Re: [PATCH 13/14] x86: BHI stubs

[Andrew Cooper]

On 30/09/2024 11:38 pm, Josh Poimboeuf wrote:
> On Mon, Sep 30, 2024 at 11:23:38PM +0100, Andrew Cooper wrote:
>> On 30/09/2024 10:30 pm, Josh Poimboeuf wrote:
>>> On Fri, Sep 27, 2024 at 09:49:09PM +0200, Peter Zijlstra wrote:
>>>> +SYM_INNER_LABEL(__bhi_args_0, SYM_L_LOCAL)
>>>> +	UNWIND_HINT_FUNC
>>>> +	cmovne %r10, %rdi
>>> IIUC, this works because if the "jz" in the CFI preamble mispredicts to
>>> the __bhi_args_* code, "cmovne" will zero out the speculative value of
>>> rdi.
>>>
>>> Why use %r10 instead of a literal $0?  Also how do you know %r10 is 0?
>> There's no encoding for CMOVcc which takes an $imm.
> Ah.
>
>> %r10 is guaranteed zero after the FineIBT prologue
> If the "jz" in the FineIBT prologue mispredicts, isn't %r10 non-zero by
> definition?

FineIBT uses SUB (and not CMP) to destroy the hash in %r10.

This makes it marginally harder to leak an unknown hash; you can't
trivially deference it, but there is a linear function if you know the
hash of the caller side.

In the bad speculation path, you're still overwriting pointers with a
number that is < 2^32, so will be stalled by SMAP if dereferenced.

>
>> , but I don't see
>> anything in patch 11 which makes this true in the !FineIBT case.
> I thought this code is only used by FineIBT?

Erm pass.  Peter?

~Andrew