| Message ID | 20220420222834.5478-2-dpsmith@apertussolutions.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | Adds starting the idle domain privileged | expand |
On 21.04.2022 00:28, Daniel P. Smith wrote: > There are now instances where internal hypervisor logic needs to make resource > allocation calls that are protectd by XSM checks. The internal hypervisor logic > is represented a number of system domains which by designed are represented by > non-privileged struct domain instances. To enable these logic blocks to > function correctly but in a controlled manner, this commit changes the idle > domain to be created as a privileged domain under the default policy, which is > inherited by the SILO policy, and demoted before transitioning to running. A > new XSM hook, xsm_transition_running, is introduced to allow each XSM policy > type to demote the idle domain appropriately for that policy type. > > For flask a stub is added to ensure that flask policy system will function > correctly with this patch until flask is extended with support for starting the > idle domain privileged and properly demoting it on the call to > xsm_transtion_running. > > Signed-off-by: Daniel P. Smith <dpsmith@apertussolutions.com> Looks okay to me, but I'm not sure in how far agreement was reached on taking this route. Just one nit: > --- a/xen/include/xsm/dummy.h > +++ b/xen/include/xsm/dummy.h > @@ -101,6 +101,18 @@ static always_inline int xsm_default_action( > } > } > > +static XSM_INLINE void cf_check xsm_transition_running(void) > +{ > + struct domain *d = current->domain; > + > + if ( d->domain_id != DOMID_IDLE ) > + panic("xsm_transition_running should only be called by idle domain\n"); > + > + d->is_privileged = false; > + > + return; > +} Please omit such return statements. Jan
On Wed, Apr 20, 2022 at 06:28:33PM -0400, Daniel P. Smith wrote: > There are now instances where internal hypervisor logic needs to make resource > allocation calls that are protectd by XSM checks. The internal hypervisor logic > is represented a number of system domains which by designed are represented by > non-privileged struct domain instances. To enable these logic blocks to > function correctly but in a controlled manner, this commit changes the idle > domain to be created as a privileged domain under the default policy, which is > inherited by the SILO policy, and demoted before transitioning to running. A > new XSM hook, xsm_transition_running, is introduced to allow each XSM policy > type to demote the idle domain appropriately for that policy type. > > For flask a stub is added to ensure that flask policy system will function > correctly with this patch until flask is extended with support for starting the > idle domain privileged and properly demoting it on the call to > xsm_transtion_running. > > Signed-off-by: Daniel P. Smith <dpsmith@apertussolutions.com> > --- > xen/arch/arm/setup.c | 6 ++++++ > xen/arch/x86/setup.c | 6 ++++++ > xen/common/sched/core.c | 7 ++++++- > xen/include/xsm/dummy.h | 12 ++++++++++++ > xen/include/xsm/xsm.h | 6 ++++++ > xen/xsm/dummy.c | 1 + > xen/xsm/flask/hooks.c | 15 +++++++++++++++ > 7 files changed, 52 insertions(+), 1 deletion(-) > > diff --git a/xen/arch/arm/setup.c b/xen/arch/arm/setup.c > index d5d0792ed4..763835aeb5 100644 > --- a/xen/arch/arm/setup.c > +++ b/xen/arch/arm/setup.c > @@ -1048,6 +1048,12 @@ void __init start_xen(unsigned long boot_phys_offset, > /* Hide UART from DOM0 if we're using it */ > serial_endboot(); > > + xsm_transition_running(); Could we put depriv or dipriviledge somewhere here? 'transition' seem to ambiguous IMO (but I'm not a native speaker). xsm_{depriv,demote}_current(); > + > + /* Ensure idle domain was not left privileged */ > + if ( current->domain->is_privileged ) > + panic("idle domain did not properly transition from setup privilege\n"); > + > system_state = SYS_STATE_active; > > for_each_domain( d ) > diff --git a/xen/arch/x86/setup.c b/xen/arch/x86/setup.c > index 6f20e17892..72695dcb07 100644 > --- a/xen/arch/x86/setup.c > +++ b/xen/arch/x86/setup.c > @@ -621,6 +621,12 @@ static void noreturn init_done(void) > void *va; > unsigned long start, end; > > + xsm_transition_running(); > + > + /* Ensure idle domain was not left privileged */ > + if ( current->domain->is_privileged ) > + panic("idle domain did not properly transition from setup privilege\n"); > + > system_state = SYS_STATE_active; > > domain_unpause_by_systemcontroller(dom0); > diff --git a/xen/common/sched/core.c b/xen/common/sched/core.c > index 19ab678181..22a619e260 100644 > --- a/xen/common/sched/core.c > +++ b/xen/common/sched/core.c > @@ -3021,7 +3021,12 @@ void __init scheduler_init(void) > sched_ratelimit_us = SCHED_DEFAULT_RATELIMIT_US; > } > > - idle_domain = domain_create(DOMID_IDLE, NULL, 0); > + /* > + * idle dom is created privileged to ensure unrestricted access during > + * setup and will be demoted by xsm_transition_running when setup is > + * complete > + */ > + idle_domain = domain_create(DOMID_IDLE, NULL, CDF_privileged); > BUG_ON(IS_ERR(idle_domain)); > BUG_ON(nr_cpu_ids > ARRAY_SIZE(idle_vcpu)); > idle_domain->vcpu = idle_vcpu; > diff --git a/xen/include/xsm/dummy.h b/xen/include/xsm/dummy.h > index 58afc1d589..b33f0ec672 100644 > --- a/xen/include/xsm/dummy.h > +++ b/xen/include/xsm/dummy.h > @@ -101,6 +101,18 @@ static always_inline int xsm_default_action( > } > } > > +static XSM_INLINE void cf_check xsm_transition_running(void) > +{ > + struct domain *d = current->domain; > + > + if ( d->domain_id != DOMID_IDLE ) > + panic("xsm_transition_running should only be called by idle domain\n"); Could you also add a check that d->is_privileged == true? Thanks, Roger.
On 4/21/22 05:20, Jan Beulich wrote: > On 21.04.2022 00:28, Daniel P. Smith wrote: >> There are now instances where internal hypervisor logic needs to make resource >> allocation calls that are protectd by XSM checks. The internal hypervisor logic >> is represented a number of system domains which by designed are represented by >> non-privileged struct domain instances. To enable these logic blocks to >> function correctly but in a controlled manner, this commit changes the idle >> domain to be created as a privileged domain under the default policy, which is >> inherited by the SILO policy, and demoted before transitioning to running. A >> new XSM hook, xsm_transition_running, is introduced to allow each XSM policy >> type to demote the idle domain appropriately for that policy type. >> >> For flask a stub is added to ensure that flask policy system will function >> correctly with this patch until flask is extended with support for starting the >> idle domain privileged and properly demoting it on the call to >> xsm_transtion_running. >> >> Signed-off-by: Daniel P. Smith <dpsmith@apertussolutions.com> > > Looks okay to me, but I'm not sure in how far agreement was reached on > taking this route. Just one nit: Thank you. As for the approach, Jason suggested it and Roger stated that if this approach was adopted it would resolve his concerns over __init. If you have a concern with this approach, please let me know so I can ensure it is addressed to the best of my ability. >> --- a/xen/include/xsm/dummy.h >> +++ b/xen/include/xsm/dummy.h >> @@ -101,6 +101,18 @@ static always_inline int xsm_default_action( >> } >> } >> >> +static XSM_INLINE void cf_check xsm_transition_running(void) >> +{ >> + struct domain *d = current->domain; >> + >> + if ( d->domain_id != DOMID_IDLE ) >> + panic("xsm_transition_running should only be called by idle domain\n"); >> + >> + d->is_privileged = false; >> + >> + return; >> +} > > Please omit such return statements. Ack. v/r, dps
On 4/21/22 05:53, Roger Pau Monné wrote: > On Wed, Apr 20, 2022 at 06:28:33PM -0400, Daniel P. Smith wrote: >> There are now instances where internal hypervisor logic needs to make resource >> allocation calls that are protectd by XSM checks. The internal hypervisor logic >> is represented a number of system domains which by designed are represented by >> non-privileged struct domain instances. To enable these logic blocks to >> function correctly but in a controlled manner, this commit changes the idle >> domain to be created as a privileged domain under the default policy, which is >> inherited by the SILO policy, and demoted before transitioning to running. A >> new XSM hook, xsm_transition_running, is introduced to allow each XSM policy >> type to demote the idle domain appropriately for that policy type. >> >> For flask a stub is added to ensure that flask policy system will function >> correctly with this patch until flask is extended with support for starting the >> idle domain privileged and properly demoting it on the call to >> xsm_transtion_running. >> >> Signed-off-by: Daniel P. Smith <dpsmith@apertussolutions.com> >> --- >> xen/arch/arm/setup.c | 6 ++++++ >> xen/arch/x86/setup.c | 6 ++++++ >> xen/common/sched/core.c | 7 ++++++- >> xen/include/xsm/dummy.h | 12 ++++++++++++ >> xen/include/xsm/xsm.h | 6 ++++++ >> xen/xsm/dummy.c | 1 + >> xen/xsm/flask/hooks.c | 15 +++++++++++++++ >> 7 files changed, 52 insertions(+), 1 deletion(-) >> >> diff --git a/xen/arch/arm/setup.c b/xen/arch/arm/setup.c >> index d5d0792ed4..763835aeb5 100644 >> --- a/xen/arch/arm/setup.c >> +++ b/xen/arch/arm/setup.c >> @@ -1048,6 +1048,12 @@ void __init start_xen(unsigned long boot_phys_offset, >> /* Hide UART from DOM0 if we're using it */ >> serial_endboot(); >> >> + xsm_transition_running(); > > Could we put depriv or dipriviledge somewhere here? 'transition' seem to > ambiguous IMO (but I'm not a native speaker). > > xsm_{depriv,demote}_current(); Let me say this explanation is not to say no but to give context to the concerns. Forms of deprive/demote were considered though when considering the concept proposed was to change the security model where the hypervisor/idle domain were now explicitly being give a new security context, is_privileged and xenboot_t, under which setup is being run. This new xsm hook is to provide a transition point for the XSM policies to set what the running security context should be for the hypervisor/idle domain. The name xsm_transition_running() clearly denotes when/where this hook should be used, where as the name xsm_depriv_current() is more generic and another developer may attempt to use it in a manner it was not intended. It is possible to consider creating an xsm_depriv_current() that functions in a more generic manner but will likely be more complicated to support general usage, especially for flask where a flask specific "lower" security context must be provided. If there is still a preference towards xsm_depriv_current() while maintaining the current mechanics as it makes more sense for the majority, I have no issue with that. >> + >> + /* Ensure idle domain was not left privileged */ >> + if ( current->domain->is_privileged ) >> + panic("idle domain did not properly transition from setup privilege\n"); >> + >> system_state = SYS_STATE_active; >> >> for_each_domain( d ) >> diff --git a/xen/arch/x86/setup.c b/xen/arch/x86/setup.c >> index 6f20e17892..72695dcb07 100644 >> --- a/xen/arch/x86/setup.c >> +++ b/xen/arch/x86/setup.c >> @@ -621,6 +621,12 @@ static void noreturn init_done(void) >> void *va; >> unsigned long start, end; >> >> + xsm_transition_running(); >> + >> + /* Ensure idle domain was not left privileged */ >> + if ( current->domain->is_privileged ) >> + panic("idle domain did not properly transition from setup privilege\n"); >> + >> system_state = SYS_STATE_active; >> >> domain_unpause_by_systemcontroller(dom0); >> diff --git a/xen/common/sched/core.c b/xen/common/sched/core.c >> index 19ab678181..22a619e260 100644 >> --- a/xen/common/sched/core.c >> +++ b/xen/common/sched/core.c >> @@ -3021,7 +3021,12 @@ void __init scheduler_init(void) >> sched_ratelimit_us = SCHED_DEFAULT_RATELIMIT_US; >> } >> >> - idle_domain = domain_create(DOMID_IDLE, NULL, 0); >> + /* >> + * idle dom is created privileged to ensure unrestricted access during >> + * setup and will be demoted by xsm_transition_running when setup is >> + * complete >> + */ >> + idle_domain = domain_create(DOMID_IDLE, NULL, CDF_privileged); >> BUG_ON(IS_ERR(idle_domain)); >> BUG_ON(nr_cpu_ids > ARRAY_SIZE(idle_vcpu)); >> idle_domain->vcpu = idle_vcpu; >> diff --git a/xen/include/xsm/dummy.h b/xen/include/xsm/dummy.h >> index 58afc1d589..b33f0ec672 100644 >> --- a/xen/include/xsm/dummy.h >> +++ b/xen/include/xsm/dummy.h >> @@ -101,6 +101,18 @@ static always_inline int xsm_default_action( >> } >> } >> >> +static XSM_INLINE void cf_check xsm_transition_running(void) >> +{ >> + struct domain *d = current->domain; >> + >> + if ( d->domain_id != DOMID_IDLE ) >> + panic("xsm_transition_running should only be called by idle domain\n"); > > Could you also add a check that d->is_privileged == true? Are you thinking along the lines of, if ( (!d->is_privileged) || (d->domain_id != DOMID_IDLE) panic("some message\n"); or is your concern more of, if ( !d->is_privileged ) return; In my mind the former is legitimate because execution should only arrive here with current->domain being the idle domain and is_privileged set to true. The latter check I feel is extraneous because 1) this hook should only ever be called under the idle domain, thus it should be checked first and should absolutely panic if another domain context is in place. Which leads to, 2) checking if it is not false before setting to false is only protecting against resetting to false for which there could be no side effects this guard would be protecting against. v/r, dps
On Thu, Apr 21, 2022 at 10:14:18AM -0400, Daniel P. Smith wrote: > On 4/21/22 05:53, Roger Pau Monné wrote: > > On Wed, Apr 20, 2022 at 06:28:33PM -0400, Daniel P. Smith wrote: > >> There are now instances where internal hypervisor logic needs to make resource > >> allocation calls that are protectd by XSM checks. The internal hypervisor logic > >> is represented a number of system domains which by designed are represented by > >> non-privileged struct domain instances. To enable these logic blocks to > >> function correctly but in a controlled manner, this commit changes the idle > >> domain to be created as a privileged domain under the default policy, which is > >> inherited by the SILO policy, and demoted before transitioning to running. A > >> new XSM hook, xsm_transition_running, is introduced to allow each XSM policy > >> type to demote the idle domain appropriately for that policy type. > >> > >> For flask a stub is added to ensure that flask policy system will function > >> correctly with this patch until flask is extended with support for starting the > >> idle domain privileged and properly demoting it on the call to > >> xsm_transtion_running. > >> > >> Signed-off-by: Daniel P. Smith <dpsmith@apertussolutions.com> > >> --- > >> xen/arch/arm/setup.c | 6 ++++++ > >> xen/arch/x86/setup.c | 6 ++++++ > >> xen/common/sched/core.c | 7 ++++++- > >> xen/include/xsm/dummy.h | 12 ++++++++++++ > >> xen/include/xsm/xsm.h | 6 ++++++ > >> xen/xsm/dummy.c | 1 + > >> xen/xsm/flask/hooks.c | 15 +++++++++++++++ > >> 7 files changed, 52 insertions(+), 1 deletion(-) > >> > >> diff --git a/xen/arch/arm/setup.c b/xen/arch/arm/setup.c > >> index d5d0792ed4..763835aeb5 100644 > >> --- a/xen/arch/arm/setup.c > >> +++ b/xen/arch/arm/setup.c > >> @@ -1048,6 +1048,12 @@ void __init start_xen(unsigned long boot_phys_offset, > >> /* Hide UART from DOM0 if we're using it */ > >> serial_endboot(); > >> > >> + xsm_transition_running(); > > > > Could we put depriv or dipriviledge somewhere here? 'transition' seem to > > ambiguous IMO (but I'm not a native speaker). > > > > xsm_{depriv,demote}_current(); > > Let me say this explanation is not to say no but to give context to the > concerns. Forms of deprive/demote were considered though when > considering the concept proposed was to change the security model where > the hypervisor/idle domain were now explicitly being give a new security > context, is_privileged and xenboot_t, under which setup is being run. > This new xsm hook is to provide a transition point for the XSM policies > to set what the running security context should be for the > hypervisor/idle domain. The name xsm_transition_running() clearly > denotes when/where this hook should be used, where as the name > xsm_depriv_current() is more generic and another developer may attempt > to use it in a manner it was not intended. Hm, I see. I (wrongly) originally understood it was related to making a transition in the running context, rather than the context being changed to the running state. Maybe xsm_{transition_,set_,}system_active() to better match the system_state? Albeit now that I understand it's purpose it doesn't feel so weird. > >> diff --git a/xen/common/sched/core.c b/xen/common/sched/core.c > >> index 19ab678181..22a619e260 100644 > >> --- a/xen/common/sched/core.c > >> +++ b/xen/common/sched/core.c > >> @@ -3021,7 +3021,12 @@ void __init scheduler_init(void) > >> sched_ratelimit_us = SCHED_DEFAULT_RATELIMIT_US; > >> } > >> > >> - idle_domain = domain_create(DOMID_IDLE, NULL, 0); > >> + /* > >> + * idle dom is created privileged to ensure unrestricted access during > >> + * setup and will be demoted by xsm_transition_running when setup is > >> + * complete > >> + */ > >> + idle_domain = domain_create(DOMID_IDLE, NULL, CDF_privileged); > >> BUG_ON(IS_ERR(idle_domain)); > >> BUG_ON(nr_cpu_ids > ARRAY_SIZE(idle_vcpu)); > >> idle_domain->vcpu = idle_vcpu; > >> diff --git a/xen/include/xsm/dummy.h b/xen/include/xsm/dummy.h > >> index 58afc1d589..b33f0ec672 100644 > >> --- a/xen/include/xsm/dummy.h > >> +++ b/xen/include/xsm/dummy.h > >> @@ -101,6 +101,18 @@ static always_inline int xsm_default_action( > >> } > >> } > >> > >> +static XSM_INLINE void cf_check xsm_transition_running(void) > >> +{ > >> + struct domain *d = current->domain; > >> + > >> + if ( d->domain_id != DOMID_IDLE ) > >> + panic("xsm_transition_running should only be called by idle domain\n"); > > > > Could you also add a check that d->is_privileged == true? > > Are you thinking along the lines of, > > if ( (!d->is_privileged) || (d->domain_id != DOMID_IDLE) > panic("some message\n"); > > or is your concern more of, > > if ( !d->is_privileged ) > return; > > In my mind the former is legitimate because execution should only arrive > here with current->domain being the idle domain and is_privileged set to > true. I was thinking about the former, maybe adding it as a separate condition so you can print a specific panic message, or joined with the other if the panic message can be adjusted to fit both conditions. Thanks, Roger.
diff --git a/xen/arch/arm/setup.c b/xen/arch/arm/setup.c index d5d0792ed4..763835aeb5 100644 --- a/xen/arch/arm/setup.c +++ b/xen/arch/arm/setup.c @@ -1048,6 +1048,12 @@ void __init start_xen(unsigned long boot_phys_offset, /* Hide UART from DOM0 if we're using it */ serial_endboot(); + xsm_transition_running(); + + /* Ensure idle domain was not left privileged */ + if ( current->domain->is_privileged ) + panic("idle domain did not properly transition from setup privilege\n"); + system_state = SYS_STATE_active; for_each_domain( d ) diff --git a/xen/arch/x86/setup.c b/xen/arch/x86/setup.c index 6f20e17892..72695dcb07 100644 --- a/xen/arch/x86/setup.c +++ b/xen/arch/x86/setup.c @@ -621,6 +621,12 @@ static void noreturn init_done(void) void *va; unsigned long start, end; + xsm_transition_running(); + + /* Ensure idle domain was not left privileged */ + if ( current->domain->is_privileged ) + panic("idle domain did not properly transition from setup privilege\n"); + system_state = SYS_STATE_active; domain_unpause_by_systemcontroller(dom0); diff --git a/xen/common/sched/core.c b/xen/common/sched/core.c index 19ab678181..22a619e260 100644 --- a/xen/common/sched/core.c +++ b/xen/common/sched/core.c @@ -3021,7 +3021,12 @@ void __init scheduler_init(void) sched_ratelimit_us = SCHED_DEFAULT_RATELIMIT_US; } - idle_domain = domain_create(DOMID_IDLE, NULL, 0); + /* + * idle dom is created privileged to ensure unrestricted access during + * setup and will be demoted by xsm_transition_running when setup is + * complete + */ + idle_domain = domain_create(DOMID_IDLE, NULL, CDF_privileged); BUG_ON(IS_ERR(idle_domain)); BUG_ON(nr_cpu_ids > ARRAY_SIZE(idle_vcpu)); idle_domain->vcpu = idle_vcpu; diff --git a/xen/include/xsm/dummy.h b/xen/include/xsm/dummy.h index 58afc1d589..b33f0ec672 100644 --- a/xen/include/xsm/dummy.h +++ b/xen/include/xsm/dummy.h @@ -101,6 +101,18 @@ static always_inline int xsm_default_action( } } +static XSM_INLINE void cf_check xsm_transition_running(void) +{ + struct domain *d = current->domain; + + if ( d->domain_id != DOMID_IDLE ) + panic("xsm_transition_running should only be called by idle domain\n"); + + d->is_privileged = false; + + return; +} + static XSM_INLINE void cf_check xsm_security_domaininfo( struct domain *d, struct xen_domctl_getdomaininfo *info) { diff --git a/xen/include/xsm/xsm.h b/xen/include/xsm/xsm.h index 3e2b7fe3db..a5c06804ab 100644 --- a/xen/include/xsm/xsm.h +++ b/xen/include/xsm/xsm.h @@ -52,6 +52,7 @@ typedef enum xsm_default xsm_default_t; * !!! WARNING !!! */ struct xsm_ops { + void (*transition_running)(void); void (*security_domaininfo)(struct domain *d, struct xen_domctl_getdomaininfo *info); int (*domain_create)(struct domain *d, uint32_t ssidref); @@ -208,6 +209,11 @@ extern struct xsm_ops xsm_ops; #ifndef XSM_NO_WRAPPERS +static inline void xsm_transition_running(void) +{ + alternative_vcall(xsm_ops.transition_running); +} + static inline void xsm_security_domaininfo( struct domain *d, struct xen_domctl_getdomaininfo *info) { diff --git a/xen/xsm/dummy.c b/xen/xsm/dummy.c index 8c044ef615..66f26c6909 100644 --- a/xen/xsm/dummy.c +++ b/xen/xsm/dummy.c @@ -14,6 +14,7 @@ #include <xsm/dummy.h> static const struct xsm_ops __initconst_cf_clobber dummy_ops = { + .transition_running = xsm_transition_running, .security_domaininfo = xsm_security_domaininfo, .domain_create = xsm_domain_create, .getdomaininfo = xsm_getdomaininfo, diff --git a/xen/xsm/flask/hooks.c b/xen/xsm/flask/hooks.c index 0bf63ffa84..decebc8231 100644 --- a/xen/xsm/flask/hooks.c +++ b/xen/xsm/flask/hooks.c @@ -186,6 +186,20 @@ static int cf_check flask_domain_alloc_security(struct domain *d) return 0; } +static void cf_check flask_transition_running(void) +{ + struct domain *d = current->domain; + + if ( d->domain_id != DOMID_IDLE ) + panic("xsm_transition_running should only be called by idle domain\n"); + + /* + * While is_privileged has no significant meaning under flask, + * set to false for the consistency check(s) in the setup code. + */ + d->is_privileged = false; +} + static void cf_check flask_domain_free_security(struct domain *d) { struct domain_security_struct *dsec = d->ssid; @@ -1766,6 +1780,7 @@ static int cf_check flask_argo_send( #endif static const struct xsm_ops __initconst_cf_clobber flask_ops = { + .transition_running = flask_transition_running, .security_domaininfo = flask_security_domaininfo, .domain_create = flask_domain_create, .getdomaininfo = flask_getdomaininfo,
There are now instances where internal hypervisor logic needs to make resource allocation calls that are protectd by XSM checks. The internal hypervisor logic is represented a number of system domains which by designed are represented by non-privileged struct domain instances. To enable these logic blocks to function correctly but in a controlled manner, this commit changes the idle domain to be created as a privileged domain under the default policy, which is inherited by the SILO policy, and demoted before transitioning to running. A new XSM hook, xsm_transition_running, is introduced to allow each XSM policy type to demote the idle domain appropriately for that policy type. For flask a stub is added to ensure that flask policy system will function correctly with this patch until flask is extended with support for starting the idle domain privileged and properly demoting it on the call to xsm_transtion_running. Signed-off-by: Daniel P. Smith <dpsmith@apertussolutions.com> --- xen/arch/arm/setup.c | 6 ++++++ xen/arch/x86/setup.c | 6 ++++++ xen/common/sched/core.c | 7 ++++++- xen/include/xsm/dummy.h | 12 ++++++++++++ xen/include/xsm/xsm.h | 6 ++++++ xen/xsm/dummy.c | 1 + xen/xsm/flask/hooks.c | 15 +++++++++++++++ 7 files changed, 52 insertions(+), 1 deletion(-)