From: <fc...@gn...> - 2002-10-06 16:20:42
|
When loading CLLIB, CLISP 2.30 segfaults on a Debian GNU/Linux with 375Mb of memory, unlimited stack size. Can anyone reproduce this with Debian or other OSes? I've tried this with the latest CLISP package from Debian/unstable, so it could be a problem specific to that version. This happens under a standard (latest CVS) CLOCC installation, by running (MK:OOS 'CLLIB 'LOAD) in CLISP. I tried loading the files individually, and some crash, others don't. log.lisp is one, rng.lisp also dies horribly. [1]> (load (translate-logical-pathname "clocc:src;cllib;rng")) ;; Loading file /home/fc/werk/lisp/clocc/clocc/src/cllib/rng.fas ... ;; Loading file /home/fc/werk/lisp/clocc/clocc/src/cllib/base.fas ... ;; Loading file /home/fc/werk/lisp/clocc/clocc/src/port/ext.fas ... ;; Loaded file /home/fc/werk/lisp/clocc/clocc/src/port/ext.fas ;; Loading file /home/fc/werk/lisp/clocc/clocc/src/port/sys.fas ... ;; Loading file /home/fc/werk/lisp/clocc/clocc/src/port/path.fas ... ;; Loaded file /home/fc/werk/lisp/clocc/clocc/src/port/path.fas ;; Loaded file /home/fc/werk/lisp/clocc/clocc/src/port/sys.fas ;; Loaded file /home/fc/werk/lisp/clocc/clocc/src/cllib/base.fas ;; Loading file /home/fc/werk/lisp/clocc/clocc/src/cllib/withtype.fas ... ;; Loaded file /home/fc/werk/lisp/clocc/clocc/src/cllib/withtype.fas *** - handle_fault error2 ! address = 0x2045313C not in [0x20204000,0x203BCC48) ! SIGSEGV cannot be cured. Fault address = 0x2045313C. Segmentation fault Strangely, on the particular case of log.lisp, when I do a (load (translate-logical-pathname "clocc:src;cllib;log")) under CLISP's prompt, it loads fine. When I put it into .CLISPRC it crashes. Thanks! fc -- http://raw.no-ip.com/~fc/ |