Wed, 12 Feb 2014 11:10:13 +0000 Fixes in java.io,Win32FileSystem - added listDrives0() native. development
Jan Vrany <jan.vrany@fit.cvut.cz> [Wed, 12 Feb 2014 11:10:13 +0000] rev 3024
Fixes in java.io,Win32FileSystem - added listDrives0() native.
Wed, 12 Feb 2014 10:30:32 +0000 Workaround in JavaRelease::OpenJDK7>>searchForSourcePath for Debian boxes. development
Jan Vrany <jan.vrany@fit.cvut.cz> [Wed, 12 Feb 2014 10:30:32 +0000] rev 3023
Workaround in JavaRelease::OpenJDK7>>searchForSourcePath for Debian boxes. Kludge for Debian packages. On Debian boxes package openjdk-7-source installs src.zip into `/usr/lib/jvm/java-7-openjdk-i386` instead of into `/usr/lib/jvm/java-7-openjdk-common`. This commit workarounds this by looking into i386 directory too (if using 64bit JDK)
Tue, 11 Feb 2014 21:52:42 +0000 Reduced dimensions on MULTIANEWARRAY#test_incomplete_02 to avoid 2-pass compressing GC. development
Jan Vrany <jan.vrany@fit.cvut.cz> [Tue, 11 Feb 2014 21:52:42 +0000] rev 3022
Reduced dimensions on MULTIANEWARRAY#test_incomplete_02 to avoid 2-pass compressing GC. Due to a known bug in 2-pass compressing GC the VM crashed sice create array was way to big. Reducing dimensions made it smaller and thus running test does not trigger 2-pass compress.
Tue, 11 Feb 2014 20:21:33 +0000 Yet another cleanup of java.io.FileSystem mess. development
Jan Vrany <jan.vrany@fit.cvut.cz> [Tue, 11 Feb 2014 20:21:33 +0000] rev 3021
Yet another cleanup of java.io.FileSystem mess.
Tue, 11 Feb 2014 16:29:28 +0000 Fixed tests'run-all.rb to run under Windows as well. development
Jan Vrany <jan.vrany@fit.cvut.cz> [Tue, 11 Feb 2014 16:29:28 +0000] rev 3020
Fixed tests'run-all.rb to run under Windows as well.
Fri, 07 Feb 2014 09:49:00 +0100 Natives cleaned up a bit. development
Jan Vrany <jan.vrany@fit.cvut.cz> [Fri, 07 Feb 2014 09:49:00 +0100] rev 3019
Natives cleaned up a bit. Some natives has been backported frm OpenJDK7 to OpenJDK6, they have been moved to a superclass
Fri, 07 Feb 2014 09:07:22 +0100 Fixes in natives for non-TRADITIONAL_STACK machines. development
Jan Vrany <jan.vrany@fit.cvut.cz> [Fri, 07 Feb 2014 09:07:22 +0100] rev 3018
Fixes in natives for non-TRADITIONAL_STACK machines. On some machines stc accesses method arguments through context. To do so, it #defines argumets as simething __context[1]. If then the structure/union member has the same name as parameter, preprocessor expands it and therefore the C code is no longer valid. Solutions is to rename structure/union members - if possible.
Mon, 10 Feb 2014 00:12:30 +0100 throw ExceptionInInitializerError when anything bad happens during class init development
Marcel Hlopko <marcel.hlopko@gmail.com> [Mon, 10 Feb 2014 00:12:30 +0100] rev 3017
throw ExceptionInInitializerError when anything bad happens during class init
Sun, 09 Feb 2014 17:15:54 +0100 merge heads development
Marcel Hlopko <marcel.hlopko@gmail.com> [Sun, 09 Feb 2014 17:15:54 +0100] rev 3016
merge heads
Sun, 09 Feb 2014 16:52:47 +0100 teach JUnitTestCaseProxy to fail when expected exception is not raised development
Marcel Hlopko <marcel.hlopko@gmail.com> [Sun, 09 Feb 2014 16:52:47 +0100] rev 3015
teach JUnitTestCaseProxy to fail when expected exception is not raised
(0) -3000 -1000 -300 -100 -10 +10 +100 +300 tip