From: "H. Peter Anvin" Subject: Re: Performance overhead of paravirt_ops on native identified Date: Fri, 22 May 2009 15:47:05 -0700 Message-ID: <4A172B69.2000709@zytor.com> References: <4A0B62F7.5030802@goop.org> <20090521184233.3c3e97ad@dhcp-100-2-144.bos.redhat.com> <4A15DA4E.2090505@goop.org> <4A1629BA.9070309@goop.org> <4A16D3DF.5000600@zytor.com> <4A172ADE.3010702@goop.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <4A172ADE.3010702@goop.org> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Jeremy Fitzhardinge Cc: Nick Piggin , "Xin, Xiaohui" , Xen-devel , Chuck Ebbert , Linux Kernel Mailing List , "Li, Xin" , "Nakajima, Jun" , Ingo Molnar List-Id: xen-devel@lists.xenproject.org Jeremy Fitzhardinge wrote: > > I did a quick experiment to see how many sites this optimisation could > actually affect. Firstly, it does absolutely nothing with frame > pointers enabled. Arranging for no frame pointers is quite tricky, > since it means disabling all debugging, tracing and other things. > > With no frame pointers, its about 26 of 5400 indirect calls are > immediately followed by ret (not all of those sites are pvops calls). > With preempt disabled, this goes up to 45 sites. > > I haven't done any actual runtime tests, but a quick survey of the > affected sites shows that only a couple are performance-sensitive; > _spin_lock and _spin_lock_irq and _spin_lock_irqsave are the most obvious. > OK, that doesn't seem like a very productive avenue. Problem still remains, obviously. -hpa