Re: Time constraint in Lua pattern functions

[Coda Highland <chighland@...>]

On Wed, Mar 30, 2016 at 10:50 AM, Roberto Ierusalimschy
<roberto@inf.puc-rio.br> wrote:
>> Given that the code is not yet released, I can't inspect it myself. ;)
>
> The code did not change (except for the type of 'hookmask' that,
> as I mentioned, went from 'lu_byte' to 'l_signalT'). I simply added
> an explanation of why the code as it is should work.
>
>
>> But are you using compiler intrinsic atomics when they're available?
>
> No.
>
> -- Roberto
>

I suppose that's not that big of a deal, the more I think about it;
true atomics are only necessary when the process is multithreaded,
which isn't true for the reference shell.

/s/ Adam