Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!swrinde!emory!att!princeton!phoenix.princeton.edu!doug From: doug@clarity.Princeton.EDU (Douglas Merrill) Newsgroups: comp.lang.lisp Subject: Running out of memory in Common Lisp 4.0 Message-ID: Date: 20 Jun 91 20:45:59 GMT Sender: news@idunno.Princeton.EDU Distribution: comp.lang.lisp Organization: Princeton University Cognitive Science Lab Lines: 46 Nntp-Posting-Host: clarity.princeton.edu We just started running Lucid Common Lisp 4.0 under SunOS 4.1.1, and have been having problems with running out of memory, garbage collection failures, and bus errors. Our network includes a Sun 4/280 fileserver and several SparcStation2's -- recently upgraded. Each SparcStation has about 32Mb of swap space. When running lisp code which formerly (under Lucid CL 3.0) worked for even very large datasets, we are finding that it now runs out of memory even with smaller ones. Usually it terminates with a message like this: ;;; Failed to re-enable EGC after the memory manager disabled it for memory ;;; reorganization: ;;; EGC could not be enabled because the dynamic GC is disabled. >>Error: GC and EGC disabled: not enough storage after GC. CONSers are now using the storage normally reserved for copying currently allocated dynamic storage. and sometimes even: ;;; Ran out of Memory during GC Process lisp stopped (signal) Our best guess is that garbage collection is simply failing to get rid of things it needs to get rid of, because 1) We've run bigger things on the earlier version 2) It shouldn't break with this kind of fatal error if it is working correctly. Has anyone else encountered a similar problem? Has a bug fix been published to correct this problem? Please email any suggestions or answers to ghn@clarity.Princeton.EDU (I'm posting this for him.) If there's enough interest, I'll post a summary to the net. Douglas Merrill doug@clarity.princeton.edu Princeton Univ. Cognitive Science Lab