| Tue, 10 Apr 2012 15:48:54 +0200 |
Gilles Duboscq |
Fix for peeling : fix a bug in the Virtual chain reconstruction
|
| Tue, 10 Apr 2012 14:06:02 +0200 |
Gilles Duboscq |
Merge
|
| Tue, 10 Apr 2012 14:05:14 +0200 |
Gilles Duboscq |
Estimate size of loop and peel only if resulting graph is not too big
|
| Tue, 10 Apr 2012 12:38:13 +0200 |
Doug Simon |
configured test source directory for graal.graph project
|
| Tue, 10 Apr 2012 12:22:46 +0200 |
Doug Simon |
Merge.
|
| Sun, 08 Apr 2012 00:09:10 +0200 |
Doug Simon |
replace monkey patch with subclassing instead
|
| Mon, 09 Apr 2012 20:43:30 +0200 |
Gilles Duboscq |
Peeling loop depending on their entry probability
|
| Mon, 09 Apr 2012 20:30:41 +0200 |
Gilles Duboscq |
Add options for OptLoopTransform and OptSafepointElimination
|
| Mon, 09 Apr 2012 19:59:01 +0200 |
Gilles Duboscq |
Loop peeling
|
| Mon, 09 Apr 2012 19:56:10 +0200 |
Gilles Duboscq |
We can use a FloatingRead to get the arrayClass in storechecks
|
| Mon, 09 Apr 2012 19:51:43 +0200 |
Gilles Duboscq |
Fix StoreIndexedNode lowering : use a FloatinReadNode instead of Floating a ReadNode
|
| Mon, 09 Apr 2012 19:41:59 +0200 |
Gilles Duboscq |
Fix for getModifiers intrinsic : use a floating read instead of floating a normal read
|
| Mon, 09 Apr 2012 19:15:41 +0200 |
Gilles Duboscq |
Loop-closed form GraphBuidling
|
| Fri, 06 Apr 2012 17:58:00 +0200 |
Gilles Duboscq |
Merge
|
| Fri, 06 Apr 2012 17:51:35 +0200 |
Gilles Duboscq |
Allow VirtualPhis in FrameStates, now the Virtual state in a LoopBegin's FrameState is the VirtualLoopPhi
|
