Re: [RFC PATCH v2 16/20] x86: Check for memory encryption on the APs

[Tom Lendacky]

On 09/12/2016 07:17 AM, Borislav Petkov wrote:
> On Mon, Aug 22, 2016 at 05:38:29PM -0500, Tom Lendacky wrote:
>> Add support to check if memory encryption is active in the kernel and that
>> it has been enabled on the AP. If memory encryption is active in the kernel
>> but has not been enabled on the AP then do not allow the AP to continue
>> start up.
>>
>> Signed-off-by: Tom Lendacky <thomas.lendacky@xxxxxxx>
>> ---
>>  arch/x86/include/asm/msr-index.h     |    2 ++
>>  arch/x86/include/asm/realmode.h      |   12 ++++++++++++
>>  arch/x86/realmode/init.c             |    4 ++++
>>  arch/x86/realmode/rm/trampoline_64.S |   19 +++++++++++++++++++
>>  4 files changed, 37 insertions(+)
> 
> ...
> 
>> diff --git a/arch/x86/realmode/rm/trampoline_64.S b/arch/x86/realmode/rm/trampoline_64.S
>> index dac7b20..94e29f4 100644
>> --- a/arch/x86/realmode/rm/trampoline_64.S
>> +++ b/arch/x86/realmode/rm/trampoline_64.S
>> @@ -30,6 +30,7 @@
>>  #include <asm/msr.h>
>>  #include <asm/segment.h>
>>  #include <asm/processor-flags.h>
>> +#include <asm/realmode.h>
>>  #include "realmode.h"
>>  
>>  	.text
>> @@ -92,6 +93,23 @@ ENTRY(startup_32)
>>  	movl	%edx, %fs
>>  	movl	%edx, %gs
>>  
>> +	/* Check for memory encryption support */
>> +	bt	$TH_FLAGS_SME_ENABLE_BIT, pa_tr_flags
>> +	jnc	.Ldone
>> +	movl	$MSR_K8_SYSCFG, %ecx
>> +	rdmsr
>> +	bt	$MSR_K8_SYSCFG_MEM_ENCRYPT_BIT, %eax
>> +	jc	.Ldone
>> +
>> +	/*
>> +	 * Memory encryption is enabled but the MSR has not been set on this
>> +	 * CPU so we can't continue
> 
> Hmm, let me try to parse this correctly: BSP has SME enabled but the
> BIOS might not've set this on the AP? Really? Is that even possible?

Anything is possible, although it's highly unlikely.

> 
> Because if SME is enabled, that means that MSR_K8_SYSCFG[23] on the BSP
> is set, right?

Correct.

> 
> Also, I want to rule out here simple BIOS idiocy: if the only problem
> with the bit not being set in the AP is because some BIOS monkey forgot
> to do so, then we should try to set it ourselves and not die for no real
> reason.

Yes, we can do that.  I was debating on which way to go with this. Most
likely this would never happen, but if it did...  I can change this to
set the MSR bit and continue.

Thanks,
Tom

> 
> Or is there another issue?
>