Re: [tip:x86/mm] x86/mm: Add support for early encryption/decryption of memory

[Tom Lendacky]

On 10/25/2017 12:34 PM, Dave Hansen wrote:
> On 07/18/2017 03:51 AM, tip-bot for Tom Lendacky wrote:
> > +/*
> > + * This routine does not change the underlying encryption setting of the
> > + * page(s) that map this memory. It assumes that eventually the memory is
> > + * meant to be accessed as either encrypted or decrypted but the contents
> > + * are currently not in the desired state.
> > + *
> > + * This routine follows the steps outlined in the AMD64 Architecture
> > + * Programmer's Manual Volume 2, Section 7.10.8 Encrypt-in-Place.
> > + */
> > +static void __init __sme_early_enc_dec(resource_size_t paddr,
> > +				       unsigned long size, bool enc)
> > +{
> > +	void *src, *dst;
> > +	size_t len;
> > +
> > +	if (!sme_me_mask)
> > +		return;
> > +
> > +	local_flush_tlb();
> > +	wbinvd();
>
> What is this TLB flush for?  WBINVD is mentioned in "Volume 2, Section
> 7.10.8", but TLB flushes are not.  Is it OK that this does not flush
> global pages?

The local_flush_tlb() is not needed and can be removed. I had that in
there from early development and testing and just missed removing it.
I can queue up a patch to delete it and remove the ambiguity with this
routine and the APM document.

Thanks,
Tom

>