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

Re: [Xen-devel] about vmxassist


  • To: yushang <yusunn@xxxxxxxxx>
  • From: Keir Fraser <keir.fraser@xxxxxxxxxxxxx>
  • Date: Sat, 08 Mar 2008 17:01:09 +0000
  • Cc: xen-devel@xxxxxxxxxxxxxxxxxxx
  • Delivery-date: Sat, 08 Mar 2008 09:02:25 -0800
  • List-id: Xen developer discussion <xen-devel.lists.xensource.com>
  • Thread-index: AciBPgFPP5tZpu0xEdyqtQAX8io7RQ==
  • Thread-topic: [Xen-devel] about vmxassist

Yes, that's why it replaces vmxassist! It more accurately simulates the
real-mode environment than vmxassist was able to, hence more real-mode code
runs correctly under full emulation (e.g., a wider range of bootloaders).

 -- Keir

On 8/3/08 16:55, "yushang" <yusunn@xxxxxxxxx> wrote:

> Thanks . I see . It seems the real-mode emulator is more advanced than
> vmxassist ?
> 
> 2008/3/9, Keir Fraser <keir.fraser@xxxxxxxxxxxxx>:
>> On 8/3/08 07:11, "yushang" <yusunn@xxxxxxxxx> wrote:
>> 
>>>>> I find the code path
>>>>> will enter the native real mode if do not use vmxassist , not a
>>>>> emulation one .
>>> Sorry but I mean the code path of hvmloader , just at the beginning of
>>> hvmloader, if do not use vmxassist execution will return from main and
>>> enter real-address mode then jmp to bios , so bios will run in
>>> real-address mode , is this true ?
>> 
>> 
>> Yes, as soon as real mode is entered (CR0 bit 0 is cleared) Xen notices that
>>  and runs the guest through the full real-mode emulator until CR0 bit 0 is
>>  set again.
>> 
>> 
>>   -- Keir
>> 
>> 
>> 



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

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