| 1 | <?xml version="1.0" encoding="UTF-8" ?> |
| 2 | <!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.3//EN" |
| 3 | "/usr/share/sgml/docbook/dtd/4.3/xdocbook.dtd"> |
| 4 | <!--<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN" >--> |
| 5 | |
| 6 | <book> |
| 7 | |
| 8 | <bookinfo> |
| 9 | <title>Linux Trace Toolkit Viewer User Guide</title> |
| 10 | <authorgroup> |
| 11 | <author> |
| 12 | <firstname>Mathieu</firstname> |
| 13 | <surname>Desnoyers</surname> |
| 14 | </author> |
| 15 | </authorgroup> |
| 16 | |
| 17 | <date>11/01/2006</date> |
| 18 | <releaseinfo>1.00.02</releaseinfo> |
| 19 | |
| 20 | <abstract> |
| 21 | <para> |
| 22 | This document describes how to install <application>Linux Trace |
| 23 | Toolkit Viewer</application> and how to use it. |
| 24 | |
| 25 | </para> |
| 26 | </abstract> |
| 27 | |
| 28 | <keywordset> |
| 29 | <keyword>Linux Trace Toolkit Viewer</keyword> |
| 30 | <keyword>Linux Trace Toolkit</keyword> |
| 31 | <keyword>tracing</keyword> |
| 32 | <keyword>Linux</keyword> |
| 33 | <keyword>visualization</keyword> |
| 34 | <keyword>operating system</keyword> |
| 35 | </keywordset> |
| 36 | |
| 37 | </bookinfo> |
| 38 | |
| 39 | <chapter> |
| 40 | <title>Introduction</title> |
| 41 | <para> |
| 42 | Linux Trace Toolkit (LTT) is a tracing tool that permits to get all the possible |
| 43 | execution information from the Linux Kernel. It is based on kernel |
| 44 | instrumentation and a high-speed relay file system to copy the information from |
| 45 | the kernel space to the user space. |
| 46 | </para> |
| 47 | |
| 48 | <para> |
| 49 | Linux Trace Toolkit Viewer (LTTV) is the second generation of visualization |
| 50 | tool. It is based on a trace format (the files where the data is recorded on |
| 51 | disk) written by the LTTng tracer. |
| 52 | </para> |
| 53 | |
| 54 | <para> |
| 55 | This document explains all the steps that are necessary in order to record a |
| 56 | trace with LTT and view it with LTTV. |
| 57 | </para> |
| 58 | </chapter> |
| 59 | |
| 60 | <chapter> |
| 61 | <title>Getting started</title> |
| 62 | |
| 63 | <sect1 id="install"> |
| 64 | <title>Installing LTTng and LTTV</title> |
| 65 | <para> |
| 66 | Follow the QUICKSTART guide found at |
| 67 | <ulink url="http://ltt.polymtl.ca">ltt.polymtl.ca</ulink>. |
| 68 | </para> |
| 69 | |
| 70 | <!-- |
| 71 | <title>Installing LTTV</title> |
| 72 | <para> |
| 73 | First, you must download the latests version of LTTV. You should get it from |
| 74 | this site : <ulink url="http://ltt.polymtl.ca">ltt.polymtl.ca</ulink>. |
| 75 | I suggest that you get it from the "Packages" section. |
| 76 | </para> |
| 77 | |
| 78 | <para> |
| 79 | You need a recent gcc compiler to compile the project. You might want to use gcc |
| 80 | 3.2 or newer. |
| 81 | You will also need some libraries in order to compile it. They are described in |
| 82 | the README of the LTTV package. These are GTK 2.0, GLIB 2.0, "popt" and Pango 1.0. |
| 83 | Install them if they are not on your system. Remember that if you use a package |
| 84 | manager from you favourite Linux distribution, you will need to specifically |
| 85 | install the librairies'development packages. |
| 86 | </para> |
| 87 | |
| 88 | |
| 89 | <para> |
| 90 | Then, you are ready to compile LTTV. Extract and untar the file you previously |
| 91 | downloaded : |
| 92 | </para> |
| 93 | |
| 94 | <screen> |
| 95 | <prompt>$</prompt> <userinput>tar -xvzof LinuxTraceToolkitViewer-x.x-dddddddd.tar.bz2</userinput> |
| 96 | </screen> |
| 97 | |
| 98 | <para> |
| 99 | Then, go to the directory newly created, and type : |
| 100 | </para> |
| 101 | |
| 102 | <screen> |
| 103 | <prompt>$</prompt> <userinput>./configure</userinput> |
| 104 | <prompt>$</prompt> <userinput>make</userinput> |
| 105 | <prompt>#</prompt> <userinput>make install</userinput> (as root) |
| 106 | </screen> |
| 107 | --> |
| 108 | <para> |
| 109 | At this point, LTTV is installed in the default directory. You may find the |
| 110 | lttv executable in /usr/local/bin and the librairies in /usr/local/lib. You will |
| 111 | also notice the presence of the convert executable in /usr/local/bin. This tool |
| 112 | will be used later to convert from the Linux Trace Toolkit trace format to the |
| 113 | LTTV format. |
| 114 | </para> |
| 115 | <!-- |
| 116 | <para> |
| 117 | You are now ready to go to the next step : installing the LTT kernel tracer. |
| 118 | </para> |
| 119 | |
| 120 | |
| 121 | </sect1> |
| 122 | |
| 123 | |
| 124 | |
| 125 | <sect1 id="install-tracer"> |
| 126 | <title>Installing LTT kernel tracer</title> |
| 127 | <para> |
| 128 | The goal of this guide is not to describe the Linux Trace Toolkit project in |
| 129 | details, as it is a |
| 130 | seperate project for now. It just gives pointers to the basic steps you must |
| 131 | take in order to generate a trace suitable for conversion. |
| 132 | </para> |
| 133 | |
| 134 | <para> |
| 135 | First, go to the <ulink url="http://ltt.polymtl.ca">ltt.polymtl.ca</ulink> |
| 136 | website, in the "Patches for the Official LTT" section. Use the latest version |
| 137 | of patches available. The file name convention used goes like this : |
| 138 | aaaaaa-x.x\-\-bbbbb-y.y.patch. That means a patch made for aaaaa, release x.x, |
| 139 | that adds bbbbb, release y.y to it. Notice the presence of the \-\- sign that |
| 140 | separates the "from" field from the name of the patch applied. This way, it's |
| 141 | impossible to be mixed up on the specific sequence of patch application. I |
| 142 | suggest that you use the "relayfs", "ltt" and then "md" patches. The "md" patch |
| 143 | adds events useful to LTTV that are not in the official LTT. |
| 144 | </para> |
| 145 | |
| 146 | <para> |
| 147 | Once you have the patches you need, get the matching Linux kernel version, apply |
| 148 | the patches on it, configure it, install it, reboot with the new kernel. You then |
| 149 | have an instrumented kernel ready for tracing. If you have problems during this phase, |
| 150 | please refer to <ulink |
| 151 | url="http://www.opersys.com/ltt">www.opersys.com/ltt</ulink>. If you need |
| 152 | instructions about how to recompile a kernel, see |
| 153 | <ulink url="http://www.tldp.org/HOWTO/Kernel-HOWTO/">Kernel-HOWTO</ulink>. |
| 154 | </para> |
| 155 | |
| 156 | </sect1> |
| 157 | |
| 158 | <sect1 id="install-daemon"> |
| 159 | <title>Installing LTT trace recording daemon</title> |
| 160 | <para> |
| 161 | In order to install the LTT trace recording daemon, you should get the latest |
| 162 | TraceToolkit (or ltt) package from the LTT ftp site. |
| 163 | Use the link "Official Linux Trace Toolkit Packages" on the |
| 164 | <ulink url="http://ltt.polymtl.ca">ltt.polymtl.ca</ulink> webpage to access it. |
| 165 | As of November 30, 2004, the most recent version is 0.9.6-pre3. |
| 166 | </para> |
| 167 | <para> |
| 168 | Then, you should apply the TraceToolkit patches from the LTTV website related |
| 169 | to the package version. Get them from the "Patches for the Official LTT" |
| 170 | section. |
| 171 | </para> |
| 172 | <para> |
| 173 | You are now ready to install the daemon in your system. Please refer to the |
| 174 | documentation in the package for details. |
| 175 | </para> |
| 176 | <para> |
| 177 | You may now use the following command to record a sample 30 seconds trace in |
| 178 | your current directory. Command line switches are described on the official |
| 179 | LTT website. |
| 180 | </para> |
| 181 | <screen> |
| 182 | <prompt>#</prompt><userinput>tracedaemon -ts30 sample.out sample.proc (as root) userinput></userinput> |
| 183 | </screen> |
| 184 | </sect1> |
| 185 | |
| 186 | |
| 187 | |
| 188 | <sect1 id="convert"> |
| 189 | <title>Conversion from LTT to LTTV trace format</title> |
| 190 | <para> |
| 191 | If you used the default directory for installation, you should find the |
| 192 | conversion tool in /usr/local/bin/convert. Before using it, some other files are |
| 193 | necessary. You will find them in |
| 194 | /usr/local/share/LinuxTraceToolkitViewer/convert/. Those are sysInfo and |
| 195 | core.xml. |
| 196 | </para> |
| 197 | <para> |
| 198 | sysInfo is a script that get information about the traced computer. It should |
| 199 | be invoked like this : |
| 200 | </para> |
| 201 | <screen> |
| 202 | <prompt>$</prompt> <userinput>sh /usr/local/LinuxTraceToolkitViewer/convert/sysInfo</userinput> |
| 203 | </screen> |
| 204 | <para> |
| 205 | It creates a file named sysInfo.out. This file has to be present in the current |
| 206 | directory where the convert tool will be executed. I suggest that you choose a |
| 207 | destination directory where will be written converted traces right now, put sysInfo.out in it, at |
| 208 | use it as current directory for running the convert tool. |
| 209 | </para> |
| 210 | <para> |
| 211 | Once the sysInfo.out file is ready and you have a trace ready for conversion, |
| 212 | you should invoke convert like the following example. This is for a uniprocessor |
| 213 | computer. If you whish to get detailed explanation on the parameters, simply |
| 214 | execute the convert tool without any option. You may also wish to see the |
| 215 | /usr/local/LinuxTraceToolkitViewer/convert/README file. |
| 216 | </para> |
| 217 | <screen> |
| 218 | <prompt>$</prompt> <userinput>/usr/local/bin/convert sample.proc 1 sample.trace sample.converted</userinput> |
| 219 | </screen> |
| 220 | <para> |
| 221 | You must then copy the core event definition file to the converted trace directory : |
| 222 | </para> |
| 223 | <screen> |
| 224 | <prompt>$</prompt> <userinput>cp /usr/local/share/LinuxTraceToolkitViewer/convert/core.xml sample.converted/</userinput> |
| 225 | </screen> |
| 226 | <para> |
| 227 | You now have a converted trace ready for visualization in LTTV. Congratulations! |
| 228 | </para> |
| 229 | --> |
| 230 | </sect1> |
| 231 | |
| 232 | <sect1 id="running"> |
| 233 | <title>Running the executable with basic libraries</title> |
| 234 | <para> |
| 235 | Starting the graphical mode with the basic viewer activated is as simple as : |
| 236 | </para> |
| 237 | <screen> |
| 238 | <prompt>$</prompt> <userinput>lttv-gui</userinput> |
| 239 | </screen> |
| 240 | <para> |
| 241 | Using the text mode is very simple too. Look in /usr/local/lib/lttv/plugins for |
| 242 | the list of modules. You may use the --help switch to get basic help on the |
| 243 | command line parameters of every loaded modules. To simply output the events of |
| 244 | a trace in a text file, try the textDump module. The batchAnalysis module |
| 245 | permits to do batch mode analysis (state and statistics calculation ) on a |
| 246 | trace. |
| 247 | </para> |
| 248 | <screen> |
| 249 | <prompt>$</prompt> <userinput>lttv -L /usr/local/lib/lttv/plugins -m textDump --help</userinput> |
| 250 | </screen> |
| 251 | </sect1> |
| 252 | </chapter> |
| 253 | |
| 254 | <chapter> |
| 255 | <title>Using LTTV graphical interface</title> |
| 256 | |
| 257 | <sect1 id="mainwindow"> |
| 258 | <title>LTTV main window</title> |
| 259 | <para> |
| 260 | This section describes the main functionnalities that are provided by the LTTV |
| 261 | GUI and how to use them. |
| 262 | </para> |
| 263 | <para> |
| 264 | By default, when the lttv GUI starts with all the graphical modules loaded, |
| 265 | it loads the statistics viewer, the control flow viewer, and the detailed event |
| 266 | list inside a tab. Other viewers can be added later to this tab by interacting |
| 267 | with the main window. Let's describe the operations available on the window : |
| 268 | </para> |
| 269 | <screenshot> |
| 270 | <mediaobject> |
| 271 | <imageobject> |
| 272 | <imagedata srccredit="Mathieu Desnoyers, 2004" fileref="lttv-numbered-5.png" |
| 273 | format="PNG" align="center"/> |
| 274 | </imageobject> |
| 275 | <imageobject> |
| 276 | <imagedata srccredit="Mathieu Desnoyers, 2004" |
| 277 | fileref="lttv-numbered-5.eps" |
| 278 | format="EPS" align="center"/> |
| 279 | </imageobject> |
| 280 | <!--<imagedata srccredit="Mathieu Desnoyers, 2004" fileref="lttv-numbered-6.svg" |
| 281 | format="SVG" align="center" scalefit="1"/> |
| 282 | </imageobject>--> |
| 283 | <caption><para>Linux Trace Toolkit Viewer GUI</para></caption> |
| 284 | </mediaobject> |
| 285 | </screenshot> |
| 286 | <orderedlist> |
| 287 | <listitem> |
| 288 | <para> |
| 289 | This toolbar allows you to navigate through the basic functionnalities of LTTV. |
| 290 | The first button opens a new window and the second one, a new tab. You can leave |
| 291 | your mouse over the buttons to read the information provided by the tooltips. |
| 292 | </para> |
| 293 | </listitem> |
| 294 | <listitem> |
| 295 | <para> |
| 296 | This notebook, containing different tabs, lets you select the "Trace Set" you |
| 297 | want to interact with. A trace set is an aggregation of traces, synchronised in |
| 298 | time. You may also want to use one tab per viewer by simply cloning the traceset |
| 299 | to a new tab. This way, you can have vertically stacked viewers in one tab, as |
| 300 | well as different viewers, independant from the time interval. Note that once |
| 301 | the Trace Set cloning is done, each trace set becomes completely independant. |
| 302 | For Traceset cloning, see the File Menu. |
| 303 | </para> |
| 304 | </listitem> |
| 305 | <listitem> |
| 306 | <para> |
| 307 | These buttons let you control the computation in progress on a trace. As |
| 308 | sometimes the computation may last for a while, you may want to stop it, restart |
| 309 | it from the beginning or simply to continue from where you stopped. This is |
| 310 | exactly what those three buttons offer you. |
| 311 | </para> |
| 312 | </listitem> |
| 313 | <listitem> |
| 314 | <para> |
| 315 | Buttons on the right side of the last spacer are semantically different from the |
| 316 | others. While the other buttons at the left side of the bar are built in the |
| 317 | lttv program and let you operate the basic functionnalities, the buttons at the |
| 318 | right side let you add a viewer to the active Tab. They belong to the |
| 319 | viewers themselves. The number of buttons that appears there should directly |
| 320 | depend on the number of viewer's modules loaded. |
| 321 | </para> |
| 322 | </listitem> |
| 323 | <listitem> |
| 324 | <para> |
| 325 | This is a tree representing the multiple statistics available for the current |
| 326 | traceset. This is shown by the guistatistics viewer. |
| 327 | </para> |
| 328 | </listitem> |
| 329 | <listitem> |
| 330 | <para> |
| 331 | This is the Y axis of the guicontrolflow viewer. It shows the process list of |
| 332 | the traced system. You may notice that it grows : it dynamically adds |
| 333 | process when they appear in the trace. |
| 334 | </para> |
| 335 | </listitem> |
| 336 | <listitem> |
| 337 | <para> |
| 338 | This is a (missing) time bar for the X axis. Maybe will it be used for viewer |
| 339 | specific buttons eventually. Work in progress. |
| 340 | </para> |
| 341 | </listitem> |
| 342 | <listitem> |
| 343 | <para> |
| 344 | The is the current time selected. The concept of current event and current time |
| 345 | selected is synchronised in a Tab for all the viewers. The control flow viewer |
| 346 | shows it a vertical white dotted line. You move this marker by clicking on the |
| 347 | background of the process state graph. This graph shows evolution of each |
| 348 | process's state through time. The meaning of the colors will be explained later. |
| 349 | </para> |
| 350 | </listitem> |
| 351 | <listitem> |
| 352 | <para> |
| 353 | This is the details event list. It shown the detailed information about each |
| 354 | event of the trace. It is synchronised with the current time and current event, |
| 355 | so selecting an event changes other viewer's current time and reciprocally. |
| 356 | </para> |
| 357 | </listitem> |
| 358 | <listitem> |
| 359 | <para> |
| 360 | You can enter the values of start time and end time you wish to see on the |
| 361 | screen here. It also supports pasting time as text input, simply by clicking of |
| 362 | the "Time Frame", "start" or "end:" fields. A valid entry consists of any |
| 363 | digital input separated by any quantity of non digital characters. For example : |
| 364 | "I start at 356247.124626 and stop at 724524.453455" would be a valid input |
| 365 | for the "Time Frame" field. |
| 366 | </para> |
| 367 | </listitem> |
| 368 | <listitem> |
| 369 | <para> |
| 370 | This horizontal scrollbar modifies the window of time shown by all the viewers |
| 371 | in the tab. It is linked with the fields below it (described at number 10 and |
| 372 | 12). Another way to modify the time shown is to use the zoom buttons of the |
| 373 | toolbar (yes, the ones that looks like magnifying glasses). |
| 374 | </para> |
| 375 | </listitem> |
| 376 | <listitem> |
| 377 | <para> |
| 378 | This field works just like the "Time Frame" field. It modifies the current time |
| 379 | selected by the viewers. For example, changing its value will change the current |
| 380 | event selected by the detailed events list and the current time selected by the |
| 381 | control flow viewer. |
| 382 | </para> |
| 383 | </listitem> |
| 384 | </orderedlist> |
| 385 | </sect1> |
| 386 | |
| 387 | <sect1 id="ControlFlowColors"> |
| 388 | <title>Control Flow View Colors</title> |
| 389 | <screenshot> |
| 390 | <mediaobject> |
| 391 | <imageobject> |
| 392 | <imagedata srccredit="Mathieu Desnoyers, 2004" fileref="lttv-color-list.png" |
| 393 | format="PNG" align="center"/> |
| 394 | </imageobject> |
| 395 | <imageobject> |
| 396 | <imagedata srccredit="Mathieu Desnoyers, 2004" |
| 397 | fileref="lttv-color-list.eps" |
| 398 | format="EPS" align="center"/> |
| 399 | </imageobject> |
| 400 | <!--<imagedata srccredit="Mathieu Desnoyers, 2004" fileref="lttv-numbered-6.svg" |
| 401 | format="SVG" align="center" scalefit="1"/> |
| 402 | </imageobject>--> |
| 403 | <caption><para>Control Flow View Color Legend</para></caption> |
| 404 | </mediaobject> |
| 405 | </screenshot> |
| 406 | |
| 407 | <para> |
| 408 | Here is a description of the colors used in the control flow view. Each color |
| 409 | represents a state of the process at a given time. |
| 410 | </para> |
| 411 | |
| 412 | <itemizedlist> |
| 413 | <listitem> |
| 414 | <para> |
| 415 | White : this color is used for process from which state is not known. It may |
| 416 | happen when you seek quickly at a far time in the trace just after it has been |
| 417 | launched. At that moment, the precomputed state information is incomplete. The |
| 418 | "unknown" state is used to identify this. Note that the viewer gets refreshed |
| 419 | once the precomputation ends. |
| 420 | </para> |
| 421 | </listitem> |
| 422 | <listitem> |
| 423 | <para> |
| 424 | Green : This color is only used for process when they are running in user mode. |
| 425 | That includes execution of all the source code of an executable as well as the |
| 426 | libraries it uses. |
| 427 | </para> |
| 428 | </listitem> |
| 429 | <listitem> |
| 430 | <para> |
| 431 | Pale blue : A process is doing a system call to the kernel, and the mode is |
| 432 | switched from process limited rights to super user mode. Only code from the |
| 433 | kernel (including modules) should be run in that state. |
| 434 | </para> |
| 435 | </listitem> |
| 436 | <listitem> |
| 437 | <para> |
| 438 | Yellow : The kernel is running a trap that services a fault. The most frequent |
| 439 | trap is the memory page fault trap : it is called every time a page is missing |
| 440 | from physical memory. |
| 441 | </para> |
| 442 | </listitem> |
| 443 | <listitem> |
| 444 | <para> |
| 445 | Orange : IRQ servicing routine is running. It interrupts the currently running |
| 446 | process. As the IRQ does not change the currently running process (on some |
| 447 | architectures it uses the same stack as the process), the IRQ state is shown in |
| 448 | the state of the process. IRQ can be nested : a higher priority interrupt can |
| 449 | interrupt a lower priority interrupt. |
| 450 | </para> |
| 451 | </listitem> |
| 452 | <listitem> |
| 453 | <para> |
| 454 | Pink : SoftIRQ handler is running. A SoftIRQ is normally triggered by an |
| 455 | interrupt that whishes to have some work done very soon, but not "now". This is |
| 456 | especially useful, for example, to have the longest part of the network stack |
| 457 | traversal done : a too long computation in the interrupt handler would increase |
| 458 | the latency of the system. Therefore, doing the long part of the computation in |
| 459 | a softirq that will be run just after the IRQ handler exits will permits to do |
| 460 | this work while interrupts are enabled, without increasing the system latency. |
| 461 | </para> |
| 462 | </listitem> |
| 463 | <listitem> |
| 464 | <para> |
| 465 | Dark red : A process in that state is waiting for an input/output operation to |
| 466 | complete before it can continue its execution. |
| 467 | </para> |
| 468 | </listitem> |
| 469 | <listitem> |
| 470 | <para> |
| 471 | Dark yellow : A process is ready to run, but waiting to get the CPU (a schedule |
| 472 | in event). |
| 473 | </para> |
| 474 | </listitem> |
| 475 | <listitem> |
| 476 | <para> |
| 477 | Dark purple : A process in zombie state. This state happens when a process |
| 478 | exits and then waits for the parent to wait for it (wait() or waitpid()). |
| 479 | </para> |
| 480 | </listitem> |
| 481 | <listitem> |
| 482 | <para> |
| 483 | Dark green : A process has just been created by its parent and is waiting for |
| 484 | first scheduling. |
| 485 | </para> |
| 486 | </listitem> |
| 487 | <listitem> |
| 488 | <para> |
| 489 | Magenta : The process has exited, but still has the control of the CPU. It may |
| 490 | happend if it has some tasks to do in the exit system call. |
| 491 | </para> |
| 492 | </listitem> |
| 493 | </itemizedlist> |
| 494 | </sect1> |
| 495 | </chapter> |
| 496 | |
| 497 | <chapter> |
| 498 | <title>Using LTTV text modules</title> |
| 499 | <sect1 id="batchAnalysis"> |
| 500 | <title>The batch analysis module</title> |
| 501 | <para> |
| 502 | This batch analysis module can be invoked like this : |
| 503 | </para> |
| 504 | <screen> |
| 505 | <prompt>$</prompt> <userinput>lttv -L path/to/lib/plugins -m batchAnalysis\ |
| 506 | -t trace1 -t trace2 ...</userinput> |
| 507 | </screen> |
| 508 | <para> |
| 509 | It permits to call any registered action to perform in batch mode on all the |
| 510 | trace set, which consists of the traces loaded on the command line. Actions that |
| 511 | are built in the batchAnalysis module are statistics computation. They can be |
| 512 | triggered by using the -s (--stats) switch. |
| 513 | </para> |
| 514 | <para> |
| 515 | However, the batchAnalysis module is mostly a backend for every other text |
| 516 | module that does batch computation over a complete trace set. |
| 517 | </para> |
| 518 | </sect1> |
| 519 | <sect1 id="textDump"> |
| 520 | <title>The text dump module</title> |
| 521 | <para> |
| 522 | The goal of this module is to convert the binary data of the traces into |
| 523 | a formatted text file. |
| 524 | </para> |
| 525 | <para> |
| 526 | The text dump module is a good example of a usage of the batch analysis module |
| 527 | backend. In fact, the text dump module depends on it. You don't need to |
| 528 | explicitly load the batchAnalysis module though, as lttv offers a rich module |
| 529 | backend that deals with the dependencies, loading the module automatically if |
| 530 | needed. |
| 531 | </para> |
| 532 | <para> |
| 533 | The text dump module is invoked just like the batchAnalysis module. It adds more |
| 534 | options that can be specified in argument. You may specify the -o switch for the |
| 535 | output file name of the text dump. You can enable the output of the field names |
| 536 | (the identifier of the fields) with the -l switch. The -s switch, for process |
| 537 | states, is very useful to indicate the state in which the process is when the |
| 538 | event happens. |
| 539 | </para> |
| 540 | <para> |
| 541 | If you use the --help option on the textDump module, you will see all the detail |
| 542 | about the switches that can be used to show per cpu statistics and per process |
| 543 | statistics. You will notice that you can use both the switches for the |
| 544 | batchAnalysis module and those for textDump. You will also notice that the |
| 545 | options --process_state (from textDump) and --stats (from batchAnalysis) has the |
| 546 | same short name "-s". If you choose to invoke this option using the short name, |
| 547 | it will use the option of the last module loaded just before the -s switch. |
| 548 | </para> |
| 549 | <para> |
| 550 | For exemple, if you load the textDump module with -m textDump, it will first |
| 551 | load the batchAnalysis module, and then load itself. As it is the last module |
| 552 | loaded, the -s switch used after it will signify --process_stats. On the other |
| 553 | hand, if you choose to specify explicitly the loading of both modules like this |
| 554 | : |
| 555 | </para> |
| 556 | <screen> |
| 557 | <prompt>$</prompt> <userinput>lttv -L path/to/lib/plugins -m batchAnalysis -s\ |
| 558 | -m textDump -s -t trace</userinput> |
| 559 | </screen> |
| 560 | <para> |
| 561 | The first "-s" will invoke batchAnalysis --stats and the second "-s" will invoke |
| 562 | textDump --process_state. The list of options generated by --help follows the |
| 563 | order of registration of the options by the modules, therefore the invocation |
| 564 | order of the modules. |
| 565 | </para> |
| 566 | </sect1> |
| 567 | <sect1 id="formattedDump"> |
| 568 | <title>The formatted dump module</title> |
| 569 | <para> |
| 570 | Like text dump the goal of this module is to convert the binary data of the |
| 571 | traces into a desired formatted text file. This module allows for a flexible |
| 572 | text output, where you specify which information you want. It also provides |
| 573 | several preset format. |
| 574 | </para> |
| 575 | <para> |
| 576 | Formatted dump plugin prints a formatted output of each events in a trace. |
| 577 | The output format is defined as a parameter. It provides a default easy |
| 578 | to read format, a "strace-like" format and the original textDump format for |
| 579 | backward compatibility. Like text dump, formatted dump uses the batch analysis |
| 580 | module. That's why you may specify the -o switch for the output file name of |
| 581 | the formatted dump. |
| 582 | </para> |
| 583 | <para> |
| 584 | The formatted dump module is invoked just like the text module. It adds more |
| 585 | options that can be specified as argument. You may specify the -F switch for |
| 586 | the format of the output followed by your output format between quotes. The -T |
| 587 | switch activates the text dump format while -S provides the "strace-like" format. |
| 588 | If there is no argument or option following formattedDump, it will provides the |
| 589 | default format, which is easier to read than textDump. |
| 590 | </para> |
| 591 | <para> |
| 592 | If you pass the --help (or -h) option to the formattedDump module, you will see |
| 593 | all the detail about the switches that can be used. You will notice that you |
| 594 | can use both the switches for batchAnalysis and those for formattedDump. You |
| 595 | will also remark that all interpreted sequences follow the FORMAT description. |
| 596 | </para> |
| 597 | <para> |
| 598 | For example, if you load the formattedDump module with -m formattedDump, you |
| 599 | have to add the -t switch from batchAnalysis followed by the path to trace. |
| 600 | Then you can add the formatted dump options like this: |
| 601 | </para> |
| 602 | <screen> |
| 603 | <prompt>$</prompt> <userinput>lttv -m formattedDump -t path/to/trace -F |
| 604 | "event name:%e timestamp:%t process name:%p"</userinput> |
| 605 | </screen> |
| 606 | <para> |
| 607 | With the same example, you can add the -o switch followed by the name of the |
| 608 | file where you want store the output: |
| 609 | <screen> |
| 610 | <prompt>$</prompt> <userinput>lttv -m formattedDump -t path/to/trace -F |
| 611 | "event name:%e timestamp:%t process name:%p -o output.txt"</userinput> |
| 612 | </screen> |
| 613 | </para> |
| 614 | <para>The get the full list of options for the format, use the help option on |
| 615 | the command line. |
| 616 | <screen> |
| 617 | <prompt>$</prompt> <userinput>lttv -m formattedDump -h</userinput> |
| 618 | </screen></para> |
| 619 | </sect1> |
| 620 | <sect1 id="textFilter"> |
| 621 | <title>The text filter module</title> |
| 622 | <para> |
| 623 | The text filter module provides the ability to filter the events of a trace. |
| 624 | Unlike formattedDump which let you select which information to display, |
| 625 | textFilter let you filter complete events and not simply their information. |
| 626 | It should be used with either textDump or formattedDump to display the |
| 627 | remaining events. |
| 628 | </para> |
| 629 | <para> |
| 630 | The possibilities of filtering are numerous through this module. You can filter |
| 631 | events by: event name, channel name, time(current, creation, insertion,...), |
| 632 | pid, ppid, process name, cpu... and many other opportunities are provided. All |
| 633 | possible fields and operators are listed in the switch -l (or --list) of |
| 634 | textFilter. |
| 635 | </para> |
| 636 | <para> |
| 637 | The text filter module is invoked just like all other text modules of lttv |
| 638 | except that you should specify if you want use textDump or formattedDump for |
| 639 | displaying your trace. You also need describe your filtering expression with -e |
| 640 | switch followed by your filter request between quotes. |
| 641 | </para> |
| 642 | <para> |
| 643 | If you use the --help (or -h) option on the textFilter module, you will see |
| 644 | all the detail about the switches that can be used. Remember that all possible |
| 645 | filtering field are listed in --list switch. |
| 646 | </para> |
| 647 | <para> |
| 648 | This module is an addition of textDump and formattedDump. The combo textFilter |
| 649 | and formattedDump provides a complete filter of a trace: event and event's |
| 650 | information. |
| 651 | </para> |
| 652 | <para> |
| 653 | If you load the textFilter module with -m textFilter, you should first add the |
| 654 | filtering expression, then load formattedDump (or textDump) module and use it |
| 655 | with its normal way. For exemple you can use textFilter and formattedDump like |
| 656 | this: |
| 657 | </para> |
| 658 | <screen> |
| 659 | <prompt>$</prompt> <userinput>lttv -m textFilter -e "channel.name=kernel" |
| 660 | -m formattedDump -t path/to/trace -F "channel name:%c timestamp:%t process |
| 661 | name:%p"</userinput> |
| 662 | </screen> |
| 663 | <para> |
| 664 | If you want use textDump instead of formattedDump with textFilter the same |
| 665 | principle apply, you just have to write: |
| 666 | <screen> |
| 667 | <prompt>$</prompt> <userinput>lttv -m textFilter -e |
| 668 | "event.name=kernel.syscall_entry" -m textDump -t path/to/trace</userinput> |
| 669 | </screen> |
| 670 | </para> |
| 671 | <para>To get the full list of options and filtering expressions, use the help |
| 672 | option followed by -l switch on the command line. |
| 673 | <screen> |
| 674 | <prompt>$</prompt> <userinput>lttv -m textFilter -h -l</userinput> |
| 675 | </screen></para> |
| 676 | </sect1> |
| 677 | </chapter> |
| 678 | |
| 679 | |
| 680 | </book> |