Jan,
Thanks for reminder. I ever noticed this problem, but I forgot to
modify the code in that version.
Attached is the updated one.
Best Regards,
-- Dongxiao
Jan Beulich wrote:
>>>> "Xu, Dongxiao" <dongxiao.xu@xxxxxxxxx> 11.12.09 12:53 >>>
>> Add RDTSCP instruction support for HVM VMX guest.
>> - RDTSCP is introduced in Nehalem processor on Intel platform. Like
>> RDTSC, RDTSCP will return the TSC value, besides, it will return the
>> low 32bit of TSC_AUX MSR. Currently Linux kernel will write (node_id
>> << 12 | process_id) into that MSR, so that when guest execs RDTSCP,
>> it will also get processor information. - This instruction is
>> supported for HVM only when the hardware has this capability
>> (indicated by cpuid).
>
> This
>
>> --- a/xen/arch/x86/hvm/vmx/vmx.c Tue Dec 08 14:14:27 2009 +0000
>> +++ b/xen/arch/x86/hvm/vmx/vmx.c Fri Dec 11 19:10:53 2009 +0800
>> @@ -138,7 +138,7 @@ static DEFINE_PER_CPU(struct vmx_msr_sta
>>
>> static u32 msr_index[VMX_MSR_COUNT] =
>> {
>> - MSR_LSTAR, MSR_STAR, MSR_SYSCALL_MASK
>> + MSR_LSTAR, MSR_STAR, MSR_SYSCALL_MASK, MSR_TSC_AUX };
>>
>> static void vmx_save_host_msrs(void)
>> @@ -146,8 +146,11 @@ static void vmx_save_host_msrs(void)
>> struct vmx_msr_state *host_msr_state =
>> &this_cpu(host_msr_state); int i;
>>
>> - for ( i = 0; i < VMX_MSR_COUNT; i++ )
>> + for ( i = 0; i < VMX_MSR_COUNT - 1; i++ )
>> rdmsrl(msr_index[i], host_msr_state->msrs[i]); +
>> + if ( cpu_has_rdtscp )
>> + rdmsrl(MSR_TSC_AUX,
>> host_msr_state->msrs[VMX_INDEX_MSR_TSC_AUX]); }
>>
>> #define WRITE_MSR(address)
>> \
>
> seems to be calling for future problems - if anyone adds an element to
> msr_index[] before MSR_TSC_AUX, the logic will silently break. A
> BUILD_BUG_ON() and a comment would be the minimum thing to add
> in my opinion.
>
> Jan
02-add-hvm-rdtscp-support.patch
Description: 02-add-hvm-rdtscp-support.patch
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
Home