From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.linutronix.de (193.142.43.55:993) by crypto-ml.lab.linutronix.de with IMAP4-SSL for ; 07 Nov 2019 20:11:37 -0000 Received: from p5b06da22.dip0.t-ipconnect.de ([91.6.218.34] helo=nanos) by Galois.linutronix.de with esmtpsa (TLS1.2:DHE_RSA_AES_256_CBC_SHA256:256) (Exim 4.80) (envelope-from ) id 1iSo80-00059R-2D for speck@linutronix.de; Thu, 07 Nov 2019 21:11:36 +0100 Date: Thu, 7 Nov 2019 21:11:34 +0100 (CET) From: Thomas Gleixner Subject: Re: The Annnoying Affair has No eXcuses - Bulletin #3 In-Reply-To: <900659f3-b0d2-71ad-ae61-60d325efa709@redhat.com> Message-ID: References: <900659f3-b0d2-71ad-ae61-60d325efa709@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit To: speck@linutronix.de List-ID: Paolo, On Thu, 7 Nov 2019, speck for Paolo Bonzini wrote: > On 07/11/19 16:16, speck for Thomas Gleixner wrote: > > The TAA fix from Josh is now applied to master and to the stable branches, > > which are still waiting for Paolos NX updates. Hint, hint... > > Looks like I have No Excuses for procrastinating. :) Bundles attached. First of all thanks for providing that! But now I really have to ask whether you actually ever looked how backports into stable kernels are done? Obviously not, because your series is just buggered: 1) All backport commits have to have commit LOOOONGSHAAAAAA upstream Right at the top of the changelog, not some randomly defined (cherrypicked ....) thingy in the middle of the SOB chain 2) All backport commits have to be the same as the upstream commit including change logs, except for eventual conflicts. Instead of using the commits from the repo or the bundle you picked your V9 version and backported it and of course you still have the wrong default in the documentation of the command line parameter along with different change logs. 3) All backport commits have to preserve the SOB/tag chain of the upstream commit, plus your own if the committer was not yourself. Of course all the ones which are in the master branch do not have that because you used your V9 version as a base. Aside of missing my SOB's the broken SOB chain from 'kvm: mmu: ITLB_MULTIHIT mitigation' which I fixed up when merging the lot is still in your backports. 4) Random 'RHEL:' tags which are nowhere in the upstream commit have really no place in stable kernels. Groan, you are doing kernel development for how long now? Sorry for my lousy mood, but after staring into futex code 3 days straight, this was really what I needed - NOT. I really was hoping just to do a pull and write a merge message and not wasting an hour for untinkering. I fixed it all up and I'm going to send out a bulletin with complete TAA/NX bundles in a few. Yours grumpy tglx