Re: [Xen-devel] [PATCH v4] x86/hvm: Allow guest_request vm_events coming from userspace

[Razvan Cojocaru <rcojocaru@xxxxxxxxxxxxxxx>]

On 08/05/2017 04:32 AM, Tamas K Lengyel wrote:
> 
> 
> On Fri, Aug 4, 2017 at 5:32 AM, Alexandru Isaila
> <aisaila@xxxxxxxxxxxxxxx <mailto:aisaila@xxxxxxxxxxxxxxx>> wrote:
> 
>     In some introspection usecases, an in-guest agent needs to communicate
>     with the external introspection agent.  An existing mechanism is
>     HVMOP_guest_request_vm_event, but this is restricted to kernel usecases
>     like all other hypercalls.
> 
>     Introduce a mechanism whereby the introspection agent can whitelist the
>     use of HVMOP_guest_request_vm_event directly from userspace.
> 
>     Signed-off-by: Alexandru Isaila <aisaila@xxxxxxxxxxxxxxx
>     <mailto:aisaila@xxxxxxxxxxxxxxx>>
> 
>     ---
>     Changes since V3:
>             - Changed commit message
>             - Added new lines
>             - Indent the maximum space on the defines
>             - Chaned the name of the define/function name/struct member
>               from vmcall to event
>     ---
>      tools/libxc/include/xenctrl.h |  1 +
>      tools/libxc/xc_monitor.c      | 14 ++++++++++++++
>      xen/arch/x86/hvm/hypercall.c  |  5 +++++
>      xen/common/monitor.c          | 14 ++++++++++++++
>      xen/include/public/domctl.h   | 21 +++++++++++----------
>      xen/include/xen/sched.h       |  5 +++--
>      6 files changed, 48 insertions(+), 12 deletions(-)
> 
>     diff --git a/tools/libxc/include/xenctrl.h
>     b/tools/libxc/include/xenctrl.h
>     index bde8313..90a056f 100644
>     --- a/tools/libxc/include/xenctrl.h
>     +++ b/tools/libxc/include/xenctrl.h
>     @@ -2022,6 +2022,7 @@ int xc_monitor_descriptor_access(xc_interface
>     *xch, domid_t domain_id,
>                                       bool enable);
>      int xc_monitor_guest_request(xc_interface *xch, domid_t domain_id,
>                                   bool enable, bool sync);
>     +int xc_allow_guest_userspace_event(xc_interface *xch, domid_t
>     domain_id, bool enable);
>      int xc_monitor_debug_exceptions(xc_interface *xch, domid_t domain_id,
>                                      bool enable, bool sync);
>      int xc_monitor_cpuid(xc_interface *xch, domid_t domain_id, bool
>     enable);
>     diff --git a/tools/libxc/xc_monitor.c b/tools/libxc/xc_monitor.c
>     index b44ce93..6064c39 100644
>     --- a/tools/libxc/xc_monitor.c
>     +++ b/tools/libxc/xc_monitor.c
>     @@ -161,6 +161,20 @@ int xc_monitor_guest_request(xc_interface *xch,
>     domid_t domain_id, bool enable,
>          return do_domctl(xch, &domctl);
>      }
> 
>     +int xc_allow_guest_userspace_event(xc_interface *xch, domid_t
>     domain_id, bool enable)
> 
> 
> This function should be prefixed with "xc_monitor_" like all the rest of
> the functions here.
That one was my suggestion, as I thought xc_monitor_-prefixed functions
are meant to toggle monitoring somehow, whereas this function only
toggles userspace use of guest request VMCALLs.

But re-adding the prefix is fine - since toggling it only influences in
the end how a vm_event is sent, it is after all monitor-related.


Thanks for the review,
Razvan

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel