Re: [Sbcl-devel] sbcl crashes on bignum operations

[William H. N. <wil...@ai...>]

On Fri, Oct 08, 2004 at 04:55:48PM +0200, Bruno Haible wrote:
> SBCL 0.8.13 crashes on a simple sequence of bignum operations:
> 
> * (integer-length (sqrt (ash 1 1000000)))
> 
> debugger invoked on a SIMPLE-TYPE-ERROR in thread 8303:
>   Too large to be represented as a DOUBLE-FLOAT:
> Argh! gc_find_freeish_pages failed (restart_page), nbytes=110112.
>    Gen Boxed Unboxed LB   LUB  !move  Alloc  Waste   Trig    WP  GCs Mem-age
>    0:     3     0  1053     0     0  4305304 20072  2000000    0   0  0.0000
>    1:     2     0     0 12746 12748 50956368 1259440  2000000    0   0  0.5006
>    2:    11     1     0 117227 117237 471929536 8281408 346429680    1   1  0.7298
>    3:     0     0     0     0     0        0     0  2000000    0   0  0.0000
>    4:     0     0     0     0     0        0     0  2000000    0   0  0.0000
>    5:     0     0     0     0     0        0     0  2000000    0   0  0.0000
>    6:     0     0     0     0     0        0     0        0    0   0  0.0000
>    Total bytes allocated=527191208
> fatal error encountered in SBCL pid 8303
> The system is too badly corrupted or confused to continue at the Lisp
> level. If the system had been compiled with the SB-LDB feature, we'd drop
> into the LDB low-level debugger now. But there's no LDB in this build, so
> we can't really do anything but just exit, sorry.
> $ 

Yikes.

The same failure occurs in my 0.8.15.8 here, and even on just bare
  (SQRT (ASH 1 1000000))
or 
  (ASH 1 1000000)
without the INTEGER-LENGTH or SQRT. The problem seems to be in the
printer; 
  (DEFPARAMETER *FOO* (ASH 1 1000000))
is OK.

Unless someone cleverly comes up with a fix before I get around to it
(hopefully soon, certainly before the next release), I will at least
log it in the BUGS file. Thank you for pointing it out.

-- 
William Harold Newman <wil...@ai...>
Such a new mode would surely count as "the next really big enormous thing."
  -- <http://hanson.gmu.edu/bigthing.html>