You are viewing a plain text version of this content. The canonical link for it is here.
Posted to bcel-dev@jakarta.apache.org by bu...@apache.org on 2008/02/01 16:56:33 UTC
DO NOT REPLY [Bug 44340] New: - BCEL is unnecessarily slow
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG�
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://issues.apache.org/bugzilla/show_bug.cgi?id=44340>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND�
INSERTED IN THE BUG DATABASE.
http://issues.apache.org/bugzilla/show_bug.cgi?id=44340
Summary: BCEL is unnecessarily slow
Product: BCEL
Version: 5.2
Platform: Other
OS/Version: All
Status: NEW
Severity: normal
Priority: P3
Component: Main
AssignedTo: bcel-dev@jakarta.apache.org
ReportedBy: markus.gaisbauer@gmail.com
Profiling BCEL showed that a lot of time is spent in very little code. I have
tracked down some of those problems, fixed them and saw a performance gain of
more than 100%. Later I have also found out that the AspectJ team and probably a
few other people had done the same a long time ago. Why these changes were never
incorporated into BCEL trunk, I do not know.
I understand, that from a performance perspective, I should use ASM anyway
(which is crazily optimized:
http://asm.objectweb.org/doc/developer-guide.html#optimizations).
Nevertheless some improvement probably would not hurt.
Here is the log that shows the performance gains of all the optimisations I
applied:
// SVN
ClassParser.parse: 0.680119516 s
ClassGen.init: 0.596259385 s
MethodGen.init: 7.627294121 s
MethodGen.getMethod: 2.415617407 s
ClassGen.getJavaClass.getBytes: 0.573408268 s
Total: 12.358684516 s
ClassParser.parse: 0.486115354 s
ClassGen.init: 0.60449182 s
MethodGen.init: 7.111806585 s
MethodGen.getMethod: 2.137157344 s
ClassGen.getJavaClass.getBytes: 0.514757783 s
Total: 11.262602001 s
ClassParser.parse: 0.513113271 s
ClassGen.init: 0.596378125 s
MethodGen.init: 7.09741055 s
MethodGen.getMethod: 2.095376224 s
ClassGen.getJavaClass.getBytes: 0.472800383 s
Total: 11.161931208 s
// Tune Instruction.readInstruction: Replace Reflection with SWITCH
ClassParser.parse: 0.675918174 s
ClassGen.init: 0.644816324 s
MethodGen.init: 2.37364546 s
MethodGen.getMethod: 2.350236018 s
ClassGen.getJavaClass.getBytes: 0.556582776 s
Total: 7.04360494 s
ClassParser.parse: 0.476201456 s
ClassGen.init: 0.579783983 s
MethodGen.init: 2.099810171 s
MethodGen.getMethod: 2.181126984 s
ClassGen.getJavaClass.getBytes: 0.488192476 s
Total: 6.227110003 s
ClassParser.parse: 0.477742278 s
ClassGen.init: 0.624694793 s
MethodGen.init: 2.130978382 s
MethodGen.getMethod: 2.002330381 s
ClassGen.getJavaClass.getBytes: 0.465331297 s
Total: 6.066029749 s
// Tune ConstantPoolGen.init: Use StringBuffer
ClassParser.parse: 0.659908237 s
ClassGen.init: 0.519261101 s
MethodGen.init: 2.305224708 s
MethodGen.getMethod: 2.148664254 s
ClassGen.getJavaClass.getBytes: 0.617371665 s
Total: 6.710080325 s
ClassParser.parse: 0.466696673 s
ClassGen.init: 0.451812126 s
MethodGen.init: 2.017315203 s
MethodGen.getMethod: 1.995585551 s
ClassGen.getJavaClass.getBytes: 0.495333748 s
Total: 5.812121207 s
ClassParser.parse: 0.446558571 s
ClassGen.init: 0.409762396 s
MethodGen.init: 1.937347861 s
MethodGen.getMethod: 2.118172599 s
ClassGen.getJavaClass.getBytes: 0.453922767 s
Total: 5.715940347 s
// Tune InstructionList.findHandle: Search directly
ClassParser.parse: 0.654898115 s
ClassGen.init: 0.55573093 s
MethodGen.init: 1.670222455 s
MethodGen.getMethod: 2.061828405 s
ClassGen.getJavaClass.getBytes: 0.540762239 s
Total: 5.888046098 s
ClassParser.parse: 0.452568062 s
ClassGen.init: 0.423769764 s
MethodGen.init: 1.369177618 s
MethodGen.getMethod: 2.015685293 s
ClassGen.getJavaClass.getBytes: 0.435557417 s
Total: 5.059064222 s
ClassParser.parse: 0.494708779 s
ClassGen.init: 0.455453825 s
MethodGen.init: 1.24245897 s
MethodGen.getMethod: 1.937969884 s
ClassGen.getJavaClass.getBytes: 0.441652383 s
Total: 4.911468856 s
// Tune InvokeInstruction: Determine size directly
ClassParser.parse: 0.660422314 s
ClassGen.init: 0.515718571 s
MethodGen.init: 1.522720755 s
MethodGen.getMethod: 1.321707714 s
ClassGen.getJavaClass.getBytes: 0.542893948 s
Total: 4.963233706 s
ClassParser.parse: 0.447217948 s
ClassGen.init: 0.402592348 s
MethodGen.init: 1.30743097 s
MethodGen.getMethod: 1.242607624 s
ClassGen.getJavaClass.getBytes: 0.445710399 s
Total: 4.194830316 s
ClassParser.parse: 0.442218102 s
ClassGen.init: 0.449956988 s
MethodGen.init: 1.335540618 s
MethodGen.getMethod: 1.136129872 s
ClassGen.getJavaClass.getBytes: 0.441781442 s
Total: 4.132813855 s
--
Configure bugmail: http://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
---------------------------------------------------------------------
To unsubscribe, e-mail: bcel-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: bcel-dev-help@jakarta.apache.org
DO NOT REPLY [Bug 44340] - BCEL is unnecessarily slow
Posted by bu...@apache.org.
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG�
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://issues.apache.org/bugzilla/show_bug.cgi?id=44340>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND�
INSERTED IN THE BUG DATABASE.
http://issues.apache.org/bugzilla/show_bug.cgi?id=44340
tcurdt@apache.org changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution| |FIXED
------- Additional Comments From tcurdt@apache.org 2008-02-02 13:06 -------
Applied ....thanks!
--
Configure bugmail: http://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
---------------------------------------------------------------------
To unsubscribe, e-mail: bcel-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: bcel-dev-help@jakarta.apache.org