On 10/14/15 13:37, Paolo Bonzini wrote:
> On 13/10/2015 15:26, Laszlo Ersek wrote:
>>>    //
>>> +  // The write to the control register is synchronous and only affects the
>>> +  // current CPU, so bring in the APs first.  The SMI handler expects that
>>> +  // all APs will rendez-vous within one PcdCpuSmmApSyncTimeout (though it
>>> +  // helpfully tries sending SMI IPIs to the missing processors if there 
>>> are
>>> +  // any).
>>> +  //
>>> +  SendSmiIpiAllExcludingSelf ();
>>> +
>>> +  //
> 
> Nevermind, this patch is unfortunately broken.
> 
> SendSmiIpiAllExcludingSelf does not work after ExitBootServices, because
> it hardcodes the physical address of the APIC.  While that could be
> fixed in LocalApicLib, it turns out that using SmmSyncModeRelaxedAp is a
> trivial addition to my series that introduces a SmmCpuFeaturesLib
> implementation specific to OvmfPkg.  Therefore, I'm withdrawing this
> patch and submitting another in that thread.

Thank you.

If you create the SmmCpuFeaturesLib instance as a copy, initially,
please make sure that you use, as basis, the CRLF-converted patches of
Mike that I pushed last night. So that the copy come from a
CRLF-terminated source, and your further customizations be similar.

Thank you!
Laszlo

_______________________________________________
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel

Reply via email to