Fancy plot for branch predictor.

bin/parseHPMC.py

@@ -2,6 +2,7 @@
 
 import os
 import sys
+import matplotlib.pyplot as plt
 
 def ComputeCPI(benchmark):
     'Computes and inserts CPI into benchmark stats.'
@@ -46,6 +47,15 @@ def ComputeDCacheMissRate(benchmark):
     DCacheMR = 100.0 * int(dataDict['D Cache Miss']) / int(dataDict['D Cache Access'])
     dataDict['DCacheMR'] = DCacheMR
 
+def ComputeAll(benchmarks):
+    for benchmark in benchmarks:
+        ComputeCPI(benchmark)
+        ComputeBranchDirMissRate(benchmark)
+        ComputeBranchTargetMissRate(benchmark)
+        ComputeRASMissRate(benchmark)
+        ComputeInstrClassMissRate(benchmark)
+        ComputeICacheMissRate(benchmark)
+        ComputeDCacheMissRate(benchmark)
     
 def printStats(benchmark):
     (nameString, opt, dataDict) = benchmark
@@ -64,45 +74,92 @@ def printStats(benchmark):
     print('D Cache Miss Rate  %1.4f' % dataDict['DCacheMR'])
     print()
 
+def ProcessFile(fileName):
+    '''Extract preformance counters from a modelsim log.  Outputs a list of tuples for each test/benchmark.
+    The tuple contains the test name, optimization characteristics, and dictionary of performance counters.'''
+    # 1 find lines with Read memfile and extract test name
+    # 2 parse counters into a list of (name, value) tuples (dictionary maybe?)
+    benchmarks = []
+    transcript = open(fileName, 'r')
+    HPMClist = { }
+    testName = ''
+    for line in transcript.readlines():
+        lineToken = line.split()
+        if(len(lineToken) > 3 and lineToken[1] == 'Read' and lineToken[2] == 'memfile'):
+            opt = lineToken[3].split('/')[-4]
+            testName = lineToken[3].split('/')[-1].split('.')[0]
+            HPMClist = { }
+        elif(len(lineToken) > 4 and lineToken[1][0:3] == 'Cnt'):
+            countToken = line.split('=')[1].split()
+            value = countToken[0]
+            name = ' '.join(countToken[1:])
+            HPMClist[name] = value
+        elif ('is done' in line):
+            benchmarks.append((testName, opt, HPMClist))
+    return benchmarks
 
-# 1 find lines with Read memfile and extract test name
+def FormatToPlot(currBenchmark):
-# 2 parse counters into a list of (name, value) tuples (dictionary maybe?)
+    names = []
-# 3 process into useful data
+    values = []
-  # cache hit rates
+    for config in currBenchmark:
-  # cache fill time
+        print ('config' , config)
-  # branch predictor status
+        names.append(config[0])
-  # hazard counts
+        values.append(config[1])
-  # CPI
+    return (names, values)
-  # instruction distribution
 
-# steps 1 and 2
+if(sys.argv[1] == '-b'):
-benchmarks = []
+    configList = []
-transcript = open(sys.argv[1], 'r')
+    for config in sys.argv[2::]:
-HPMClist = { }
+        benchmarks = ProcessFile(config)
-testName = ''
+        ComputeAll(benchmarks)
-for line in transcript.readlines():
+        configList.append((config.split('.')[0], benchmarks))
-    lineToken = line.split()
-    if(len(lineToken) > 3 and lineToken[1] == 'Read' and lineToken[2] == 'memfile'):
-        opt = lineToken[3].split('/')[-4]
-        testName = lineToken[3].split('/')[-1].split('.')[0]
-        HPMClist = { }
-    elif(len(lineToken) > 4 and lineToken[1][0:3] == 'Cnt'):
-        countToken = line.split('=')[1].split()
-        value = countToken[0]
-        name = ' '.join(countToken[1:])
-        HPMClist[name] = value
-    elif ('is done' in line):
-        benchmarks.append((testName, opt, HPMClist))
 
-#print(benchmarks[0])
+    # Merge all configruations into a single list
+    benchmarkAll = []
+    for (config, benchmarks) in configList:
+        print(config)
+        for benchmark in benchmarks:
+            (nameString, opt, dataDict) = benchmark
+            benchmarkAll.append((nameString, opt, config, dataDict))
 
-for benchmark in benchmarks:
+    # now extract all branch prediction direction miss rates for each
-    ComputeCPI(benchmark)
+    # namestring + opt, config
-    ComputeBranchDirMissRate(benchmark)
+    benchmarkDict = { }
-    ComputeBranchTargetMissRate(benchmark)
+    for benchmark in benchmarkAll:
-    ComputeRASMissRate(benchmark)
+        (name, opt, config, dataDict) = benchmark
-    ComputeInstrClassMissRate(benchmark)
+        if name+'_'+opt in benchmarkDict:
-    ComputeICacheMissRate(benchmark)
+            benchmarkDict[name+'_'+opt].append((config, dataDict['BDMR']))
-    ComputeDCacheMissRate(benchmark)
+        else:
-    printStats(benchmark)
+            benchmarkDict[name+'_'+opt] = [(config, dataDict['BDMR'])]
+
+    size = len(benchmarkDict)
+    index = 1
+    print('Number of plots', size)
+    for benchmarkName in benchmarkDict:
+        currBenchmark = benchmarkDict[benchmarkName]
+        (names, values) = FormatToPlot(currBenchmark)
+        print(names, values)
+        plt.subplot(6, 7, index)
+        plt.bar(names, values)
+        plt.title(benchmarkName)
+        plt.ylabel('BR Dir Miss Rate (%)')
+        #plt.xlabel('Predictor')
+        index += 1
+    #plt.tight_layout()
+    plt.show()
+    
+            
+else:
+    # steps 1 and 2
+    benchmarks = ProcessFile(sys.argv[1])
+    # 3 process into useful data
+    # cache hit rates
+    # cache fill time
+    # branch predictor status
+    # hazard counts
+    # CPI
+    # instruction distribution
+    ComputeAll(benchmarks)
+    for benchmark in benchmarks:
+        printStats(benchmark)