File size: 8,600 Bytes
6705032 0cea2c9 6705032 0cea2c9 6705032 0cea2c9 6705032 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 |
import os, sys,string,re,glob
import json
import csv
import copy
import pathlib
import time
folder = str(pathlib.Path(__file__).parent.resolve())
#Total_encas
func_num_dic = {
"riscv" : 568,
"pulp" : 698,
"xcore" : 188
}
wrong_lis_all = []
wrong_stmt = []
err_def_dic = {}
def get_wrong_list():
global wrong_stmt
global wrong_lis_all
global err_def_dic
with open(folder+"/wrong_func_list_def.csv", 'r', encoding='utf-8') as fcsv:
reader = csv.reader(fcsv)
for row in reader:
if row[0] == "idx":
continue
wrong_stmt.append(row[0].strip().lower() + " " + row[1].strip().lower() + " " + row[2].strip().lower())
wrong_lis_all.append(" ".join(row))
if " ".join([row[2], row[3]]) not in err_def_dic.keys():
err_def_dic[" ".join([row[2], row[3]])]= 1
else:
err_def_dic[" ".join([row[2], row[3]])] += 1
def calculate_accuracy():
func_res = {}
stable_stmt_dic = {}
all_func_lis = []
global wrong_stmt
global wrong_lis_all
global err_def_dic
total_dic = {}
wrong_dic = {}
asm_file = []
for line in open(folder+"/result.jsonl", 'r', encoding="utf-8"):
dic = json.loads(line)
all_func_lis.append(dic["File"].strip().lower() + " " + dic["Func"].strip().lower() + " " + dic["Target"].strip().lower())
if int(dic["vega_pre"]) == 1:
if " ".join([dic["Target"], dic["Module"], dic["File"], dic["Func"]]) not in func_res.keys():
func_res[" ".join([dic["Target"], dic["Module"], dic["File"], dic["Func"]])] = [dic["vega_code"].replace("zmtarzm", dic["Target"])]
else:
func_res[" ".join([dic["Target"], dic["Module"], dic["File"], dic["Func"]])].append(dic["vega_code"].replace("zmtarzm", dic["Target"]))
if " ".join([dic["Target"], dic["Module"]]) not in stable_stmt_dic.keys():
stable_stmt_dic[" ".join([dic["Target"], dic["Module"]])] = [0, 0]
if dic["Stable"].lower() == "true":
stable_stmt_dic[" ".join([dic["Target"], dic["Module"]])][0] += 1
stable_stmt_dic[" ".join([dic["Target"], dic["Module"]])][1] += 1
else:
stable_stmt_dic[" ".join([dic["Target"], dic["Module"]])][1] += 1
if dic["Target"] + " " + dic["Module"] not in total_dic.keys():
total_dic[dic["Target"] + " " + dic["Module"]] = [dic["File"].strip() + " " + dic["Func"].strip() + " " + dic["Target"].strip()]
wrong_dic[dic["Target"] + " " + dic["Module"]] = []
else:
total_dic[dic["Target"] + " " + dic["Module"]].append(dic["File"].strip() + " " + dic["Func"].strip() + " " + dic["Target"].strip())
if dic["File"].strip().lower() + " " + dic["Func"].strip().lower() + " " + dic["Target"].strip().lower() in wrong_stmt:
#print(dic["File"].strip() + " " + dic["Func"].strip() + " " + dic["Target"].strip())
wrong_dic[dic["Target"] + " " + dic["Module"]].append(dic["File"].strip() + " " + dic["Func"].strip() + " " + dic["Target"].strip())
if dic["ans_code"].replace(" ", "") != dic["vega_code"].replace(" ", "") or dic["ans_pre"] != dic["vega_pre"]:
wrong_dic[dic["Target"] + " " + dic["Module"]].append(dic["File"].strip() + " " + dic["Func"].strip() + " " + dic["Target"].strip())
if dic["ans_code"] != dic["vega_code"]:
wrong_lis_all.append(" ".join([dic["File"], dic["Func"], dic["Target"], dic["Module"], "Err_V"]))
if dic["ans_pre"] != dic["vega_pre"]:
wrong_lis_all.append(" ".join([dic["File"], dic["Func"], dic["Target"], dic["Module"], "Err_CS"]))
all_func_lis = list(set(all_func_lis))
with open(folder+"/Fig8_Acc.csv", 'a', encoding='utf-8', newline="") as f:
f_csv = csv.writer(f)
avg_dic = {}
all_dic = {}
for k in total_dic.keys():
Correct_Func_Num = len(list(set(total_dic[k])))-len(list(set(wrong_dic[k])))
Total_Func_Num = len(list(set(total_dic[k])))
Accuracy_Func = 1-round(len(list(set(wrong_dic[k]))) * 1.0 / len(list(set(total_dic[k]))), 3)
Wrong_Func_Percentage = round(len(list(set(wrong_dic[k]))) * 1.0 / len(list(set(total_dic[k]))), 3)
Pre_Equal_1_Stmt_Percentage = round(stable_stmt_dic[k][0]/stable_stmt_dic[k][1], 3)
Pre_Less_1_Stmt_Percentage = 1 - round(stable_stmt_dic[k][0]/stable_stmt_dic[k][1], 3)
if k.split(" ")[0] not in avg_dic.keys():
avg_dic[k.split(" ")[0]] = Accuracy_Func
all_dic[k.split(" ")[0]] = Correct_Func_Num
else:
avg_dic[k.split(" ")[0]] += Accuracy_Func
all_dic[k.split(" ")[0]] += Correct_Func_Num
tem_k = k.replace("PULP", "RI5CY")
f_csv.writerow(tem_k.split(" ") + [Correct_Func_Num, Total_Func_Num, Accuracy_Func, Wrong_Func_Percentage, Pre_Equal_1_Stmt_Percentage, Pre_Less_1_Stmt_Percentage])
for k in avg_dic:
if k.lower() == "riscv":
f_csv.writerow([k, "AVG", round(avg_dic[k] / 7.0, 3)])
f_csv.writerow([k, "ALL", round(all_dic[k] / func_num_dic[k.lower()], 3)])
elif k.lower() == "pulp":
f_csv.writerow(["RI5CY", "AVG", round(avg_dic[k] / 7.0, 3)])
f_csv.writerow(["RI5CY", "ALL", round(all_dic[k] / func_num_dic[k.lower()], 3)])
else:
f_csv.writerow([k, "AVG", round(avg_dic[k] / 6.0, 3)])
f_csv.writerow([k, "ALL", round(all_dic[k] / func_num_dic[k.lower()], 3)])
for k in func_res.keys():
Tar_Path = folder + "/../ForkFlow/VEGA_Code/" + "/".join(k.split(" ")) + ".cpp"
Tar_Path = Tar_Path.replace("enum/NodeType", "enum NodeType")
Tar_Path = Tar_Path.replace("enum/CondCode", "enum CondCode")
Tar_Path = Tar_Path.replace("ExpandSSRInsts/ExpandPseudo", "ExpandSSRInsts/ExpandSSRInsts")
if os.path.exists(Tar_Path):
with open(Tar_Path, 'w') as file:
for idx, l in enumerate(func_res[k]):
if idx < len(func_res[k])-1:
file.write(l)
file.write("\n")
else:
file.write(l)
else:
print(Tar_Path)
return total_dic
if __name__ == '__main__':
get_wrong_list()
with open(folder+"/Fig8_Acc.csv", 'w', encoding='utf-8', newline="") as f:
f_csv = csv.writer(f)
f_csv.writerow(["Target", "Module", "Correct", "Total", "Accurate", "Inaccurate", "Confidence Score≈1.00", "Confidence Score in [0.50, 1.00)"])
total_dic = calculate_accuracy()
wrong_lis_all = list(set(wrong_lis_all))
with open(folder+"/wrong_list_all.csv", 'w', encoding='utf-8', newline="") as f:
f_csv = csv.writer(f)
for err in wrong_lis_all:
f_csv.writerow(err.split(" "))
with open(folder+"/../ForkFlow/wrong_list_all.csv", 'w', encoding='utf-8', newline="") as f:
f_csv = csv.writer(f)
for err in wrong_lis_all:
f_csv.writerow(err.split(" "))
wrong_dic = {}
with open(folder+"/wrong_list_all.csv", 'r', encoding='utf-8') as f:
f_csv = csv.reader(f)
for row in f_csv:
if " ".join([row[-3].lower(), row[-1].lower()]) not in wrong_dic.keys():
wrong_dic[" ".join([row[-3].lower(), row[-1].lower()])] = 1
else:
wrong_dic[" ".join([row[-3].lower(), row[-1].lower()])] += 1
#print(wrong_dic)
target_func_num_dic = {}
for k in total_dic:
if k.split(" ")[0].lower() not in target_func_num_dic:
target_func_num_dic[k.split(" ")[0].lower()] = len(list(set(total_dic[k])))
else:
target_func_num_dic[k.split(" ")[0].lower()] += len(list(set(total_dic[k])))
with open(folder+"/Table2.csv", 'w', encoding='utf-8', newline = "") as f:
f_csv = csv.writer(f)
for k in target_func_num_dic:
#print(target_func_num_dic[k])
for err_type in ["err_v", "err_cs", "err_def"]:
if k + " " + err_type in wrong_dic.keys():
f_csv.writerow([k.replace("pulp", "ri5cy"), err_type, round(float(wrong_dic[k + " " + err_type]) / float(target_func_num_dic[k]), 3)])
else:
f_csv.writerow([k.replace("pulp", "ri5cy"), err_type, 0])
|