Path: csiph.com!news.swapon.de!newsreader4.netcologne.de!news.netcologne.de!.POSTED!not-for-mail From: Beloumi Newsgroups: comp.lang.java.security Subject: Entropy for PRNGs Date: Sat, 31 Oct 2015 16:54:58 +0100 Organization: news.netcologne.de Lines: 16 Distribution: world Message-ID: References: NNTP-Posting-Host: cable-78-34-11-61.netcologne.de Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Trace: newsreader4.netcologne.de 1446310544 31850 78.34.11.61 (31 Oct 2015 16:55:44 GMT) X-Complaints-To: abuse@netcologne.de NNTP-Posting-Date: Sat, 31 Oct 2015 16:55:44 +0000 (UTC) User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.3.0 In-Reply-To: Xref: csiph.com comp.lang.java.security:319 I have written a collector of entropy from GUIs (key strokes, mouse events) and - this is the main source of entropy - the Java thread schedule. https://github.com/Beloumi/EntropyPool (Comments, suggestions and criticism are welcome) The thread schedule as source of entropy is also used in ThreadedSeedGenerator (used in Bouncy Castles Crypto API, OpenJDK and others) and Oracles SpinnerEntropySource. What I am interested in is: As far as I know, the behavior of the Java Thread Scheduler depends on the implementation of the JVM and is not part of the specification of the JVM. So the thread behavior might vary for different system and JVMs. Does anybody know differences for threads of the same priority? Beloumi