src/Pure/Concurrent/future.ML
Mon, 17 Oct 2016 16:58:39 +0200 wenzelm eliminated unused argument;
Mon, 05 Sep 2016 23:11:00 +0200 wenzelm clarified modules;
Sat, 09 Apr 2016 14:17:50 +0200 wenzelm tuned signature;
Sat, 09 Apr 2016 14:11:31 +0200 wenzelm tuned signature -- closer to Exn.Interrupt.expose in Scala;
Sat, 09 Apr 2016 14:00:23 +0200 wenzelm clarified bootstrap;
Wed, 06 Apr 2016 16:33:33 +0200 wenzelm clarified modules;
Sat, 02 Apr 2016 23:29:05 +0200 wenzelm prefer infix operations;
Sat, 02 Apr 2016 22:38:26 +0200 wenzelm tuned signature;
Sat, 02 Apr 2016 21:10:07 +0200 wenzelm careful export of type-dependent functions, without losing their special status;
Fri, 18 Mar 2016 16:26:35 +0100 wenzelm clarified modules;
Thu, 03 Mar 2016 15:23:02 +0100 wenzelm clarified modules;
Thu, 18 Feb 2016 23:10:28 +0100 wenzelm unconditional Multithreading;
Tue, 03 Nov 2015 13:54:34 +0100 wenzelm clarified modules;
Tue, 01 Sep 2015 23:10:23 +0200 wenzelm thread context for exceptions from forks, e.g. relevant when printing errors;
Wed, 12 Aug 2015 01:39:31 +0200 wenzelm default ML context for forks, e.g. relevant for debugging and toplevel pretty-printing;
Tue, 21 Jul 2015 14:12:45 +0200 wenzelm more explicit thread identification;
Mon, 29 Jun 2015 20:55:46 +0200 wenzelm improved scheduling for urgent tasks, using farm of replacement threads (may lead to factor 2 overloading, but CPUs are usually hyperthreaded);
Thu, 29 Jan 2015 15:21:16 +0100 wenzelm explicit threads_stack_limit (for recent Poly/ML SVN versions), which leads to soft interrupt instead of exhaustion of virtual memory, which is particularly relevant for the bigger address space of x86_64;
Thu, 29 Jan 2015 13:58:02 +0100 wenzelm unused;
Thu, 29 Jan 2015 13:49:03 +0100 wenzelm clarified worker_wait;
Sat, 10 Jan 2015 14:28:41 +0100 wenzelm explicit shutdown of scheduler and worker thread farm, assuming Session.shutdown() before saving heap image;
Sat, 10 Jan 2015 12:21:27 +0100 wenzelm discontinued worker_trend: prefer constant number of active + reserve threads;
Fri, 09 Jan 2015 20:12:42 +0100 wenzelm permissive worker_start: failure to fork thread is deferred to later attempt to provide missing threads, without crashing scheduler;
Mon, 22 Dec 2014 20:40:37 +0100 wenzelm discontinued central critical sections: NAMED_CRITICAL / CRITICAL;
Wed, 26 Nov 2014 11:43:51 +0100 wenzelm load simple_thread.ML later, such that it benefits from redefined print_exception_trace;
Mon, 23 Jun 2014 12:54:48 +0200 wenzelm more on "Futures";
Mon, 31 Mar 2014 10:28:08 +0200 wenzelm support bulk messages consisting of small string segments, which are more healthy to the Poly/ML RTS and might prevent spurious GC crashes such as MTGCProcessMarkPointers::ScanAddressesInObject;
Thu, 05 Dec 2013 17:58:03 +0100 wenzelm merged, resolving obvious conflicts in NEWS and src/Pure/System/isabelle_process.ML;
Thu, 28 Nov 2013 12:54:39 +0100 wenzelm more official task context via Task_Queue.enroll, which is required to participate in group cancellation (e.g. to terminate command exec);
Wed, 20 Nov 2013 22:10:45 +0100 wenzelm register actual group of nested worker context -- relevant for insulated cancellation of exec_ids (see also 78693e46a237, e0169f13bd37);
less more (0) -100 -50 -30 tip