[Bug-mit-scheme] [bug #36429] Compiler mysteriously runs out of stack space

[anonymous]

URL:
  <http://savannah.gnu.org/bugs/?36429>

                 Summary: Compiler mysteriously runs out of stack space
                 Project: MIT/GNU Scheme
            Submitted by: None
            Submitted on: Mon 07 May 2012 09:21:01 PM UTC
                Category: compiler
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: None
                  Status: None
                 Privacy: Public
             Assigned to: None
         Originator Name: Alexey Radul
        Originator Email: address@hidden
             Open/Closed: Open
         Discussion Lock: Any
                Keywords: 

    _______________________________________________________

Details:

Hi all,

I stumbled across a bug that causes the compiler to exhaust its
recursion depth.  The enclosed file produces the following bizarre
interaction in a fresh MIT Scheme:

MIT/GNU Scheme running under GNU/Linux
Type `^C' (control-C) followed by `H' to obtain information about interrupts.

Copyright (C) 2011 Massachusetts Institute of Technology
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Image saved on Tuesday November 8, 2011 at 10:45:46 PM
  Release 9.1.1     || Microcode 15.3 || Runtime 15.7 || SF 4.41
  LIAR/x86-64 4.118 || Edwin 3.116

1 ]=> (cf "foo")

;Generating SCode for file: "foo.scm" => "foo.bin"...
;Warning: Unreferenced defined variable: face-color-list
;... done
;Compiling file: "foo.bin" => "foo.com"... aborted
;Aborting!: maximum recursion depth exceeded

1 ]=> 

I would not expect reasonably normal code to make the compiler run
out of stack.

Now, the enclosed file does contain a mistake: one of the procedures
defined therein always recursively calls itself (and even not in
tail position).  However, my attempts to produce a small example
using this to cause the compiler to run out of stack failed.  I also
got very confused when I tried to reduce the present example,
because commenting out a form I would not have expected to have
anything to do with anything caused the misbehavior to disappear.

Cheers,
~Alexey




    _______________________________________________________

File Attachments:


-------------------------------------------------------
Date: Mon 07 May 2012 09:21:01 PM UTC  Name: foo.scm  Size: 3kB   By: None

<http://savannah.gnu.org/bugs/download.php?file_id=25815>

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?36429>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/