{ "cells": [ { "cell_type": "markdown", "metadata": { "run_control": { "frozen": false, "read_only": false } }, "source": [ "# Analysis of PELT signal error due to skip last window update" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "collapsed": false, "run_control": { "frozen": false, "marked": false, "read_only": false } }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2017-07-25 18:34:56,088 INFO : root : Using LISA logging configuration:\n", "2017-07-25 18:34:56,090 INFO : root : /home/joelaf/repo/lisa-aosp/external/lisa/logging.conf\n" ] } ], "source": [ "import logging\n", "from conf import LisaLogging\n", "LisaLogging.setup()" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "collapsed": false, "run_control": { "frozen": false, "marked": false, "read_only": false } }, "outputs": [], "source": [ "# Generate plots inline\n", "%matplotlib inline\n", "import json\n", "import os\n", "from trace import Trace\n", "import numpy\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "import trappy\n", "\n", "path_to_dat = \"/home/joelaf/repo/lisa-aosp/external/lisa/ipynb/scratchpad/pelt-error/trace.dat\"" ] }, { "cell_type": "markdown", "metadata": { "run_control": { "frozen": false, "read_only": false } }, "source": [ "## Parse Trace and Profiling Data" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "collapsed": false, "run_control": { "frozen": false, "marked": false, "read_only": false } }, "outputs": [], "source": [ "trace = Trace(None, path_to_dat, events=[ 'sched_switch', 'pelt_update' ])" ] }, { "cell_type": "markdown", "metadata": { "collapsed": true, "run_control": { "frozen": false, "read_only": false } }, "source": [ "## Trace visualization" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "collapsed": false, "run_control": { "frozen": false, "marked": false, "read_only": false } }, "outputs": [ { "data": { "text/html": [ "\n", "
\n", "\n", "\n", "\n", " \n", "
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "trappy.plotter.plot_trace(trace.ftrace)" ] }, { "cell_type": "markdown", "metadata": { "run_control": { "frozen": false, "read_only": false } }, "source": [ "## Latency DataFrames" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "Index([u'__comm', u'__cpu', u'__line', u'__pid', u'__tgid', u'acc_load_avg',\n", " u'acc_util_avg', u'cfs_rq', u'delta_us', u'load_avg', u'load_err',\n", " u'load_sum', u'sum_err', u'util_avg', u'util_err'],\n", " dtype='object')" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df = trace.data_frame.trace_event('pelt_update')\n", "rq_df = df[df.cfs_rq == 1]\n", "rq_df.columns" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "UTIL ERROR\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", "\n", "\n", "\n", "\n", "
\n", "\n", "\n", "\n", "\n", "\n", " \n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LOAD ERROR\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", "\n", "\n", "\n", "\n", "
\n", "\n", "\n", "\n", "\n", "\n", " \n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plot the accurate and the actual signals for the RQ\n", "print 'UTIL ERROR'\n", "trappy.ILinePlot(trace.ftrace,\n", " signals = [\n", " 'pelt_update:util_avg',\n", " 'pelt_update:acc_util_avg',\n", " ]).view()\n", "\n", "print 'LOAD ERROR'\n", "trappy.ILinePlot(trace.ftrace,\n", " signals = [\n", " 'pelt_update:load_avg',\n", " 'pelt_update:acc_load_avg',\n", " ]).view()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Data showing util/load errors on the RQ (cpu 1) when thread0 running\n", "-------------------------------\n", "Note that, as expected, the error exists only for cases where delta is < 1ms (now - last_update_time)" ] }, { "cell_type": "code", "execution_count": 31, "metadata": { "collapsed": false, "scrolled": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "number of errors: 11\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "/home/joelaf/anaconda2/lib/python2.7/site-packages/ipykernel/__main__.py:3: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n", " app.launch_new_instance()\n", "/home/joelaf/anaconda2/lib/python2.7/site-packages/ipykernel/__main__.py:8: FutureWarning: sort(columns=....) is deprecated, use sort_values(by=.....)\n" ] }, { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
__commacc_load_avgutil_avgacc_util_avgutil_errdelta_usload_sumsum_errload_avgload_errutil_err_pc
Time
6.986053thread091768913454356860-46080761517.105263
3.749162thread0887587125194219977-531456761216.000000
3.749187thread088758712284248649-28672761216.000000
6.986014thread0907688125904310780-604160761415.789474
4.355859thread0857383104354059408-445440741113.698630
4.355883thread085738310274087056-27648741113.698630
6.075973thread086758510274125064-27648751113.333333
5.982620thread02272112251468410844301-700416212156.635071
0.736533thread04013803981885419171804-874496383184.736842
4.983089thread03983853961154519047561-558080387112.857143
4.983090thread0398385396111190475610387112.857143
\n", "
" ], "text/plain": [ " __comm acc_load_avg util_avg acc_util_avg util_err delta_us \\\n", "Time \n", "6.986053 thread0 91 76 89 13 45 \n", "3.749162 thread0 88 75 87 12 519 \n", "3.749187 thread0 88 75 87 12 28 \n", "6.986014 thread0 90 76 88 12 590 \n", "4.355859 thread0 85 73 83 10 435 \n", "4.355883 thread0 85 73 83 10 27 \n", "6.075973 thread0 86 75 85 10 27 \n", "5.982620 thread0 227 211 225 14 684 \n", "0.736533 thread0 401 380 398 18 854 \n", "4.983089 thread0 398 385 396 11 545 \n", "4.983090 thread0 398 385 396 11 1 \n", "\n", " load_sum sum_err load_avg load_err util_err_pc \n", "Time \n", "6.986053 4356860 -46080 76 15 17.105263 \n", "3.749162 4219977 -531456 76 12 16.000000 \n", "3.749187 4248649 -28672 76 12 16.000000 \n", "6.986014 4310780 -604160 76 14 15.789474 \n", "4.355859 4059408 -445440 74 11 13.698630 \n", "4.355883 4087056 -27648 74 11 13.698630 \n", "6.075973 4125064 -27648 75 11 13.333333 \n", "5.982620 10844301 -700416 212 15 6.635071 \n", "0.736533 19171804 -874496 383 18 4.736842 \n", "4.983089 19047561 -558080 387 11 2.857143 \n", "4.983090 19047561 0 387 11 2.857143 " ] }, "execution_count": 31, "metadata": {}, "output_type": "execute_result" } ], "source": [ "errpc_fn = (lambda row: (row['util_err'] * 100.0 / row['util_avg']))\n", "\n", "err_df = rq_df[(rq_df.util_err > 10) | (rq_df.load_err > 10)][rq_df['__comm'] == 'thread0']\n", "err_df = err_df[['__comm', 'acc_load_avg', 'util_avg', 'acc_util_avg', \\\n", " 'util_err', 'delta_us', 'load_sum', 'sum_err', 'load_avg', 'load_err']]\n", "\n", "err_df['util_err_pc'] = err_df.apply(errpc_fn ,axis=1)\n", "\n", "err_df = err_df.sort(columns=['util_err_pc'], ascending=False)\n", "print 'number of errors: ' + str(len(err_df))\n", "\n", "err_df.head(40)\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Summary of issues\n", "\n", "### * At 5.98s, there is a 6% error in util_avg 225 vs 211) - this causes a glitch and makes the signal less smooth\n", "----\n", "![glitch](pelt-signal-glitch-error-6pc.png)\n", "\n", "\n", "### * At 3.06s, there is a 3% error in util_avg - causing lowered peak of util_avg (397 -> 387) with delta ~450us\n", "![lower-peak](pelt-3percent-error.png)\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "

\n", "# Histogram of Errors before and after fix\n", "\n", "## BEFORE: util_avg and load_avg occurences of errors" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "collapsed": false, "scrolled": false }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/joelaf/anaconda2/lib/python2.7/site-packages/ipykernel/__main__.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n", " if __name__ == '__main__':\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAA38AAAGrCAYAAABwhy6fAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBJQ/AACAEVD+AAAARkD5AwAAGIH/HxwhhzDVW9h+AAAAAElFTkSuQmCC\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "df = rq_df[(rq_df.util_err > 0) | (rq_df.load_err > 0)][rq_df['__comm'] == 'thread0']\n", "df = df[(df.util_err > 0) | (df.load_err > 0)]\n", "df = df[['util_err', 'load_err']].plot(kind='hist', figsize=(15,7), bins=60, xlim=(0, 20), xticks=range(0,20), stacked=True, title = 'Occurence of errors > 10 counts', )" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## AFTER fix: util_avg and load_avg occurences of errors" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "![fixed](pelt-hist-fixed.png)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [] } ], "metadata": { "_draft": { "nbviewer_url": "https://gist.github.com/ec38b4edb2da1ef21e2aa9b1d6c64f65" }, "gist": { "data": { "description": "TraceAnalysis_TasksLatencies.ipynb", "public": false }, "id": "ec38b4edb2da1ef21e2aa9b1d6c64f65" }, "hide_input": false, "kernelspec": { "display_name": "Python 2", "language": "python", "name": "python2" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 2 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython2", "version": "2.7.13" }, "toc": { "colors": { "hover_highlight": "#DAA520", "running_highlight": "#FF0000", "selected_highlight": "#FFD700" }, "moveMenuLeft": true, "nav_menu": { "height": "296px", "width": "252px" }, "navigate_menu": true, "number_sections": true, "sideBar": true, "threshold": 4, "toc_cell": false, "toc_number_sections": true, "toc_section_display": "block", "toc_threshold": 6, "toc_window_display": false } }, "nbformat": 4, "nbformat_minor": 0 }