Re: [PATCH v2] x86: bring back rep movsq for user access on CPUs without ERMS

[Mateusz Guzik]

On 9/3/23, Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
> On Sun, 3 Sept 2023 at 14:06, Mateusz Guzik <mjguzik@xxxxxxxxx> wrote:
>>
>> I don't think it is *that* bad. I did a quick sanity check on that
>> front by rolling with bpftrace on cases which pass AT_EMPTY_PATH *and*
>> provide a path.
>
> I guess you are right - nobody sane would use AT_EMPTY_PATH except if
> they don't have a path.
>
> Of course, the only reason we're discussing this in the first place is
> because people are doing insane things, which makes _that_ particular
> argument very weak indeed...
>

I put blame on whoever allowed non-NULL path and AT_EMPTY_PATH as a
valid combination, forcing the user buf to be accessed no matter what.
But I'm not going to go digging for names. ;)

-- 
Mateusz Guzik <mjguzik gmail.com>