[PATCH v8 RESEND 27/38] iommu/amd: Allow the AMD IOMMU to work with memory encryption

[Tom Lendacky]

On 6/28/2017 4:36 AM, Joerg Roedel wrote:
> Hi Tom,

Hi Joerg,

> 
> On Tue, Jun 27, 2017 at 10:12:30AM -0500, Tom Lendacky wrote:
>> ---
>>   drivers/iommu/amd_iommu.c       |   30 ++++++++++++++++--------------
>>   drivers/iommu/amd_iommu_init.c  |   34 ++++++++++++++++++++++++++++------
>>   drivers/iommu/amd_iommu_proto.h |   10 ++++++++++
>>   drivers/iommu/amd_iommu_types.h |    2 +-
>>   4 files changed, 55 insertions(+), 21 deletions(-)
> 
> Looks like a straightforward change. Just one nit below.
> 
>> +static bool amd_iommu_supports_sme(void)
>> +{
>> +	if (!sme_active() || (boot_cpu_data.x86 != 0x17))
>> +		return true;
>> +
>> +	/* For Fam17h, a specific level of support is required */
>> +	if (boot_cpu_data.microcode >= 0x08001205)
>> +		return true;
>> +
>> +	if ((boot_cpu_data.microcode >= 0x08001126) &&
>> +	    (boot_cpu_data.microcode <= 0x080011ff))
>> +		return true;
>> +
>> +	pr_notice("AMD-Vi: IOMMU not currently supported when SME is active\n");
>> +
>> +	return false;
>> +}
> 
> The name of the function is misleading. It checks whether the IOMMU can
> be enabled when SME is active. But the name suggests that it checks
> whether the iommu hardware supports SME.
> 
> How about renaming it to amd_iommu_sme_check()?

Can do.

Thanks,
Tom

> 
> With that change the patch is:
> 
> 	Acked-by: Joerg Roedel <jroedel at suse.de>
>