Re: [PATCH 01/13] x86/HV: Initialize GHCB page in Isolation VM

[Tianyu Lan <ltykernel@gmail.com>]

Hi Joerg:
      Thanks for your review.


On 8/2/2021 7:53 PM, Joerg Roedel wrote:
> On Wed, Jul 28, 2021 at 10:52:16AM -0400, Tianyu Lan wrote:
>> +static int hyperv_init_ghcb(void)
>> +{
>> +	u64 ghcb_gpa;
>> +	void *ghcb_va;
>> +	void **ghcb_base;
>> +
>> +	if (!ms_hyperv.ghcb_base)
>> +		return -EINVAL;
>> +
>> +	rdmsrl(MSR_AMD64_SEV_ES_GHCB, ghcb_gpa);
>> +	ghcb_va = memremap(ghcb_gpa, HV_HYP_PAGE_SIZE, MEMREMAP_WB);
> 
> This deserves a comment. As I understand it, the GHCB pa is set by
> Hyper-V or the paravisor, so the page does not need to be allocated by
> Linux.
> And it is not mapped unencrypted because the GHCB page is allocated
> above the VTOM boundary?

You are right. The ghdb page is allocated by paravisor and its physical 
address is above VTOM boundary. Will add a comment to describe this.
Thanks for suggestion.

> 
>> @@ -167,6 +190,31 @@ static int hv_cpu_die(unsigned int cpu)
>>   {
>>   	struct hv_reenlightenment_control re_ctrl;
>>   	unsigned int new_cpu;
>> +	unsigned long flags;
>> +	void **input_arg;
>> +	void *pg;
>> +	void **ghcb_va = NULL;
>> +
>> +	local_irq_save(flags);
>> +	input_arg = (void **)this_cpu_ptr(hyperv_pcpu_input_arg);
>> +	pg = *input_arg;
> 
> Pg is never used later on, why is it set?

Sorry for noise. This should be removed during rebase and will fix in 
the next version.