From mboxrd@z Thu Jan 1 00:00:00 1970 From: Konrad Rzeszutek Wilk Subject: Re: [PATCH v11 for-xen-4.5 13/20] x86/VPMU: Save VPMU state for PV guests during context switch Date: Tue, 23 Sep 2014 11:18:18 -0400 Message-ID: <20140923151818.GL3007@laptop.dumpdata.com> References: <1411430281-6132-1-git-send-email-boris.ostrovsky@oracle.com> <1411430281-6132-14-git-send-email-boris.ostrovsky@oracle.com> <20140923150131.GI3007@laptop.dumpdata.com> <54218D6F.9050808@oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <54218D6F.9050808@oracle.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: Boris Ostrovsky Cc: kevin.tian@intel.com, keir@xen.org, jbeulich@suse.com, jun.nakajima@intel.com, andrew.cooper3@citrix.com, tim@xen.org, dietmar.hahn@ts.fujitsu.com, xen-devel@lists.xen.org, Aravind.Gopalakrishnan@amd.com, suravee.suthikulpanit@amd.com List-Id: xen-devel@lists.xenproject.org On Tue, Sep 23, 2014 at 11:10:39AM -0400, Boris Ostrovsky wrote: > On 09/23/2014 11:01 AM, Konrad Rzeszutek Wilk wrote: > >On Mon, Sep 22, 2014 at 07:57:54PM -0400, Boris Ostrovsky wrote: > >>Save VPMU state during context switch for both HVM and PV(H) guests. > >> > >>A subsequent patch will make it possible for vpmu_switch_to() to call > >Could you say what is the name of the 'subsequent patch' in the commit > >description please? > > It's patch 19 where NMI PMU interrupts are added. "x86/VPMU: NMI-based VPMU support" would look great in the commit description. > > > > >>vmx_vmcs_try_enter()->vcpu_pause() which needs is_running to be correctly > >>set/cleared. To prepare for that, call context_saved() before vpmu_switch_to() > >>is executed. > >And what ramifications does that have to the guests if one only applied > >patches up to this point? > > None. Which is why I decided to make this change in this patch and not wait > until patch 19. > OK, please do spell it out in the patch so that Anxious Release Managers don't have to sweat about regressions. And with those changes: Reviewed-by: Konrad Rzeszutek Wilk > -boris > > > > >>Signed-off-by: Boris Ostrovsky > >>--- > >> xen/arch/x86/domain.c | 20 +++++++++----------- > >> 1 file changed, 9 insertions(+), 11 deletions(-) > >> > >>diff --git a/xen/arch/x86/domain.c b/xen/arch/x86/domain.c > >>index 5e492d8..d7fb99d 100644 > >>--- a/xen/arch/x86/domain.c > >>+++ b/xen/arch/x86/domain.c > >>@@ -1495,16 +1495,13 @@ void context_switch(struct vcpu *prev, struct vcpu *next) > >> } > >> if ( prev != next ) > >>- _update_runstate_area(prev); > >>- > >>- if ( is_hvm_vcpu(prev) ) > >> { > >>- if (prev != next) > >>- vpmu_switch_from(prev, next); > >>+ _update_runstate_area(prev); > >>+ vpmu_switch_from(prev, next); > >>+ } > >>- if ( !list_empty(&prev->arch.hvm_vcpu.tm_list) ) > >>+ if ( is_hvm_vcpu(prev) && !list_empty(&prev->arch.hvm_vcpu.tm_list) ) > >> pt_save_timer(prev); > >>- } > >> local_irq_disable(); > >>@@ -1543,15 +1540,16 @@ void context_switch(struct vcpu *prev, struct vcpu *next) > >> !is_hardware_domain(next->domain)); > >> } > >>- if ( is_hvm_vcpu(prev) && (prev != next) ) > >>- /* Must be done with interrupts enabled */ > >>- vpmu_switch_to(prev, next); > >>- > >> context_saved(prev); > >> if ( prev != next ) > >>+ { > >> _update_runstate_area(next); > >>+ /* Must be done with interrupts enabled */ > >>+ vpmu_switch_to(prev, next); > >>+ } > >>+ > >> /* Ensure that the vcpu has an up-to-date time base. */ > >> update_vcpu_system_time(next); > >>-- > >>1.8.1.4 > >> > >> > >>_______________________________________________ > >>Xen-devel mailing list > >>Xen-devel@lists.xen.org > >>http://lists.xen.org/xen-devel >