Wed, 20 Mar 2013 20:13:08 +0100 |
Christos Kotselidis |
-Create push patch
|
Mon, 04 Mar 2013 16:46:37 +0100 |
Christos Kotselidis |
-More progress in G1 WBs
|
Wed, 27 Feb 2013 15:15:04 +0100 |
Christos Kotselidis |
-More porting to G1 WBs
|
Wed, 20 Feb 2013 16:15:43 +0100 |
Christos Kotselidis |
Towards porting G1 WBs
|
Tue, 19 Feb 2013 17:57:14 +0100 |
Christos Kotselidis |
-Implement WB primitive stub call for G1's slow path later
|
Tue, 26 Mar 2013 12:01:42 +0100 |
Doug Simon |
made method substitutions for Thread be interpretable
|
Wed, 20 Mar 2013 22:23:14 +0100 |
Doug Simon |
rename packages: *snippets* -> *replacements*
|
Wed, 20 Mar 2013 20:23:23 +0100 |
Doug Simon |
renamings as part of a move away from the overloading of the "snippet" term for various code replacement/substitution mechanisms towards it being solely for the concept of IR lowering
|
Fri, 15 Mar 2013 20:06:19 +0100 |
Doug Simon |
moved AMD64ConvertSnippets to (new) project com.oracle.graal.snippets.amd64
|
Fri, 15 Mar 2013 15:38:27 +0100 |
Doug Simon |
AMD64 HotSpot backend now models RBP as an incoming parameter that must be preserved until the end of the method. For non-leaf methods, the value is preserved in the special stack slot required by the HotSpot runtime for walking/inspecting frames of such methods.
|
Mon, 11 Mar 2013 13:06:14 +0100 |
Gilles Duboscq |
Use a DeoptimizeOp to avoid spilling before the deoptimize runtime call
|
Fri, 08 Mar 2013 14:33:33 +0100 |
Doug Simon |
replaced AMD64Arithmetic.ConvertSlowPath with a snippet (GRAAL-140)
|
Wed, 06 Mar 2013 16:37:18 +0100 |
Thomas Wuerthinger |
Create new way of handling unwind that recalculates rbp from rsp. Remove unused instanceof slow path stub.
|
Tue, 05 Mar 2013 21:03:20 +0100 |
Thomas Wuerthinger |
Removed unused imports.
|
Tue, 05 Mar 2013 21:01:19 +0100 |
Thomas Wuerthinger |
Remove debug stub.
|
Tue, 05 Mar 2013 19:20:05 +0100 |
Thomas Wuerthinger |
Remove setDeoptInfo functionality. Start getting rid of scratch register usages.
|
Mon, 18 Feb 2013 21:08:42 -0800 |
Thomas Wuerthinger |
Fix compiler warnings.
|
Mon, 18 Feb 2013 20:21:09 -0800 |
Thomas Wuerthinger |
Introduce marker interface InvokeTarget. Remove method lookupRuntimeCall from CodeCacheProvider.
|
Wed, 06 Feb 2013 10:24:39 +0100 |
Roland Schatz |
move AMD64HotSpotRuntime.getTarget up to HotSpotRuntime
|
Tue, 05 Feb 2013 13:54:18 +0100 |
Roland Schatz |
expose TargetDescription through CodeCacheProvider API
|
Tue, 22 Jan 2013 22:00:34 +0100 |
Doug Simon |
disabled auto-formatting of manually formatted code
|
Wed, 16 Jan 2013 10:19:09 +0100 |
Christian Haeubl |
windows-specific bugfix
|
Tue, 15 Jan 2013 21:11:32 +0100 |
Doug Simon |
added com.sun.crypto.provider.CipherBlockChaining intrinsifications
|
Mon, 14 Jan 2013 14:06:37 +0100 |
Doug Simon |
added support for making calls with the native/C ABI
|
Mon, 14 Jan 2013 00:14:54 +0100 |
Doug Simon |
AESCrypt intrinsification - disabled by default as it doesn't yet work
|
Fri, 14 Dec 2012 13:26:19 +0100 |
Christian Haeubl |
OS-specific identity hashcode bugfix
|
Wed, 12 Dec 2012 15:05:21 +0100 |
Christian Haeubl |
bugfixes for compiler intrinsics
|
Tue, 11 Dec 2012 09:10:50 +0100 |
Christian Haeubl |
fixes after merge
|
Tue, 11 Dec 2012 08:48:12 +0100 |
Christian Haeubl |
Merge.
|
Tue, 11 Dec 2012 08:28:00 +0100 |
Christian Haeubl |
added more compiler intrinsics
|
Fri, 07 Dec 2012 23:41:51 +0100 |
Doug Simon |
removed declaration of temps (no longer) used by the monitorenter/monitorexits stubs
|
Fri, 07 Dec 2012 23:38:09 +0100 |
Doug Simon |
added TLAB fast refill stub for array allocation
|
Fri, 07 Dec 2012 18:26:26 +0100 |
Doug Simon |
consolidated new_type_array and new_object_array stubs into one as there no difference between them
|
Fri, 07 Dec 2012 15:12:05 +0100 |
Doug Simon |
added support for writing stubs in Java and wrote the TLAB fast refill stub
|
Mon, 03 Dec 2012 15:32:17 +0100 |
Doug Simon |
decoupled C++ Graal runtime from C1
|
Wed, 14 Nov 2012 11:28:02 +0100 |
Doug Simon |
modifications to support non-perm-gen changes in HotSpot
|
Tue, 30 Oct 2012 11:12:35 -0700 |
Christian Wimmer |
CodeCacheProvider.callKillsRegisters is not necessary because this information can be inferred from the register configuration
|
Thu, 25 Oct 2012 14:50:14 +0200 |
Doug Simon |
removed Architecture.isX86() and Architecture.isSPARC()
|
Thu, 25 Oct 2012 12:52:54 +0200 |
Doug Simon |
moved AMD64 into its own (new) project: com.oracle.graal.amd64
|
Mon, 08 Oct 2012 19:34:32 -0700 |
Christian Wimmer |
SnippetTemplate must not depend on CodeCacheProvider (only on MetaAccessProvider). This means that the hasSideEffect flag of a RuntimeCall must be already fixed in the RuntimeCall.Descriptor to avoid a lookup of the actual RuntimeCall using the CodeCacheProvider.
|
Mon, 08 Oct 2012 17:36:51 +0200 |
Doug Simon |
renamed package: com.oracle.graal.compiler.target.amd64 -> com.oracle.graal.compiler.amd64
|
Mon, 08 Oct 2012 17:35:03 +0200 |
Doug Simon |
renamed packages: com.oracle.max.asm... -> com.oracle.graal.asm...
|
Fri, 05 Oct 2012 17:55:12 +0200 |
Doug Simon |
consolidated framework for runtime and stub calls and moved declaration of descriptors for such calls to the source file in which they are used
|
Wed, 03 Oct 2012 01:18:03 +0200 |
Doug Simon |
moved AMD64 specific HotSpot code in com.oracle.graal.hotspot.amd64 project
base
graal/com.oracle.graal.hotspot/src/com/oracle/graal/hotspot/target/amd64/AMD64HotSpotRuntime.java@5d40d4cc6842
|