Re: [PATCH] x86: use WRITE_ONCE() when setting PTEs

[Dave Hansen <dave.hansen@intel.com>]

On 09/02/2018 11:14 AM, Nadav Amit wrote:
> When page-table entries are set, the compiler might optimize their
> assignment by using multiple instructions to set the PTE. This might
> turn into a security hazard if the user somehow manages to use the
> interim PTE. L1TF does not make our lives easier, making even an interim
> non-present PTE a security hazard.
> 
> Using WRITE_ONCE() to set PTEs and friends should prevent this potential
> security hazard.

But, our types are already 64-bit, and we're doing a 64-bit pointer
write.  Our WRITE_ONCE() implementation boils down to:

static __always_inline void __write_once_size(...
{
        switch (size) {
	...
        case 8: *(volatile __u64 *)p = *(__u64 *)res; break;


For 64-bit types, which is precisely the same thing.  Right?

So, while I agree that it's nice to document the need for a full 64-bit
"atomicity" of the PTE set, I don't see a practical problem here with
our implementation.

There's probably a massive number of things that would break if we
assumed sane 64-bit writes can be observed piecemeal.