From mboxrd@z Thu Jan 1 00:00:00 1970 From: Shuai Ruan Subject: Re: XSAVE flavors Date: Thu, 4 Feb 2016 14:49:35 +0800 Message-ID: <32826.261901255$1454568793@news.gmane.org> References: <56A791BA02000078000CB1C9@prv-mh.provo.novell.com> <56A79AE402000078000CB224@prv-mh.provo.novell.com> <20160202063110.GB3036@shuai.ruan@linux.intel.com> <56B0881E02000078000CD58A@prv-mh.provo.novell.com> Reply-To: Shuai Ruan Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from mail6.bemta3.messagelabs.com ([195.245.230.39]) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1aRDlf-0000ZF-6C for xen-devel@lists.xenproject.org; Thu, 04 Feb 2016 06:51:51 +0000 Content-Disposition: inline In-Reply-To: <56B0881E02000078000CD58A@prv-mh.provo.novell.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Jan Beulich Cc: xen-devel , Kevin Tian , Jun Nakajima List-Id: xen-devel@lists.xenproject.org On Tue, Feb 02, 2016 at 02:42:38AM -0700, Jan Beulich wrote: > >> > With this another question then is whether, when both XSAVEC > >> > and XSAVEOPT are available, it is indeed always better to use > >> > XSAVEC (as the code is doing after your enabling). > > Yes. > > But current no machine only support xsavec not support xsaves. > > I enable xsavec for "xsavec is a feature". > > But this shouldn't preclude the code being in reasonable shape > also for the case where a CPU has XSAVEC but no XSAVES. The > more that right now we don't really need XSAVES (since we don't > yet allow any bit to get set in XSS). > Actually, when I enable xsaves/xsavec, I have put xsavec into consideration. If xsavec is used we also need to guarntee that xcomp_bv never has any bits clear which are set in xstate_bv and the compaction bit is set. Those guarntee and xsavec specific code in my patch is always behind "if( cpu_has_xsavec )" or " if ( cpu_has_xsaves || cpu_has_xsavec )". Please remind me if there is some other things I am not aware. > >> And I'm afraid there's yet one more issue: If my reading of the > >> SDM is right, then the offsets at which components get saved > >> by XSAVEC / XSAVES aren't fixed, but depend on RFBM (as that's > >> what gets stored into xcomp_bv[62:0]). xstate_comp_offsets[], > >> otoh, gets computed based on all available features, irrespective > >> of vcpu_xsave_mask() returning four different values depending > >> on current guest state. I can't see how get_xsave_addr() can > >> work correctly without honoring xcomp_bv. Nor can I convince > >> myself that state can't get corrupted / lost, e.g. when a save > >> with v->fpu_dirtied set is followed by one with v->fpu_dirtied > >> clear. > >> > >> Am I misunderstanding what the SDM writes? > >> > > Yes. you are right. This is a issue. I will find a way to solve > > this. > > Thanks. For xstate_comp_offsets is only used in get_xsave_addr when performing migration. I intend to recaculte xstate_comp_offsets based on the vcpu->arch.xsavec_area.save_hdr.xcomp_bv before get_xsave_addr is called. The patch will be sent out after Chinese New Year holiday. Thanks > > Jan > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel >