6.45) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES256-SHA encrypted) ESMTPS; Thu, 11 Feb 2016 19:33:32 +0000 Received: from EUSAAHC004.ericsson.se (Unknown_Domain [147.117.188.84]) by usplmg20.ericsson.net (Symantec Mail Security) with SMTP id F5.AA.12433.D7EDCB65; Thu, 11 Feb 2016 20:18:21 +0100 (CET) Received: from [142.133.110.144] (147.117.188.8) by smtp-am.internal.ericsson.com (147.117.188.86) with Microsoft SMTP Server id 14.3.248.2; Thu, 11 Feb 2016 14:33:30 -0500 Subject: Re: [PATCH] Support 'make check-parallel' in gdb's build dir To: Pedro Alves , References: <1455207502-11058-1-git-send-email-palves@redhat.com> <56BCBA68.6000906@ericsson.com> <56BCBFE3.3080904@redhat.com> From: Simon Marchi Message-ID: <56BCE209.6010901@ericsson.com> Date: Thu, 11 Feb 2016 19:33:00 -0000 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.5.1 MIME-Version: 1.0 In-Reply-To: <56BCBFE3.3080904@redhat.com> Content-Type: text/plain; charset="windows-1252" Content-Transfer-Encoding: 7bit X-IsSubscribed: yes X-SW-Source: 2016-02/txt/msg00387.txt.bz2 On 16-02-11 12:07 PM, Pedro Alves wrote: >> I understand why you couldn't run parallel tests against a single, bare-metal >> target, but I don't understand why having a remote host would limit that >> (assuming you can connect multiple times simultaneously to that remote host). > > I think that it's because with remote host testing, unless you assume > the build and host machines share the same file system, sources/binaries > of all tests end up downloaded to/produced in the same remote host dir, and > then thus tests clash if run in parallel. Ahh it makes sense, the same thing happens with a remote target (the same procedures in Dejagnu are used to upload the files). So there could also be some clash when testing with a remote target, if there are two tests with the same name, or tests that use the same external files/libraries. That's why I found that odd, why a remote host in particular is dangerous, and not a remote target. Otherwise, the README LGTM.