|VM-friendly GC email@example.com (1992-02-08)|
|Re: VM-friendly GC firstname.lastname@example.org.COM (1992-02-13)|
|APL Compilation (WAS: VM-friendly GC) email@example.com (1992-02-14)|
|From:||firstname.lastname@example.org.COM (Robert Bernecky)|
|Organization:||Snake Island Research Inc,Toronto|
|Date:||Thu, 13 Feb 1992 12:11:49 -0500|
In article 92-02-039 email@example.com (Steve Boswell) writes:
>Has there been any work in VM-friendly garbage collection? My guess
>would be to do reference counting and a FIFO freespace list, as a
>start, and to split memory up into sub-heaps so marking doesn't cause
>all memory to be paged in.
See the latest issue of the IBM Systems Journal. It is totally devoted to
APL (It being the 25th anniversary of APL implementations), and there is a
fairly well-written article on a new storage manager which is used in
IBM's APL2 product. The manager was written specifically to cope with
problems inherent in manipulating the large sorts of arrays which you
encounter in image processing, etc. 128 megabytes and stuff like that.
Robert Bernecky firstname.lastname@example.org email@example.com
Snake Island Research Inc (416) 368-6944 FAX: (416) 360-4694
18 Fifth Street, Ward's Island, Toronto, Ontario M5J 2B9, Canada
[The exact reference is R. Trimble, "Storage Management in APL Systems,"
IBM Sys J 30:4, 1991, pp. 456-468, order number G321-5446. What I found
most surprising about the whole issue is that IBM seems to have done no
work at all on compiling APL, just more and better interpreters. -John]
Return to the
Search the comp.compilers archives again.