lkml.org 
[lkml]   [2010]   [Jun]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: [PATCH 12/17] Add helper function get_kernel_ns
    On Wed, Jun 16, 2010 at 04:12:46PM +0800, Jason Wang wrote:
    > Zachary Amsden wrote:
    > > On 06/14/2010 10:41 PM, Avi Kivity wrote:
    > >
    > >> On 06/15/2010 10:34 AM, Zachary Amsden wrote:
    > >>
    > >>> Add a helper function for the multiple places this is used. Note
    > >>> that it
    > >>> must not be called in preemptible context, as that would mean the kernel
    > >>> could enter software suspend state, which would cause non-atomic
    > >>> operation
    > >>> of the monotonic_to_bootbased computation.
    > >>>
    > >>> Open question: should the KVM_SET_CLOCK / KVM_GET_CLOCK ioctls use this
    > >>> as well? Currently, they are not bootbased (but perhaps should be).
    > >>>
    > >>> Signed-off-by: Zachary Amsden<zamsden@redhat.com>
    > >>> ---
    > >>> arch/x86/kvm/x86.c | 26 +++++++++++++-------------
    > >>> 1 files changed, 13 insertions(+), 13 deletions(-)
    > >>>
    > >>> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
    > >>> index 703ea43..15c7317 100644
    > >>> --- a/arch/x86/kvm/x86.c
    > >>> +++ b/arch/x86/kvm/x86.c
    > >>> @@ -915,6 +915,16 @@ static void kvm_get_time_scale(uint32_t
    > >>> scaled_khz, uint32_t base_khz,
    > >>> __func__, base_khz, scaled_khz, shift, *pmultiplier);
    > >>> }
    > >>>
    > >>> +static inline u64 get_kernel_ns(void)
    > >>> +{
    > >>> + struct timespec ts;
    > >>> +
    > >>> + WARN_ON(preemptible());
    > >>> + ktime_get_ts(&ts);
    > >>> + monotonic_to_bootbased(&ts);
    > >>> + return timespec_to_ns(&ts);
    > >>> +}
    > >>> +
    > >>> static DEFINE_PER_CPU(unsigned long, cpu_tsc_khz);
    > >>>
    > >>>
    > >> Isn't something like this a candidate for the time infrastructure?
    > >>
    > >>
    > >
    > > Should it be? It certainly seems reasonable.
    > >
    > >
    > > Also, should we be using it for the cases KVM_GET_CLOCK /
    > > KVM_SET_CLOCK? It seems global kvmclock_offset for the VM ignores the
    > > bootbased conversion.
    > >
    > >
    > Yes we should.
    indeed. Also shows the importance of doing it in a central place.
    Have we had an acessor like that before, we would probably not have
    made this mistake.





    \
     
     \ /
      Last update: 2010-06-16 16:07    [W:0.024 / U:0.908 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site