[#290] — Florian Frank <flori@...>
Hi all,
5 messages
2002/08/03
[#297] GC longjmp macros — Michal Rokos <m.rokos@...>
Hi,
5 messages
2002/08/05
[#308] Q: OSSL in std. distr? — Michal Rokos <m.rokos@...>
Hi,
4 messages
2002/08/08
[#326] Implications of a #force_free method in Object? — Matthew Bloch <mattbee@...>
Hello;
8 messages
2002/08/19
[#328] Int vs Long — Michal Rokos <m.rokos@...>
Hi,
7 messages
2002/08/21
[#337] Int vs Long (2nd part) — Michal Rokos <m.rokos@...>
Hi,
7 messages
2002/08/22
[#340] Int vs Long #3 — Michal Rokos <m.rokos@...>
Hi,
9 messages
2002/08/22
[#344] Re: [Cleanup] Int vs Long #3
— nobu.nokada@...
2002/08/22
Hi,
[#348] Re: [Cleanup] Int vs Long #3
— Michal Rokos <m.rokos@...>
2002/08/23
Hello,
[#353] File (struct stat handling) — Michal Rokos <m.rokos@...>
Hello,
6 messages
2002/08/23
[#358] node.h for eval.c — Michal Rokos <m.rokos@...>
Hi,
5 messages
2002/08/23
[#372] rb_class_path — Michal Rokos <m.rokos@...>
Hello,
7 messages
2002/08/27
[#382] Port match to new dup, clone framework — Michal Rokos <m.rokos@...>
Hi,
5 messages
2002/08/28
[#393] in dln.c — Michal Rokos <m.rokos@...>
Hi,
14 messages
2002/08/30
[#398] Re: [MemLeak] in dln.c
— nobu.nokada@...
2002/08/31
Hi,
[#403] Re: [MemLeak] in dln.c
— Michal Rokos <m.rokos@...>
2002/09/02
Hello,
Re: Implications of a #force_free method in Object?
From:
nobu.nokada@...
Date:
2002-08-19 12:25:46 UTC
List:
ruby-core #327
Hi, At Mon, 19 Aug 2002 20:29:05 +0900, Matthew Bloch wrote: > big_image.force_free > > Which would simply free the memory used by that object and nilify the > reference, no questions asked. It sounds very dangerous, and the method wouldn't be able to change the reference itself. I guess it requires new syntax. Rather, if single object, big_image consumes huge memory, it should have a method to free its resouces? > Obviously users of this method would have the > burden of ensuring that no further references to this object exist, but the > garbage collector could be used to spot such 'dangling reference' errors, and > of course the runtime will fault it if such a reference is used. It would need to scan whole space as GC to detect such 'dangling reference', and should be expensive equivalently to your current method, nilfying and starting GC. -- Nobu Nakada