JDK-6561457 : merge-o in fix to 6516101 bogusly sets the null_seen bit.
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 7
  • Priority: P2
  • Status: Resolved
  • Resolution: Fixed
  • OS: generic
  • CPU: generic
  • Submitted: 2007-05-23
  • Updated: 2010-04-02
  • Resolved: 2007-06-20
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
JDK 6 JDK 7 Other
6u4Fixed 7Fixed hs10Fixed
Related Reports
Relates :  
Relates :  
Description
The following line was introduced twice in each of two files:
    set_mdp_flag_at(mdp, BitData::null_seen_byte_constant());

The files are:
    src/cpu/amd64/vm/interp_masm_amd64.cpp
    src/cpu/i486/vm/interp_masm_i486.cpp

The lines were apparently introduced by re-application of a patch.
Their contexts are extremely similar, but one context is the correct
location and the other is incorrect.  The bug is that the null_seen
bit is set in the MDO on code paths generated by both
profile_null_seen and profile_typecheck in InterpreterMacroAssembler.
Only the former should set that bit.  The result is that the
server compiler always performs a careful explicit null check
on the casted oop, instead of (what should be the usual case)
an implicit null check.

Fix is to remove the duplicate lines, one from each file.

Comments
EVALUATION Couldn't have said it better myself.
23-05-2007