From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by mail.openembedded.org (Postfix) with ESMTP id 23CD57825F for ; Fri, 9 Jun 2017 09:13:13 +0000 (UTC) Received: from orsmga001.jf.intel.com ([10.7.209.18]) by orsmga102.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 09 Jun 2017 02:13:15 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.39,317,1493708400"; d="scan'208";a="1139569990" Received: from kanavin-desktop.fi.intel.com (HELO [10.237.68.161]) ([10.237.68.161]) by orsmga001.jf.intel.com with ESMTP; 09 Jun 2017 02:13:14 -0700 To: Leonardo Sandoval References: <1496956519.26945.171.camel@linux.intel.com> From: Alexander Kanavin Message-ID: Date: Fri, 9 Jun 2017 12:13:15 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.8.0 MIME-Version: 1.0 In-Reply-To: <1496956519.26945.171.camel@linux.intel.com> Cc: openembedded-core@lists.openembedded.org Subject: Re: [PATCHv2 0/2] Multi-threaded RPM support X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Patches and discussions about the oe-core layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 09 Jun 2017 09:13:14 -0000 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit On 06/09/2017 12:15 AM, Leonardo Sandoval wrote: > On Thu, 2017-06-08 at 17:42 +0300, Alexander Kanavin wrote: >> Changes in version 2: use openmp for multi-processing instead of NSPR library, as >> requested by upstream. Add a link to the upstream submission to patches. >> >> These two patches add multi-threaded features to RPM to speed up the do_package_write_rpm task. >> >> Specifically: >> >> 1) Multi-threaded xz compressor is used instead of default single-threaded gz. This has the most >> dramatic effect when a recipe produces a smaller number of large-sized packages. >> >> 2) Packages creation is run in multiple threads via thread pools. This is most beneficial when >> a recipe produces a large amount of small packages. >> >> Some not very scientific benchmarks for time and .rpm sizes >> (time is measured for do_package_write_rpm tasks only): >> > > Tried v2 again and I am getting this problem (segmentation faults): > > http://errors.yoctoproject.org/Errors/Build/38812/ > > > The idea I have is to get some numbers from buildstats and have better > insight of the impact of this series. BTW, I tried core-image-sato. Now I see the issue locally, even though it does not crash, but I think it's the same issue :) Bear with me, it's my first openmp adventure ever. Alex