From mboxrd@z Thu Jan 1 00:00:00 1970 X-GM-THRID: 6919450737082630144 X-Received: by 2002:a17:906:b219:: with SMTP id p25mr6125387ejz.430.1611156394822; Wed, 20 Jan 2021 07:26:34 -0800 (PST) X-BeenThere: isar-users@googlegroups.com Received: by 2002:a17:906:cb0f:: with SMTP id lk15ls2943378ejb.6.gmail; Wed, 20 Jan 2021 07:26:33 -0800 (PST) X-Google-Smtp-Source: ABdhPJyFk6fRWifAytd1FUPo0ov+LY/rfaV2wToEKmfbwhq0AcmQmshpQl6B1nWYAvD0y4bBnJb9 X-Received: by 2002:a17:906:ae91:: with SMTP id md17mr6520802ejb.302.1611156393880; Wed, 20 Jan 2021 07:26:33 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1611156393; cv=none; d=google.com; s=arc-20160816; b=wJmOz2E9UyHwgDNojslqANo6UyeqX+fwUr8V1B/FVlOeYP1oPNMyYDyJurdz689UPv Zf8+qIemYVfxkfVCVcehi4ny0zqCCRvQ6ldXY+ABZPWjCbOvjxKqAcKSuqJSeK4GNHfy rWVhk2NkCn3WPKriAqC2WfZXIwJrFFG1IIGFrWFzWD6x5yN+pXS7C8q7WMD3ErS2tBLu v9l/1H3eaefUCB8GaBKw2h20zbB2D1t5A4KSF4EOo/9e3aaGUraga5boMu9RsoVObTad RYb3eI+slN9fl6ntNDYtyS0oUEoEUeuAKSCVqgU9CvoLBIM9HZv52NT30hwg599r7dB+ IcvQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:subject:cc:to:from:date; bh=MM944ckOj8RO3NEs+Sdp65OFW1aa8/hQlSOEwlL/Bxk=; b=OCQxWYp1v+YmvSyMZxxkPUZAMuQOcNN5P07JAJKQ9nSspDa6PkGtf5d0VZbKNtkG7Q YCs9lzaDTTygRCtC2bgL5MmVZiyGNoCsBjBU7l28tuQqKtAQNlUo4kfOkcoTaWXsEd+S dKiYY/Rc91vr4AP8eQHtMB067CZdnfpzEvuScodTPsluHesuwQ9fn+RqzRIDTsSBJpm+ yc7F1vlYt4lpA1KpY0zxLRfoBOk06zTDYQfiQICdgH0nhcZXH9uqBGKIGaeT6Rlml7yC ndyM/MKp/V1cvFAm1Zyp8Yo0O+v5HXbN2zqMfJSO4nU4rlMkXtpHWruwoMdPbkO6Dxd2 Oe2g== ARC-Authentication-Results: i=1; gmr-mx.google.com; spf=pass (google.com: domain of henning.schild@siemens.com designates 192.35.17.2 as permitted sender) smtp.mailfrom=henning.schild@siemens.com; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=siemens.com Return-Path: Received: from thoth.sbs.de (thoth.sbs.de. [192.35.17.2]) by gmr-mx.google.com with ESMTPS id d2si182278edo.5.2021.01.20.07.26.33 for (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Wed, 20 Jan 2021 07:26:33 -0800 (PST) Received-SPF: pass (google.com: domain of henning.schild@siemens.com designates 192.35.17.2 as permitted sender) client-ip=192.35.17.2; Authentication-Results: gmr-mx.google.com; spf=pass (google.com: domain of henning.schild@siemens.com designates 192.35.17.2 as permitted sender) smtp.mailfrom=henning.schild@siemens.com; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=siemens.com Received: from mail2.sbs.de (mail2.sbs.de [192.129.41.66]) by thoth.sbs.de (8.15.2/8.15.2) with ESMTPS id 10KFQX2B016707 (version=TLSv1.2 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Wed, 20 Jan 2021 16:26:33 +0100 Received: from md1za8fc.ad001.siemens.net ([139.22.120.228]) by mail2.sbs.de (8.15.2/8.15.2) with ESMTP id 10KFQWFf012126; Wed, 20 Jan 2021 16:26:33 +0100 Date: Wed, 20 Jan 2021 16:26:31 +0100 From: Henning Schild To: "Moessbauer, Felix (T RDA IOT SES-DE)" Cc: Baurzhan Ismagulov , "isar-users@googlegroups.com" , "Schmidt, Adriaan (T RDA IOT SES-DE)" , "Kiszka, Jan (T RDA IOT)" Subject: Re: [PATCH 1/1] RFC: Package perf from linux kernel tools Message-ID: <20210120162631.01281c23@md1za8fc.ad001.siemens.net> In-Reply-To: References: <20210119124157.12662-1-felix.moessbauer@siemens.com> <20210119124157.12662-2-felix.moessbauer@siemens.com> <20210119163107.GW22444@yssyq.m.ilbers.de> X-Mailer: Claws Mail 3.17.8 (GTK+ 2.24.32; x86_64-pc-linux-gnu) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-TUID: XQJlQxKyrI1i Am Wed, 20 Jan 2021 16:23:18 +0100 schrieb "Moessbauer, Felix (T RDA IOT SES-DE)" : > Hi, > > today I implemented the packaging in the kernel recipe and fixed some > nasty aspects like exact version numbers in the patches. > I'll send this patch to the mailing list as well. > > One issue is still that we need per-kernel patches. > If anything in the original makefile of perf changed, our patches do > not apply. That lead me to the initial approach with having a > dedicated package with multiple recipes, to have different patch-sets > per kernel version (and put it into the responsibility of the user). > > Another issue with the kernel-recipe approach is that it's no longer > possible to use the Debian kernel but customize / fix the > linux-perf- package. Maybe we need both. But i guess debian has perf packages for kernels taken from them. So this would only be a problem if one had a need to add python binding support to such a kernel. In which case one could always rebuild the same sources+config under another name ... Henning > @Schild, Henning (T RDA IOT SES-DE): > Having the Python stuff in a dedicated commit is definitely a good > idea. It's already in its own patchset. I'll send a v2 to the list as > well. > > Best regards, > Felix > > > -----Original Message----- > > From: Baurzhan Ismagulov > > Sent: Tuesday, January 19, 2021 5:31 PM > > To: Moessbauer, Felix (T RDA IOT SES-DE) > > Cc: isar-users@googlegroups.com; > > Schmidt, Adriaan (T RDA IOT SES-DE) ; > > Schild, Henning (T RDA IOT SES-DE) ; > > Kiszka, Jan (T RDA IOT) > > Subject: Re: [PATCH 1/1] RFC: Package perf from linux kernel tools > > > > Hello Felix, > > > > On Tue, Jan 19, 2021 at 01:41:58PM +0100, Felix Moessbauer wrote: > > > This patch adds the linux-perf-4.19 package for a kernel that is > > > build with ISAR. > > > As the perf binaries have to be version specific, we patch the > > > corresponding files from the kernel tools source to incorporate > > > the major and minor version number. > > > These patches are copied over from the official debianization of > > > perf that is done in the src:linux package. > > > Additionally, we package the perf python bindings as well and > > > provide them as module "perf_4_19" (in the > > > tools-perf-python.patch). > > > > > > As this debianization is not part of the kernel receipt, the perf > > > > > > > Suggest "receipt" -> "recipe". > > > > > > > package can be based on a different patch version of the kernel. > > > While this might be intended, it is now the responibility of the > > > user to ensure that both the kernel and the linux-perf- > > > packages are compatible. > > ... > > > .../linux-perf/files/tools-perf-install.patch | 58 +++++++++ > > > .../linux-perf/files/tools-perf-python.patch | 55 ++++++++ > > > .../linux-perf/files/tools-perf-version.patch | 119 > > > ++++++++++++++++++ > > > > The patches look common sense to me; ideally, it would be better to > > handle that in the kernel and have a single kernel source package. > > I wonder why Debian hasn't upstreamed that yet -- whether there > > were any technical problems, or just no one happened to look at > > that. That said, I personally could live with this till we have a > > better solution. > > > > If nothing works, maybe we could relax the goal of being able to > > build any given kernel version and experiment with conditional > > execution paths. > > > > > > With kind regards, > > Baurzhan.