{
"metadata": {
"name": "",
"signature": "sha256:4e50c9159c1a001f1a581f813b7e533f0ef5d8ec4220406180bad11eb9cdba5c"
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "heading",
"level": 1,
"metadata": {},
"source": [
"Registering arbitrary trace events"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"`trappy` knows about some trace events. You can add your own in the notebook without having to change any code in `trappy`. After the trace is registered, the next time you parse a trace file that information will be part of the trace object as a pandas `DataFrame` and can be analyzed like the other `DataFrame`s in the `FTrace` class.\n",
"\n",
"The trace event must follow the following format:\n",
"\n",
" title: key0=value0 key1=value1 key2=value2 ...\n",
"\n",
"The title should be something that's unique in the trace. For example, you can generate trace with the following `trace_printk()`:\n",
"\n",
" trace_printk(\"thermal_gpu_power_get: frequency=%u load=%d\\n\", freq, load);\n",
"\n",
"which will appear in the `trace.txt` as:\n",
"\n",
" kworker/6:1-457 [006] 144.439566: bprint: 0xc042f8a0f: thermal_gpu_power_get: frequency=177 load=0\n",
"\n",
"You can add this event to the trace instance using `register_dynamic_ftrace()`"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"First import trappy"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"import sys\n",
"sys.path.append(\"..\")\n",
"%matplotlib inline\n",
"import trappy"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 1
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Register it. The first argument is the name under which you will find it in the trace instance. It will be changed to lower_case_with_underscores_to_separate_words. The second argument is some unique text in the trace, usually the title:"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"trappy.register_dynamic_ftrace(\"gpu_power_in\", \"thermal_gpu_power_get\")"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 2,
"text": [
"trappy.dynamic.gpu_power_in"
]
}
],
"prompt_number": 2
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we can parse the trace"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"trace = trappy.FTrace(\"/path/to/trace\")"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 3
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"`trace` now has a `gpu_power_in` member that contains the information. We can see the first few lines of the generated dataframe:"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"trace.gpu_power_in.data_frame.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" frequency | \n",
" load | \n",
"
\n",
" \n",
" | Time | \n",
" | \n",
" | \n",
"
\n",
" \n",
" \n",
" \n",
" | 109.999957 | \n",
" 480 | \n",
" 96 | \n",
"
\n",
" \n",
" | 110.099925 | \n",
" 480 | \n",
" 100 | \n",
"
\n",
" \n",
" | 110.199930 | \n",
" 480 | \n",
" 100 | \n",
"
\n",
" \n",
" | 110.299938 | \n",
" 480 | \n",
" 100 | \n",
"
\n",
" \n",
" | 110.399924 | \n",
" 480 | \n",
" 100 | \n",
"
\n",
" \n",
"
\n",
"