From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gleb Natapov Subject: Re: [PATCHv2 6/9] perf: expose perf capability to other modules. Date: Mon, 7 Nov 2011 18:22:07 +0200 Message-ID: <20111107162207.GK8670@redhat.com> References: <1320323618-10375-1-git-send-email-gleb@redhat.com> <1320323618-10375-7-git-send-email-gleb@redhat.com> <1320674870.18053.37.camel@twins> <20111107155336.GI8670@redhat.com> <1320681671.17809.5.camel@twins> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: kvm@vger.kernel.org, avi@redhat.com, mtosatti@redhat.com, linux-kernel@vger.kernel.org, mingo@elte.hu, acme@ghostprotocols.net To: Peter Zijlstra Return-path: Content-Disposition: inline In-Reply-To: <1320681671.17809.5.camel@twins> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On Mon, Nov 07, 2011 at 05:01:11PM +0100, Peter Zijlstra wrote: > On Mon, 2011-11-07 at 17:53 +0200, Gleb Natapov wrote: > > I removed branch-miss-retired here because for perf user it exists. Perf > > approximates it by other event but perf user shouldn't know that. A > > guest is not always runs with exactly same cpu model number as a host, > > so if we will not drop the bit here if guest will see cpu model other > > than the one that has this quirk it will not be able to use the event. > > Right, so what model number do you expose? Anyway I don't really mind Depends on what management wants. You can specify -cpu Nehalem or -cpu Conroe or even override model manually by doing -cpu host,model=15. > masking the thing as long as we grow an ebx iteration, > > > BTW why perf does not check event mask to see if architectural event is > > available before programming it? > > No idea why not.. just one of those things nobody noticed/got around to > doing or so. -- Gleb.