[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH v9 00/20] x86/PMU: Xen PMU PV(H) support

>>> On 08.08.14 at 18:55, <boris.ostrovsky@xxxxxxxxxx> wrote:
> Here is the ninth version of PV(H) PMU patches.
> Changes in v9:
> * Restore VPMU context after context_saved() is called in
>   context_switch(). This is needed because vpmu_load() may end up
>   calling vmx_vmcs_try_enter()->vcpu_pause() and that needs is_running
>   to be correctly set/cleared. (patch 18, dropped review acks)
> * Added patch 2 to properly manage VPMU_CONTEXT_LOADED
> * Addressed most of Jan's comments.
>   o Keep track of time in vpmu_force_context_switch() to properly break
>     out of a loop when using hypercall continuations
>   o Fixed logic in calling vpmu_do_msr() in emulate_privileged_op()
>   o Cleaned up vpmu_interrupt() wrt vcpu variable names to (hopefully)
>     make it more clear which vcpu we are using
>   o Cleaned up vpmu_do_wrmsr()
>   o Did *not* replace sizeof(uint64_t) with sizeof(variable) in
>     amd_vpmu_initialise(): throughout the code registers are declared as
>     uint64_t and if we are to add a new type (e.g. reg_t) this should be
>     done in a separate patch, unrelated to this series.

Suspicious, but I hopefully won't forget to comment on this when
reaching the actual patch.

In any event, I think you continue to send this series to an
incomplete list of people (possibly because you didn't update it
after the initial version): Both SVM and VMX have one more
maintainer than you Cc (Aravind and Eddie respectively; oddly
enough you send the patch directly to two of the others).


Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.