Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cornell!rochester!pt.cs.cmu.edu!pt!dld From: dld@F.GP.CS.CMU.EDU (David Detlefs) Newsgroups: comp.lang.c++ Subject: RE: Garbage Collection Message-ID: Date: 13 Feb 89 23:30:30 GMT Organization: CMU CS Department Lines: 42 << Kevin Sullivan asks about garbage collection for C++. >> I've been doing a literature search in this area recently, and have found the following relevant references: Boehm & Weiser: "Garbage Collection in an Uncooperative Environment" Software Practice & Experience, September 88 Caplinger: "A Memory Allocator with Garbage Collection for C" Winter 88 Usenix Both of these present mark & sweep algorithms fairly compatible with the existing malloc/free/etc interface. Bartlett: "Compacting Garbage Collection with Ambiguous Roots" DEC WRL Research Report 88/2 Outlines an approach applicable to a copying collector for C. (To get, send a message to wrl-techreports@decwrl.dec.com, with body "help.") All of these methods are applicable, mutas mutandis, to C++. There is also a rumor that Chris Torek of Maryland is working on something along these lines (Chris, if you read this, could you confirm or deny to me, along with a pointer to a reference, or a description of what you're doing? Thanks...) Hopefully I've done a service. Perhaps the collective consciousness out there can help me in return. I would like to find some emprical evidence on just how much more reliable garbage collection makes programs. The ideal reference would be one that states "in project X, we found Y bugs, Z% of which were caused by improper explicit storage management. These took W% of the time to debug..." Any pointers to such a paper would be greatly appreciated. Thanks! -- Dave Detlefs Any correlation between my employer's opinion Carnegie-Mellon CS and my own is statistical rather than causal, dld@cs.cmu.edu except in those cases where I have helped to form my employer's opinion. (Null disclaimer.) --