From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qt1-f196.google.com (mail-qt1-f196.google.com [209.85.160.196]) by mail.openembedded.org (Postfix) with ESMTP id E7E866BC7D for ; Wed, 2 Jan 2019 18:18:24 +0000 (UTC) Received: by mail-qt1-f196.google.com with SMTP id y20so34345316qtm.13 for ; Wed, 02 Jan 2019 10:18:26 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=YXVEMin20VBu6/1NM4N8782XZ3827P9DICM7TL5AWkk=; b=Ysx+85dVq9cvAuNuqz5OSKZahUWCMmBONpSOpNKPhR2sMKR1fA2bzxMDaQ0I+fHwma n3iWuXSV71lLxLLMrlfPhVm6FDzRZFybNrcn2KFhS6yWf8d7L1mCdAf0p3OOrd7qpal+ wGQRy+0OygFimkLLx0bISrolWqLDUyXahSPb5obAi5L/DzH+J9bCklly7vt+zVKW0Lev DJ/CcrsVoXXwy5thjDtcRhqXCN20QmUq5KD5BqK9sOYaHPfqxutdj3yI3yMgdaoYdW0B WCgkSBVq4+/BxgcERBBVWasl0iVkuT+xvmS7KdCaqXpcnteh62Hn+xsdv0s8fLH+TUbp 2JnA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=YXVEMin20VBu6/1NM4N8782XZ3827P9DICM7TL5AWkk=; b=RKrN7umq5xOiRKtqFXbiLHIvq5kfetjrDXddKrRAIsEgC1hIiD8JOvylD1Ap+92WrM 5v1v32FeV/Gfnk3eVEIdqIKKqzbsHCDDv76ySpKtd2bUHdAoyb3Mh88KCq01aL7lB/J+ LuMZtR+xpkbFTDMh8CyrEDtkjrbN5rv8bfdPhkvlprd9cRMtaq3W1MA3EVF3k/T6BQ/V u31Vy/oposHimxz3haekSPBRO7y+NsK8I8huNl//J+fp2qI/x6UJu1wJEECPvtc056A9 UR19FPba38hiQHz3kXCJLUfej+G+XQGS1nFL+rh7/3417WSVIHwJj2AMzoFryIx0BlF9 88Jw== X-Gm-Message-State: AA+aEWbK39yk2p4HJelOQufo40z+BHD9xIJWM9GiAmq8QZURq2/c/6zI WingKuMwSTLOToRDaWrbef8bimXFirpWr+Ca+Do= X-Google-Smtp-Source: AFSGD/WU/AKmbOfPXBVICpUj4bgKuhNiD8C7g3fMwCApN2l6GZleW0JNnMvW43DPCznqtUC2g3yDrLVxMQSJ2p0G0kc= X-Received: by 2002:ac8:518d:: with SMTP id c13mr42140697qtn.254.1546453105540; Wed, 02 Jan 2019 10:18:25 -0800 (PST) MIME-Version: 1.0 References: <20181230165033.8850-1-raj.khem@gmail.com> <07876792-0a06-c91c-fb4d-4efb033ef5c5@windriver.com> In-Reply-To: <07876792-0a06-c91c-fb4d-4efb033ef5c5@windriver.com> From: Khem Raj Date: Wed, 2 Jan 2019 10:17:59 -0800 Message-ID: To: Randy MacLeod Cc: openembeded-devel Subject: Re: [meta-oe][PATCH 1/3] guider: add a new runtime profiler package X-BeenThere: openembedded-devel@lists.openembedded.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Using the OpenEmbedded metadata to build Distributions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 02 Jan 2019 18:18:25 -0000 Content-Type: text/plain; charset="UTF-8" On Wed, Jan 2, 2019 at 8:59 AM Randy MacLeod wrote: > > On 1/2/19 9:00 AM, Peace Lee wrote: > > Thanks for your advice :) > > > > Should I submit a new patch again after changing a below patch? > > I am not familiar with this process ;) > > I expect that Khem will do that once he wakes up. > > > It couldn't be reached to openembedded-devel members When I sent patch > > first so that I sent github pull-request to Khem Raj. > > We send patches to the oe-devel list, people comment, the gatekeeper > merges at some point. The process is documented here: > > https://www.openembedded.org/wiki/How_to_submit_a_patch_to_OpenEmbedded > Yes please prepare a v2 and its ok if you want to update your current pull on GH I can bring it into ml. > > > > Anyway, I agree with your comments (fixing typos, moving the recipe to > > meta-tool, dropping last some lines). > > But what does "_Shouldn't the packages that guider uses be recommended > > at least?_" mean? > > I thought that you were forking top, perf, etc and then screen-scrapping > the output but it seems that you have reimplemented(parts of) > the functionality of these tools so we don't need to add a > DEPENDS/RECOMMENDS to the recipe to ensure that top for example > is added to the system image. > > > Guider only requires "python-ctypes, python-shell, python-json, > > python-subprocess" package :) > > Khem's recipe has that covered then: > > +RDEPENDS_${PN} = "python-ctypes python-shell \ > + python-json python-subprocess" > > > Some kernel configurations and patches are needed to trace user-level > > functions but it depends on users. > > The yocto-kenrel-cache: > > http://git.yoctoproject.org/cgit/cgit.cgi/yocto-kernel-cache/tree/00-README > > has the concept of features: > > http://git.yoctoproject.org/cgit/cgit.cgi/yocto-kernel-cache/tree/features > such as: > > http://git.yoctoproject.org/cgit/cgit.cgi/yocto-kernel-cache/tree/features/lttng > > since this recipe is for meta-openembedded rather than oe-core, > we'd leave it up to users as you say. meta-openembedded depends on > oe-core but the two projects are organized by different people. > > > > Do you have a document that explains your motivation for creating guider > and perhaps documents a few cases where it's better than the standard > tools? I likely just need to play with guider a bit more but it seems > like a use case that walked people though a performance issue that you > resolved more easily with guider than with the plethora of linux tools: > http://www.brendangregg.com/linuxperf.html > would be good marketing. > > > ../Randy > > > > > Happy new year! > > > > On Tue, Jan 1, 2019 at 4:30 AM Randy MacLeod > > > wrote: > > > > Neat tool! I'll have to play with it a bit more. > > > > On 12/30/18 11:50 AM, Khem Raj wrote: > > > From: Peace Lee > > > > > > > guider [1] is a system-wide performance analyzer for platforms based > > > > Well, it seems to be more of a meta-tool that uses standard tools. > > In addition, it improves usability at the cost of defining policy > > such as immediately analyzing some data collection on target. > > It would be nice to succinctly summarize that if you can. > > > on linux kernel. it can measure vary system resource usage and > > s/it/It/ > > s/vary/a variety of/ > > > trace thread/process activities. and it reports vary meaningful > > > output to analyze system status. > > It reports very meaningful output to analyze system status. > > > > > > [1] https://github.com/iipeace/guider > > > > > > Signed-off-by: Peace Lee > > > > > Signed-off-by: Khem Raj > > > > > --- > > > .../recipes-devtools/guider/guider_3.9.3.bb > > | 34 +++++++++++++++++++ > > > 1 file changed, 34 insertions(+) > > > create mode 100644 > > meta-oe/recipes-devtools/guider/guider_3.9.3.bb > > > > > > diff --git a/meta-oe/recipes-devtools/guider/guider_3.9.3.bb > > > > b/meta-oe/recipes-devtools/guider/guider_3.9.3.bb > > > > > new file mode 100644 > > > index 0000000000..96d6c62ae2 > > > --- /dev/null > > > +++ b/meta-oe/recipes-devtools/guider/guider_3.9.3.bb > > > > > @@ -0,0 +1,34 @@ > > > +SUMMARY = "runtime performance analyzer" > > > +HOMEPAGE = "https://github.com/iipeace/guider" > > > +BUGTRACKER = "https://github.com/iipeace/guider/issues" > > > +AUTHOR = "Peace Lee >" > > > + > > > +LICENSE = "GPLv2+" > > > +LIC_FILES_CHKSUM = > > "file://LICENSE;md5=2c1c00f9d3ed9e24fa69b932b7e7aff2" > > > + > > > +SRC_URI = "git://github.com/iipeace/${BPN} > > " > > > +SRCREV = "ac05a2ce0ab480957d642d947d8f617bd224031d" > > > + > > > +S = "${WORKDIR}/git" > > > +R = "${RECIPE_SYSROOT_NATIVE}" > > > + > > > +inherit distutils > > > + > > > +GUIDER_OBJ = "guider.pyc" > > > +GUIDER_SCRIPT = "guider" > > > + > > > +do_install() { > > > + python ${S}/setup.py install > > > + > > > + install -d ${D}${bindir} > > > + install -v -m 0755 ${R}/${bindir}/${GUIDER_SCRIPT} > > ${D}${bindir}/${GUIDER_SCRIPT} > > > + > > > + install -d ${D}${datadir}/${BPN} > > > + install -v -m 0755 ${R}/${datadir}/${BPN}/${GUIDER_OBJ} > > ${D}${datadir}/${BPN}/${GUIDER_OBJ} > > > +} > > > + > > > +RDEPENDS_${PN} = "python-ctypes python-shell \ > > > + python-json python-subprocess" > > > > Shouldn't the packages that guider uses be recommended at least? > > I'm not familiar with the app so maybe it can do useful things without > > top, perf, etc. > > > > Should we add a kernel cfg fragment? > > > > > + > > > +FILES_SOLIBSDEV = "" > > > +#FILES_${PN} = "usr" > > Drop that line? > > > > > > > > > > > > > -- > > # Randy MacLeod > > # Wind River Linux > > > > > -- > # Randy MacLeod > # Wind River Linux