From: Tim Bates <tim@...>
Date: 2005-03-19T09:29:55+09:00
Subject: Re: Maximum stack depth

Csaba Henk wrote:
> In my ruby installations under gentoo, there is no "stack level too
> deep", ruby just eats up all memory and then the OS gracefully kills
> it (it happens when I don't link it against pthread).

I get the same thing (also on Gentoo), except with 1GB of RAM and 2GB of 
swap my system slows to a crawl long before the operating system kills 
it and my patience runs out. It can take some minutes to kill the 
program once I realise this is happening - which is generally when my 
swap space usage goes above 0%. I think I might actually prefer a stack 
level too deep error. Usually by the time my system starts swapping, the 
stack is over 200,000 levels deep.

> It seems that function data is stored on the heap. Is it possible? Under
> what circumstances can this phenomenon -- ie., that ruby doesn't runs
> out of stack upon a deep recursion -- occur (OS, kernel (OS, compilation
> options, compilation options, ...) ?

After a bit of research, I've discovered that Gentoo by default doesn't 
have a stack size limit. Watch: (using the bash builtin `ulimit`)

$ ulimit -a
<snip>
stack size            (kbytes, -s) unlimited
<snip>

$ ulimit -s 1024
$ ruby -e 'def t(i) t(i+1) end;t 0'
-e:1:in `t': stack level too deep (SystemStackError)
         from -e:1:in `t'
         from -e:1:in `t'
         from -e:1:in `t'
         from -e:1:in `t'
         from -e:1:in `t'
         from -e:1:in `t'
         from -e:1:in `t'
         from -e:1:in `t'
          ... 401 levels...
         from -e:1:in `t'
         from -e:1:in `t'
         from -e:1:in `t'
         from -e:1

Tim.

-- 
Tim Bates
tim@bates.id.au