Re: RFR [8051382] Optimize java.lang.reflect.Modifier.toString()

2014-08-04 Thread Martin Buchholz
Looks good! On Mon, Aug 4, 2014 at 4:27 AM, Ivan Gerasimov wrote: > Hi Martin! > > Sorry for the pause, I had to take a break. > > Thank you for your StringJoiner rework! > I created a bug to track it: > https://bugs.openjdk.java.net/browse/JDK-8054221 > > With your implementation of StringJoi

Re: RFR [8054221] StringJoiner imlementation optimization

2014-08-04 Thread Martin Buchholz
Thanks, Ivan. On Mon, Aug 4, 2014 at 9:16 AM, Ivan Gerasimov wrote: > Hello! > > Here's the proposal by Martin Buchholz to improve implementation of > StringJoiner class: > http://cr.openjdk.java.net/~martin/webrevs/openjdk9/ > StringJoiner-optimization/ > > I think that result of concatenation

Re: RFR [8054221] StringJoiner imlementation optimization

2014-08-04 Thread Martin Buchholz
On Mon, Aug 4, 2014 at 10:04 AM, Claes Redestad wrote: > Hi, > > - is there a sufficiently good reason not to use an ArrayList > over a manually resized String[] here? It'd simplify and use > overflow-conscious code at a negligible performance cost > This is a core library - it should be fanati

Re: RFR [8054221] StringJoiner imlementation optimization

2014-08-04 Thread Claes Redestad
Hi, - is there a sufficiently good reason not to use an ArrayList over a manually resized String[] here? It'd simplify and use overflow-conscious code at a negligible performance cost - nit: elt(s) -> element(s) - nit: always use curly braces /Claes On 08/04/2014 06:16 PM, Ivan Gerasimov

RFR [8054221] StringJoiner imlementation optimization

2014-08-04 Thread Ivan Gerasimov
Hello! Here's the proposal by Martin Buchholz to improve implementation of StringJoiner class: http://cr.openjdk.java.net/~martin/webrevs/openjdk9/StringJoiner-optimization/ I think that result of concatenation in the merge() method can be saved to be reused later. This would be useful, for e

Fwd: RFR (XS) JDK-8051861 [TESTBUG] test "java/math/BigInteger/BigIntegerTest.java" does not hold Random value to have a possibility to reproduce it

2014-08-04 Thread vasily stolbov
Original Message Subject: RFR (XS) JDK-8051861 [TESTBUG] test "java/math/BigInteger/BigIntegerTest.java" does not hold Random value to have a possibility to reproduce it Date: Mon, 04 Aug 2014 17:45:02 +0400 From: vasily stolbov Organization: Oracle Corporation To:

Re: Why is finalize wrong?

2014-08-04 Thread David M. Lloyd
On 08/04/2014 08:46 AM, Jaroslav Tulach wrote: Hi. Last week we touched topic of finalization and what is wrong with it. I proposed three reasons why Object.finalize is bad. Is it correct and extensive list or would you change it or expand it? Thanks as ... # 1 - Because of automatic JDK mana

Re: Why is finalize wrong?

2014-08-04 Thread Jaroslav Tulach
Hi. Last week we touched topic of finalization and what is wrong with it. I proposed three reasons why Object.finalize is bad. Is it correct and extensive list or would you change it or expand it? Thanks as ... > > # 1 - Because of automatic JDK management thread? > > # 2 - Or because once fin

RFR (XS) JDK-8051861 [TESTBUG] test "java/math/BigInteger/BigIntegerTest.java" does not hold Random value to have a possibility to reproduce it

2014-08-04 Thread vasily stolbov
Hi, Please review and help me with integration: Problem: java.math.BigInteger.BigIntegerTest uses java.math.Random(), so we have no possibility to reproduce it. Solution: Test gets start random seed from environment variable. If this variable not exists, test write current random seed

Re: WAR == single classloader was: Benefits of activeReferenceQueue

2014-08-04 Thread David M. Lloyd
On 08/04/2014 06:27 AM, Florian Weimer wrote: On 08/01/2014 06:39 AM, Jaroslav Tulach wrote: Dne Út 29. července 2014 14:13:25, Florian Weimer napsal(a): On 07/29/2014 10:05 AM, Jaroslav Tulach wrote: Plus, because there is a single classloader which loads all the classes from a WAR, by keepin

Re: RFR [8051382] Optimize java.lang.reflect.Modifier.toString()

2014-08-04 Thread Ivan Gerasimov
Hi Martin! Sorry for the pause, I had to take a break. Thank you for your StringJoiner rework! I created a bug to track it: https://bugs.openjdk.java.net/browse/JDK-8054221 With your implementation of StringJoiner, the first version of Modifier.toString() [1] became (almost) as fast as origi

Re: WAR == single classloader was: Benefits of activeReferenceQueue

2014-08-04 Thread Florian Weimer
On 08/01/2014 06:39 AM, Jaroslav Tulach wrote: Dne Út 29. července 2014 14:13:25, Florian Weimer napsal(a): On 07/29/2014 10:05 AM, Jaroslav Tulach wrote: Plus, because there is a single classloader which loads all the classes from a WAR, by keeping the activerReferenceQueue thread alive and ho