Merge pull request #2811 from FarooqAbdulla02/perf_lock

Adding "perf_lock" Tests.

perf/perf_lock.py

@@ -0,0 +1,124 @@
+#!/usr/bin/env python
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+#
+# See LICENSE for more details.
+#
+# Copyright: 2024 IBM
+# Author: Shaik Abdulla <[email protected]>
+
+import os
+import platform
+from avocado import Test
+from avocado.utils import distro, process, dmesg
+from avocado.utils.software_manager.manager import SoftwareManager
+
+
+class perf_lock(Test):
+
+    """
+    Tests perf lock and it's options namely record, report, info, script
+    contention with all possible flags with the help of yaml file
+    :avocado: tags=perf
+    """
+
+    def setUp(self):
+        '''
+        Install the basic packages to support perf
+        '''
+
+        # Check for basic utilities
+        smm = SoftwareManager()
+        detected_distro = distro.detect()
+        self.distro_name = detected_distro.name
+
+        deps = ['gcc', 'make']
+        if self.distro_name in ['rhel', 'SuSE', 'fedora', 'centos']:
+            deps.extend(['perf'])
+        else:
+            self.cancel("Install the package for perf supported \
+                         by %s" % detected_distro.name)
+        for package in deps:
+            if not smm.check_installed(package) and not smm.install(package):
+                self.cancel('%s is needed for the test to be run' % package)
+
+        # output file to capture perf events
+        self.output_file = "perf.data"
+
+        # Check perf lock is available in the system.
+        output = process.getoutput('perf list', shell=True)
+        if 'lock:contention_begin' in output or \
+           'lock:contention_end' in output:
+            self.log.info("perf lock is available")
+        else:
+            self.cancel('perf lock is not available')
+
+        # Getting the parameters from yaml file
+        self.report = self.params.get('name', default='report')
+        self.option = self.params.get('option', default='')
+
+        # Clear the dmesg, by that we can capture the delta at
+        # the end of the test.
+        dmesg.clear_dmesg()
+
+    def run_cmd(self, cmd):
+        try:
+            process.run(cmd, ignore_status=False, sudo=True)
+        except process.CmdError as details:
+            self.fail("Command %s failed: %s" % (cmd, details))
+
+    def test_lock(self):
+        """
+        'perf lock record' records lock events and this command produces the
+        file "perf.data" which contains tracing results of lock events. The
+        perf.data file is then used as input for report and other commands.
+        """
+        # test perf lock with normal user
+        if self.report == 'normal_user_test':
+            # create temporary user
+            if process.system('useradd test', sudo=True, ignore_status=True):
+                self.log.warn('test useradd failed')
+            cmd = 'perf lock record -o %s -a -v sleep 3' % self.output_file
+            if not process.system('id test', ignore_status=True):
+                if not process.system("su -c '%s' test" % cmd, shell=True,
+                                      ignore_status=True):
+                    self.fail("normal user has access to run perf lock")
+                process.system('userdel -f test', sudo=True)
+            else:
+                self.log.warn('User test does not exist, skipping test')
+        else:
+            # Record command
+            record_cmd = "perf lock record -o %s -- " \
+                         "perf bench sched messaging" % self.output_file
+            self.run_cmd(record_cmd)
+            # Report command
+            if "-i" in self.option:
+                self.option += " " + self.output_file
+            if "--vmlinux" in self.option or "--kallsyms" in self.option:
+                if self.distro_name in ['rhel', 'fedora', 'centos']:
+                    self.option += " /boot/vmlinuz-" + platform.uname()[2]
+                elif self.distro_name in ['SuSE', 'Ubuntu']:
+                    self.option += " /boot/vmlinux-" + platform.uname()[2]
+            report_cmd = "perf lock %s %s" % (self.report, self.option)
+            # Validate the output of perf lock record
+            if os.path.exists(self.output_file):
+                if not os.stat(self.output_file).st_size:
+                    self.fail("%s sample not captured" % self.output_file)
+                else:
+                    self.run_cmd(report_cmd)
+
+        # Verify dmesg
+        dmesg.collect_errors_dmesg(['WARNING: CPU:', 'Oops', 'Segfault',
+                                    'soft lockup', 'Unable to handle'])
+
+    def tearDown(self):
+        # Delete the perf.data file
+        if os.path.isfile(self.output_file):
+            process.run('rm -f %s' % self.output_file)

perf/perf_lock.py.data/readme.txt

@@ -0,0 +1,21 @@
+Overview
+
+This automated script simplifies the usage of the perf lock command in Linux for analyzing system performance by capturing and analyzing lock contention events.
+It streamlines the process of recording and reporting lock events with below mentioned features with supported options.
+
+Features that handled by this script
+
+1. Record 
+2. Report
+3. info
+4. Script
+5. Contention
+
+
+Feature options that handled by YAML
+
+1. Record - all_cpus with verbose enabled.
+2. Report - wait_field, all_fields, lock_with_entries and vmlinux_full.
+3. info - Threads_list and map_instances.
+4. script - input_data.
+5. contention - contetion_fields,all_cpu_entries,average_wait_key,PID_trace,input_full and vmlinux_full.

perf/perf_lock.py.data/record_report.yaml

@@ -0,0 +1,51 @@
+subsystem: !mux
+    record:
+        name: record
+        variants: !mux
+          all_cpus:
+            option: -a -v sleep 3
+
+    report:
+        name: report
+        variants: !mux
+            wait_field:
+                option: -k wait_total -i
+            all_fields:
+                option: -t -F acquired,contended,avg_wait,wait_total
+            locks_with_entries:
+                option: -c -D -E 5
+            vmlinux_full:
+                option: -E 3 --kallsyms
+
+    info:
+        name: info
+        variants: !mux
+            threads_list:
+                option: -t
+            map_instances:
+                option: -m
+
+    script:
+        name: script
+        variants: !mux
+           input_data:
+                option: -f -i
+
+    contention:
+        name: contention
+        variants: !mux
+          contention_fields:
+                  option: -C 0 -E 2 -F contended,wait_total -q
+          all_cpus_entires:
+                  option: -a -t -E 1 -q
+          avg_wait_key:
+                  option: -k avg_wait -l
+          trace_PID:
+                  option:  -p 1 -x ,
+          input_full:
+                  option: -Y spinlock -q -i
+          vmlinux_full:
+                  option: -E 2 --vmlinux
+
+    normal_user_test:
+        name: normal_user_test