On Wed, May 1, 2019 at 7:45 AM Tamas K Lengyel <ta...@tklengyel.com> wrote: > > On Wed, May 1, 2019 at 1:50 AM Andrew Cooper <andrew.coop...@citrix.com> > wrote: > > > > On 01/05/2019 05:22, Tamas K Lengyel wrote: > > > Currently the gs_shadow value is only cached when the vCPU is being > > > scheduled > > > out by Xen. Reporting this (usually) stale value through vm_event is > > > incorrect, > > > since it doesn't represent the actual state of the vCPU at the time the > > > event > > > was recorded. This prevents vm_event subscribers from correctly finding > > > kernel > > > structures in the guest when it is trapped while in ring3. > > > > > > Signed-off-by: Tamas K Lengyel <ta...@tklengyel.com> > > > Cc: Razvan Cojocaru <rcojoc...@bitdefender.com> > > > Cc: Jan Beulich <jbeul...@suse.com> > > > Cc: Andrew Cooper <andrew.coop...@citrix.com> > > > Cc: Wei Liu <wei.l...@citrix.com> > > > Cc: Roger Pau Monne <roger....@citrix.com> > > > --- > > > xen/arch/x86/vm_event.c | 2 +- > > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > > > diff --git a/xen/arch/x86/vm_event.c b/xen/arch/x86/vm_event.c > > > index 51c3493b1d..4464940da7 100644 > > > --- a/xen/arch/x86/vm_event.c > > > +++ b/xen/arch/x86/vm_event.c > > > > Actually, come to think of it, the same is true for the SYSENTER > > details, which by default are read/write to the guest while it is > > scheduled. As a result, the details reported here will from the last > > vcpu context switch, and possibly stale. > > I'll look into it.
The sysenter values look fine to me because vmx_save_vmcs_ctxt calls vmx_vmcs_save, which refreshes the values from the actual VMCS. It's only shadow_gs that is not refreshed. Tamas _______________________________________________ Xen-devel mailing list Xen-devel@lists.xenproject.org https://lists.xenproject.org/mailman/listinfo/xen-devel
