kvn [Mon, 24 Aug 2009 11:13:46 -0700] rev 3687
6873800: enable compressed oops by default
Summary: enable compressed oops by default
Reviewed-by: never, ysr
cfang [Thu, 20 Aug 2009 12:42:57 -0700] rev 3686
6873116: Modify reexecute implementation to use pcDesc to record the reexecute bit
Summary: use PcDesc to keep record of the reexecute bit instead of using DebugInfoStreams
Reviewed-by: kvn, never, twisti
kvn [Wed, 19 Aug 2009 19:05:18 -0700] rev 3685
6873799: enable escape analysis by default
Summary: enable escape analysis by default
Reviewed-by: never
never [Wed, 19 Aug 2009 18:54:35 -0700] rev 3684
6873777: FPU control word optimization still performed with SSE
Reviewed-by: kvn
never [Mon, 17 Aug 2009 14:45:02 -0700] rev 3683
6795465: Crash in assembler_sparc.cpp with client compiler on solaris-sparc
Reviewed-by: twisti, cfang
cfang [Mon, 17 Aug 2009 12:11:06 -0700] rev 3682
6866651: Regression: simple int sum crashes jvm (build 1.6.0_14-b08 and 1.7.0-ea-b59)
Summary: delay dead code elimination in set_req_X to make it safe
Reviewed-by: kvn, never
cfang [Mon, 17 Aug 2009 09:48:02 -0700] rev 3681
6829127: Deoptimization Failure on Specjvm98 _227_mtrt with -XX:+DeoptimizeALot since Hs11 b01
Summary: Make sure the control word is correct in deopt_blob after restore_result_registers
Reviewed-by: kvn, never
never [Fri, 14 Aug 2009 22:11:18 -0700] rev 3680
Merge
never [Fri, 14 Aug 2009 15:55:26 -0700] rev 3679
Merge
never [Fri, 14 Aug 2009 15:53:46 -0700] rev 3678
6862863: C2 compiler fails in elide_copy()
Reviewed-by: kvn
kvn [Fri, 14 Aug 2009 12:23:11 -0700] rev 3677
6869822: assert(Universe::narrow_oop_shift() == 0,"use unscaled narrow oop")
Summary: Replace the assert with narrow_oop_shift set to 0.
Reviewed-by: never, jcoomes
never [Fri, 14 Aug 2009 00:02:12 -0700] rev 3676
6862956: PhaseIdealLoop should have a CFG verification mode
Reviewed-by: kvn, twisti
trims [Fri, 21 Aug 2009 20:39:41 -0700] rev 3675
Merge
apangin [Wed, 19 Aug 2009 15:46:22 -0700] rev 3674
Merge
apangin [Mon, 17 Aug 2009 15:03:39 -0700] rev 3673
Merge
dcubed [Tue, 11 Aug 2009 11:57:51 -0600] rev 3672
6862945: 4/3 conversion of jmethodID to methodOop in JVMTI is too expensive
Summary: Refactor JNIHandles::checked_resolve_jmethod_id() into fast and paranoid parts.
Reviewed-by: never, alanb