Re: [PATCHv3 3/3] x86/tdx: Handle load_unaligned_zeropad() page-cross to a shared page

[Sean Christopherson]

On Thu, May 26, 2022, Dave Hansen wrote:
> On 5/26/22 13:36, Kirill A. Shutemov wrote:
> > On Thu, May 26, 2022 at 09:20:56AM -0700, Dave Hansen wrote:
> >> On 5/24/22 15:10, Kirill A. Shutemov wrote:
> >>> +	/*
> >>> +	 * MMIO accesses suppose to be naturally aligned and therefore never
> >>> +	 * cross a page boundary. Seeing unaligned accesses indicates a bug or
> >>> +	 * load_unaligned_zeropad() that steps into unmapped shared page.
> >> Wait a sec though...
> >>
> >> We've been talking all along about how MMIO accesses are in some cases
> >> just plain old compiler-generated memory accesses.  It's *probably* bad
> >> code that does this, but it's not necessarily a bug.
> > Compiler-generated memory accesses tend to be aligned too. You need to do
> > something make them unalinged, like __packed or pointer trickery.
> 
> I totally agree.  But, the point is that __packed or pointer trickery is
> goofy, but it's not necessarily a bug.  This might crash the kernel on
> goofy stuff, not bugs.

Yeah, I don't think it's worth exploding on unaligned accesses, it's specifically
page splits that are a mess and are an absolutely nightmare to handle.  E.g. for
VirtIO kicks, the data and page offset are completely ignored/irrelevant, so a
multi-byte write to any random byte in the page should work, even though it's all
kinds of goofy.