A-Tune/collection/parser/perf_parser.py
Zhipeng Xie 4335408875 atune: init code
upload code to gitee

Signed-off-by: Zhipeng Xie <xiezhipeng1@huawei.com>
2019-11-13 17:14:15 +08:00

69 lines
2.8 KiB
Python

#!/usr/bin/python3
# -*- coding: utf-8 -*-
# Copyright (c) 2019 Huawei Technologies Co., Ltd.
# A-Tune is licensed under the Mulan PSL v1.
# You can use this software according to the terms and conditions of the Mulan PSL v1.
# You may obtain a copy of Mulan PSL v1 at:
# http://license.coscl.org.cn/MulanPSL
# THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY OR FIT FOR A PARTICULAR
# PURPOSE.
# See the Mulan PSL v1 for more details.
# Create: 2019-10-29
"""
The parser to parse the output of perf-cpu.sh and perf-mem.sh.
"""
from __future__ import division
from __future__ import print_function
from . import base
class PerfParser(base.Parser):
"""The parser to parse the output of perf-cpu.sh and perf-mem.sh"""
def __init__(self, raw_data_file, data_to_collect, interval=1, **kwargs):
"""Initialize a perf parser.
@param raw_data_file: the path of raw data
@param data_to_collect: list of str which represents the metrics to parse
@param interval: the interval time of data. It is only for perf-mem.sh.
If you want to parse the output of perf-cpu.sh, please
set interval to 1.
@param alias: alias name of output fields (default: "perf")
"""
base.Parser.__init__(self, raw_data_file, data_to_collect, **kwargs)
self._interval = interval
def _check_data_to_collect(self):
"""Read the first batch output of perf-cpu.sh or perf-mem.sh and check
whether or not all metrics in data_to_collect are in the output. If
not, it will raise ValueError.
"""
with open(self._raw_data_file, 'r') as raw_data_fd:
raw_data_fd.readline()
datas = raw_data_fd.readline().split()
diff_set = set(self._data_to_collect) - set(datas)
if diff_set:
raise ValueError("`{}`: Unknown data name `{}`".format(self._raw_data_file, ','.join(diff_set)))
def _get_iter(self):
"""Get the iteration of the perf parser.
@return: the iteration of the perf parser
"""
with open(self._raw_data_file, 'r') as raw_data_fd:
raw_data_fd.readline()
attrs = raw_data_fd.readline().split()
for row_num, line in enumerate(raw_data_fd, 3):
line_part = line.split()
line_data = [float(d[:-1] if d.endswith('%') else d) / self._interval for d in line_part]
if len(attrs) != len(line_data):
print("WARNING: {}: Line {}: The number of columns may be wrong."
.format(self._raw_data_file, row_num))
return
data = dict(zip(attrs, line_data))
yield [data[attr] for attr in self._data_to_collect]