(1) The usual t3GT3 bug. If no one objects, I am going to remove this from
the test, open/promote a defect, and return to it later once we have the
more important GC bugs worked out.
(2) gctest, SPECjbb, and bytecodeTest are platform-independent. Same as
AIX.
(3) threads test failed inside collector.
Perry
|---------+------------------------------------------------------->
| | Jalapeno (Michael Hind) |
| | <jalapeno@...> |
| | Sent by: |
| | jikesrvm-regression-admin@....|
| | usf.ibm.com |
| | |
| | |
| | 01/30/2003 06:37 AM |
| | Please respond to jikesrvm-regression |
|---------+------------------------------------------------------->
>----------------------------------------------------------------------------------------|
| |
| To: jikesrvm-regression@... |
| cc: |
| Subject: [Jikesrvm-regression] [AIX] regression FAILED 10 tests |
>----------------------------------------------------------------------------------------|
cvs checkout performed on Wed Jan 29 20:05:30 EST 2003
Ran 248
Passed 238
Failed 10
There were no javadoc errors.
Failure Details
---------------
BaseBaseSemiSpace: Failed in jni
FullAdaptiveSemiSpace: Failed in threads
FullAdaptiveSemiSpace: Failed in jni
BaseBaseMarkSweep: Failed in jni
BaseBaseCopyMS: Failed in jni
BaseBaseGenCopy: Failed in gctest
BaseBaseGenCopy: Failed in jni
FullOptSemiSpace: Failed in threads
FastMarkSweep: Failed in bytecodeTests
FastSemiSpace: Failed in SPECjbb2000
Output of this run is available (for 24 hours) at
http://www-124.ibm.com/developerworks/oss/jikesrvm/sanityResults/AIX.tar.gz
Sanity tests are specified in rvm/regression/config/night-sanity
Performance tests are specified in
rvm/regression/config/night-sanity-performance
Performance Summary
Results for SPECjvm98
----------------------------------
FastSemiSpace Result: 48.51
FastWatsonMarkSweep Result: 40.75
FastMarkSweep Result: 42.58
FastAdaptiveSemiSpace Result: 50.78
Results for SPECjbb2000
----------------------------------
FastWatsonMarkSweep Valid run, Score is 8444
FastAdaptiveSemiSpace Valid run, Score is 8311
FastMarkSweep Valid run, Score is 7755
SPECjvm98 Performance Details
-----------------------------
RESULT: FastAdaptiveSemiSpace SPECjvm98
Best Run
compress Time: 15.245 Ratio: 77.07
jess Time: 7.066 Ratio: 53.78
db Time: 19.289 Ratio: 26.18
javac Time: 12.184 Ratio: 34.88
mpegaudio Time: 18.272 Ratio: 60.20
mtrt Time: 4.338 Ratio: 106.04
jack Time: 12.626 Ratio: 36.04
Bottom Line: Result: 50.78
First Run
compress Time: 21.08 Ratio: 55.74
jess Time: 14.374 Ratio: 26.44
db Time: 22.898 Ratio: 22.05
javac Time: 17.769 Ratio: 23.92
mpegaudio Time: 22.088 Ratio: 49.80
mtrt Time: 6.458 Ratio: 71.23
jack Time: 14.188 Ratio: 32.07
First Run Result: 36.63
RESULT: FastSemiSpace SPECjvm98
Best Run
compress Time: 15.406 Ratio: 76.27
jess Time: 7.869 Ratio: 48.29
db Time: 22.708 Ratio: 22.24
javac Time: 12.463 Ratio: 34.10
mpegaudio Time: 18.245 Ratio: 60.29
mtrt Time: 4.393 Ratio: 104.71
jack Time: 12.69 Ratio: 35.86
Bottom Line: Result: 48.51
First Run
compress Time: 20.306 Ratio: 57.86
jess Time: 28.065 Ratio: 13.54
db Time: 25.529 Ratio: 19.78
javac Time: 53.942 Ratio: 7.88
mpegaudio Time: 26.665 Ratio: 41.25
mtrt Time: 11.622 Ratio: 39.58
jack Time: 36.477 Ratio: 12.47
First Run Result: 21.99
RESULT: FastMarkSweep SPECjvm98
Best Run
compress Time: 14.911 Ratio: 78.80
jess Time: 9.764 Ratio: 38.92
db Time: 26.735 Ratio: 18.89
javac Time: 14.751 Ratio: 28.81
mpegaudio Time: 18.288 Ratio: 60.15
mtrt Time: 5.217 Ratio: 88.17
jack Time: 15.881 Ratio: 28.65
Bottom Line: Result: 42.58
First Run
compress Time: 19.08 Ratio: 61.58
jess Time: 49.717 Ratio: 7.64
db Time: 29.975 Ratio: 16.85
javac Time: 84.047 Ratio: 5.06
mpegaudio Time: 32.527 Ratio: 33.82
mtrt Time: 17.461 Ratio: 26.34
jack Time: 87.128 Ratio: 5.22
First Run Result: 15.19
RESULT: FastWatsonMarkSweep SPECjvm98
Best Run
compress Time: 21.129 Ratio: 55.61
jess Time: 8.281 Ratio: 45.89
db Time: 29.908 Ratio: 16.89
javac Time: 13.785 Ratio: 30.83
mpegaudio Time: 21.117 Ratio: 52.09
mtrt Time: 4.406 Ratio: 104.40
jack Time: 17.607 Ratio: 25.84
Bottom Line: Result: 40.75
First Run
compress Time: 23.545 Ratio: 49.90
jess Time: 17.718 Ratio: 21.45
db Time: 30.567 Ratio: 16.52
javac Time: 21.986 Ratio: 19.33
mpegaudio Time: 23.197 Ratio: 47.42
mtrt Time: 5.824 Ratio: 78.98
jack Time: 24.14 Ratio: 18.85
First Run Result: 30.42
Opt Level 0 compilation breakdown for SPECjvm98 size 100
Compilation Subsystem Report
Comp #Meths Time bcb/ms mcb/bcb
MCKB BCKB
JNI 29 2 14.50 496.00
14.0 0.0
Base 64 35 1104.26 12.35
466.0 37.7
Opt 1780 14393 11.81 4.63
768.7 166.0
Explanation of Metrics
#Meths: Total number of methods compiled by the
compiler
Time: Total compilation time in milliseconds
bcb/ms: Number of bytecode bytes complied per
millisecond
mcb/bcb: Ratio of machine code bytes to bytecode bytes
MCKB: Total number of machine code bytes generated in
kilobytes
BCKB: Total number of bytecode bytes compiled in
kilobytes
Optimizing Compiler SubSystem
Phase
Time
(ms)
(0fTotal)
Convert Bytecodes to HIR
Generate HIR 1412 10.85%
Branch Optimizations 125 0.92%
Adjust Branch Probabilities 26 0.15%
TOTAL 1563 12.00%
CFG Structural Analysis
Build LST 157 1.15%
Yield Point Insertion 20 0.15%
Estimate Block Frequencies 109 0.77 %
Infrequent BBs
TOTAL 287 2.15%
Field Analysis 33 0.23%
Convert HIR to LIR
Expand Runtime Services 188 1.38 %
Infrequent RS calls
HIR Operator Expansion 221 1.69%
Branch Optimizations 112 0.85%
Adjust Branch Probabilities 32 0.23%
TOTAL 554 4.23%
Basic Block Frequency Estimation
Build LST 236 1.77%
Estimate Block Frequencies 149 1.08 %
Infrequent BBs
TOTAL 386 2.92%
Code Reordering 36 0.23%
Convert LIR to MIR
SplitBasicBlock 28 0.15%
Instruction Selection
Reduce Operators 43 0.31%
ConvertALUOps 16 0.08%
Normalize Constants 189 1.38%
Live Handlers 16 0.08%
DepGraph & BURS 2583 19.85 %
Infrequent BBs
Complex Operators 249 1.85%
TOTAL 3099 23.77%
TOTAL 3127 24.00%
NullCheckCombining 115 0.85%
Register Mapping
Expand Calling Convention 135 1.00%
Live Analysis 1550 11.92%
Register Allocation
Register Allocation Prep 154 1.15%
Linear Scan Composite Ph
Interval Analysis 445 3.38%
Register Restriction 2432 18.69%
Linear Scan 2110 16.23%
Update GCMaps 1 32 0.23%
Spill Code 332 2.54%
Update GCMaps 2 37 0.23%
TOTAL 5391 41.46%
TOTAL 5545 42.62%
Insert Prologue/Epilogue 157 1.15%
TOTAL 7388 56.77%
Generate Machine Code
Final MIR Expansion 86 0.62%
Assembler Driver 171 1.31%
TOTAL 257 1.92%
TOTAL COMPILATION TIME 13751
Opt Level 1 compilation breakdown for SPECjvm98 size 100
Compilation Subsystem Report
Comp #Meths Time bcb/ms mcb/bcb
MCKB BCKB
JNI 29 3 9.67 496.00 14.0
0.0
Base 64 36 1073.58 12.35
466.0 37.7
Opt 1514 39585 4.28 9.02 1493.0
165.5
Explanation of Metrics
#Meths: Total number of methods compiled by the
compiler
Time: Total compilation time in milliseconds
bcb/ms: Number of bytecode bytes complied per
millisecond
mcb/bcb: Ratio of machine code bytes to bytecode bytes
MCKB: Total number of machine code bytes generated in
kilobytes
BCKB: Total number of bytecode bytes compiled in
kilobytes
Optimizing Compiler SubSystem
Phase
Time
(ms)
(0fTotal)
Convert Bytecodes to HIR
Generate HIR 2267 5.95%
Branch Optimizations 240 0.63%
Adjust Branch Probabilities 32 0.08%
TOTAL 2539 6.66%
Tail Recursion Elimination 29 0.05%
CFG Structural Analysis
Build LST 245 0.63%
Yield Point Insertion 19 0.03%
Estimate Block Frequencies 142 0.37 %
Infrequent BBs
TOTAL 407 1.05%
Simple Opts 374 0.97%
Escape Transformations 394 1.03%
Branch Optimizations 97 0.24%
Static Splitting 34 0.08%
Simple Opts 248 0.63%
Local CopyProp 104 0.26 %
Infrequent BBs
Local ConstantProp 45 0.11 %
Infrequent BBs
Local CSE 563 1.47 %
Infrequent BBs
Field Analysis 32 0.08%
Convert HIR to LIR
Expand Runtime Services 6502 17.11 %
Infrequent RS calls
Branch Optimizations 271 0.71%
Local Cast Optimizations 70 0.18 % Infrequent
BBs
HIR Operator Expansion 302 0.79%
Branch Optimizations 477 1.24%
Adjust Branch Probabilities 69 0.16%
TOTAL 7694 20.24%
Local CopyProp 514 1.34 % Infrequent
BBs
Local ConstantProp 132 0.34 % Infrequent
BBs
Local CSE 819 2.13 % Infrequent
BBs
Simple Opts 805 2.11%
Basic Block Frequency Estimation
Build LST 769 2.00%
Estimate Block Frequencies 388 1.00 %
Infrequent BBs
TOTAL 1158 3.03%
Code Reordering 1210 3.18%
Branch Optimizations 556 1.45%
Convert LIR to MIR
SplitBasicBlock 46 0.11%
Instruction Selection
Reduce Operators 74 0.18%
ConvertALUOps 13 0.03%
Normalize Constants 396 1.03%
Live Handlers 14 0.03%
DepGraph & BURS 4928 12.95 %
Infrequent BBs
Complex Operators 531 1.39%
TOTAL 5959 15.66%
TOTAL 6005 15.79%
NullCheckCombining 229 0.58%
Register Mapping
Expand Calling Convention 740 1.95%
Live Analysis 3568 9.37%
Register Allocation
Register Allocation Prep 151 0.39%
Linear Scan Composite Ph
Interval Analysis 971 2.55%
Register Restriction 3120 8.21%
Linear Scan 3892 10.24%
Update GCMaps 1 45 0.11%
Spill Code 1027 2.68%
Update GCMaps 2 72 0.18%
TOTAL 9130 24.03%
TOTAL 9282 24.42%
Insert Prologue/Epilogue 201 0.53%
TOTAL 13792 36.29%
Branch Optimizations 235 0.61%
Generate Machine Code
Final MIR Expansion 121 0.32%
Assembler Driver 498 1.29%
TOTAL 620 1.63%
TOTAL COMPILATION TIME 38647
Opt Level 2 compilation breakdown for SPECjvm98 size 100
Compilation Subsystem Report
Comp #Meths Time bcb/ms mcb/bcb
MCKB BCKB
JNI 30 3 10.00 496.00
14.5 0.0
Base 64 36 1073.58 12.35
466.0 37.7
Opt 1514 118758 1.43 9.65 1596.7
165.5
Explanation of Metrics
#Meths: Total number of methods compiled by the
compiler
Time: Total compilation time in milliseconds
bcb/ms: Number of bytecode bytes complied per
millisecond
mcb/bcb: Ratio of machine code bytes to bytecode bytes
MCKB: Total number of machine code bytes generated in
kilobytes
BCKB: Total number of bytecode bytes compiled in
kilobytes
Optimizing Compiler SubSystem
Phase
Time
(ms)
(0fTotal)
Convert Bytecodes to HIR
Generate HIR 1946 1.69%
Branch Optimizations 218 0.18%
Adjust Branch Probabilities 32 0.03%
TOTAL 2197 1.90%
Tail Recursion Elimination 30 0.03%
CFG Structural Analysis
Build LST 221 0.19%
Yield Point Insertion 19 0.01%
Estimate Block Frequencies 139 0.11 %
Infrequent BBs
TOTAL 379 0.32%
Simple Opts 356 0.30%
Escape Transformations 368 0.31%
Branch Optimizations 94 0.08%
Static Splitting 37 0.03%
CFGTransformations 357 0.30%
Simple Opts 236 0.20%
Loop Unrolling 653 0.57%
SSA
Basic Block Frequency Estima
Build LST 275 0.23%
Estimate Block Frequenci 177 0.15 %
Infrequent BBs
TOTAL 453 0.39%
HIR SSA transformations
Local CopyProp 116 0.10 %
Infrequent BBs
Local ConstantProp 50 0.04 %
Infrequent BBs
Branch Optimizations 120 0.10%
Dominators + LpStrTree 360 0.31%
Dominance Frontier 616 0.53%
Load Elimination
GVN Preparation 18 0.01%
Enter SSA 1927 1.67%
Global Value Number 1688 1.46%
Load Elimination Pre 105 0.09%
Enter SSA 2878 2.50%
Index Propagation 2637 2.29%
Load Eliminator 152 0.13%
TOTAL 9408 8.17%
Load Elimination
GVN Preparation 17 0.01%
Enter SSA 1108 0.96%
Global Value Number 1300 1.13%
Load Elimination Pre 92 0.08%
Enter SSA 2280 1.98%
Index Propagation 2081 1.81%
Load Eliminator 93 0.08%
TOTAL 6974 6.06%
Load Elimination
GVN Preparation 17 0.01%
Enter SSA 239 0.20%
Global Value Number 282 0.24%
Load Elimination Pre 90 0.08%
Enter SSA 420 0.37%
Index Propagation 325 0.28%
Load Eliminator 31 0.03%
TOTAL 1406 1.22%
Load Elimination
GVN Preparation 16 0.01%
Enter SSA 22 0.02%
Global Value Number 16 0.01%
Load Elimination Pre 84 0.07%
Enter SSA 22 0.02%
Index Propagation 16 0.01%
Load Eliminator 16 0.01%
TOTAL 195 0.17%
Load Elimination
GVN Preparation 17 0.01%
Enter SSA 21 0.02%
Global Value Number 16 0.01%
Load Elimination Pre 83 0.07%
Enter SSA 21 0.02%
Index Propagation 16 0.01%
Load Eliminator 16 0.01%
TOTAL 193 0.17%
RedundantBranchEliminati
Ensure SSA 1278 1.10%
Global Value Number 1508 1.30%
RBE Transform 39 0.03%
TOTAL 2826 2.45%
SSA Tune Up
SSA Tune UpPreparati 18 0.01%
Enter SSA 1267 1.10%
Simple Opts 259 0.22%
SSA Expression Foldi 205 0.17%
TOTAL 1751 1.52%
SSA Tune Up
SSA Tune UpPreparati 17 0.01%
Enter SSA 1246 1.08%
Simple Opts 256 0.22%
SSA Expression Foldi 199 0.17%
TOTAL 1719 1.49%
Global Code Placement
GCP Preparation 18 0.01%
Enter SSA 2129 1.84%
Global CSE 1225 1.06%
Enter SSA 2068 1.79%
LICM 5751 5.00%
GCP Finalization 18 0.01%
TOTAL 11211 9.75%
Leave SSA 1745 1.51%
TOTAL 38700 33.65%
Coalesce Moves 1386 1.20%
Post SSA cleanup
Local CopyProp 39 0.03 %
Infrequent BBs
Local ConstantProp 59 0.04 %
Infrequent BBs
Simple Opts 238 0.20%
Escape Transformations 385 0.33%
Branch Optimizations 150 0.13%
TOTAL 872 0.76%
TOTAL 41412 36.01%
Local CopyProp 34 0.03 %
Infrequent BBs
Local ConstantProp 39 0.03 %
Infrequent BBs
Local CSE 531 0.46 %
Infrequent BBs
Field Analysis 35 0.03%
Convert HIR to LIR
Expand Runtime Services 5904 5.13 %
Infrequent RS calls
Branch Optimizations 264 0.23%
Local Cast Optimizations 65 0.05 % Infrequent
BBs
HIR Operator Expansion 295 0.25%
Branch Optimizations 476 0.41%
Adjust Branch Probabilities 74 0.06%
TOTAL 7081 6.16%
SSA
Basic Block Frequency Estima
Build LST 765 0.66%
Estimate Block Frequenci 401 0.35 %
Infrequent BBs
TOTAL 1166 1.01%
LIR SSA transformations
CFGTransformations 2396 2.08%
Dominators + LpStrTree 961 0.83%
Dominance Frontier 3657 3.17%
Global Code Placement
GCP Preparation 17 0.01%
Enter SSA 1988 1.72%
Global CSE 2398 2.08%
Enter SSA 1373 1.19%
LICM 10779 9.37%
GCP Finalization 17 0.01%
TOTAL 16575 14.41%
Leave SSA 4598 3.99%
TOTAL 28190 24.51%
Coalesce Moves 4426 3.84%
Post SSA cleanup
Local CopyProp 143 0.12 %
Infrequent BBs
Local ConstantProp 111 0.10 %
Infrequent BBs
Simple Opts 713 0.62%
Branch Optimizations 483 0.42%
TOTAL 1451 1.26%
TOTAL 35235 30.63%
Local CopyProp 82 0.07 %
Infrequent BBs
Local ConstantProp 96 0.08 %
Infrequent BBs
Local CSE 742 0.64 %
Infrequent BBs
Simple Opts 649 0.56%
Basic Block Frequency Estimation
Build LST 687 0.59%
Estimate Block Frequencies 364 0.31 %
Infrequent BBs
TOTAL 1051 0.91%
Code Reordering 1021 0.89%
Branch Optimizations 529 0.45%
Convert LIR to MIR
SplitBasicBlock 44 0.03%
Instruction Selection
Reduce Operators 71 0.06%
ConvertALUOps 14 0.01%
Normalize Constants 390 0.34%
Live Handlers 2690 2.34%
DepGraph & BURS 4732 4.11 %
Infrequent BBs
Complex Operators 498 0.43%
TOTAL 8397 7.30%
TOTAL 8442 7.34%
NullCheckCombining 231 0.20%
Register Mapping
Expand Calling Convention 658 0.57%
Live Analysis 3231 2.81%
Register Allocation
Register Allocation Prep 138 0.11%
Linear Scan Composite Ph
Interval Analysis 865 0.75%
Register Restriction 3154 2.74%
Linear Scan 3559 3.09%
Update GCMaps 1 47 0.03%
Spill Code 901 0.78%
Update GCMaps 2 66 0.05%
TOTAL 8593 7.47%
TOTAL 8731 7.59%
Insert Prologue/Epilogue 184 0.16%
TOTAL 12805 11.13%
Branch Optimizations 168 0.14%
Generate Machine Code
Final MIR Expansion 114 0.10%
Assembler Driver 500 0.43%
TOTAL 614 0.53%
TOTAL COMPILATION TIME 115517
_______________________________________________
Jikesrvm-regression mailing list
Jikesrvm-regression@...
http://www-124.ibm.com/developerworks/oss/mailman/listinfo/jikesrvm-regression
|