bug#13074: VM Segfaults with Bad `Call' Instruction

[Noah Lavine]

On Wed, Dec 5, 2012 at 5:14 PM, Ludovic Courtès <address@hidden> wrote:

Noah Lavine <address@hidden> skribis:

> That makes sense. I hit this error in debugging a CPS->GLIL compiler (which
> I hope will become Guile's compiler, but that's another story). However,
> once the debugging is done, I suppose it won't make a difference.

Oooh, make sure to post about your plans.

I will post more when I have more code to show, but basically it's the same idea as the CPS-to-RTL experiment earlier. The difference is that in that post I said that adding two new things at the same time (CPS and RTL) was probably a bad idea. Now I'm doing something about it, by making the CPS compiler generate GLIL instead. I hope this will be an easier path towards a nicer compiler.

 

> What do you think about enabling it only in the debug VM, or something like
> that? Then if there's some way for me to run my code in debug mode, I can
> get the better output without slowing down most things.

I’m inclined to leave it as is, because it’s only hit when generating
wrong code.  How strongly do you feel about it?  :-)

Well, I just fixed the bug, so I feel fine right now. :-)

In general, I do think there should at least be an option for having full error-checking in the VM. It would have been much, much harder for me to find this without having patched the VM, because it would have taken me a very long time to try each new thing I tried, because I would have had to restart Guile. I am happy for it not to be on the regular code-path, though. I also realize that writing a compiler is an unusual application, so maybe it should even be a compile-time option for users who prefer their Guile slow. How does that sound?

Noah