* [PATCH v4 1/3] perf test: Add metric value validation test
2023-06-18 17:28 [PATCH v4 0/3] Add metric value validation test Weilin Wang
@ 2023-06-18 17:28 ` Weilin Wang
2023-06-18 17:28 ` [PATCH v4 2/3] perf test: Add skip list for metrics known would fail Weilin Wang
2023-06-18 17:28 ` [PATCH v4 3/3] perf test: Rerun failed metrics with longer workload Weilin Wang
2 siblings, 0 replies; 4+ messages in thread
From: Weilin Wang @ 2023-06-18 17:28 UTC (permalink / raw)
To: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo, Jiri Olsa,
Namhyung Kim, Adrian Hunter, Ian Rogers, linux-perf-users,
linux-kernel
Cc: Weilin Wang, Kan Liang, Samantha Alt, Perry Taylor, Caleb Biggers,
ravi.bangoria
Add metric value validation test to check if metric values are with in
correct value ranges. There are three types of tests included: 1)
positive-value test checks if all the metrics collected are non-negative;
2) single-value test checks if the list of metrics have values in given
value ranges; 3) relationship test checks if multiple metrics follow a
given relationship, e.g. memory_bandwidth_read + memory_bandwidth_write =
memory_bandwidth_total.
Signed-off-by: Weilin Wang <weilin.wang@intel.com>
---
.../tests/shell/lib/perf_metric_validation.py | 514 ++++++++++++++++++
.../lib/perf_metric_validation_rules.json | 387 +++++++++++++
tools/perf/tests/shell/stat_metrics_values.sh | 30 +
3 files changed, 931 insertions(+)
create mode 100644 tools/perf/tests/shell/lib/perf_metric_validation.py
create mode 100644 tools/perf/tests/shell/lib/perf_metric_validation_rules.json
create mode 100755 tools/perf/tests/shell/stat_metrics_values.sh
diff --git a/tools/perf/tests/shell/lib/perf_metric_validation.py b/tools/perf/tests/shell/lib/perf_metric_validation.py
new file mode 100644
index 000000000000..81bd2bf38b67
--- /dev/null
+++ b/tools/perf/tests/shell/lib/perf_metric_validation.py
@@ -0,0 +1,514 @@
+#SPDX-License-Identifier: GPL-2.0
+import re
+import csv
+import json
+import argparse
+from pathlib import Path
+import subprocess
+
+class Validator:
+ def __init__(self, rulefname, reportfname='', t=5, debug=False, datafname='', fullrulefname='', workload='true', metrics=''):
+ self.rulefname = rulefname
+ self.reportfname = reportfname
+ self.rules = None
+ self.collectlist=metrics
+ self.metrics = set()
+ self.tolerance = t
+
+ self.workloads = [x for x in workload.split(",") if x]
+ self.wlidx = 0 # idx of current workloads
+ self.allresults = dict() # metric results of all workload
+ self.allignoremetrics = dict() # metrics with no results or negative results
+ self.allfailtests = dict()
+ self.alltotalcnt = dict()
+ self.allpassedcnt = dict()
+ self.allerrlist = dict()
+
+ self.results = dict() # metric results of current workload
+ # vars for test pass/failure statistics
+ self.ignoremetrics= set() # metrics with no results or negative results, neg result counts as a failed test
+ self.failtests = dict()
+ self.totalcnt = 0
+ self.passedcnt = 0
+ # vars for errors
+ self.errlist = list()
+
+ # vars for Rule Generator
+ self.pctgmetrics = set() # Percentage rule
+
+ # vars for debug
+ self.datafname = datafname
+ self.debug = debug
+ self.fullrulefname = fullrulefname
+
+ def read_json(self, filename: str) -> dict:
+ try:
+ with open(Path(filename).resolve(), "r") as f:
+ data = json.loads(f.read())
+ except OSError as e:
+ print(f"Error when reading file {e}")
+ sys.exit()
+
+ return data
+
+ def json_dump(self, data, output_file):
+ parent = Path(output_file).parent
+ if not parent.exists():
+ parent.mkdir(parents=True)
+
+ with open(output_file, "w+") as output_file:
+ json.dump(data,
+ output_file,
+ ensure_ascii=True,
+ indent=4)
+
+ def get_results(self, idx:int = 0):
+ return self.results[idx]
+
+ def get_bounds(self, lb, ub, error, alias={}, ridx:int = 0) -> list:
+ """
+ Get bounds and tolerance from lb, ub, and error.
+ If missing lb, use 0.0; missing ub, use float('inf); missing error, use self.tolerance.
+
+ @param lb: str/float, lower bound
+ @param ub: str/float, upper bound
+ @param error: float/str, error tolerance
+ @returns: lower bound, return inf if the lower bound is a metric value and is not collected
+ upper bound, return -1 if the upper bound is a metric value and is not collected
+ tolerance, denormalized base on upper bound value
+ """
+ # init ubv and lbv to invalid values
+ def get_bound_value (bound, initval, ridx):
+ val = initval
+ if isinstance(bound, int) or isinstance(bound, float):
+ val = bound
+ elif isinstance(bound, str):
+ if bound == '':
+ val = float("inf")
+ elif bound in alias:
+ vall = self.get_value(alias[ub], ridx)
+ if vall:
+ val = vall[0]
+ elif bound.replace('.', '1').isdigit():
+ val = float(bound)
+ else:
+ print("Wrong bound: {0}".format(bound))
+ else:
+ print("Wrong bound: {0}".format(bound))
+ return val
+
+ ubv = get_bound_value(ub, -1, ridx)
+ lbv = get_bound_value(lb, float('inf'), ridx)
+ t = get_bound_value(error, self.tolerance, ridx)
+
+ # denormalize error threshold
+ denormerr = t * ubv / 100 if ubv != 100 and ubv > 0 else t
+
+ return lbv, ubv, denormerr
+
+ def get_value(self, name:str, ridx:int = 0) -> list:
+ """
+ Get value of the metric from self.results.
+ If result of this metric is not provided, the metric name will be added into self.ignoremetics and self.errlist.
+ All future test(s) on this metric will fail.
+
+ @param name: name of the metric
+ @returns: list with value found in self.results; list is empty when not value found.
+ """
+ results = []
+ data = self.results[ridx] if ridx in self.results else self.results[0]
+ if name not in self.ignoremetrics:
+ if name in data:
+ results.append(data[name])
+ elif name.replace('.', '1').isdigit():
+ results.append(float(name))
+ else:
+ self.errlist.append("Metric '%s' is not collected or the value format is incorrect"%(name))
+ self.ignoremetrics.add(name)
+ return results
+
+ def check_bound(self, val, lb, ub, err):
+ return True if val <= ub + err and val >= lb - err else False
+
+ # Positive Value Sanity check
+ def pos_val_test(self):
+ """
+ Check if metrics value are non-negative.
+ One metric is counted as one test.
+ Failure: when metric value is negative or not provided.
+ Metrics with negative value will be added into the self.failtests['PositiveValueTest'] and self.ignoremetrics.
+ """
+ negmetric = set()
+ missmetric = set()
+ pcnt = 0
+ tcnt = 0
+ for name, val in self.get_results().items():
+ if val is None or val == '':
+ missmetric.add(name)
+ self.errlist.append("Metric '%s' is not collected"%(name))
+ elif val < 0:
+ negmetric.add("{0}(={1:.4f})".format(name, val))
+ else:
+ pcnt += 1
+ tcnt += 1
+
+ self.failtests['PositiveValueTest']['Total Tests'] = tcnt
+ self.failtests['PositiveValueTest']['Passed Tests'] = pcnt
+ if len(negmetric) or len(missmetric)> 0:
+ self.ignoremetrics.update(negmetric)
+ self.ignoremetrics.update(missmetric)
+ self.failtests['PositiveValueTest']['Failed Tests'].append({'NegativeValue':list(negmetric), 'MissingValue':list(missmetric)})
+
+ return
+
+ def evaluate_formula(self, formula:str, alias:dict, ridx:int = 0):
+ """
+ Evaluate the value of formula.
+
+ @param formula: the formula to be evaluated
+ @param alias: the dict has alias to metric name mapping
+ @returns: value of the formula is success; -1 if the one or more metric value not provided
+ """
+ stack = []
+ b = 0
+ errs = []
+ sign = "+"
+ f = str()
+
+ #TODO: support parenthesis?
+ for i in range(len(formula)):
+ if i+1 == len(formula) or formula[i] in ('+', '-', '*', '/'):
+ s = alias[formula[b:i]] if i+1 < len(formula) else alias[formula[b:]]
+ v = self.get_value(s, ridx)
+ if not v:
+ errs.append(s)
+ else:
+ f = f + "{0}(={1:.4f})".format(s, v[0])
+ if sign == "*":
+ stack[-1] = stack[-1] * v
+ elif sign == "/":
+ stack[-1] = stack[-1] / v
+ elif sign == '-':
+ stack.append(-v[0])
+ else:
+ stack.append(v[0])
+ if i + 1 < len(formula):
+ sign = formula[i]
+ f += sign
+ b = i + 1
+
+ if len(errs) > 0:
+ return -1, "Metric value missing: "+','.join(errs)
+
+ val = sum(stack)
+ return val, f
+
+ # Relationships Tests
+ def relationship_test(self, rule: dict):
+ """
+ Validate if the metrics follow the required relationship in the rule.
+ eg. lower_bound <= eval(formula)<= upper_bound
+ One rule is counted as ont test.
+ Failure: when one or more metric result(s) not provided, or when formula evaluated outside of upper/lower bounds.
+
+ @param rule: dict with metric name(+alias), formula, and required upper and lower bounds.
+ """
+ alias = dict()
+ for m in rule['Metrics']:
+ alias[m['Alias']] = m['Name']
+ lbv, ubv, t = self.get_bounds(rule['RangeLower'], rule['RangeUpper'], rule['ErrorThreshold'], alias, ridx=rule['RuleIndex'])
+ val, f = self.evaluate_formula(rule['Formula'], alias, ridx=rule['RuleIndex'])
+ if val == -1:
+ self.failtests['RelationshipTest']['Failed Tests'].append({'RuleIndex': rule['RuleIndex'], 'Description':f})
+ elif not self.check_bound(val, lbv, ubv, t):
+ lb = rule['RangeLower']
+ ub = rule['RangeUpper']
+ if isinstance(lb, str):
+ if lb in alias:
+ lb = alias[lb]
+ if isinstance(ub, str):
+ if ub in alias:
+ ub = alias[ub]
+ self.failtests['RelationshipTest']['Failed Tests'].append({'RuleIndex': rule['RuleIndex'], 'Formula':f,
+ 'RangeLower': lb, 'LowerBoundValue': self.get_value(lb),
+ 'RangeUpper': ub, 'UpperBoundValue':self.get_value(ub),
+ 'ErrorThreshold': t, 'CollectedValue': val})
+ else:
+ self.passedcnt += 1
+ self.failtests['RelationshipTest']['Passed Tests'] += 1
+ self.totalcnt += 1
+ self.failtests['RelationshipTest']['Total Tests'] += 1
+
+ return
+
+
+ # Single Metric Test
+ def single_test(self, rule:dict):
+ """
+ Validate if the metrics are in the required value range.
+ eg. lower_bound <= metrics_value <= upper_bound
+ One metric is counted as one test in this type of test.
+ One rule may include one or more metrics.
+ Failure: when the metric value not provided or the value is outside the bounds.
+ This test updates self.total_cnt and records failed tests in self.failtest['SingleMetricTest'].
+
+ @param rule: dict with metrics to validate and the value range requirement
+ """
+ lbv, ubv, t = self.get_bounds(rule['RangeLower'], rule['RangeUpper'], rule['ErrorThreshold'])
+ metrics = rule['Metrics']
+ passcnt = 0
+ totalcnt = 0
+ faillist = []
+ for m in metrics:
+ totalcnt += 1
+ result = self.get_value(m['Name'])
+ if len(result) > 0 and self.check_bound(result[0], lbv, ubv, t):
+ passcnt += 1
+ else:
+ faillist.append({'MetricName':m['Name'], 'CollectedValue':result})
+
+ self.totalcnt += totalcnt
+ self.passedcnt += passcnt
+ self.failtests['SingleMetricTest']['Total Tests'] += totalcnt
+ self.failtests['SingleMetricTest']['Passed Tests'] += passcnt
+ if len(faillist) != 0:
+ self.failtests['SingleMetricTest']['Failed Tests'].append({'RuleIndex':rule['RuleIndex'],
+ 'RangeLower': rule['RangeLower'],
+ 'RangeUpper': rule['RangeUpper'],
+ 'ErrorThreshold':rule['ErrorThreshold'],
+ 'Failure':faillist})
+
+ return
+
+ def create_report(self):
+ """
+ Create final report and write into a JSON file.
+ """
+ alldata = list()
+ for i in range(0, len(self.workloads)):
+ reportstas = {"Total Rule Count": self.alltotalcnt[i], "Passed Rule Count": self.allpassedcnt[i]}
+ data = {"Metric Validation Statistics": reportstas, "Tests in Category": self.allfailtests[i],
+ "Errors":self.allerrlist[i]}
+ alldata.append({"Workload": self.workloads[i], "Report": data})
+
+ json_str = json.dumps(alldata, indent=4)
+ print("Test validation finished. Final report: ")
+ print(json_str)
+
+ if self.debug:
+ allres = [{"Workload": self.workloads[i], "Results": self.allresults[i]} for i in range(0, len(self.workloads))]
+ self.json_dump(allres, self.datafname)
+
+ def check_rule(self, testtype, metric_list):
+ """
+ Check if the rule uses metric(s) that not exist in current platform.
+
+ @param metric_list: list of metrics from the rule.
+ @return: False when find one metric out in Metric file. (This rule should not skipped.)
+ True when all metrics used in the rule are found in Metric file.
+ """
+ if testtype == "RelationshipTest":
+ for m in metric_list:
+ if m['Name'] not in self.metrics:
+ return False
+ return True
+
+ # Start of Collector and Converter
+ def convert(self, data: list, idx: int):
+ """
+ Convert collected metric data from the -j output to dict of {metric_name:value}.
+ """
+ for json_string in data:
+ try:
+ result =json.loads(json_string)
+ if "metric-unit" in result and result["metric-unit"] != "(null)" and result["metric-unit"] != "":
+ name = result["metric-unit"].split(" ")[1] if len(result["metric-unit"].split(" ")) > 1 \
+ else result["metric-unit"]
+ if idx not in self.results: self.results[idx] = dict()
+ self.results[idx][name.lower()] = float(result["metric-value"])
+ except ValueError as error:
+ continue
+ return
+
+ def collect_perf(self, data_file: str, workload: str):
+ """
+ Collect metric data with "perf stat -M" on given workload with -a and -j.
+ """
+ self.results = dict()
+ tool = 'perf'
+ print(f"Starting perf collection")
+ print(f"Workload: {workload}")
+ collectlist = dict()
+ if self.collectlist != "":
+ collectlist[0] = {x for x in self.collectlist.split(",")}
+ else:
+ collectlist[0] = set(list(self.metrics))
+ # Create metric set for relationship rules
+ for rule in self.rules:
+ if rule["TestType"] == "RelationshipTest":
+ metrics = [m["Name"] for m in rule["Metrics"]]
+ if not any(m not in collectlist[0] for m in metrics):
+ collectlist[rule["RuleIndex"]] = set(metrics)
+
+ for idx, metrics in collectlist.items():
+ if idx == 0: wl = "sleep 0.5".split()
+ else: wl = workload.split()
+ for metric in metrics:
+ command = [tool, 'stat', '-j', '-M', f"{metric}", "-a"]
+ command.extend(wl)
+ cmd = subprocess.run(command, stderr=subprocess.PIPE, encoding='utf-8')
+ data = [x+'}' for x in cmd.stderr.split('}\n') if x]
+ self.convert(data, idx)
+ # End of Collector and Converter
+
+ # Start of Rule Generator
+ def parse_perf_metrics(self):
+ """
+ Read and parse perf metric file:
+ 1) find metrics with '1%' or '100%' as ScaleUnit for Percent check
+ 2) create metric name list
+ """
+ command = ['perf', 'list', '-j', '--details', 'metrics']
+ cmd = subprocess.run(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE, encoding='utf-8')
+ try:
+ data = json.loads(cmd.stdout)
+ for m in data:
+ if 'MetricName' not in m:
+ print("Warning: no metric name")
+ continue
+ name = m['MetricName']
+ self.metrics.add(name)
+ if 'ScaleUnit' in m and (m['ScaleUnit'] == '1%' or m['ScaleUnit'] == '100%'):
+ self.pctgmetrics.add(name.lower())
+ except ValueError as error:
+ print(f"Error when parsing metric data")
+ sys.exit()
+
+ return
+
+ def create_rules(self):
+ """
+ Create full rules which includes:
+ 1) All the rules from the "relationshi_rules" file
+ 2) SingleMetric rule for all the 'percent' metrics
+
+ Reindex all the rules to avoid repeated RuleIndex
+ """
+ self.rules = self.read_json(self.rulefname)['RelationshipRules']
+ pctgrule = {'RuleIndex':0,
+ 'TestType':'SingleMetricTest',
+ 'RangeLower':'0',
+ 'RangeUpper': '100',
+ 'ErrorThreshold': self.tolerance,
+ 'Description':'Metrics in percent unit have value with in [0, 100]',
+ 'Metrics': [{'Name': m} for m in self.pctgmetrics]}
+ self.rules.append(pctgrule)
+
+ # Re-index all rules to avoid repeated RuleIndex
+ idx = 1
+ for r in self.rules:
+ r['RuleIndex'] = idx
+ idx += 1
+
+ if self.debug:
+ #TODO: need to test and generate file name correctly
+ data = {'RelationshipRules':self.rules, 'SupportedMetrics': [{"MetricName": name} for name in self.metrics]}
+ self.json_dump(data, self.fullrulefname)
+
+ return
+ # End of Rule Generator
+
+ def _storewldata(self, key):
+ '''
+ Store all the data of one workload into the corresponding data structure for all workloads.
+ @param key: key to the dictionaries (index of self.workloads).
+ '''
+ self.allresults[key] = self.results
+ self.allignoremetrics[key] = self.ignoremetrics
+ self.allfailtests[key] = self.failtests
+ self.alltotalcnt[key] = self.totalcnt
+ self.allpassedcnt[key] = self.passedcnt
+ self.allerrlist[key] = self.errlist
+
+ #Initialize data structures before data validation of each workload
+ def _init_data(self):
+
+ testtypes = ['PositiveValueTest', 'RelationshipTest', 'SingleMetricTest']
+ self.results = dict()
+ self.ignoremetrics= set()
+ self.errlist = list()
+ self.failtests = {k:{'Total Tests':0, 'Passed Tests':0, 'Failed Tests':[]} for k in testtypes}
+ self.totalcnt = 0
+ self.passedcnt = 0
+
+ def test(self):
+ '''
+ The real entry point of the test framework.
+ This function loads the validation rule JSON file and Standard Metric file to create rules for
+ testing and namemap dictionaries.
+ It also reads in result JSON file for testing.
+
+ In the test process, it passes through each rule and launch correct test function bases on the
+ 'TestType' field of the rule.
+
+ The final report is written into a JSON file.
+ '''
+ self.parse_perf_metrics()
+ self.create_rules()
+ for i in range(0, len(self.workloads)):
+ self._init_data()
+ self.collect_perf(self.datafname, self.workloads[i])
+ # Run positive value test
+ self.pos_val_test()
+ for r in self.rules:
+ # skip rules that uses metrics not exist in this platform
+ testtype = r['TestType']
+ if not self.check_rule(testtype, r['Metrics']):
+ continue
+ if testtype == 'RelationshipTest':
+ self.relationship_test(r)
+ elif testtype == 'SingleMetricTest':
+ self.single_test(r)
+ else:
+ print("Unsupported Test Type: ", testtype)
+ self.errlist.append("Unsupported Test Type from rule: " + r['RuleIndex'])
+ self._storewldata(i)
+ print("Workload: ", self.workloads[i])
+ print("Total metrics collected: ", self.failtests['PositiveValueTest']['Total Tests'])
+ print("Non-negative metric count: ", self.failtests['PositiveValueTest']['Passed Tests'])
+ print("Total Test Count: ", self.totalcnt)
+ print("Passed Test Count: ", self.passedcnt)
+
+ self.create_report()
+ return sum(self.alltotalcnt.values()) != sum(self.allpassedcnt.values())
+# End of Class Validator
+
+
+def main() -> None:
+ parser = argparse.ArgumentParser(description="Launch metric value validation")
+
+ parser.add_argument("-rule", help="Base validation rule file", required=True)
+ parser.add_argument("-output_dir", help="Path for validator output file, report file", required=True)
+ parser.add_argument("-debug", help="Debug run, save intermediate data to files", action="store_true", default=False)
+ parser.add_argument("-wl", help="Workload to run while data collection", default="true")
+ parser.add_argument("-m", help="Metric list to validate", default="")
+ args = parser.parse_args()
+ outpath = Path(args.output_dir)
+ reportf = Path.joinpath(outpath, 'perf_report.json')
+ fullrule = Path.joinpath(outpath, 'full_rule.json')
+ datafile = Path.joinpath(outpath, 'perf_data.json')
+
+ validator = Validator(args.rule, reportf, debug=args.debug,
+ datafname=datafile, fullrulefname=fullrule, workload=args.wl,
+ metrics=args.m)
+ ret = validator.test()
+
+ return ret
+
+
+if __name__ == "__main__":
+ import sys
+ sys.exit(main())
+
+
+
diff --git a/tools/perf/tests/shell/lib/perf_metric_validation_rules.json b/tools/perf/tests/shell/lib/perf_metric_validation_rules.json
new file mode 100644
index 000000000000..debaa910da9f
--- /dev/null
+++ b/tools/perf/tests/shell/lib/perf_metric_validation_rules.json
@@ -0,0 +1,387 @@
+{
+ "RelationshipRules": [
+ {
+ "RuleIndex": 1,
+ "Formula": "a+b",
+ "TestType": "RelationshipTest",
+ "RangeLower": "c",
+ "RangeUpper": "c",
+ "ErrorThreshold": 5.0,
+ "Description": "Intel(R) Optane(TM) Persistent Memory(PMEM) bandwidth total includes Intel(R) Optane(TM) Persistent Memory(PMEM) read bandwidth and Intel(R) Optane(TM) Persistent Memory(PMEM) write bandwidth",
+ "Metrics": [
+ {
+ "Name": "pmem_memory_bandwidth_read",
+ "Alias": "a"
+ },
+ {
+ "Name": "pmem_memory_bandwidth_write",
+ "Alias": "b"
+ },
+ {
+ "Name": "pmem_memory_bandwidth_total",
+ "Alias": "c"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 2,
+ "Formula": "a+b",
+ "TestType": "RelationshipTest",
+ "RangeLower": "c",
+ "RangeUpper": "c",
+ "ErrorThreshold": 5.0,
+ "Description": "DDR memory bandwidth total includes DDR memory read bandwidth and DDR memory write bandwidth",
+ "Metrics": [
+ {
+ "Name": "memory_bandwidth_read",
+ "Alias": "a"
+ },
+ {
+ "Name": "memory_bandwidth_write",
+ "Alias": "b"
+ },
+ {
+ "Name": "memory_bandwidth_total",
+ "Alias": "c"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 3,
+ "Formula": "a+b",
+ "TestType": "RelationshipTest",
+ "RangeLower": "100",
+ "RangeUpper": "100",
+ "ErrorThreshold": 5.0,
+ "Description": "Total memory read accesses includes memory reads from last level cache (LLC) addressed to local DRAM and memory reads from the last level cache (LLC) addressed to remote DRAM.",
+ "Metrics": [
+ {
+ "Name": "numa_reads_addressed_to_local_dram",
+ "Alias": "a"
+ },
+ {
+ "Name": "numa_reads_addressed_to_remote_dram",
+ "Alias": "b"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 4,
+ "Formula": "a",
+ "TestType": "SingleMetricTest",
+ "RangeLower": "0.125",
+ "RangeUpper": "",
+ "ErrorThreshold": "",
+ "Description": "",
+ "Metrics": [
+ {
+ "Name": "cpi",
+ "Alias": "a"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 5,
+ "Formula": "",
+ "TestType": "SingleMetricTest",
+ "RangeLower": "0",
+ "RangeUpper": "1",
+ "ErrorThreshold": 5.0,
+ "Description": "Ratio values should be within value range [0,1)",
+ "Metrics": [
+ {
+ "Name": "loads_per_instr",
+ "Alias": ""
+ },
+ {
+ "Name": "stores_per_instr",
+ "Alias": ""
+ },
+ {
+ "Name": "l1d_mpi",
+ "Alias": ""
+ },
+ {
+ "Name": "l1d_demand_data_read_hits_per_instr",
+ "Alias": ""
+ },
+ {
+ "Name": "l1_i_code_read_misses_with_prefetches_per_instr",
+ "Alias": ""
+ },
+ {
+ "Name": "l2_demand_data_read_hits_per_instr",
+ "Alias": ""
+ },
+ {
+ "Name": "l2_mpi",
+ "Alias": ""
+ },
+ {
+ "Name": "l2_demand_data_read_mpi",
+ "Alias": ""
+ },
+ {
+ "Name": "l2_demand_code_mpi",
+ "Alias": ""
+ }
+ ]
+ },
+ {
+ "RuleIndex": 6,
+ "Formula": "a+b+c+d",
+ "TestType": "RelationshipTest",
+ "RangeLower": "100",
+ "RangeUpper": "100",
+ "ErrorThreshold": 5.0,
+ "Description": "Sum of TMA level 1 metrics should be 100%",
+ "Metrics": [
+ {
+ "Name": "tma_frontend_bound",
+ "Alias": "a"
+ },
+ {
+ "Name": "tma_bad_speculation",
+ "Alias": "b"
+ },
+ {
+ "Name": "tma_backend_bound",
+ "Alias": "c"
+ },
+ {
+ "Name": "tma_retiring",
+ "Alias": "d"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 7,
+ "Formula": "a+b",
+ "TestType": "RelationshipTest",
+ "RangeLower": "c",
+ "RangeUpper": "c",
+ "ErrorThreshold": 5.0,
+ "Description": "Sum of the level 2 children should equal level 1 parent",
+ "Metrics": [
+ {
+ "Name": "tma_fetch_latency",
+ "Alias": "a"
+ },
+ {
+ "Name": "tma_fetch_bandwidth",
+ "Alias": "b"
+ },
+ {
+ "Name": "tma_frontend_bound",
+ "Alias": "c"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 8,
+ "Formula": "a+b",
+ "TestType": "RelationshipTest",
+ "RangeLower": "c",
+ "RangeUpper": "c",
+ "ErrorThreshold": 5.0,
+ "Description": "Sum of the level 2 children should equal level 1 parent",
+ "Metrics": [
+ {
+ "Name": "tma_branch_mispredicts",
+ "Alias": "a"
+ },
+ {
+ "Name": "tma_machine_clears",
+ "Alias": "b"
+ },
+ {
+ "Name": "tma_bad_speculation",
+ "Alias": "c"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 9,
+ "Formula": "a+b",
+ "TestType": "RelationshipTest",
+ "RangeLower": "c",
+ "RangeUpper": "c",
+ "ErrorThreshold": 5.0,
+ "Description": "Sum of the level 2 children should equal level 1 parent",
+ "Metrics": [
+ {
+ "Name": "tma_memory_bound",
+ "Alias": "a"
+ },
+ {
+ "Name": "tma_core_bound",
+ "Alias": "b"
+ },
+ {
+ "Name": "tma_backend_bound",
+ "Alias": "c"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 10,
+ "Formula": "a+b",
+ "TestType": "RelationshipTest",
+ "RangeLower": "c",
+ "RangeUpper": "c",
+ "ErrorThreshold": 5.0,
+ "Description": "Sum of the level 2 children should equal level 1 parent",
+ "Metrics": [
+ {
+ "Name": "tma_light_operations",
+ "Alias": "a"
+ },
+ {
+ "Name": "tma_heavy_operations",
+ "Alias": "b"
+ },
+ {
+ "Name": "tma_retiring",
+ "Alias": "c"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 11,
+ "Formula": "a+b+c",
+ "TestType": "RelationshipTest",
+ "RangeLower": "100",
+ "RangeUpper": "100",
+ "ErrorThreshold": 5.0,
+ "Description": "The all_requests includes the memory_page_empty, memory_page_misses, and memory_page_hits equals.",
+ "Metrics": [
+ {
+ "Name": "memory_page_empty_vs_all_requests",
+ "Alias": "a"
+ },
+ {
+ "Name": "memory_page_misses_vs_all_requests",
+ "Alias": "b"
+ },
+ {
+ "Name": "memory_page_hits_vs_all_requests",
+ "Alias": "c"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 12,
+ "Formula": "a-b",
+ "TestType": "RelationshipTest",
+ "RangeLower": "0",
+ "RangeUpper": "",
+ "ErrorThreshold": 5.0,
+ "Description": "CPU utilization in kernel mode should always be <= cpu utilization",
+ "Metrics": [
+ {
+ "Name": "cpu_utilization",
+ "Alias": "a"
+ },
+ {
+ "Name": "cpu_utilization_in_kernel_mode",
+ "Alias": "b"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 13,
+ "Formula": "a-b",
+ "TestType": "RelationshipTest",
+ "RangeLower": "0",
+ "RangeUpper": "",
+ "ErrorThreshold": 5.0,
+ "Description": "Total L2 misses per instruction should be >= L2 demand data read misses per instruction",
+ "Metrics": [
+ {
+ "Name": "l2_mpi",
+ "Alias": "a"
+ },
+ {
+ "Name": "l2_demand_data_read_mpi",
+ "Alias": "b"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 14,
+ "Formula": "a-b",
+ "TestType": "RelationshipTest",
+ "RangeLower": "0",
+ "RangeUpper": "",
+ "ErrorThreshold": 5.0,
+ "Description": "Total L2 misses per instruction should be >= L2 demand code misses per instruction",
+ "Metrics": [
+ {
+ "Name": "l2_mpi",
+ "Alias": "a"
+ },
+ {
+ "Name": "l2_demand_code_mpi",
+ "Alias": "b"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 15,
+ "Formula": "b+c+d",
+ "TestType": "RelationshipTest",
+ "RangeLower": "a",
+ "RangeUpper": "a",
+ "ErrorThreshold": 5.0,
+ "Description": "L3 data read, rfo, code misses per instruction equals total L3 misses per instruction.",
+ "Metrics": [
+ {
+ "Name": "llc_mpi",
+ "Alias": "a"
+ },
+ {
+ "Name": "llc_data_read_mpi_demand_plus_prefetch",
+ "Alias": "b"
+ },
+ {
+ "Name": "llc_rfo_read_mpi_demand_plus_prefetch",
+ "Alias": "c"
+ },
+ {
+ "Name": "llc_code_read_mpi_demand_plus_prefetch",
+ "Alias": "d"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 16,
+ "Formula": "a",
+ "TestType": "SingleMetricTest",
+ "RangeLower": "0",
+ "RangeUpper": "8",
+ "ErrorThreshold": 0.0,
+ "Description": "Setting generous range for allowable frequencies",
+ "Metrics": [
+ {
+ "Name": "uncore_freq",
+ "Alias": "a"
+ }
+ ]
+ },
+ {
+ "RuleIndex": 17,
+ "Formula": "a",
+ "TestType": "SingleMetricTest",
+ "RangeLower": "0",
+ "RangeUpper": "8",
+ "ErrorThreshold": 0.0,
+ "Description": "Setting generous range for allowable frequencies",
+ "Metrics": [
+ {
+ "Name": "cpu_operating_frequency",
+ "Alias": "a"
+ }
+ ]
+ }
+ ]
+}
\ No newline at end of file
diff --git a/tools/perf/tests/shell/stat_metrics_values.sh b/tools/perf/tests/shell/stat_metrics_values.sh
new file mode 100755
index 000000000000..db57a4761b72
--- /dev/null
+++ b/tools/perf/tests/shell/stat_metrics_values.sh
@@ -0,0 +1,30 @@
+#!/bin/bash
+# perf metrics value validation
+# SPDX-License-Identifier: GPL-2.0
+if [ "x$PYTHON" == "x" ]
+then
+ if which python3 > /dev/null
+ then
+ PYTHON=python3
+ else
+ echo Skipping test, python3 not detected please set environment variable PYTHON.
+ exit 2
+ fi
+fi
+
+grep -q GenuineIntel /proc/cpuinfo || (echo Skipping non-Intel; exit 2)
+
+pythonvalidator=$(dirname $0)/lib/perf_metric_validation.py
+rulefile=$(dirname $0)/lib/perf_metric_validation_rules.json
+tmpdir=$(mktemp -d /tmp/__perf_test.program.XXXXX)
+workload="perf bench futex hash -r 2 -s"
+
+# Add -debug, save data file and full rule file
+echo "Launch python validation script $pythonvalidator"
+echo "Output will be stored in: $tmpdir"
+$PYTHON $pythonvalidator -rule $rulefile -output_dir $tmpdir -wl "${workload}"
+ret=$?
+rm -rf $tmpdir
+
+exit $ret
+
--
2.39.1
^ permalink raw reply related [flat|nested] 4+ messages in thread* [PATCH v4 3/3] perf test: Rerun failed metrics with longer workload
2023-06-18 17:28 [PATCH v4 0/3] Add metric value validation test Weilin Wang
2023-06-18 17:28 ` [PATCH v4 1/3] perf test: " Weilin Wang
2023-06-18 17:28 ` [PATCH v4 2/3] perf test: Add skip list for metrics known would fail Weilin Wang
@ 2023-06-18 17:28 ` Weilin Wang
2 siblings, 0 replies; 4+ messages in thread
From: Weilin Wang @ 2023-06-18 17:28 UTC (permalink / raw)
To: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo, Jiri Olsa,
Namhyung Kim, Adrian Hunter, Ian Rogers, linux-perf-users,
linux-kernel
Cc: Weilin Wang, Kan Liang, Samantha Alt, Perry Taylor, Caleb Biggers,
ravi.bangoria
Rerun failed metrics with longer workload to avoid false failure because
sometimes metric value test fails when running in very short amount of
time. Skip rerun if equal to or more than 20 metrics fail.
Signed-off-by: Weilin Wang <weilin.wang@intel.com>
---
.../tests/shell/lib/perf_metric_validation.py | 129 +++++++++++-------
1 file changed, 83 insertions(+), 46 deletions(-)
diff --git a/tools/perf/tests/shell/lib/perf_metric_validation.py b/tools/perf/tests/shell/lib/perf_metric_validation.py
index 3c3a9b4f8b82..50a34a9cc040 100644
--- a/tools/perf/tests/shell/lib/perf_metric_validation.py
+++ b/tools/perf/tests/shell/lib/perf_metric_validation.py
@@ -11,8 +11,9 @@ class Validator:
self.rulefname = rulefname
self.reportfname = reportfname
self.rules = None
- self.collectlist=metrics
- self.metrics = set(metrics)
+ self.collectlist:str = metrics
+ self.metrics = self.__set_metrics(metrics)
+ self.skiplist = set()
self.tolerance = t
self.workloads = [x for x in workload.split(",") if x]
@@ -41,6 +42,12 @@ class Validator:
self.debug = debug
self.fullrulefname = fullrulefname
+ def __set_metrics(self, metrics=''):
+ if metrics != '':
+ return set(metrics.split(","))
+ else:
+ return set()
+
def read_json(self, filename: str) -> dict:
try:
with open(Path(filename).resolve(), "r") as f:
@@ -113,7 +120,7 @@ class Validator:
All future test(s) on this metric will fail.
@param name: name of the metric
- @returns: list with value found in self.results; list is empty when not value found.
+ @returns: list with value found in self.results; list is empty when value is not found.
"""
results = []
data = self.results[ridx] if ridx in self.results else self.results[0]
@@ -123,7 +130,6 @@ class Validator:
elif name.replace('.', '1').isdigit():
results.append(float(name))
else:
- self.errlist.append("Metric '%s' is not collected or the value format is incorrect"%(name))
self.ignoremetrics.add(name)
return results
@@ -138,27 +144,32 @@ class Validator:
Failure: when metric value is negative or not provided.
Metrics with negative value will be added into the self.failtests['PositiveValueTest'] and self.ignoremetrics.
"""
- negmetric = set()
- missmetric = set()
+ negmetric = dict()
pcnt = 0
tcnt = 0
+ rerun = list()
for name, val in self.get_results().items():
- if val is None or val == '':
- missmetric.add(name)
- self.errlist.append("Metric '%s' is not collected"%(name))
- elif val < 0:
- negmetric.add("{0}(={1:.4f})".format(name, val))
- self.collectlist[0].append(name)
+ if val < 0:
+ negmetric[name] = val
+ rerun.append(name)
else:
pcnt += 1
tcnt += 1
+ if len(rerun) > 0 and len(rerun) < 20:
+ second_results = dict()
+ self.second_test(rerun, second_results)
+ for name, val in second_results.items():
+ if name not in negmetric: continue
+ if val >= 0:
+ del negmetric[name]
+ pcnt += 1
self.failtests['PositiveValueTest']['Total Tests'] = tcnt
self.failtests['PositiveValueTest']['Passed Tests'] = pcnt
- if len(negmetric) or len(missmetric)> 0:
- self.ignoremetrics.update(negmetric)
- self.ignoremetrics.update(missmetric)
- self.failtests['PositiveValueTest']['Failed Tests'].append({'NegativeValue':list(negmetric), 'MissingValue':list(missmetric)})
+ if len(negmetric.keys()):
+ self.ignoremetrics.update(negmetric.keys())
+ negmessage = ["{0}(={1:.4f})".format(name, val) for name, val in negmetric.items()]
+ self.failtests['PositiveValueTest']['Failed Tests'].append({'NegativeValue': negmessage})
return
@@ -259,21 +270,36 @@ class Validator:
metrics = rule['Metrics']
passcnt = 0
totalcnt = 0
- faillist = []
+ faillist = list()
+ failures = dict()
+ rerun = list()
for m in metrics:
totalcnt += 1
result = self.get_value(m['Name'])
- if len(result) > 0 and self.check_bound(result[0], lbv, ubv, t):
+ if len(result) > 0 and self.check_bound(result[0], lbv, ubv, t) or m['Name'] in self.skiplist:
passcnt += 1
else:
- faillist.append({'MetricName':m['Name'], 'CollectedValue':result})
- self.collectlist[0].append(m['Name'])
+ failures[m['Name']] = result
+ rerun.append(m['Name'])
+
+ if len(rerun) > 0 and len(rerun) < 20:
+ second_results = dict()
+ self.second_test(rerun, second_results)
+ for name, val in second_results.items():
+ if name not in failures: continue
+ if self.check_bound(val, lbv, ubv, t):
+ passcnt += 1
+ del failures[name]
+ else:
+ failures[name] = val
+ self.results[0][name] = val
self.totalcnt += totalcnt
self.passedcnt += passcnt
self.failtests['SingleMetricTest']['Total Tests'] += totalcnt
self.failtests['SingleMetricTest']['Passed Tests'] += passcnt
- if len(faillist) != 0:
+ if len(failures.keys()) != 0:
+ faillist = [{'MetricName':name, 'CollectedValue':val} for name, val in failures.items()]
self.failtests['SingleMetricTest']['Failed Tests'].append({'RuleIndex':rule['RuleIndex'],
'RangeLower': rule['RangeLower'],
'RangeUpper': rule['RangeUpper'],
@@ -316,7 +342,7 @@ class Validator:
return True
# Start of Collector and Converter
- def convert(self, data: list, idx: int):
+ def convert(self, data: list, metricvalues:dict):
"""
Convert collected metric data from the -j output to dict of {metric_name:value}.
"""
@@ -326,20 +352,29 @@ class Validator:
if "metric-unit" in result and result["metric-unit"] != "(null)" and result["metric-unit"] != "":
name = result["metric-unit"].split(" ")[1] if len(result["metric-unit"].split(" ")) > 1 \
else result["metric-unit"]
- if idx not in self.results: self.results[idx] = dict()
- self.results[idx][name.lower()] = float(result["metric-value"])
+ metricvalues[name.lower()] = float(result["metric-value"])
except ValueError as error:
continue
return
- def collect_perf(self, data_file: str, workload: str):
+ def _run_perf(self, metric, workload: str):
+ tool = 'perf'
+ command = [tool, 'stat', '-j', '-M', f"{metric}", "-a"]
+ wl = workload.split()
+ command.extend(wl)
+ print(" ".join(command))
+ cmd = subprocess.run(command, stderr=subprocess.PIPE, encoding='utf-8')
+ data = [x+'}' for x in cmd.stderr.split('}\n') if x]
+ return data
+
+
+ def collect_perf(self, workload: str):
"""
Collect metric data with "perf stat -M" on given workload with -a and -j.
"""
self.results = dict()
- tool = 'perf'
print(f"Starting perf collection")
- print(f"Workload: {workload}")
+ print(f"Long workload: {workload}")
collectlist = dict()
if self.collectlist != "":
collectlist[0] = {x for x in self.collectlist.split(",")}
@@ -353,17 +388,20 @@ class Validator:
collectlist[rule["RuleIndex"]] = [",".join(list(set(metrics)))]
for idx, metrics in collectlist.items():
- if idx == 0: wl = "sleep 0.5".split()
- else: wl = workload.split()
+ if idx == 0: wl = "true"
+ else: wl = workload
for metric in metrics:
- command = [tool, 'stat', '-j', '-M', f"{metric}", "-a"]
- command.extend(wl)
- print(" ".join(command))
- cmd = subprocess.run(command, stderr=subprocess.PIPE, encoding='utf-8')
- data = [x+'}' for x in cmd.stderr.split('}\n') if x]
- self.convert(data, idx)
- self.collectlist = dict()
- self.collectlist[0] = list()
+ data = self._run_perf(metric, wl)
+ if idx not in self.results: self.results[idx] = dict()
+ self.convert(data, self.results[idx])
+ return
+
+ def second_test(self, collectlist, second_results):
+ workload = self.workloads[self.wlidx]
+ for metric in collectlist:
+ data = self._run_perf(metric, workload)
+ self.convert(data, second_results)
+
# End of Collector and Converter
# Start of Rule Generator
@@ -381,7 +419,7 @@ class Validator:
if 'MetricName' not in m:
print("Warning: no metric name")
continue
- name = m['MetricName']
+ name = m['MetricName'].lower()
self.metrics.add(name)
if 'ScaleUnit' in m and (m['ScaleUnit'] == '1%' or m['ScaleUnit'] == '100%'):
self.pctgmetrics.add(name.lower())
@@ -391,14 +429,12 @@ class Validator:
return
- def remove_unsupported_rules(self, rules, skiplist: set = None):
- for m in skiplist:
- self.metrics.discard(m)
+ def remove_unsupported_rules(self, rules):
new_rules = []
for rule in rules:
add_rule = True
for m in rule["Metrics"]:
- if m["Name"] not in self.metrics:
+ if m["Name"] in self.skiplist or m["Name"] not in self.metrics:
add_rule = False
break
if add_rule:
@@ -415,15 +451,15 @@ class Validator:
"""
data = self.read_json(self.rulefname)
rules = data['RelationshipRules']
- skiplist = set(data['SkipList'])
- self.rules = self.remove_unsupported_rules(rules, skiplist)
+ self.skiplist = set([name.lower() for name in data['SkipList']])
+ self.rules = self.remove_unsupported_rules(rules)
pctgrule = {'RuleIndex':0,
'TestType':'SingleMetricTest',
'RangeLower':'0',
'RangeUpper': '100',
'ErrorThreshold': self.tolerance,
'Description':'Metrics in percent unit have value with in [0, 100]',
- 'Metrics': [{'Name': m} for m in self.pctgmetrics]}
+ 'Metrics': [{'Name': m.lower()} for m in self.pctgmetrics]}
self.rules.append(pctgrule)
# Re-index all rules to avoid repeated RuleIndex
@@ -479,8 +515,9 @@ class Validator:
self.parse_perf_metrics()
self.create_rules()
for i in range(0, len(self.workloads)):
+ self.wlidx = i
self._init_data()
- self.collect_perf(self.datafname, self.workloads[i])
+ self.collect_perf(self.workloads[i])
# Run positive value test
self.pos_val_test()
for r in self.rules:
--
2.39.1
^ permalink raw reply related [flat|nested] 4+ messages in thread