Re: [Xen-devel] [PATCH RFC V8 1/5] xen: Emulate with no writes

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 27.08.14 at 16:01, <rcojocaru@xxxxxxxxxxxxxxx> wrote:
> Added support for emulating an instruction with no memory writes.
> Additionally, introduced hvm_emulate_one_full(), which inspects
> possible return values from the hvm_emulate_one() functions
> (EXCEPTION, UNHANDLEABLE) and acts on them.
> 
> Changes since V1:
>  - Removed the Linux code that computes the length of an instruction.
>  - Unused function parameters are no longer marked.
>  - Refactored the code to eliminate redundancy.
>  - Made the exception passed on to the guest by hvm_emulate_one_full()
>    configurable.
> 
> Changes since V2:
>  - Renamed hvmemul_write_dummy() to hvmemul_write_discard().
>  - Fixed a comment (Xen coding style rules).
>  - Renamed hvm_emulate_one_with_ops() to _hvm_emulate_one().
>  - Changed stack variable hvm_emulate_ops_no_write into a static const.
>  - Modified struct hvm_emulate_ctxt ctx initialization syntax.
>  - Renamed unhandleable_trapnr to trapnr and unhandleable_errcode
>    to errcode.
>  - Changed errcode's type to unsigned int.
>  - Removed hvm_emulate_one() loop that went on until it returned
>    something other than X86EMUL_RETRY (to prevent potential
>    blocking against the time calibration rendezvous).
> 
> Changes since V3:
>  - The rep_ins, rep_movs and cmpxchg handlers are
>    now also inactive.
> 
> Changes since V4:
>  - Also discarding IO reads (dummy read_io handler).
> 
> Changes since V5:
>  - Added discard versions of write_io and rep_outs.
> 
> Changes since V6:
>  - Added discard versions of write_msr and wbinvd.
>  - Now calling hvm_emulate_writeback() for the X86EMUL_UNHANDLEABLE
>    case as well.
> 
> Changes since V7:
>  - Renamed hvm_emulate_one_full() to hvm_mem_event_emulate_one().
>  - Added a comment explaining how X86EMUL_RETRY is being treated.
> 
> Signed-off-by: Razvan Cojocaru <rcojocaru@xxxxxxxxxxxxxxx>

Acked-by: Jan Beulich <jbeulich@xxxxxxxx>

But btw., please get used to putting the version history after the
first --- separator. And commonly such history is more useful if
sorted in reverse chronological order.

Jan


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