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 09A656EF9F for ; Fri, 21 Feb 2014 04:01:07 +0000 (UTC) Received: from orsmga002.jf.intel.com ([10.7.209.21]) by orsmga102.jf.intel.com with ESMTP; 20 Feb 2014 19:56:48 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.97,516,1389772800"; d="scan'208";a="486977003" Received: from unknown (HELO [10.255.14.1]) ([10.255.14.1]) by orsmga002.jf.intel.com with ESMTP; 20 Feb 2014 20:01:05 -0800 Message-ID: <5306CF81.4040203@linux.intel.com> Date: Thu, 20 Feb 2014 20:01:05 -0800 From: Saul Wold User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.2.0 MIME-Version: 1.0 To: Jason Plum , "openembedded-core@lists.openembedded.org" References: <1390245778-9252-1-git-send-email-jplum@devonit.com> <52E7FB3B.8000000@linux.intel.com> <52E91639.6010606@devonit.com> <52F878DE.6000609@linux.intel.com> In-Reply-To: <52F878DE.6000609@linux.intel.com> Subject: Re: [PATCH v3] meta/oe/lib/path.py - check mount points, not device 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, 21 Feb 2014 04:01:11 -0000 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit On 02/09/2014 10:59 PM, Saul Wold wrote: > On 01/29/2014 06:54 AM, Jason Plum wrote: >> >> On 01/28/14 13:47, Saul Wold wrote: >>> On 01/20/2014 11:22 AM, Jason Plum wrote: >>>> Signed-off-by: Jason Plum >>>> --- >>>> meta/lib/oe/path.py | 13 ++++++++++++- >>>> 1 file changed, 12 insertions(+), 1 deletion(-) >>>> >>> This patch is still not applying cleanly to master, I am also still a >>> little concerned if this will affect performance since it's used often >>> in sstate code. >>> >>> Have you compared build times with and without this patch? >>> >>> Sau! >> Saul, >> >> I pulled a fresh copy of master when I made this particular send-email >> patch, so it must keep changing from under me. I have the last commit to >> master on this checkout as 9c8d9781794ed0886a79c8ce4544ba98be0ff858 >> >> The performance difference should be quite minimal, but I have not >> extensively tested with timing metrics in place. > > Can you describe how much "quite minimal" really is, Richard is still > concerned with the impact this might has (see his comment to my > Consolidated Pull request). > Jason, As mentioned before, there is some concern about this patch, if you could run before and after performance tests with the profile (-P) option and review the results, that would help us feel better about this patch. Also an expanded commit messages explaining more about why this patch is needed would help. Thanks Sau! > Sau! > >>>> diff --git a/meta/lib/oe/path.py b/meta/lib/oe/path.py >>>> index 46783f8..dab20fc 100644 >>>> --- a/meta/lib/oe/path.py >>>> +++ b/meta/lib/oe/path.py >>>> @@ -90,7 +90,7 @@ def copyhardlinktree(src, dst): >>>> if os.path.isdir(src) and not len(os.listdir(src)): >>>> return >>>> >>>> - if (os.stat(src).st_dev == os.stat(dst).st_dev): >>>> + if (getmount(src) == getmount(dst)): >>>> # Need to copy directories only with tar first since cp >>>> will error if two >>>> # writers try and create a directory at the same time >>>> cmd = 'cd %s; find . -type d -print | tar -cf - -C %s -p >>>> --files-from - --no-recursion | tar -xf - -C %s' % (src, src, dst) >>>> @@ -257,3 +257,14 @@ def realpath(file, root, use_physdir = True, >>>> loop_cnt = 100, assume_dir = False) >>>> raise >>>> >>>> return file >>>> + >>>> +def getmount(path): >>>> + """ getmount takes a path parameter, uses os.path.realpath with >>>> os.path.abspath to >>>> + handle symlink resolution, and returns the path to the mount >>>> point of the input path >>>> + """ >>>> + path = os.path.realpath(os.path.abspath(path)) >>>> + while path != os.path.sep: >>>> + if os.path.ismount(path): >>>> + return path >>>> + path = os.path.abspath(os.path.join(path, os.pardir)) >>>> + return path >>>> \ No newline at end of file >>>> >> > _______________________________________________ > Openembedded-core mailing list > Openembedded-core@lists.openembedded.org > http://lists.openembedded.org/mailman/listinfo/openembedded-core > >