Re: LuaJIT 1.1.5 issue/bug?

[Tim Mensch <tim-lua-l@...>]

Mike Pall wrote:
> Tim Mensch wrote:
> > I'm using LuaJIT 1.1.5 embedded in a Windows/Mac game SDK (Playground),  
> > and it's be working fine for me.
>
> Ah, good to know.
>   
Too bad I can't type, though. "It's be working..."  Ah well.
> But one thing: I'd like to ask you to make it clear for your users
> that you're using LuaJIT 1.x and not 2.x. Most of the talk on the
> Lua mailing list or elsewhere now focuses on 2.x features and
> performance. That could be a potential source of confusion for
> your users.
>   
OK, I'll be sure to include that in the docs. I will consider shifting 
to 2.x later, though I'm not sure that it will actually work as well for 
us: The alternate C stacks in LuaJIT 1.x mean that people can suspend 
their coroutines from within their C functions, and with C functions on 
the stack. I assume that won't work in LuaJIT 2.x because of the new way 
things work? Or am I wrong?
> > One of the SDK users, though, pointed  
> > out a problem reported when using the Microsoft Application Verifier.  
> > (http://bit.ly/SalYW)
>
> Congratulations! He found the first real bug since two years. :-)
>   
That's what made be doubt my own eyes. And his bug report at first, 
frankly. My gut was that the Application Verifier was broken with 
respect to JIT code, but then I realized this was happening before the 
JIT code was even generated, so I dug deeper.
> That's the correct fix.
>   
Glad I was able to help. :)

Tim


__________ Information from ESET NOD32 Antivirus, version of virus signature database 4969 (20100323) __________

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com