| Wed, 05 Feb 2014 10:38:12 +0100 |
Matthias Grimmer |
GNFI: rename project from .ffi.amd64 to .nfi.hotspot.amd64
|
| Thu, 30 Jan 2014 15:50:46 +0100 |
Matthias Grimmer |
GNFI fixes
|
| Thu, 30 Jan 2014 09:19:45 +0100 |
Matthias Grimmer |
Add the Graal Native Function Interface
|
| Thu, 30 Jan 2014 22:49:14 +0100 |
Doug Simon |
fixed or removed incorrect assertions
|
| Mon, 06 Jan 2014 17:19:18 -0800 |
Tom Rodriguez |
Embed compressed constants when possible and use more efficient patterns for encoding
|
| Fri, 13 Dec 2013 14:05:48 +0100 |
Doug Simon |
moved emitting code for LIR and queries about whether an edge goes to its lexical successor "inside" CompilationResultBuilder
|
| Fri, 13 Dec 2013 13:12:58 +0100 |
Doug Simon |
added subqWide to AMD64Assembler
|
| Thu, 05 Dec 2013 15:55:19 +0100 |
Doug Simon |
force long encoding of frame push instruction in a method without a stack bang; removed -G:StackShadowPages option
|
| Wed, 04 Dec 2013 16:01:30 +0100 |
Doug Simon |
minor renamings
|
| Tue, 03 Dec 2013 18:02:11 +0100 |
Doug Simon |
made the frameContext of a CompilationResultBuilder always non-null and added FrameContext.hasFrame() to determine if a frame is actually generated
|
| Tue, 03 Dec 2013 16:33:41 +0100 |
Doug Simon |
added CompilationResultBuilderFactory to support peep-hole instrumentation of methods as their code is emitted
|
| Tue, 03 Dec 2013 10:51:16 +0100 |
Doug Simon |
renamed TargetMethodAssembler to CompilationResultBuilder
|
| Mon, 02 Dec 2013 17:03:22 +0100 |
Doug Simon |
ensure instruction at verified entry point is safely patchable (GRAAL-605)
|
| Fri, 25 Oct 2013 23:50:05 +0200 |
Doug Simon |
cleaned up mechanism for saving registers and updating the callee save maps in stubs
|
| Fri, 25 Oct 2013 01:24:19 +0200 |
Doug Simon |
refactored all deferred initialization of backends into HotSpotBackend
|
| Thu, 24 Oct 2013 15:36:39 +0200 |
Roland Schatz |
Make AMD64HotSpotTruffleBackend a decorator for the default backend.
|
| Mon, 21 Oct 2013 17:42:42 +0200 |
Doug Simon |
added support code for Truffle to inject special tail-call code into the prefix of OptimizedCallTarget.call(PackedFrame, Arguments)
|
| Tue, 15 Oct 2013 22:21:01 +0200 |
Doug Simon |
rename: graalRuntime -> runtime, getGraalRuntime -> getRuntime
|
| Tue, 15 Oct 2013 14:14:06 +0200 |
Doug Simon |
split HotSpotRuntime into separate provider implementations
|
| Sat, 12 Oct 2013 01:03:47 +0200 |
Doug Simon |
moved lookupForeignCall() from CodeCacheProvider to ForeignCallsProvider (GRAAL-511)
|
| Fri, 11 Oct 2013 17:53:29 +0200 |
Gilles Duboscq |
Merge
|
| Thu, 10 Oct 2013 18:26:22 +0200 |
Gilles Duboscq |
Adapt compressed pointers implementation for last HotSpot changes
|
| Thu, 10 Oct 2013 23:20:04 +0200 |
Doug Simon |
rename: lookupRegisterConfig -> getRegisterConfig (part 2)
|
| Thu, 10 Oct 2013 20:50:54 +0200 |
Doug Simon |
made GraalCodeCacheProvider independent of CodeCacheProvider and renamed the former to LoweringProvider (GRAAL-511)
|
| Thu, 10 Oct 2013 16:14:55 +0200 |
Doug Simon |
made CodeCacheProvider independent of MetaAccessProvider (GRAAL-511)
|
| Mon, 30 Sep 2013 11:16:36 +0200 |
Bernhard Urban |
AMD64HotSpot: use conditional jump for IC_MISS_HANDLER
|
| Mon, 30 Sep 2013 09:32:18 +0200 |
Bernhard Urban |
AMD64HotSpot: emit jump to IC miss handler directly instead of emitting it inside verfied_entry
|
| Tue, 24 Sep 2013 14:24:47 -0400 |
Morris Meyer |
Finished PTX assembler and Register -> Variable conversion
|
| Thu, 05 Sep 2013 16:36:13 -0700 |
Christian Wimmer |
Fixed registers that flow into a block need to be inputs of the LabelOp, otherwise the register allocator can insert spill moves before the definition of a fixed register.
|
| Fri, 30 Aug 2013 13:51:22 +0200 |
Christos Kotselidis |
Move compressed pointers' logic to HotSpot specific move
|
| Wed, 28 Aug 2013 13:27:05 +0200 |
Christos Kotselidis |
Refactor heap base register
|
| Wed, 31 Jul 2013 14:04:24 -0700 |
twisti |
added architecture specific frame maps
|
| Mon, 15 Jul 2013 16:53:35 +0200 |
Doug Simon |
cannot omit frame for compiled methods that make a foreign call (GRAAL-362)
|
| Fri, 12 Jul 2013 14:17:42 +0200 |
Christos Kotselidis |
Add compressed oop support in HotSpot Backend
|
| Tue, 09 Jul 2013 11:14:36 +0200 |
Roland Schatz |
Allow foreign call stubs to be declared as leaf methods.
|
| Tue, 04 Jun 2013 17:33:14 +0200 |
Doug Simon |
converted remaining options in GraalOptions to new system (GRAAL-27)
|
| Thu, 16 May 2013 17:12:17 +0200 |
Doug Simon |
rename: RuntimeCallTarget -> ForeignCallLinkage
|
| Wed, 08 May 2013 22:55:17 +0200 |
Doug Simon |
introduced installedCodeOwner parameter of type ResolvedJavaMethod to GraalCompiler.compileGraph to properly distinguish the source method of a graph from the method under which the code compiled for the graph will be installed
|
| Wed, 08 May 2013 14:58:18 +0200 |
Doug Simon |
removed 'method' parameter from GraalCompiler.compiledMethod() and renamed the latter to 'compileGraph'.
|
| Wed, 08 May 2013 12:14:03 +0200 |
Doug Simon |
removed LIRGenerator.method field
|
| Tue, 07 May 2013 23:38:53 +0200 |
Doug Simon |
compilation is explicitly given a CallingConvention for the graph being compiled instead of deriving it from a method
|
| Thu, 02 May 2013 06:08:02 +0200 |
Doug Simon |
a compiled stub can now specify whether it needs to preserve registers. If a stub does not preserve registers and assertions are enabled, then all non-temporary registers are zapped after a C runtime call from the stub.
|
| Tue, 30 Apr 2013 20:56:43 +0200 |
Doug Simon |
Merge.
|
| Tue, 30 Apr 2013 20:28:33 +0200 |
Doug Simon |
moved declaration of EXCEPTION_HANDLER, DEOPT_HANDLER and IC_MISS_HANDLER descriptors to platform independent code
|
| Tue, 30 Apr 2013 19:11:14 +0200 |
Doug Simon |
stack overflow check and deopt/exception handler entry points are omitted from compiled stubs
|
| Tue, 30 Apr 2013 19:24:00 +0200 |
Roland Schatz |
Make assembler creation in backend more extensible.
|
| Mon, 29 Apr 2013 18:26:39 +0200 |
Doug Simon |
simplified register preservation in compiled stubs to avoid the need for using ParametersOp
|
| Fri, 26 Apr 2013 18:36:41 +0200 |
Doug Simon |
added support from compiled stubs to be installed as RuntimeStubs and to be able to directly call C/C++ runtime functions (GRAAL-81)
|
| Sun, 21 Apr 2013 21:41:09 +0200 |
Doug Simon |
temporaries are recorded for compiled stubs, allowing for better register allocation around calls to such stubs (GRAAL-210)
|
| Tue, 19 Mar 2013 14:08:54 +0100 |
Doug Simon |
removed unnecessary MethodEndBreakpointGuards option
|
| Tue, 19 Mar 2013 09:58:04 +0100 |
Doug Simon |
rename: HotSpotAMD64LIRGenerator -> AMD64HotSpotLIRGenerator
|
| Sun, 17 Mar 2013 23:53:19 +0100 |
Thomas Wuerthinger |
Put exception handler entry back.
|
| Sun, 17 Mar 2013 23:46:58 +0100 |
Thomas Wuerthinger |
Remove last usages of scratch register. Separate runtime calls from normal method calls in LIR.
|
| Fri, 15 Mar 2013 18:16:15 +0100 |
Doug Simon |
refactored HotSpotAMD64LIRGenerator into its own file
|
| 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.
|
| Thu, 14 Mar 2013 14:35:53 +0100 |
Doug Simon |
moved logic for reserving a special area/slot in a frame (e.g., for use during deoptimization) out of FrameMap and into platform specific backend class
|
| Thu, 14 Mar 2013 14:08:59 +0100 |
Doug Simon |
LIRGenerator is transmitted across backend passes instead of just the LIR so that backend-specific, per-compilation information can be attached to the backend-specific LIRGenerator object
|
| Mon, 11 Mar 2013 23:28:34 +0100 |
Thomas Wuerthinger |
Draft speculation log.
|
| Mon, 11 Mar 2013 13:06:14 +0100 |
Gilles Duboscq |
Use a DeoptimizeOp to avoid spilling before the deoptimize runtime call
|
| Mon, 11 Mar 2013 11:32:08 +0100 |
Thomas Wuerthinger |
Remove stubs as a parameter to TargetMethodAssembler.
|
