ia64/xen-unstable
changeset 18062:64d8a24d28f6
x86: No need to vcpu_update_system_time() on map_vcpu_info().
Also, ensure update_vcpu_system_time() triggers at least once for any
given vcpu, even if the TSC stamp is zero.
Signed-off-by: Keir Fraser <keir.fraser@citrix.com>
Also, ensure update_vcpu_system_time() triggers at least once for any
given vcpu, even if the TSC stamp is zero.
Signed-off-by: Keir Fraser <keir.fraser@citrix.com>
author | Keir Fraser <keir.fraser@citrix.com> |
---|---|
date | Tue Jul 15 17:56:25 2008 +0100 (2008-07-15) |
parents | 675fb031df88 |
children | 64b55f15a055 |
files | xen/arch/x86/domain.c |
line diff
1.1 --- a/xen/arch/x86/domain.c Tue Jul 15 16:13:46 2008 +0100 1.2 +++ b/xen/arch/x86/domain.c Tue Jul 15 17:56:25 2008 +0100 1.3 @@ -286,6 +286,10 @@ int vcpu_initialise(struct vcpu *v) 1.4 1.5 v->arch.flags = TF_kernel_mode; 1.6 1.7 + /* Ensure that update_vcpu_system_time() fires at least once. */ 1.8 + if ( !is_idle_domain(d) ) 1.9 + vcpu_info(v, time).tsc_timestamp = ~0ull; 1.10 + 1.11 #if defined(__i386__) 1.12 mapcache_vcpu_init(v); 1.13 #endif 1.14 @@ -812,14 +816,6 @@ map_vcpu_info(struct vcpu *v, unsigned l 1.15 for ( i = 0; i < BITS_PER_GUEST_LONG(d); i++ ) 1.16 set_bit(i, &vcpu_info(v, evtchn_pending_sel)); 1.17 1.18 - /* 1.19 - * Only bother to update time for the current vcpu. If we're 1.20 - * operating on another vcpu, then it had better not be running at 1.21 - * the time. 1.22 - */ 1.23 - if ( v == current ) 1.24 - update_vcpu_system_time(v); 1.25 - 1.26 return 0; 1.27 } 1.28