JDK-4758432 : JVM 1.4.2 regression on Sierra EE startup compared to JVM 1.3.1 with -Xcomp
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 1.4.2
  • Priority: P4
  • Status: Closed
  • Resolution: Duplicate
  • OS: solaris_8
  • CPU: sparc
  • Submitted: 2002-10-04
  • Updated: 2002-10-05
  • Resolved: 2002-10-05
Related Reports
Duplicate :  
Relates :  
Description
JVM 1.4.2 (client) has a regression on Sierra EE RC4 startup compared to JVM 1.3.1 with -Xcomp HotSpot option (full compilation): 73 sec vs 43 sec of  user CPU time (70% regression)
To reproduce:
-Install JDK 1.3.1
-Install Sun ONE Studio Sierra Release Candidate 4 using Installer with JDK 1.3.1 
-Run command: /bin/time <sierra_with_jdk1.3.1_install_path>/bin/runide.sh -J-Dnetbeans.close=true -J-Xcomp 
-Install JDK 1.4.2
-Install Sun ONE Studio Sierra Release Candidate 4 using Installer with JDK 1.4.2
-Run command: /bin/time <sierra_with_jdk1.4.2_install_path>/bin/runide.sh -J-Dnetbeans.close=true -J-Xcomp
Note: second and following startups were always measured.
###@###.### 2002-10-04 

Comments
WORK AROUND
11-06-2004

SUGGESTED FIX
11-06-2004

PUBLIC COMMENTS
10-06-2004

EVALUATION HotSpot is not designed to be benchmarked in -Xcomp mode. Among other non-compile-speed differences between -Xcomp and -Xmixed mode, substantially more deoptimization and patching occurs in -Xcomp runs. ###@###.### has been working on compile speed and has achieved substantial speedups which should be present in 1.4.2. For this reason I'm closing this bug as a duplicate of the RFE he has open for this work, 4741220. ###@###.### 2002-10-04
04-10-2002