njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 1 | <?xml version="1.0"?> <!-- -*- sgml -*- --> |
| 2 | <!DOCTYPE chapter PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN" |
sewardj | 7aeb10f | 2006-12-10 02:59:16 +0000 | [diff] [blame] | 3 | "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd" |
| 4 | [ <!ENTITY % vg-entities SYSTEM "../../docs/xml/vg-entities.xml"> %vg-entities; ]> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 5 | |
philippe | e6bce13 | 2013-03-10 16:29:02 +0000 | [diff] [blame] | 6 | <!-- Referenced from both the manual and manpage --> |
| 7 | <chapter id="&vg-cg-manual-id;" xreflabel="&vg-cg-manual-label;"> |
njn | 05a8917 | 2009-07-29 02:36:21 +0000 | [diff] [blame] | 8 | <title>Cachegrind: a cache and branch-prediction profiler</title> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 9 | |
| 10 | <para>To use this tool, you must specify |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 11 | <option>--tool=cachegrind</option> on the |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 12 | Valgrind command line.</para> |
| 13 | |
njn | 05a8917 | 2009-07-29 02:36:21 +0000 | [diff] [blame] | 14 | <sect1 id="cg-manual.overview" xreflabel="Overview"> |
| 15 | <title>Overview</title> |
| 16 | |
| 17 | <para>Cachegrind simulates how your program interacts with a machine's cache |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 18 | hierarchy and (optionally) branch predictor. It simulates a machine with |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 19 | independent first-level instruction and data caches (I1 and D1), backed by a |
| 20 | unified second-level cache (L2). This exactly matches the configuration of |
| 21 | many modern machines.</para> |
| 22 | |
florian | ea84cd5 | 2012-10-07 21:03:27 +0000 | [diff] [blame] | 23 | <para>However, some modern machines have three or four levels of cache. For these |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 24 | machines (in the cases where Cachegrind can auto-detect the cache |
florian | ea84cd5 | 2012-10-07 21:03:27 +0000 | [diff] [blame] | 25 | configuration) Cachegrind simulates the first-level and last-level caches. |
| 26 | The reason for this choice is that the last-level cache has the most influence on |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 27 | runtime, as it masks accesses to main memory. Furthermore, the L1 caches |
| 28 | often have low associativity, so simulating them can detect cases where the |
| 29 | code interacts badly with this cache (eg. traversing a matrix column-wise |
| 30 | with the row length being a power of 2).</para> |
| 31 | |
| 32 | <para>Therefore, Cachegrind always refers to the I1, D1 and LL (last-level) |
| 33 | caches.</para> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 34 | |
| 35 | <para> |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 36 | Cachegrind gathers the following statistics (abbreviations used for each statistic |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 37 | is given in parentheses):</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 38 | <itemizedlist> |
| 39 | <listitem> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 40 | <para>I cache reads (<computeroutput>Ir</computeroutput>, |
| 41 | which equals the number of instructions executed), |
| 42 | I1 cache read misses (<computeroutput>I1mr</computeroutput>) and |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 43 | LL cache instruction read misses (<computeroutput>ILmr</computeroutput>). |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 44 | </para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 45 | </listitem> |
| 46 | <listitem> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 47 | <para>D cache reads (<computeroutput>Dr</computeroutput>, which |
| 48 | equals the number of memory reads), |
| 49 | D1 cache read misses (<computeroutput>D1mr</computeroutput>), and |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 50 | LL cache data read misses (<computeroutput>DLmr</computeroutput>). |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 51 | </para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 52 | </listitem> |
| 53 | <listitem> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 54 | <para>D cache writes (<computeroutput>Dw</computeroutput>, which equals |
| 55 | the number of memory writes), |
| 56 | D1 cache write misses (<computeroutput>D1mw</computeroutput>), and |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 57 | LL cache data write misses (<computeroutput>DLmw</computeroutput>). |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 58 | </para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 59 | </listitem> |
sewardj | 8badbaa | 2007-05-08 09:20:25 +0000 | [diff] [blame] | 60 | <listitem> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 61 | <para>Conditional branches executed (<computeroutput>Bc</computeroutput>) and |
| 62 | conditional branches mispredicted (<computeroutput>Bcm</computeroutput>). |
| 63 | </para> |
sewardj | 8badbaa | 2007-05-08 09:20:25 +0000 | [diff] [blame] | 64 | </listitem> |
| 65 | <listitem> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 66 | <para>Indirect branches executed (<computeroutput>Bi</computeroutput>) and |
| 67 | indirect branches mispredicted (<computeroutput>Bim</computeroutput>). |
| 68 | </para> |
sewardj | 8badbaa | 2007-05-08 09:20:25 +0000 | [diff] [blame] | 69 | </listitem> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 70 | </itemizedlist> |
| 71 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 72 | <para>Note that D1 total accesses is given by |
| 73 | <computeroutput>D1mr</computeroutput> + |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 74 | <computeroutput>D1mw</computeroutput>, and that LL total |
| 75 | accesses is given by <computeroutput>ILmr</computeroutput> + |
| 76 | <computeroutput>DLmr</computeroutput> + |
| 77 | <computeroutput>DLmw</computeroutput>. |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 78 | </para> |
| 79 | |
njn | 05a8917 | 2009-07-29 02:36:21 +0000 | [diff] [blame] | 80 | <para>These statistics are presented for the entire program and for each |
| 81 | function in the program. You can also annotate each line of source code in |
| 82 | the program with the counts that were caused directly by it.</para> |
| 83 | |
njn | c8cccb1 | 2005-07-25 23:30:24 +0000 | [diff] [blame] | 84 | <para>On a modern machine, an L1 miss will typically cost |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 85 | around 10 cycles, an LL miss can cost as much as 200 |
sewardj | 8badbaa | 2007-05-08 09:20:25 +0000 | [diff] [blame] | 86 | cycles, and a mispredicted branch costs in the region of 10 |
| 87 | to 30 cycles. Detailed cache and branch profiling can be very useful |
njn | 05a8917 | 2009-07-29 02:36:21 +0000 | [diff] [blame] | 88 | for understanding how your program interacts with the machine and thus how |
| 89 | to make it faster.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 90 | |
| 91 | <para>Also, since one instruction cache read is performed per |
| 92 | instruction executed, you can find out how many instructions are |
njn | 05a8917 | 2009-07-29 02:36:21 +0000 | [diff] [blame] | 93 | executed per line, which can be useful for traditional profiling.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 94 | |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 95 | </sect1> |
| 96 | |
| 97 | |
| 98 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 99 | <sect1 id="cg-manual.profile" |
| 100 | xreflabel="Using Cachegrind, cg_annotate and cg_merge"> |
| 101 | <title>Using Cachegrind, cg_annotate and cg_merge</title> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 102 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 103 | <para>First off, as for normal Valgrind use, you probably want to |
| 104 | compile with debugging info (the |
njn | a331164 | 2009-08-10 01:29:14 +0000 | [diff] [blame] | 105 | <option>-g</option> option). But by contrast with |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 106 | normal Valgrind use, you probably do want to turn |
| 107 | optimisation on, since you should profile your program as it will |
| 108 | be normally run.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 109 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 110 | <para>Then, you need to run Cachegrind itself to gather the profiling |
| 111 | information, and then run cg_annotate to get a detailed presentation of that |
| 112 | information. As an optional intermediate step, you can use cg_merge to sum |
njn | 69d495d | 2010-06-30 05:23:34 +0000 | [diff] [blame] | 113 | together the outputs of multiple Cachegrind runs into a single file which |
| 114 | you then use as the input for cg_annotate. Alternatively, you can use |
mjw | 3a09084 | 2013-07-03 10:00:17 +0000 | [diff] [blame] | 115 | cg_diff to difference the outputs of two Cachegrind runs into a single file |
njn | 69d495d | 2010-06-30 05:23:34 +0000 | [diff] [blame] | 116 | which you then use as the input for cg_annotate.</para> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 117 | |
| 118 | |
| 119 | <sect2 id="cg-manual.running-cachegrind" xreflabel="Running Cachegrind"> |
| 120 | <title>Running Cachegrind</title> |
| 121 | |
| 122 | <para>To run Cachegrind on a program <filename>prog</filename>, run:</para> |
| 123 | <screen><![CDATA[ |
| 124 | valgrind --tool=cachegrind prog |
| 125 | ]]></screen> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 126 | |
| 127 | <para>The program will execute (slowly). Upon completion, |
| 128 | summary statistics that look like this will be printed:</para> |
| 129 | |
| 130 | <programlisting><![CDATA[ |
| 131 | ==31751== I refs: 27,742,716 |
| 132 | ==31751== I1 misses: 276 |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 133 | ==31751== LLi misses: 275 |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 134 | ==31751== I1 miss rate: 0.0% |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 135 | ==31751== LLi miss rate: 0.0% |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 136 | ==31751== |
| 137 | ==31751== D refs: 15,430,290 (10,955,517 rd + 4,474,773 wr) |
| 138 | ==31751== D1 misses: 41,185 ( 21,905 rd + 19,280 wr) |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 139 | ==31751== LLd misses: 23,085 ( 3,987 rd + 19,098 wr) |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 140 | ==31751== D1 miss rate: 0.2% ( 0.1% + 0.4%) |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 141 | ==31751== LLd miss rate: 0.1% ( 0.0% + 0.4%) |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 142 | ==31751== |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 143 | ==31751== LL misses: 23,360 ( 4,262 rd + 19,098 wr) |
| 144 | ==31751== LL miss rate: 0.0% ( 0.0% + 0.4%)]]></programlisting> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 145 | |
| 146 | <para>Cache accesses for instruction fetches are summarised |
| 147 | first, giving the number of fetches made (this is the number of |
| 148 | instructions executed, which can be useful to know in its own |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 149 | right), the number of I1 misses, and the number of LL instruction |
| 150 | (<computeroutput>LLi</computeroutput>) misses.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 151 | |
| 152 | <para>Cache accesses for data follow. The information is similar |
| 153 | to that of the instruction fetches, except that the values are |
| 154 | also shown split between reads and writes (note each row's |
| 155 | <computeroutput>rd</computeroutput> and |
| 156 | <computeroutput>wr</computeroutput> values add up to the row's |
| 157 | total).</para> |
| 158 | |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 159 | <para>Combined instruction and data figures for the LL cache |
| 160 | follow that. Note that the LL miss rate is computed relative to the total |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 161 | number of memory accesses, not the number of L1 misses. I.e. it is |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 162 | <computeroutput>(ILmr + DLmr + DLmw) / (Ir + Dr + Dw)</computeroutput> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 163 | not |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 164 | <computeroutput>(ILmr + DLmr + DLmw) / (I1mr + D1mr + D1mw)</computeroutput> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 165 | </para> |
| 166 | |
| 167 | <para>Branch prediction statistics are not collected by default. |
njn | a331164 | 2009-08-10 01:29:14 +0000 | [diff] [blame] | 168 | To do so, add the option <option>--branch-sim=yes</option>.</para> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 169 | |
| 170 | </sect2> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 171 | |
| 172 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 173 | <sect2 id="cg-manual.outputfile" xreflabel="Output File"> |
| 174 | <title>Output File</title> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 175 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 176 | <para>As well as printing summary information, Cachegrind also writes |
| 177 | more detailed profiling information to a file. By default this file is named |
| 178 | <filename>cachegrind.out.<pid></filename> (where |
| 179 | <filename><pid></filename> is the program's process ID), but its name |
| 180 | can be changed with the <option>--cachegrind-out-file</option> option. This |
| 181 | file is human-readable, but is intended to be interpreted by the |
| 182 | accompanying program cg_annotate, described in the next section.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 183 | |
njn | 374a36d | 2007-11-23 01:41:32 +0000 | [diff] [blame] | 184 | <para>The default <computeroutput>.<pid></computeroutput> suffix |
de | 7e109d1 | 2005-11-18 22:09:58 +0000 | [diff] [blame] | 185 | on the output file name serves two purposes. Firstly, it means you |
| 186 | don't have to rename old log files that you don't want to overwrite. |
| 187 | Secondly, and more importantly, it allows correct profiling with the |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 188 | <option>--trace-children=yes</option> option of |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 189 | programs that spawn child processes.</para> |
| 190 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 191 | <para>The output file can be big, many megabytes for large applications |
| 192 | built with full debugging information.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 193 | |
| 194 | </sect2> |
| 195 | |
| 196 | |
| 197 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 198 | <sect2 id="cg-manual.running-cg_annotate" xreflabel="Running cg_annotate"> |
| 199 | <title>Running cg_annotate</title> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 200 | |
njn | 374a36d | 2007-11-23 01:41:32 +0000 | [diff] [blame] | 201 | <para>Before using cg_annotate, |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 202 | it is worth widening your window to be at least 120-characters |
| 203 | wide if possible, as the output lines can be quite long.</para> |
| 204 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 205 | <para>To get a function-by-function summary, run:</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 206 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 207 | <screen>cg_annotate <filename></screen> |
| 208 | |
| 209 | <para>on a Cachegrind output file.</para> |
| 210 | |
| 211 | </sect2> |
| 212 | |
| 213 | |
| 214 | <sect2 id="cg-manual.the-output-preamble" xreflabel="The Output Preamble"> |
| 215 | <title>The Output Preamble</title> |
| 216 | |
| 217 | <para>The first part of the output looks like this:</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 218 | |
| 219 | <programlisting><![CDATA[ |
| 220 | -------------------------------------------------------------------------------- |
| 221 | I1 cache: 65536 B, 64 B, 2-way associative |
| 222 | D1 cache: 65536 B, 64 B, 2-way associative |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 223 | LL cache: 262144 B, 64 B, 8-way associative |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 224 | Command: concord vg_to_ucode.c |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 225 | Events recorded: Ir I1mr ILmr Dr D1mr DLmr Dw D1mw DLmw |
| 226 | Events shown: Ir I1mr ILmr Dr D1mr DLmr Dw D1mw DLmw |
| 227 | Event sort order: Ir I1mr ILmr Dr D1mr DLmr Dw D1mw DLmw |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 228 | Threshold: 99% |
| 229 | Chosen for annotation: |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 230 | Auto-annotation: off |
| 231 | ]]></programlisting> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 232 | |
| 233 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 234 | <para>This is a summary of the annotation options:</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 235 | |
| 236 | <itemizedlist> |
| 237 | |
| 238 | <listitem> |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 239 | <para>I1 cache, D1 cache, LL cache: cache configuration. So |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 240 | you know the configuration with which these results were |
| 241 | obtained.</para> |
| 242 | </listitem> |
| 243 | |
| 244 | <listitem> |
| 245 | <para>Command: the command line invocation of the program |
| 246 | under examination.</para> |
| 247 | </listitem> |
| 248 | |
| 249 | <listitem> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 250 | <para>Events recorded: which events were recorded.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 251 | |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 252 | </listitem> |
| 253 | |
| 254 | <listitem> |
sewardj | 08e31e2 | 2007-05-23 21:58:33 +0000 | [diff] [blame] | 255 | <para>Events shown: the events shown, which is a subset of the events |
| 256 | gathered. This can be adjusted with the |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 257 | <option>--show</option> option.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 258 | </listitem> |
| 259 | |
| 260 | <listitem> |
| 261 | <para>Event sort order: the sort order in which functions are |
| 262 | shown. For example, in this case the functions are sorted |
| 263 | from highest <computeroutput>Ir</computeroutput> counts to |
| 264 | lowest. If two functions have identical |
| 265 | <computeroutput>Ir</computeroutput> counts, they will then be |
| 266 | sorted by <computeroutput>I1mr</computeroutput> counts, and |
| 267 | so on. This order can be adjusted with the |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 268 | <option>--sort</option> option.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 269 | |
| 270 | <para>Note that this dictates the order the functions appear. |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 271 | It is <emphasis>not</emphasis> the order in which the columns |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 272 | appear; that is dictated by the "events shown" line (and can |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 273 | be changed with the <option>--show</option> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 274 | option).</para> |
| 275 | </listitem> |
| 276 | |
| 277 | <listitem> |
njn | 374a36d | 2007-11-23 01:41:32 +0000 | [diff] [blame] | 278 | <para>Threshold: cg_annotate |
sewardj | 08e31e2 | 2007-05-23 21:58:33 +0000 | [diff] [blame] | 279 | by default omits functions that cause very low counts |
| 280 | to avoid drowning you in information. In this case, |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 281 | cg_annotate shows summaries the functions that account for |
| 282 | 99% of the <computeroutput>Ir</computeroutput> counts; |
| 283 | <computeroutput>Ir</computeroutput> is chosen as the |
| 284 | threshold event since it is the primary sort event. The |
| 285 | threshold can be adjusted with the |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 286 | <option>--threshold</option> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 287 | option.</para> |
| 288 | </listitem> |
| 289 | |
| 290 | <listitem> |
| 291 | <para>Chosen for annotation: names of files specified |
| 292 | manually for annotation; in this case none.</para> |
| 293 | </listitem> |
| 294 | |
| 295 | <listitem> |
| 296 | <para>Auto-annotation: whether auto-annotation was requested |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 297 | via the <option>--auto=yes</option> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 298 | option. In this case no.</para> |
| 299 | </listitem> |
| 300 | |
| 301 | </itemizedlist> |
| 302 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 303 | </sect2> |
| 304 | |
| 305 | |
| 306 | <sect2 id="cg-manual.the-global" |
| 307 | xreflabel="The Global and Function-level Counts"> |
| 308 | <title>The Global and Function-level Counts</title> |
| 309 | |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 310 | <para>Then follows summary statistics for the whole |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 311 | program:</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 312 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 313 | <programlisting><![CDATA[ |
| 314 | -------------------------------------------------------------------------------- |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 315 | Ir I1mr ILmr Dr D1mr DLmr Dw D1mw DLmw |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 316 | -------------------------------------------------------------------------------- |
| 317 | 27,742,716 276 275 10,955,517 21,905 3,987 4,474,773 19,280 19,098 PROGRAM TOTALS]]></programlisting> |
| 318 | |
| 319 | <para> |
| 320 | These are similar to the summary provided when Cachegrind finishes running. |
| 321 | </para> |
| 322 | |
| 323 | <para>Then comes function-by-function statistics:</para> |
| 324 | |
| 325 | <programlisting><![CDATA[ |
| 326 | -------------------------------------------------------------------------------- |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 327 | Ir I1mr ILmr Dr D1mr DLmr Dw D1mw DLmw file:function |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 328 | -------------------------------------------------------------------------------- |
| 329 | 8,821,482 5 5 2,242,702 1,621 73 1,794,230 0 0 getc.c:_IO_getc |
| 330 | 5,222,023 4 4 2,276,334 16 12 875,959 1 1 concord.c:get_word |
| 331 | 2,649,248 2 2 1,344,810 7,326 1,385 . . . vg_main.c:strcmp |
| 332 | 2,521,927 2 2 591,215 0 0 179,398 0 0 concord.c:hash |
| 333 | 2,242,740 2 2 1,046,612 568 22 448,548 0 0 ctype.c:tolower |
| 334 | 1,496,937 4 4 630,874 9,000 1,400 279,388 0 0 concord.c:insert |
| 335 | 897,991 51 51 897,831 95 30 62 1 1 ???:??? |
| 336 | 598,068 1 1 299,034 0 0 149,517 0 0 ../sysdeps/generic/lockfile.c:__flockfile |
| 337 | 598,068 0 0 299,034 0 0 149,517 0 0 ../sysdeps/generic/lockfile.c:__funlockfile |
| 338 | 598,024 4 4 213,580 35 16 149,506 0 0 vg_clientmalloc.c:malloc |
| 339 | 446,587 1 1 215,973 2,167 430 129,948 14,057 13,957 concord.c:add_existing |
| 340 | 341,760 2 2 128,160 0 0 128,160 0 0 vg_clientmalloc.c:vg_trap_here_WRAPPER |
| 341 | 320,782 4 4 150,711 276 0 56,027 53 53 concord.c:init_hash_table |
| 342 | 298,998 1 1 106,785 0 0 64,071 1 1 concord.c:create |
| 343 | 149,518 0 0 149,516 0 0 1 0 0 ???:tolower@@GLIBC_2.0 |
| 344 | 149,518 0 0 149,516 0 0 1 0 0 ???:fgetc@@GLIBC_2.0 |
| 345 | 95,983 4 4 38,031 0 0 34,409 3,152 3,150 concord.c:new_word_node |
| 346 | 85,440 0 0 42,720 0 0 21,360 0 0 vg_clientmalloc.c:vg_bogus_epilogue]]></programlisting> |
| 347 | |
| 348 | <para>Each function |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 349 | is identified by a |
| 350 | <computeroutput>file_name:function_name</computeroutput> pair. If |
| 351 | a column contains only a dot it means the function never performs |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 352 | that event (e.g. the third row shows that |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 353 | <computeroutput>strcmp()</computeroutput> contains no |
| 354 | instructions that write to memory). The name |
mjw | 2be5122 | 2013-04-05 13:19:12 +0000 | [diff] [blame] | 355 | <computeroutput>???</computeroutput> is used if the file name |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 356 | and/or function name could not be determined from debugging |
| 357 | information. If most of the entries have the form |
| 358 | <computeroutput>???:???</computeroutput> the program probably |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 359 | wasn't compiled with <option>-g</option>.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 360 | |
sewardj | 08e31e2 | 2007-05-23 21:58:33 +0000 | [diff] [blame] | 361 | <para>It is worth noting that functions will come both from |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 362 | the profiled program (e.g. <filename>concord.c</filename>) |
| 363 | and from libraries (e.g. <filename>getc.c</filename>)</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 364 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 365 | </sect2> |
| 366 | |
| 367 | |
| 368 | <sect2 id="cg-manual.line-by-line" xreflabel="Line-by-line Counts"> |
| 369 | <title>Line-by-line Counts</title> |
| 370 | |
| 371 | <para>There are two ways to annotate source files -- by specifying them |
| 372 | manually as arguments to cg_annotate, or with the |
| 373 | <option>--auto=yes</option> option. For example, the output from running |
| 374 | <filename>cg_annotate <filename> concord.c</filename> for our example |
| 375 | produces the same output as above followed by an annotated version of |
| 376 | <filename>concord.c</filename>, a section of which looks like:</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 377 | |
| 378 | <programlisting><![CDATA[ |
| 379 | -------------------------------------------------------------------------------- |
| 380 | -- User-annotated source: concord.c |
| 381 | -------------------------------------------------------------------------------- |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 382 | Ir I1mr ILmr Dr D1mr DLmr Dw D1mw DLmw |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 383 | |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 384 | . . . . . . . . . void init_hash_table(char *file_name, Word_Node *table[]) |
| 385 | 3 1 1 . . . 1 0 0 { |
| 386 | . . . . . . . . . FILE *file_ptr; |
| 387 | . . . . . . . . . Word_Info *data; |
| 388 | 1 0 0 . . . 1 1 1 int line = 1, i; |
| 389 | . . . . . . . . . |
| 390 | 5 0 0 . . . 3 0 0 data = (Word_Info *) create(sizeof(Word_Info)); |
| 391 | . . . . . . . . . |
| 392 | 4,991 0 0 1,995 0 0 998 0 0 for (i = 0; i < TABLE_SIZE; i++) |
| 393 | 3,988 1 1 1,994 0 0 997 53 52 table[i] = NULL; |
| 394 | . . . . . . . . . |
| 395 | . . . . . . . . . /* Open file, check it. */ |
| 396 | 6 0 0 1 0 0 4 0 0 file_ptr = fopen(file_name, "r"); |
| 397 | 2 0 0 1 0 0 . . . if (!(file_ptr)) { |
| 398 | . . . . . . . . . fprintf(stderr, "Couldn't open '%s'.\n", file_name); |
| 399 | 1 1 1 . . . . . . exit(EXIT_FAILURE); |
| 400 | . . . . . . . . . } |
| 401 | . . . . . . . . . |
| 402 | 165,062 1 1 73,360 0 0 91,700 0 0 while ((line = get_word(data, line, file_ptr)) != EOF) |
| 403 | 146,712 0 0 73,356 0 0 73,356 0 0 insert(data->;word, data->line, table); |
| 404 | . . . . . . . . . |
| 405 | 4 0 0 1 0 0 2 0 0 free(data); |
| 406 | 4 0 0 1 0 0 2 0 0 fclose(file_ptr); |
| 407 | 3 0 0 2 0 0 . . . }]]></programlisting> |
| 408 | |
| 409 | <para>(Although column widths are automatically minimised, a wide |
| 410 | terminal is clearly useful.)</para> |
| 411 | |
| 412 | <para>Each source file is clearly marked |
| 413 | (<computeroutput>User-annotated source</computeroutput>) as |
| 414 | having been chosen manually for annotation. If the file was |
| 415 | found in one of the directories specified with the |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 416 | <option>-I</option>/<option>--include</option> option, the directory |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 417 | and file are both given.</para> |
| 418 | |
| 419 | <para>Each line is annotated with its event counts. Events not |
sewardj | 08e31e2 | 2007-05-23 21:58:33 +0000 | [diff] [blame] | 420 | applicable for a line are represented by a dot. This is useful |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 421 | for distinguishing between an event which cannot happen, and one |
| 422 | which can but did not.</para> |
| 423 | |
| 424 | <para>Sometimes only a small section of a source file is |
sewardj | 8d9fec5 | 2005-11-15 20:56:23 +0000 | [diff] [blame] | 425 | executed. To minimise uninteresting output, Cachegrind only shows |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 426 | annotated lines and lines within a small distance of annotated |
| 427 | lines. Gaps are marked with the line numbers so you know which |
| 428 | part of a file the shown code comes from, eg:</para> |
| 429 | |
| 430 | <programlisting><![CDATA[ |
| 431 | (figures and code for line 704) |
| 432 | -- line 704 ---------------------------------------- |
| 433 | -- line 878 ---------------------------------------- |
| 434 | (figures and code for line 878)]]></programlisting> |
| 435 | |
| 436 | <para>The amount of context to show around annotated lines is |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 437 | controlled by the <option>--context</option> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 438 | option.</para> |
| 439 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 440 | <para>To get automatic annotation, use the <option>--auto=yes</option> option. |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 441 | cg_annotate will automatically annotate every source file it can |
| 442 | find that is mentioned in the function-by-function summary. |
| 443 | Therefore, the files chosen for auto-annotation are affected by |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 444 | the <option>--sort</option> and |
| 445 | <option>--threshold</option> options. Each |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 446 | source file is clearly marked (<computeroutput>Auto-annotated |
| 447 | source</computeroutput>) as being chosen automatically. Any |
| 448 | files that could not be found are mentioned at the end of the |
| 449 | output, eg:</para> |
| 450 | |
| 451 | <programlisting><![CDATA[ |
| 452 | ------------------------------------------------------------------ |
| 453 | The following files chosen for auto-annotation could not be found: |
| 454 | ------------------------------------------------------------------ |
| 455 | getc.c |
| 456 | ctype.c |
| 457 | ../sysdeps/generic/lockfile.c]]></programlisting> |
| 458 | |
| 459 | <para>This is quite common for library files, since libraries are |
| 460 | usually compiled with debugging information, but the source files |
| 461 | are often not present on a system. If a file is chosen for |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 462 | annotation both manually and automatically, it |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 463 | is marked as <computeroutput>User-annotated |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 464 | source</computeroutput>. Use the |
| 465 | <option>-I</option>/<option>--include</option> option to tell Valgrind where |
| 466 | to look for source files if the filenames found from the debugging |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 467 | information aren't specific enough.</para> |
| 468 | |
| 469 | <para>Beware that cg_annotate can take some time to digest large |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 470 | <filename>cachegrind.out.<pid></filename> files, |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 471 | e.g. 30 seconds or more. Also beware that auto-annotation can |
| 472 | produce a lot of output if your program is large!</para> |
| 473 | |
| 474 | </sect2> |
| 475 | |
| 476 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 477 | <sect2 id="cg-manual.assembler" xreflabel="Annotating Assembly Code Programs"> |
| 478 | <title>Annotating Assembly Code Programs</title> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 479 | |
njn | 374a36d | 2007-11-23 01:41:32 +0000 | [diff] [blame] | 480 | <para>Valgrind can annotate assembly code programs too, or annotate |
| 481 | the assembly code generated for your C program. Sometimes this is |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 482 | useful for understanding what is really happening when an |
| 483 | interesting line of C code is translated into multiple |
| 484 | instructions.</para> |
| 485 | |
| 486 | <para>To do this, you just need to assemble your |
njn | 85a38bc | 2008-10-30 02:41:13 +0000 | [diff] [blame] | 487 | <computeroutput>.s</computeroutput> files with assembly-level debug |
njn | 7316df2 | 2009-08-04 01:16:01 +0000 | [diff] [blame] | 488 | information. You can use compile with the <option>-S</option> to compile C/C++ |
| 489 | programs to assembly code, and then assemble the assembly code files with |
| 490 | <option>-g</option> to achieve this. You can then profile and annotate the |
| 491 | assembly code source files in the same way as C/C++ source files.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 492 | |
| 493 | </sect2> |
| 494 | |
njn | 7064fb2 | 2008-05-29 23:09:52 +0000 | [diff] [blame] | 495 | <sect2 id="ms-manual.forkingprograms" xreflabel="Forking Programs"> |
| 496 | <title>Forking Programs</title> |
| 497 | <para>If your program forks, the child will inherit all the profiling data that |
| 498 | has been gathered for the parent.</para> |
| 499 | |
| 500 | <para>If the output file format string (controlled by |
| 501 | <option>--cachegrind-out-file</option>) does not contain <option>%p</option>, |
| 502 | then the outputs from the parent and child will be intermingled in a single |
| 503 | output file, which will almost certainly make it unreadable by |
| 504 | cg_annotate.</para> |
| 505 | </sect2> |
| 506 | |
| 507 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 508 | <sect2 id="cg-manual.annopts.warnings" xreflabel="cg_annotate Warnings"> |
| 509 | <title>cg_annotate Warnings</title> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 510 | |
| 511 | <para>There are a couple of situations in which |
njn | 374a36d | 2007-11-23 01:41:32 +0000 | [diff] [blame] | 512 | cg_annotate issues warnings.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 513 | |
| 514 | <itemizedlist> |
| 515 | <listitem> |
| 516 | <para>If a source file is more recent than the |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 517 | <filename>cachegrind.out.<pid></filename> file. |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 518 | This is because the information in |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 519 | <filename>cachegrind.out.<pid></filename> is only |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 520 | recorded with line numbers, so if the line numbers change at |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 521 | all in the source (e.g. lines added, deleted, swapped), any |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 522 | annotations will be incorrect.</para> |
| 523 | </listitem> |
| 524 | <listitem> |
| 525 | <para>If information is recorded about line numbers past the |
| 526 | end of a file. This can be caused by the above problem, |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 527 | i.e. shortening the source file while using an old |
| 528 | <filename>cachegrind.out.<pid></filename> file. If |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 529 | this happens, the figures for the bogus lines are printed |
| 530 | anyway (clearly marked as bogus) in case they are |
| 531 | important.</para> |
| 532 | </listitem> |
| 533 | </itemizedlist> |
| 534 | |
| 535 | </sect2> |
| 536 | |
| 537 | |
| 538 | |
sewardj | 778d783 | 2007-11-22 01:21:56 +0000 | [diff] [blame] | 539 | <sect2 id="cg-manual.annopts.things-to-watch-out-for" |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 540 | xreflabel="Unusual Annotation Cases"> |
| 541 | <title>Unusual Annotation Cases</title> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 542 | |
| 543 | <para>Some odd things that can occur during annotation:</para> |
| 544 | |
| 545 | <itemizedlist> |
| 546 | <listitem> |
| 547 | <para>If annotating at the assembler level, you might see |
| 548 | something like this:</para> |
| 549 | <programlisting><![CDATA[ |
| 550 | 1 0 0 . . . . . . leal -12(%ebp),%eax |
| 551 | 1 0 0 . . . 1 0 0 movl %eax,84(%ebx) |
| 552 | 2 0 0 0 0 0 1 0 0 movl $1,-20(%ebp) |
| 553 | . . . . . . . . . .align 4,0x90 |
| 554 | 1 0 0 . . . . . . movl $.LnrB,%eax |
| 555 | 1 0 0 . . . 1 0 0 movl %eax,-16(%ebp)]]></programlisting> |
| 556 | |
| 557 | <para>How can the third instruction be executed twice when |
| 558 | the others are executed only once? As it turns out, it |
| 559 | isn't. Here's a dump of the executable, using |
| 560 | <computeroutput>objdump -d</computeroutput>:</para> |
| 561 | <programlisting><![CDATA[ |
| 562 | 8048f25: 8d 45 f4 lea 0xfffffff4(%ebp),%eax |
| 563 | 8048f28: 89 43 54 mov %eax,0x54(%ebx) |
| 564 | 8048f2b: c7 45 ec 01 00 00 00 movl $0x1,0xffffffec(%ebp) |
| 565 | 8048f32: 89 f6 mov %esi,%esi |
| 566 | 8048f34: b8 08 8b 07 08 mov $0x8078b08,%eax |
| 567 | 8048f39: 89 45 f0 mov %eax,0xfffffff0(%ebp)]]></programlisting> |
| 568 | |
| 569 | <para>Notice the extra <computeroutput>mov |
| 570 | %esi,%esi</computeroutput> instruction. Where did this come |
| 571 | from? The GNU assembler inserted it to serve as the two |
| 572 | bytes of padding needed to align the <computeroutput>movl |
| 573 | $.LnrB,%eax</computeroutput> instruction on a four-byte |
| 574 | boundary, but pretended it didn't exist when adding debug |
| 575 | information. Thus when Valgrind reads the debug info it |
| 576 | thinks that the <computeroutput>movl |
| 577 | $0x1,0xffffffec(%ebp)</computeroutput> instruction covers the |
| 578 | address range 0x8048f2b--0x804833 by itself, and attributes |
| 579 | the counts for the <computeroutput>mov |
| 580 | %esi,%esi</computeroutput> to it.</para> |
| 581 | </listitem> |
| 582 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 583 | <!-- |
| 584 | I think this isn't true any more, not since cost centres were moved from |
| 585 | being associated with instruction addresses to being associated with |
| 586 | source line numbers. |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 587 | <listitem> |
| 588 | <para>Inlined functions can cause strange results in the |
| 589 | function-by-function summary. If a function |
| 590 | <computeroutput>inline_me()</computeroutput> is defined in |
| 591 | <filename>foo.h</filename> and inlined in the functions |
| 592 | <computeroutput>f1()</computeroutput>, |
| 593 | <computeroutput>f2()</computeroutput> and |
| 594 | <computeroutput>f3()</computeroutput> in |
| 595 | <filename>bar.c</filename>, there will not be a |
| 596 | <computeroutput>foo.h:inline_me()</computeroutput> function |
| 597 | entry. Instead, there will be separate function entries for |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 598 | each inlining site, i.e. |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 599 | <computeroutput>foo.h:f1()</computeroutput>, |
| 600 | <computeroutput>foo.h:f2()</computeroutput> and |
| 601 | <computeroutput>foo.h:f3()</computeroutput>. To find the |
| 602 | total counts for |
| 603 | <computeroutput>foo.h:inline_me()</computeroutput>, add up |
| 604 | the counts from each entry.</para> |
| 605 | |
| 606 | <para>The reason for this is that although the debug info |
njn | 7316df2 | 2009-08-04 01:16:01 +0000 | [diff] [blame] | 607 | output by GCC indicates the switch from |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 608 | <filename>bar.c</filename> to <filename>foo.h</filename>, it |
| 609 | doesn't indicate the name of the function in |
| 610 | <filename>foo.h</filename>, so Valgrind keeps using the old |
| 611 | one.</para> |
| 612 | </listitem> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 613 | --> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 614 | |
| 615 | <listitem> |
| 616 | <para>Sometimes, the same filename might be represented with |
| 617 | a relative name and with an absolute name in different parts |
| 618 | of the debug info, eg: |
| 619 | <filename>/home/user/proj/proj.h</filename> and |
| 620 | <filename>../proj.h</filename>. In this case, if you use |
| 621 | auto-annotation, the file will be annotated twice with the |
| 622 | counts split between the two.</para> |
| 623 | </listitem> |
| 624 | |
| 625 | <listitem> |
| 626 | <para>Files with more than 65,535 lines cause difficulties |
sewardj | 08e31e2 | 2007-05-23 21:58:33 +0000 | [diff] [blame] | 627 | for the Stabs-format debug info reader. This is because the line |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 628 | number in the <computeroutput>struct nlist</computeroutput> |
| 629 | defined in <filename>a.out.h</filename> under Linux is only a |
| 630 | 16-bit value. Valgrind can handle some files with more than |
| 631 | 65,535 lines correctly by making some guesses to identify |
| 632 | line number overflows. But some cases are beyond it, in |
| 633 | which case you'll get a warning message explaining that |
| 634 | annotations for the file might be incorrect.</para> |
sewardj | 08e31e2 | 2007-05-23 21:58:33 +0000 | [diff] [blame] | 635 | |
njn | 7316df2 | 2009-08-04 01:16:01 +0000 | [diff] [blame] | 636 | <para>If you are using GCC 3.1 or later, this is most likely |
| 637 | irrelevant, since GCC switched to using the more modern DWARF2 |
sewardj | 08e31e2 | 2007-05-23 21:58:33 +0000 | [diff] [blame] | 638 | format by default at version 3.1. DWARF2 does not have any such |
| 639 | limitations on line numbers.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 640 | </listitem> |
| 641 | |
| 642 | <listitem> |
| 643 | <para>If you compile some files with |
njn | 7e5d4ed | 2009-07-30 02:57:52 +0000 | [diff] [blame] | 644 | <option>-g</option> and some without, some |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 645 | events that take place in a file without debug info could be |
| 646 | attributed to the last line of a file with debug info |
| 647 | (whichever one gets placed before the non-debug-info file in |
| 648 | the executable).</para> |
| 649 | </listitem> |
| 650 | |
| 651 | </itemizedlist> |
| 652 | |
| 653 | <para>This list looks long, but these cases should be fairly |
| 654 | rare.</para> |
| 655 | |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 656 | </sect2> |
| 657 | |
| 658 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 659 | <sect2 id="cg-manual.cg_merge" xreflabel="cg_merge"> |
| 660 | <title>Merging Profiles with cg_merge</title> |
sewardj | 94dc508 | 2007-02-08 11:31:03 +0000 | [diff] [blame] | 661 | |
| 662 | <para> |
njn | 374a36d | 2007-11-23 01:41:32 +0000 | [diff] [blame] | 663 | cg_merge is a simple program which |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 664 | reads multiple profile files, as created by Cachegrind, merges them |
sewardj | 94dc508 | 2007-02-08 11:31:03 +0000 | [diff] [blame] | 665 | together, and writes the results into another file in the same format. |
| 666 | You can then examine the merged results using |
njn | 374a36d | 2007-11-23 01:41:32 +0000 | [diff] [blame] | 667 | <computeroutput>cg_annotate <filename></computeroutput>, as |
sewardj | 94dc508 | 2007-02-08 11:31:03 +0000 | [diff] [blame] | 668 | described above. The merging functionality might be useful if you |
| 669 | want to aggregate costs over multiple runs of the same program, or |
| 670 | from a single parallel run with multiple instances of the same |
| 671 | program.</para> |
| 672 | |
| 673 | <para> |
njn | 374a36d | 2007-11-23 01:41:32 +0000 | [diff] [blame] | 674 | cg_merge is invoked as follows: |
sewardj | 94dc508 | 2007-02-08 11:31:03 +0000 | [diff] [blame] | 675 | </para> |
| 676 | |
| 677 | <programlisting><![CDATA[ |
| 678 | cg_merge -o outputfile file1 file2 file3 ...]]></programlisting> |
| 679 | |
| 680 | <para> |
| 681 | It reads and checks <computeroutput>file1</computeroutput>, then read |
| 682 | and checks <computeroutput>file2</computeroutput> and merges it into |
| 683 | the running totals, then the same with |
| 684 | <computeroutput>file3</computeroutput>, etc. The final results are |
| 685 | written to <computeroutput>outputfile</computeroutput>, or to standard |
| 686 | out if no output file is specified.</para> |
| 687 | |
| 688 | <para> |
| 689 | Costs are summed on a per-function, per-line and per-instruction |
| 690 | basis. Because of this, the order in which the input files does not |
| 691 | matter, although you should take care to only mention each file once, |
| 692 | since any file mentioned twice will be added in twice.</para> |
| 693 | |
| 694 | <para> |
njn | 374a36d | 2007-11-23 01:41:32 +0000 | [diff] [blame] | 695 | cg_merge does not attempt to check |
sewardj | 94dc508 | 2007-02-08 11:31:03 +0000 | [diff] [blame] | 696 | that the input files come from runs of the same executable. It will |
| 697 | happily merge together profile files from completely unrelated |
| 698 | programs. It does however check that the |
| 699 | <computeroutput>Events:</computeroutput> lines of all the inputs are |
| 700 | identical, so as to ensure that the addition of costs makes sense. |
| 701 | For example, it would be nonsensical for it to add a number indicating |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 702 | D1 read references to a number from a different file indicating LL |
sewardj | 94dc508 | 2007-02-08 11:31:03 +0000 | [diff] [blame] | 703 | write misses.</para> |
| 704 | |
| 705 | <para> |
| 706 | A number of other syntax and sanity checks are done whilst reading the |
njn | 374a36d | 2007-11-23 01:41:32 +0000 | [diff] [blame] | 707 | inputs. cg_merge will stop and |
sewardj | 94dc508 | 2007-02-08 11:31:03 +0000 | [diff] [blame] | 708 | attempt to print a helpful error message if any of the input files |
| 709 | fail these checks.</para> |
| 710 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 711 | </sect2> |
| 712 | |
| 713 | |
njn | 69d495d | 2010-06-30 05:23:34 +0000 | [diff] [blame] | 714 | <sect2 id="cg-manual.cg_diff" xreflabel="cg_diff"> |
| 715 | <title>Differencing Profiles with cg_diff</title> |
| 716 | |
| 717 | <para> |
| 718 | cg_diff is a simple program which |
| 719 | reads two profile files, as created by Cachegrind, finds the difference |
| 720 | between them, and writes the results into another file in the same format. |
| 721 | You can then examine the merged results using |
| 722 | <computeroutput>cg_annotate <filename></computeroutput>, as |
| 723 | described above. This is very useful if you want to measure how a change to |
| 724 | a program affected its performance. |
| 725 | </para> |
| 726 | |
| 727 | <para> |
| 728 | cg_diff is invoked as follows: |
| 729 | </para> |
| 730 | |
| 731 | <programlisting><![CDATA[ |
| 732 | cg_diff file1 file2]]></programlisting> |
| 733 | |
| 734 | <para> |
| 735 | It reads and checks <computeroutput>file1</computeroutput>, then read |
| 736 | and checks <computeroutput>file2</computeroutput>, then computes the |
| 737 | difference (effectively <computeroutput>file1</computeroutput> - |
| 738 | <computeroutput>file2</computeroutput>). The final results are written to |
| 739 | standard output.</para> |
| 740 | |
| 741 | <para> |
| 742 | Costs are summed on a per-function basis. Per-line costs are not summed, |
| 743 | because doing so is too difficult. For example, consider differencing two |
| 744 | profiles, one from a single-file program A, and one from the same program A |
| 745 | where a single blank line was inserted at the top of the file. Every single |
| 746 | per-line count has changed. In comparison, the per-function counts have not |
| 747 | changed. The per-function count differences are still very useful for |
| 748 | determining differences between programs. Note that because the result is |
| 749 | the difference of two profiles, many of the counts will be negative; this |
| 750 | indicates that the counts for the relevant function are fewer in the second |
| 751 | version than those in the first version.</para> |
| 752 | |
| 753 | <para> |
| 754 | cg_diff does not attempt to check |
| 755 | that the input files come from runs of the same executable. It will |
| 756 | happily merge together profile files from completely unrelated |
| 757 | programs. It does however check that the |
| 758 | <computeroutput>Events:</computeroutput> lines of all the inputs are |
| 759 | identical, so as to ensure that the addition of costs makes sense. |
| 760 | For example, it would be nonsensical for it to add a number indicating |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 761 | D1 read references to a number from a different file indicating LL |
njn | 69d495d | 2010-06-30 05:23:34 +0000 | [diff] [blame] | 762 | write misses.</para> |
| 763 | |
| 764 | <para> |
| 765 | A number of other syntax and sanity checks are done whilst reading the |
| 766 | inputs. cg_diff will stop and |
| 767 | attempt to print a helpful error message if any of the input files |
| 768 | fail these checks.</para> |
| 769 | |
| 770 | <para> |
| 771 | Sometimes you will want to compare Cachegrind profiles of two versions of a |
| 772 | program that you have sitting side-by-side. For example, you might have |
| 773 | <computeroutput>version1/prog.c</computeroutput> and |
| 774 | <computeroutput>version2/prog.c</computeroutput>, where the second is |
| 775 | slightly different to the first. A straight comparison of the two will not |
| 776 | be useful -- because functions are qualified with filenames, a function |
| 777 | <function>f</function> will be listed as |
| 778 | <computeroutput>version1/prog.c:f</computeroutput> for the first version but |
| 779 | <computeroutput>version2/prog.c:f</computeroutput> for the second |
| 780 | version.</para> |
| 781 | |
| 782 | <para> |
| 783 | When this happens, you can use the <option>--mod-filename</option> option. |
| 784 | Its argument is a Perl search-and-replace expression that will be applied |
| 785 | to all the filenames in both Cachegrind output files. It can be used to |
| 786 | remove minor differences in filenames. For example, the option |
| 787 | <option>--mod-filename='s/version[0-9]/versionN/'</option> will suffice for |
| 788 | this case.</para> |
| 789 | |
njn | e5930da | 2010-12-17 00:45:19 +0000 | [diff] [blame] | 790 | <para> |
| 791 | Similarly, sometimes compilers auto-generate certain functions and give them |
| 792 | randomized names. For example, GCC sometimes auto-generates functions with |
| 793 | names like <function>T.1234</function>, and the suffixes vary from build to |
| 794 | build. You can use the <option>--mod-funcname</option> option to remove |
| 795 | small differences like these; it works in the same way as |
| 796 | <option>--mod-filename</option>.</para> |
| 797 | |
njn | 69d495d | 2010-06-30 05:23:34 +0000 | [diff] [blame] | 798 | </sect2> |
| 799 | |
| 800 | |
sewardj | 94dc508 | 2007-02-08 11:31:03 +0000 | [diff] [blame] | 801 | </sect1> |
| 802 | |
| 803 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 804 | |
njn | a331164 | 2009-08-10 01:29:14 +0000 | [diff] [blame] | 805 | <sect1 id="cg-manual.cgopts" xreflabel="Cachegrind Command-line Options"> |
| 806 | <title>Cachegrind Command-line Options</title> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 807 | |
| 808 | <!-- start of xi:include in the manpage --> |
| 809 | <para>Cachegrind-specific options are:</para> |
| 810 | |
| 811 | <variablelist id="cg.opts.list"> |
| 812 | |
| 813 | <varlistentry id="opt.I1" xreflabel="--I1"> |
| 814 | <term> |
| 815 | <option><![CDATA[--I1=<size>,<associativity>,<line size> ]]></option> |
| 816 | </term> |
| 817 | <listitem> |
| 818 | <para>Specify the size, associativity and line size of the level 1 |
| 819 | instruction cache. </para> |
| 820 | </listitem> |
| 821 | </varlistentry> |
| 822 | |
| 823 | <varlistentry id="opt.D1" xreflabel="--D1"> |
| 824 | <term> |
| 825 | <option><![CDATA[--D1=<size>,<associativity>,<line size> ]]></option> |
| 826 | </term> |
| 827 | <listitem> |
| 828 | <para>Specify the size, associativity and line size of the level 1 |
| 829 | data cache.</para> |
| 830 | </listitem> |
| 831 | </varlistentry> |
| 832 | |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 833 | <varlistentry id="opt.LL" xreflabel="--LL"> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 834 | <term> |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 835 | <option><![CDATA[--LL=<size>,<associativity>,<line size> ]]></option> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 836 | </term> |
| 837 | <listitem> |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 838 | <para>Specify the size, associativity and line size of the last-level |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 839 | cache.</para> |
| 840 | </listitem> |
| 841 | </varlistentry> |
| 842 | |
| 843 | <varlistentry id="opt.cache-sim" xreflabel="--cache-sim"> |
| 844 | <term> |
| 845 | <option><![CDATA[--cache-sim=no|yes [yes] ]]></option> |
| 846 | </term> |
| 847 | <listitem> |
| 848 | <para>Enables or disables collection of cache access and miss |
| 849 | counts.</para> |
| 850 | </listitem> |
| 851 | </varlistentry> |
| 852 | |
| 853 | <varlistentry id="opt.branch-sim" xreflabel="--branch-sim"> |
| 854 | <term> |
| 855 | <option><![CDATA[--branch-sim=no|yes [no] ]]></option> |
| 856 | </term> |
| 857 | <listitem> |
| 858 | <para>Enables or disables collection of branch instruction and |
| 859 | misprediction counts. By default this is disabled as it |
| 860 | slows Cachegrind down by approximately 25%. Note that you |
| 861 | cannot specify <option>--cache-sim=no</option> |
| 862 | and <option>--branch-sim=no</option> |
| 863 | together, as that would leave Cachegrind with no |
| 864 | information to collect.</para> |
| 865 | </listitem> |
| 866 | </varlistentry> |
| 867 | |
| 868 | <varlistentry id="opt.cachegrind-out-file" xreflabel="--cachegrind-out-file"> |
| 869 | <term> |
| 870 | <option><![CDATA[--cachegrind-out-file=<file> ]]></option> |
| 871 | </term> |
| 872 | <listitem> |
| 873 | <para>Write the profile data to |
| 874 | <computeroutput>file</computeroutput> rather than to the default |
| 875 | output file, |
| 876 | <filename>cachegrind.out.<pid></filename>. The |
| 877 | <option>%p</option> and <option>%q</option> format specifiers |
| 878 | can be used to embed the process ID and/or the contents of an |
| 879 | environment variable in the name, as is the case for the core |
| 880 | option <option><xref linkend="opt.log-file"/></option>. |
| 881 | </para> |
| 882 | </listitem> |
| 883 | </varlistentry> |
| 884 | |
| 885 | </variablelist> |
| 886 | <!-- end of xi:include in the manpage --> |
| 887 | |
| 888 | </sect1> |
| 889 | |
| 890 | |
| 891 | |
njn | a331164 | 2009-08-10 01:29:14 +0000 | [diff] [blame] | 892 | <sect1 id="cg-manual.annopts" xreflabel="cg_annotate Command-line Options"> |
| 893 | <title>cg_annotate Command-line Options</title> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 894 | |
njn | c206a81 | 2009-08-07 07:56:20 +0000 | [diff] [blame] | 895 | <!-- start of xi:include in the manpage --> |
| 896 | <variablelist id="cg_annotate.opts.list"> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 897 | |
| 898 | <varlistentry> |
| 899 | <term> |
| 900 | <option><![CDATA[-h --help ]]></option> |
| 901 | </term> |
| 902 | <listitem> |
| 903 | <para>Show the help message.</para> |
| 904 | </listitem> |
| 905 | </varlistentry> |
| 906 | |
| 907 | <varlistentry> |
| 908 | <term> |
| 909 | <option><![CDATA[--version ]]></option> |
| 910 | </term> |
| 911 | <listitem> |
| 912 | <para>Show the version number.</para> |
| 913 | </listitem> |
| 914 | </varlistentry> |
| 915 | |
| 916 | <varlistentry> |
| 917 | <term> |
| 918 | <option><![CDATA[--show=A,B,C [default: all, using order in |
| 919 | cachegrind.out.<pid>] ]]></option> |
| 920 | </term> |
| 921 | <listitem> |
| 922 | <para>Specifies which events to show (and the column |
| 923 | order). Default is to use all present in the |
| 924 | <filename>cachegrind.out.<pid></filename> file (and |
| 925 | use the order in the file). Useful if you want to concentrate on, for |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 926 | example, I cache misses (<option>--show=I1mr,ILmr</option>), or data |
| 927 | read misses (<option>--show=D1mr,DLmr</option>), or LL data misses |
| 928 | (<option>--show=DLmr,DLmw</option>). Best used in conjunction with |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 929 | <option>--sort</option>.</para> |
| 930 | </listitem> |
| 931 | </varlistentry> |
| 932 | |
| 933 | <varlistentry> |
| 934 | <term> |
| 935 | <option><![CDATA[--sort=A,B,C [default: order in |
| 936 | cachegrind.out.<pid>] ]]></option> |
| 937 | </term> |
| 938 | <listitem> |
| 939 | <para>Specifies the events upon which the sorting of the |
| 940 | function-by-function entries will be based.</para> |
| 941 | </listitem> |
| 942 | </varlistentry> |
| 943 | |
| 944 | <varlistentry> |
| 945 | <term> |
njn | 69d495d | 2010-06-30 05:23:34 +0000 | [diff] [blame] | 946 | <option><![CDATA[--threshold=X [default: 0.1%] ]]></option> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 947 | </term> |
| 948 | <listitem> |
| 949 | <para>Sets the threshold for the function-by-function |
njn | 69d495d | 2010-06-30 05:23:34 +0000 | [diff] [blame] | 950 | summary. A function is shown if it accounts for more than X% |
| 951 | of the counts for the primary sort event. If auto-annotating, also |
| 952 | affects which files are annotated.</para> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 953 | |
| 954 | <para>Note: thresholds can be set for more than one of the |
| 955 | events by appending any events for the |
| 956 | <option>--sort</option> option with a colon |
| 957 | and a number (no spaces, though). E.g. if you want to see |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 958 | each function that covers more than 1% of LL read misses or 1% of LL |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 959 | write misses, use this option:</para> |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 960 | <para><option>--sort=DLmr:1,DLmw:1</option></para> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 961 | </listitem> |
| 962 | </varlistentry> |
| 963 | |
| 964 | <varlistentry> |
| 965 | <term> |
| 966 | <option><![CDATA[--auto=<no|yes> [default: no] ]]></option> |
| 967 | </term> |
| 968 | <listitem> |
| 969 | <para>When enabled, automatically annotates every file that |
| 970 | is mentioned in the function-by-function summary that can be |
| 971 | found. Also gives a list of those that couldn't be found.</para> |
| 972 | </listitem> |
| 973 | </varlistentry> |
| 974 | |
| 975 | <varlistentry> |
| 976 | <term> |
| 977 | <option><![CDATA[--context=N [default: 8] ]]></option> |
| 978 | </term> |
| 979 | <listitem> |
| 980 | <para>Print N lines of context before and after each |
| 981 | annotated line. Avoids printing large sections of source |
| 982 | files that were not executed. Use a large number |
| 983 | (e.g. 100000) to show all source lines.</para> |
| 984 | </listitem> |
| 985 | </varlistentry> |
| 986 | |
| 987 | <varlistentry> |
| 988 | <term> |
| 989 | <option><![CDATA[-I<dir> --include=<dir> [default: none] ]]></option> |
| 990 | </term> |
| 991 | <listitem> |
| 992 | <para>Adds a directory to the list in which to search for |
| 993 | files. Multiple <option>-I</option>/<option>--include</option> |
| 994 | options can be given to add multiple directories.</para> |
| 995 | </listitem> |
| 996 | </varlistentry> |
| 997 | |
| 998 | </variablelist> |
njn | c206a81 | 2009-08-07 07:56:20 +0000 | [diff] [blame] | 999 | <!-- end of xi:include in the manpage --> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1000 | |
| 1001 | </sect1> |
| 1002 | |
| 1003 | |
mjw | 3a09084 | 2013-07-03 10:00:17 +0000 | [diff] [blame] | 1004 | <sect1 id="cg-manual.mergeopts" xreflabel="cg_merge Command-line Options"> |
| 1005 | <title>cg_merge Command-line Options</title> |
| 1006 | |
| 1007 | <!-- start of xi:include in the manpage --> |
| 1008 | <variablelist id="cg_merge.opts.list"> |
| 1009 | |
| 1010 | <varlistentry> |
| 1011 | <term> |
| 1012 | <option><![CDATA[-o outfile]]></option> |
| 1013 | </term> |
| 1014 | <listitem> |
| 1015 | <para>Write the profile data to <computeroutput>outfile</computeroutput> |
| 1016 | rather than to standard output. |
| 1017 | </para> |
| 1018 | </listitem> |
| 1019 | </varlistentry> |
| 1020 | |
| 1021 | </variablelist> |
| 1022 | <!-- end of xi:include in the manpage --> |
| 1023 | |
| 1024 | </sect1> |
| 1025 | |
| 1026 | |
njn | 69d495d | 2010-06-30 05:23:34 +0000 | [diff] [blame] | 1027 | <sect1 id="cg-manual.diffopts" xreflabel="cg_diff Command-line Options"> |
| 1028 | <title>cg_diff Command-line Options</title> |
| 1029 | |
| 1030 | <!-- start of xi:include in the manpage --> |
| 1031 | <variablelist id="cg_diff.opts.list"> |
| 1032 | |
| 1033 | <varlistentry> |
| 1034 | <term> |
| 1035 | <option><![CDATA[-h --help ]]></option> |
| 1036 | </term> |
| 1037 | <listitem> |
| 1038 | <para>Show the help message.</para> |
| 1039 | </listitem> |
| 1040 | </varlistentry> |
| 1041 | |
| 1042 | <varlistentry> |
| 1043 | <term> |
| 1044 | <option><![CDATA[--version ]]></option> |
| 1045 | </term> |
| 1046 | <listitem> |
| 1047 | <para>Show the version number.</para> |
| 1048 | </listitem> |
| 1049 | </varlistentry> |
| 1050 | |
| 1051 | <varlistentry> |
| 1052 | <term> |
| 1053 | <option><![CDATA[--mod-filename=<expr> [default: none]]]></option> |
| 1054 | </term> |
| 1055 | <listitem> |
| 1056 | <para>Specifies a Perl search-and-replace expression that is applied |
| 1057 | to all filenames. Useful for removing minor differences in paths |
| 1058 | between two different versions of a program that are sitting in |
| 1059 | different directories.</para> |
| 1060 | </listitem> |
| 1061 | </varlistentry> |
| 1062 | |
njn | e5930da | 2010-12-17 00:45:19 +0000 | [diff] [blame] | 1063 | <varlistentry> |
| 1064 | <term> |
| 1065 | <option><![CDATA[--mod-funcname=<expr> [default: none]]]></option> |
| 1066 | </term> |
| 1067 | <listitem> |
| 1068 | <para>Like <option>--mod-filename</option>, but for filenames. |
| 1069 | Useful for removing minor differences in randomized names of |
| 1070 | auto-generated functions generated by some compilers.</para> |
| 1071 | </listitem> |
| 1072 | </varlistentry> |
| 1073 | |
njn | 69d495d | 2010-06-30 05:23:34 +0000 | [diff] [blame] | 1074 | </variablelist> |
| 1075 | <!-- end of xi:include in the manpage --> |
| 1076 | |
| 1077 | </sect1> |
| 1078 | |
| 1079 | |
| 1080 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1081 | |
sewardj | 778d783 | 2007-11-22 01:21:56 +0000 | [diff] [blame] | 1082 | <sect1 id="cg-manual.acting-on" |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1083 | xreflabel="Acting on Cachegrind's Information"> |
| 1084 | <title>Acting on Cachegrind's Information</title> |
njn | 3a9d5dc | 2007-09-17 22:19:01 +0000 | [diff] [blame] | 1085 | <para> |
njn | a31dac2 | 2009-07-30 03:21:42 +0000 | [diff] [blame] | 1086 | Cachegrind gives you lots of information, but acting on that information |
| 1087 | isn't always easy. Here are some rules of thumb that we have found to be |
njn | 07f9656 | 2007-09-17 22:28:21 +0000 | [diff] [blame] | 1088 | useful.</para> |
njn | 3a9d5dc | 2007-09-17 22:19:01 +0000 | [diff] [blame] | 1089 | |
| 1090 | <para> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1091 | First of all, the global hit/miss counts and miss rates are not that useful. |
| 1092 | If you have multiple programs or multiple runs of a program, comparing the |
| 1093 | numbers might identify if any are outliers and worthy of closer |
| 1094 | investigation. Otherwise, they're not enough to act on.</para> |
njn | 3a9d5dc | 2007-09-17 22:19:01 +0000 | [diff] [blame] | 1095 | |
| 1096 | <para> |
njn | a31dac2 | 2009-07-30 03:21:42 +0000 | [diff] [blame] | 1097 | The function-by-function counts are more useful to look at, as they pinpoint |
| 1098 | which functions are causing large numbers of counts. However, beware that |
| 1099 | inlining can make these counts misleading. If a function |
| 1100 | <function>f</function> is always inlined, counts will be attributed to the |
| 1101 | functions it is inlined into, rather than itself. However, if you look at |
| 1102 | the line-by-line annotations for <function>f</function> you'll see the |
| 1103 | counts that belong to <function>f</function>. (This is hard to avoid, it's |
| 1104 | how the debug info is structured.) So it's worth looking for large numbers |
| 1105 | in the line-by-line annotations.</para> |
| 1106 | |
| 1107 | <para> |
njn | 07f9656 | 2007-09-17 22:28:21 +0000 | [diff] [blame] | 1108 | The line-by-line source code annotations are much more useful. In our |
| 1109 | experience, the best place to start is by looking at the |
| 1110 | <computeroutput>Ir</computeroutput> numbers. They simply measure how many |
| 1111 | instructions were executed for each line, and don't include any cache |
| 1112 | information, but they can still be very useful for identifying |
| 1113 | bottlenecks.</para> |
njn | 3a9d5dc | 2007-09-17 22:19:01 +0000 | [diff] [blame] | 1114 | |
| 1115 | <para> |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 1116 | After that, we have found that LL misses are typically a much bigger source |
njn | 3a9d5dc | 2007-09-17 22:19:01 +0000 | [diff] [blame] | 1117 | of slow-downs than L1 misses. So it's worth looking for any snippets of |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 1118 | code with high <computeroutput>DLmr</computeroutput> or |
| 1119 | <computeroutput>DLmw</computeroutput> counts. (You can use |
| 1120 | <option>--show=DLmr |
| 1121 | --sort=DLmr</option> with cg_annotate to focus just on |
| 1122 | <literal>DLmr</literal> counts, for example.) If you find any, it's still |
njn | a31dac2 | 2009-07-30 03:21:42 +0000 | [diff] [blame] | 1123 | not always easy to work out how to improve things. You need to have a |
njn | 07f9656 | 2007-09-17 22:28:21 +0000 | [diff] [blame] | 1124 | reasonable understanding of how caches work, the principles of locality, and |
| 1125 | your program's data access patterns. Improving things may require |
| 1126 | redesigning a data structure, for example.</para> |
njn | 3a9d5dc | 2007-09-17 22:19:01 +0000 | [diff] [blame] | 1127 | |
| 1128 | <para> |
njn | a31dac2 | 2009-07-30 03:21:42 +0000 | [diff] [blame] | 1129 | Looking at the <computeroutput>Bcm</computeroutput> and |
| 1130 | <computeroutput>Bim</computeroutput> misses can also be helpful. |
| 1131 | In particular, <computeroutput>Bim</computeroutput> misses are often caused |
| 1132 | by <literal>switch</literal> statements, and in some cases these |
| 1133 | <literal>switch</literal> statements can be replaced with table-driven code. |
| 1134 | For example, you might replace code like this:</para> |
| 1135 | |
| 1136 | <programlisting><![CDATA[ |
| 1137 | enum E { A, B, C }; |
| 1138 | enum E e; |
| 1139 | int i; |
| 1140 | ... |
| 1141 | switch (e) |
| 1142 | { |
tom | 270e2a3 | 2011-08-15 11:11:41 +0000 | [diff] [blame] | 1143 | case A: i += 1; break; |
| 1144 | case B: i += 2; break; |
| 1145 | case C: i += 3; break; |
njn | a31dac2 | 2009-07-30 03:21:42 +0000 | [diff] [blame] | 1146 | } |
| 1147 | ]]></programlisting> |
| 1148 | |
| 1149 | <para>with code like this:</para> |
| 1150 | |
| 1151 | <programlisting><![CDATA[ |
| 1152 | enum E { A, B, C }; |
| 1153 | enum E e; |
| 1154 | enum E table[] = { 1, 2, 3 }; |
| 1155 | int i; |
| 1156 | ... |
| 1157 | i += table[e]; |
| 1158 | ]]></programlisting> |
| 1159 | |
| 1160 | <para> |
| 1161 | This is obviously a contrived example, but the basic principle applies in a |
| 1162 | wide variety of situations.</para> |
| 1163 | |
| 1164 | <para> |
njn | 3a9d5dc | 2007-09-17 22:19:01 +0000 | [diff] [blame] | 1165 | In short, Cachegrind can tell you where some of the bottlenecks in your code |
| 1166 | are, but it can't tell you how to fix them. You have to work that out for |
| 1167 | yourself. But at least you have the information! |
| 1168 | </para> |
| 1169 | |
| 1170 | </sect1> |
sewardj | 94dc508 | 2007-02-08 11:31:03 +0000 | [diff] [blame] | 1171 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1172 | |
| 1173 | <sect1 id="cg-manual.sim-details" |
| 1174 | xreflabel="Simulation Details"> |
| 1175 | <title>Simulation Details</title> |
| 1176 | <para> |
| 1177 | This section talks about details you don't need to know about in order to |
| 1178 | use Cachegrind, but may be of interest to some people. |
| 1179 | </para> |
| 1180 | |
| 1181 | <sect2 id="cache-sim" xreflabel="Cache Simulation Specifics"> |
| 1182 | <title>Cache Simulation Specifics</title> |
| 1183 | |
| 1184 | <para>Specific characteristics of the cache simulation are as |
| 1185 | follows:</para> |
| 1186 | |
| 1187 | <itemizedlist> |
| 1188 | |
| 1189 | <listitem> |
| 1190 | <para>Write-allocate: when a write miss occurs, the block |
| 1191 | written to is brought into the D1 cache. Most modern caches |
| 1192 | have this property.</para> |
| 1193 | </listitem> |
| 1194 | |
| 1195 | <listitem> |
| 1196 | <para>Bit-selection hash function: the set of line(s) in the cache |
| 1197 | to which a memory block maps is chosen by the middle bits |
| 1198 | M--(M+N-1) of the byte address, where:</para> |
| 1199 | <itemizedlist> |
| 1200 | <listitem> |
| 1201 | <para>line size = 2^M bytes</para> |
| 1202 | </listitem> |
| 1203 | <listitem> |
| 1204 | <para>(cache size / line size / associativity) = 2^N bytes</para> |
| 1205 | </listitem> |
| 1206 | </itemizedlist> |
| 1207 | </listitem> |
| 1208 | |
| 1209 | <listitem> |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 1210 | <para>Inclusive LL cache: the LL cache typically replicates all |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1211 | the entries of the L1 caches, because fetching into L1 involves |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 1212 | fetching into LL first (this does not guarantee strict inclusiveness, |
| 1213 | as lines evicted from LL still could reside in L1). This is |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1214 | standard on Pentium chips, but AMD Opterons, Athlons and Durons |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 1215 | use an exclusive LL cache that only holds |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1216 | blocks evicted from L1. Ditto most modern VIA CPUs.</para> |
| 1217 | </listitem> |
| 1218 | |
| 1219 | </itemizedlist> |
| 1220 | |
| 1221 | <para>The cache configuration simulated (cache size, |
| 1222 | associativity and line size) is determined automatically using |
| 1223 | the x86 CPUID instruction. If you have a machine that (a) |
| 1224 | doesn't support the CPUID instruction, or (b) supports it in an |
| 1225 | early incarnation that doesn't give any cache information, then |
| 1226 | Cachegrind will fall back to using a default configuration (that |
| 1227 | of a model 3/4 Athlon). Cachegrind will tell you if this |
| 1228 | happens. You can manually specify one, two or all three levels |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 1229 | (I1/D1/LL) of the cache from the command line using the |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1230 | <option>--I1</option>, |
| 1231 | <option>--D1</option> and |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 1232 | <option>--LL</option> options. |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1233 | For cache parameters to be valid for simulation, the number |
| 1234 | of sets (with associativity being the number of cache lines in |
| 1235 | each set) has to be a power of two.</para> |
| 1236 | |
| 1237 | <para>On PowerPC platforms |
| 1238 | Cachegrind cannot automatically |
| 1239 | determine the cache configuration, so you will |
| 1240 | need to specify it with the |
| 1241 | <option>--I1</option>, |
| 1242 | <option>--D1</option> and |
njn | 2d853a1 | 2010-10-06 22:46:31 +0000 | [diff] [blame] | 1243 | <option>--LL</option> options.</para> |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1244 | |
| 1245 | |
| 1246 | <para>Other noteworthy behaviour:</para> |
| 1247 | |
| 1248 | <itemizedlist> |
| 1249 | <listitem> |
| 1250 | <para>References that straddle two cache lines are treated as |
| 1251 | follows:</para> |
| 1252 | <itemizedlist> |
| 1253 | <listitem> |
| 1254 | <para>If both blocks hit --> counted as one hit</para> |
| 1255 | </listitem> |
| 1256 | <listitem> |
| 1257 | <para>If one block hits, the other misses --> counted |
| 1258 | as one miss.</para> |
| 1259 | </listitem> |
| 1260 | <listitem> |
| 1261 | <para>If both blocks miss --> counted as one miss (not |
| 1262 | two)</para> |
| 1263 | </listitem> |
| 1264 | </itemizedlist> |
| 1265 | </listitem> |
| 1266 | |
| 1267 | <listitem> |
| 1268 | <para>Instructions that modify a memory location |
| 1269 | (e.g. <computeroutput>inc</computeroutput> and |
| 1270 | <computeroutput>dec</computeroutput>) are counted as doing |
| 1271 | just a read, i.e. a single data reference. This may seem |
| 1272 | strange, but since the write can never cause a miss (the read |
| 1273 | guarantees the block is in the cache) it's not very |
| 1274 | interesting.</para> |
| 1275 | |
| 1276 | <para>Thus it measures not the number of times the data cache |
| 1277 | is accessed, but the number of times a data cache miss could |
| 1278 | occur.</para> |
| 1279 | </listitem> |
| 1280 | |
| 1281 | </itemizedlist> |
| 1282 | |
| 1283 | <para>If you are interested in simulating a cache with different |
| 1284 | properties, it is not particularly hard to write your own cache |
| 1285 | simulator, or to modify the existing ones in |
| 1286 | <computeroutput>cg_sim.c</computeroutput>. We'd be |
| 1287 | interested to hear from anyone who does.</para> |
| 1288 | |
| 1289 | </sect2> |
| 1290 | |
| 1291 | |
| 1292 | <sect2 id="branch-sim" xreflabel="Branch Simulation Specifics"> |
| 1293 | <title>Branch Simulation Specifics</title> |
| 1294 | |
| 1295 | <para>Cachegrind simulates branch predictors intended to be |
| 1296 | typical of mainstream desktop/server processors of around 2004.</para> |
| 1297 | |
| 1298 | <para>Conditional branches are predicted using an array of 16384 2-bit |
| 1299 | saturating counters. The array index used for a branch instruction is |
| 1300 | computed partly from the low-order bits of the branch instruction's |
| 1301 | address and partly using the taken/not-taken behaviour of the last few |
| 1302 | conditional branches. As a result the predictions for any specific |
| 1303 | branch depend both on its own history and the behaviour of previous |
| 1304 | branches. This is a standard technique for improving prediction |
| 1305 | accuracy.</para> |
| 1306 | |
| 1307 | <para>For indirect branches (that is, jumps to unknown destinations) |
| 1308 | Cachegrind uses a simple branch target address predictor. Targets are |
| 1309 | predicted using an array of 512 entries indexed by the low order 9 |
| 1310 | bits of the branch instruction's address. Each branch is predicted to |
| 1311 | jump to the same address it did last time. Any other behaviour causes |
| 1312 | a mispredict.</para> |
| 1313 | |
| 1314 | <para>More recent processors have better branch predictors, in |
| 1315 | particular better indirect branch predictors. Cachegrind's predictor |
| 1316 | design is deliberately conservative so as to be representative of the |
| 1317 | large installed base of processors which pre-date widespread |
| 1318 | deployment of more sophisticated indirect branch predictors. In |
| 1319 | particular, late model Pentium 4s (Prescott), Pentium M, Core and Core |
| 1320 | 2 have more sophisticated indirect branch predictors than modelled by |
| 1321 | Cachegrind. </para> |
| 1322 | |
| 1323 | <para>Cachegrind does not simulate a return stack predictor. It |
| 1324 | assumes that processors perfectly predict function return addresses, |
| 1325 | an assumption which is probably close to being true.</para> |
| 1326 | |
| 1327 | <para>See Hennessy and Patterson's classic text "Computer |
| 1328 | Architecture: A Quantitative Approach", 4th edition (2007), Section |
| 1329 | 2.3 (pages 80-89) for background on modern branch predictors.</para> |
| 1330 | |
| 1331 | </sect2> |
| 1332 | |
| 1333 | <sect2 id="cg-manual.annopts.accuracy" xreflabel="Accuracy"> |
| 1334 | <title>Accuracy</title> |
| 1335 | |
| 1336 | <para>Valgrind's cache profiling has a number of |
| 1337 | shortcomings:</para> |
| 1338 | |
| 1339 | <itemizedlist> |
| 1340 | <listitem> |
| 1341 | <para>It doesn't account for kernel activity -- the effect of system |
| 1342 | calls on the cache and branch predictor contents is ignored.</para> |
| 1343 | </listitem> |
| 1344 | |
| 1345 | <listitem> |
| 1346 | <para>It doesn't account for other process activity. |
| 1347 | This is probably desirable when considering a single |
| 1348 | program.</para> |
| 1349 | </listitem> |
| 1350 | |
| 1351 | <listitem> |
| 1352 | <para>It doesn't account for virtual-to-physical address |
| 1353 | mappings. Hence the simulation is not a true |
| 1354 | representation of what's happening in the |
| 1355 | cache. Most caches and branch predictors are physically indexed, but |
| 1356 | Cachegrind simulates caches using virtual addresses.</para> |
| 1357 | </listitem> |
| 1358 | |
| 1359 | <listitem> |
| 1360 | <para>It doesn't account for cache misses not visible at the |
| 1361 | instruction level, e.g. those arising from TLB misses, or |
| 1362 | speculative execution.</para> |
| 1363 | </listitem> |
| 1364 | |
| 1365 | <listitem> |
| 1366 | <para>Valgrind will schedule |
| 1367 | threads differently from how they would be when running natively. |
| 1368 | This could warp the results for threaded programs.</para> |
| 1369 | </listitem> |
| 1370 | |
| 1371 | <listitem> |
| 1372 | <para>The x86/amd64 instructions <computeroutput>bts</computeroutput>, |
| 1373 | <computeroutput>btr</computeroutput> and |
| 1374 | <computeroutput>btc</computeroutput> will incorrectly be |
| 1375 | counted as doing a data read if both the arguments are |
| 1376 | registers, eg:</para> |
| 1377 | <programlisting><![CDATA[ |
| 1378 | btsl %eax, %edx]]></programlisting> |
| 1379 | |
| 1380 | <para>This should only happen rarely.</para> |
| 1381 | </listitem> |
| 1382 | |
| 1383 | <listitem> |
| 1384 | <para>x86/amd64 FPU instructions with data sizes of 28 and 108 bytes |
| 1385 | (e.g. <computeroutput>fsave</computeroutput>) are treated as |
| 1386 | though they only access 16 bytes. These instructions seem to |
| 1387 | be rare so hopefully this won't affect accuracy much.</para> |
| 1388 | </listitem> |
| 1389 | |
| 1390 | </itemizedlist> |
| 1391 | |
| 1392 | <para>Another thing worth noting is that results are very sensitive. |
mjw | 2be5122 | 2013-04-05 13:19:12 +0000 | [diff] [blame] | 1393 | Changing the size of the executable being profiled, or the sizes |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1394 | of any of the shared libraries it uses, or even the length of their |
| 1395 | file names, can perturb the results. Variations will be small, but |
| 1396 | don't expect perfectly repeatable results if your program changes at |
| 1397 | all.</para> |
| 1398 | |
| 1399 | <para>More recent GNU/Linux distributions do address space |
| 1400 | randomisation, in which identical runs of the same program have their |
| 1401 | shared libraries loaded at different locations, as a security measure. |
| 1402 | This also perturbs the results.</para> |
| 1403 | |
| 1404 | <para>While these factors mean you shouldn't trust the results to |
| 1405 | be super-accurate, they should be close enough to be useful.</para> |
| 1406 | |
| 1407 | </sect2> |
| 1408 | |
| 1409 | </sect1> |
| 1410 | |
| 1411 | |
| 1412 | |
sewardj | 778d783 | 2007-11-22 01:21:56 +0000 | [diff] [blame] | 1413 | <sect1 id="cg-manual.impl-details" |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1414 | xreflabel="Implementation Details"> |
| 1415 | <title>Implementation Details</title> |
njn | 3a9d5dc | 2007-09-17 22:19:01 +0000 | [diff] [blame] | 1416 | <para> |
njn | 534f781 | 2006-10-21 22:22:59 +0000 | [diff] [blame] | 1417 | This section talks about details you don't need to know about in order to |
| 1418 | use Cachegrind, but may be of interest to some people. |
njn | 3a9d5dc | 2007-09-17 22:19:01 +0000 | [diff] [blame] | 1419 | </para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 1420 | |
sewardj | 778d783 | 2007-11-22 01:21:56 +0000 | [diff] [blame] | 1421 | <sect2 id="cg-manual.impl-details.how-cg-works" |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1422 | xreflabel="How Cachegrind Works"> |
| 1423 | <title>How Cachegrind Works</title> |
njn | 534f781 | 2006-10-21 22:22:59 +0000 | [diff] [blame] | 1424 | <para>The best reference for understanding how Cachegrind works is chapter 3 of |
| 1425 | "Dynamic Binary Analysis and Instrumentation", by Nicholas Nethercote. It |
njn | 25ac384 | 2009-08-07 02:58:11 +0000 | [diff] [blame] | 1426 | is available on the <ulink url="&vg-pubs-url;">Valgrind publications |
njn | 011215f | 2006-10-21 23:00:59 +0000 | [diff] [blame] | 1427 | page</ulink>.</para> |
njn | 534f781 | 2006-10-21 22:22:59 +0000 | [diff] [blame] | 1428 | </sect2> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 1429 | |
sewardj | 778d783 | 2007-11-22 01:21:56 +0000 | [diff] [blame] | 1430 | <sect2 id="cg-manual.impl-details.file-format" |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1431 | xreflabel="Cachegrind Output File Format"> |
| 1432 | <title>Cachegrind Output File Format</title> |
njn | 534f781 | 2006-10-21 22:22:59 +0000 | [diff] [blame] | 1433 | <para>The file format is fairly straightforward, basically giving the |
| 1434 | cost centre for every line, grouped by files and |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1435 | functions. It's also totally generic and self-describing, in the sense that |
| 1436 | it can be used for any events that can be counted on a line-by-line basis, |
| 1437 | not just cache and branch predictor events. For example, earlier versions |
| 1438 | of Cachegrind didn't have a branch predictor simulation. When this was |
| 1439 | added, the file format didn't need to change at all. So the format (and |
| 1440 | consequently, cg_annotate) could be used by other tools.</para> |
njn | 534f781 | 2006-10-21 22:22:59 +0000 | [diff] [blame] | 1441 | |
| 1442 | <para>The file format:</para> |
| 1443 | <programlisting><![CDATA[ |
| 1444 | file ::= desc_line* cmd_line events_line data_line+ summary_line |
| 1445 | desc_line ::= "desc:" ws? non_nl_string |
| 1446 | cmd_line ::= "cmd:" ws? cmd |
| 1447 | events_line ::= "events:" ws? (event ws)+ |
| 1448 | data_line ::= file_line | fn_line | count_line |
| 1449 | file_line ::= "fl=" filename |
| 1450 | fn_line ::= "fn=" fn_name |
| 1451 | count_line ::= line_num ws? (count ws)+ |
| 1452 | summary_line ::= "summary:" ws? (count ws)+ |
| 1453 | count ::= num | "."]]></programlisting> |
| 1454 | |
| 1455 | <para>Where:</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 1456 | <itemizedlist> |
| 1457 | <listitem> |
njn | 534f781 | 2006-10-21 22:22:59 +0000 | [diff] [blame] | 1458 | <para><computeroutput>non_nl_string</computeroutput> is any |
| 1459 | string not containing a newline.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 1460 | </listitem> |
njn | 534f781 | 2006-10-21 22:22:59 +0000 | [diff] [blame] | 1461 | <listitem> |
| 1462 | <para><computeroutput>cmd</computeroutput> is a string holding the |
| 1463 | command line of the profiled program.</para> |
| 1464 | </listitem> |
| 1465 | <listitem> |
njn | 2624212 | 2007-01-22 03:21:27 +0000 | [diff] [blame] | 1466 | <para><computeroutput>event</computeroutput> is a string containing |
| 1467 | no whitespace.</para> |
| 1468 | </listitem> |
| 1469 | <listitem> |
njn | 534f781 | 2006-10-21 22:22:59 +0000 | [diff] [blame] | 1470 | <para><computeroutput>filename</computeroutput> and |
| 1471 | <computeroutput>fn_name</computeroutput> are strings.</para> |
| 1472 | </listitem> |
| 1473 | <listitem> |
| 1474 | <para><computeroutput>num</computeroutput> and |
| 1475 | <computeroutput>line_num</computeroutput> are decimal |
| 1476 | numbers.</para> |
| 1477 | </listitem> |
| 1478 | <listitem> |
| 1479 | <para><computeroutput>ws</computeroutput> is whitespace.</para> |
| 1480 | </listitem> |
| 1481 | </itemizedlist> |
| 1482 | |
| 1483 | <para>The contents of the "desc:" lines are printed out at the top |
| 1484 | of the summary. This is a generic way of providing simulation |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1485 | specific information, e.g. for giving the cache configuration for |
njn | 534f781 | 2006-10-21 22:22:59 +0000 | [diff] [blame] | 1486 | cache simulation.</para> |
| 1487 | |
| 1488 | <para>More than one line of info can be presented for each file/fn/line number. |
| 1489 | In such cases, the counts for the named events will be accumulated.</para> |
| 1490 | |
njn | 3a9d5dc | 2007-09-17 22:19:01 +0000 | [diff] [blame] | 1491 | <para>Counts can be "." to represent zero. This makes the files easier for |
| 1492 | humans to read.</para> |
njn | 534f781 | 2006-10-21 22:22:59 +0000 | [diff] [blame] | 1493 | |
| 1494 | <para>The number of counts in each |
| 1495 | <computeroutput>line</computeroutput> and the |
| 1496 | <computeroutput>summary_line</computeroutput> should not exceed |
| 1497 | the number of events in the |
| 1498 | <computeroutput>event_line</computeroutput>. If the number in |
| 1499 | each <computeroutput>line</computeroutput> is less, cg_annotate |
njn | 3a9d5dc | 2007-09-17 22:19:01 +0000 | [diff] [blame] | 1500 | treats those missing as though they were a "." entry. This saves space. |
| 1501 | </para> |
njn | 534f781 | 2006-10-21 22:22:59 +0000 | [diff] [blame] | 1502 | |
| 1503 | <para>A <computeroutput>file_line</computeroutput> changes the |
| 1504 | current file name. A <computeroutput>fn_line</computeroutput> |
| 1505 | changes the current function name. A |
| 1506 | <computeroutput>count_line</computeroutput> contains counts that |
| 1507 | pertain to the current filename/fn_name. A "fn=" |
| 1508 | <computeroutput>file_line</computeroutput> and a |
| 1509 | <computeroutput>fn_line</computeroutput> must appear before any |
| 1510 | <computeroutput>count_line</computeroutput>s to give the context |
| 1511 | of the first <computeroutput>count_line</computeroutput>s.</para> |
| 1512 | |
| 1513 | <para>Each <computeroutput>file_line</computeroutput> will normally be |
| 1514 | immediately followed by a <computeroutput>fn_line</computeroutput>. But it |
| 1515 | doesn't have to be.</para> |
| 1516 | |
njn | 3da8196 | 2009-08-07 00:18:25 +0000 | [diff] [blame] | 1517 | <para>The summary line is redundant, because it just holds the total counts |
| 1518 | for each event. But this serves as a useful sanity check of the data; if |
| 1519 | the totals for each event don't match the summary line, something has gone |
| 1520 | wrong.</para> |
njn | 3e986b2 | 2004-11-30 10:43:45 +0000 | [diff] [blame] | 1521 | |
| 1522 | </sect2> |
| 1523 | |
| 1524 | </sect1> |
| 1525 | </chapter> |