1 SimGrid-java (3.8) unstable; urgency=low
3 * Introduce org.simgrid.trace.Trace (partial implementation of JAVA bindings
5 Although all TRACE methods appear in the Trace class, please not that only
6 few of thems are really implemented at the JNI level (see ./src/jtrace.c
7 for further information).
9 SimGrid-java (3.7.1) stable; urgency=low
11 The "Java aint got to be bloated and slow" release
14 * Various internal cleanups and performance improvement
15 Simulations are expected to run up to twice faster or so
16 * Make Process.kill(process) an instance method, not a static one
17 * User processes are not java.lang.Thread subclasses.
18 This breaks the compatibility (sorry), but previous API was
19 brain-dead, making it impossible to have non-trivial
20 initializations in the process constructor.
21 * Require a full constructor per Process sub-class.
22 Kinda breaks the compatibility (sorry), but this allows a much more
23 efficient way to launch the processes at simulation startup.
24 * Do not embeed our version of semaphores, java 1.5 can be considered
25 as sufficiently prevalent for us to not dupplicate its features.
29 * Add examples for almost every part of the API
30 We spotted and fixed a lot of bugs in the process
32 * New module: asynchronous communication API
33 * New function: Process.sleep()
34 It takes milliseconds as argument, just as java.lang.Thread.sleep()
35 * New module: org.simgrid.msg.Mutex (SimGrid mutexes)
36 * New module: org.simgrid.msg.RngStream (RngStreams random generators)
38 -- 2012-06-12 Da SimGrid team <simgrid-devel@lists.gforge.inria.fr>
40 SimGrid-java (1.0 (3.6) unstable; urgency=low
43 * Split of every thing from simgrid v3.5 into a separate package.
45 -- 2011-10-05 Da SimGrid team <simgrid-devel@lists.gforge.inria.fr>