[jruby-dev] Slowness of attr sets

Our direct ivar reads/writes are now pretty fast (faster than MRI in bench_ivar_access, at least on my slowish machine), and attr reads are also pretty fast (also faster than MRI), especially since Charlie got rid of the callback. But attr writes are still painfully slow, and I'm not sure why. Her

[jruby-dev] [jira] Created: (JRUBY-1460) TCPSockets do not stay fully closed

TCPSockets do not stay fully closed --- Key: JRUBY-1460 URL: http://jira.codehaus.org/browse/JRUBY-1460 Project: JRuby Issue Type: Bug Reporter: Evan Weaver Assignee: Thomas E Enebo

[jruby-dev] [jira] Created: (JRUBY-1459) TCPSocket#closewrite has no effect

TCPSocket#closewrite has no effect -- Key: JRUBY-1459 URL: http://jira.codehaus.org/browse/JRUBY-1459 Project: JRuby Issue Type: Bug Reporter: Evan Weaver Assignee: Thomas E Enebo Not

RE: [jruby-dev] Compile to .rbj instead of .class?

I researched this further, and it looks like that there is a possible workaround using JET's Mixed Compilation Model, which is described here: http://www.excelsior-usa.com/doc/jet500/jetw009.html#0249 It basically involves running the application once and letting the JIT native code compiler cach

[jruby-dev] [jira] Created: (JRUBY-1458) ARGF.rewind blows up (and it shouldnt)

ARGF.rewind blows up (and it shouldnt) -- Key: JRUBY-1458 URL: http://jira.codehaus.org/browse/JRUBY-1458 Project: JRuby Issue Type: Bug Environment: trunk rev 4652, Ubuntu Linux Repor

[jruby-dev] [jira] Created: (JRUBY-1457) Incorrect handling of STDIN redirect by system() and friends

Incorrect handling of STDIN redirect by system() and friends Key: JRUBY-1457 URL: http://jira.codehaus.org/browse/JRUBY-1457 Project: JRuby Issue Type: Bug Affects Versions: JR

[jruby-dev] Re: [jruby-user] Bug cleanup in progress

[EMAIL PROTECTED] wrote: As you continue the triage, if you see windows-related bugs, assign them to me. I have a couple already and I might as well push through them all at once. You're a brave, brave man. It shall be done. - Charlie --

Re: [jruby-dev] test fails for trunk rev 4648

Ola Bini wrote: Alex Durgin wrote: compile-test: Compiling 82 source files to >JRuby\build\classes\test JRuby\test\org\jruby\test\MainTestSuite.java:87: cannot find symbol symbol : class TestRubyBigDecimal location: class org.jruby.test.MainTestSuite suite.addTestSuite(TestRu

Re: [jruby-dev] Compile to .rbj instead of .class?

- "Peter K Chan" <[EMAIL PROTECTED]> wrote: > It isn't so simple. You can rename all the .rbj to .class, and JET > may compile them successfully, but how do they get loaded up? The JRuby > classloader will still be expecting .rbj and doing it's own defineClass() on > bytecode, and not native c

[jruby-dev] Re: [jruby-user] Bug cleanup in progress

As you continue the triage, if you see windows-related bugs, assign them to me. I have a couple already and I might as well push through them all at once. /Nick On 10/20/07, Charles Oliver Nutter <[EMAIL PROTECTED]> wrote: > I went through about a 120 bugs this evening, fixing a bunch, applying >

Re: [jruby-dev] Bug cleanup in progress

Go Charlie, Go! On Oct 19, 2007, at 10:38 PM, Charles Oliver Nutter wrote: I went through about a 120 bugs this evening, fixing a bunch, applying patches for a bunch, closing a bunch that were already working, backporting a few fixes, assigning some to specific versions or committers, and

Re: [jruby-dev] test fails for trunk rev 4648

Alex Durgin wrote: compile-test: Compiling 82 source files to >JRuby\build\classes\test JRuby\test\org\jruby\test\MainTestSuite.java:87: cannot find symbol symbol : class TestRubyBigDecimal location: class org.jruby.test.MainTestSuite suite.addTestSuite(TestRubyBigDecimal.cla

[jruby-dev] test fails for trunk rev 4648

compile-test: Compiling 82 source files to >JRuby\build\classes\test >JRuby\test\org\jruby\test\MainTestSuite.java:87: cannot find symbol symbol : class TestRubyBigDecimal location: class org.jruby.test.MainTestSuite suite.addTestSuite(TestRubyBigDecimal.class);