From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: [PATCH 1/2] tools, perf: Add a precise event qualifier v2 Date: Wed, 24 Jul 2013 03:33:15 +0200 Message-ID: <20130724013315.GU6123@two.firstfloor.org> References: <1374501138-13496-1-git-send-email-andi@firstfloor.org> <20130723060108.GA18396@tassilo.jf.intel.com> <20130723225150.GT6123@two.firstfloor.org> <51EF222D.9070909@oracle.com> Mime-Version: 1.0 Return-path: Content-Disposition: inline In-Reply-To: <51EF222D.9070909@oracle.com> Sender: trinity-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Sasha Levin Cc: Andi Kleen , Vince Weaver , Andi Kleen , acme@infradead.org, mingo@kernel.org, linux-kernel@vger.kernel.org, Peter Zijlstra , Stephane Eranian , torvalds@linux-foundation.org, trinity@vger.kernel.org On Tue, Jul 23, 2013 at 08:39:09PM -0400, Sasha Levin wrote: > On 07/23/2013 06:51 PM, Andi Kleen wrote: > >On Tue, Jul 23, 2013 at 05:27:43PM -0400, Vince Weaver wrote: > >>> > >>>I hate having to justify why breaking the ABI is unacceptable. > >Well it's a testing ABI, so we can do changes to it. > > The testing ABI has a simple policy about changes: > > The interface can be changed to add new features, but the > current interface will not break by doing this, unless grave > errors or security problems are found in them. > > It's probably fine to change a testing ABI once in a while, but when things > like trinity start breaking that often due to ABI changes in the same exact > place, that's too much IMO. It sounds like trinity is breaking (well printing a message, not really breaking) on any addition. So if we follow that the perf sysfs interface would be completely frozen and can never be extended over today. I don't think it's a big problem that a test tool needs to be extended when the software it's testing changes. If there are enough other widely used programs that actually break from additions probably would need a v2 of the sysfs interface for extensions (with new file or directory names), and keep v1 frozen for compatibility. But I don't think that's the case today? -Andi -- ak@linux.intel.com -- Speaking for myself only.