Wed, 18 Feb 2015 01:13:17 +0100 |
Doug Simon |
made use of Graal stubs instead of equivalent HotSpot stubs optional and off by default
|
Wed, 18 Feb 2015 00:09:24 +0100 |
Doug Simon |
Merge.
|
Tue, 17 Feb 2015 23:39:41 +0100 |
Thomas Wuerthinger |
Assert that current bci points to an if Java bytecode when branch taken probability is requested.
|
Tue, 17 Feb 2015 23:31:15 +0100 |
Thomas Wuerthinger |
Undo previous change on probability cut-off for inlining. Perform experiment later.
|
Tue, 17 Feb 2015 23:08:49 +0100 |
Thomas Wuerthinger |
Use atomic integer for iterable IDs to prevent races.
|
Tue, 17 Feb 2015 20:38:13 +0100 |
Thomas Wuerthinger |
Merge.
|
Tue, 17 Feb 2015 20:37:45 +0100 |
Thomas Wuerthinger |
Correctly create IsCompilationConstantNode in FastPE mode.
|
Tue, 17 Feb 2015 17:31:48 +0100 |
Thomas Wuerthinger |
Cut off loop exit probability at 1D.
|
Tue, 17 Feb 2015 17:05:04 +0100 |
Thomas Wuerthinger |
Another fix in the cpp part.
|
Tue, 17 Feb 2015 16:51:36 +0100 |
Thomas Wuerthinger |
Fix cpp code.
|
Tue, 17 Feb 2015 16:32:58 +0100 |
Thomas Wuerthinger |
Merge.
|
Tue, 17 Feb 2015 16:32:47 +0100 |
Thomas Wuerthinger |
Do not eagerly resolve types in lookupType.
|
Tue, 17 Feb 2015 16:31:39 +0100 |
Thomas Wuerthinger |
Add a probability-based cut-off to the inlining policy.
|
Tue, 17 Feb 2015 01:50:08 +0100 |
Thomas Wuerthinger |
Fix graph dumping in the graph builder.
|
Tue, 17 Feb 2015 01:44:13 +0100 |
Thomas Wuerthinger |
Merge.
|