1<html lang="en"> 2<head> 3<title>Bug Reporting - GNU Binary Utilities</title> 4<meta http-equiv="Content-Type" content="text/html"> 5<meta name="description" content="GNU Binary Utilities"> 6<meta name="generator" content="makeinfo 4.13"> 7<link title="Top" rel="start" href="index.html#Top"> 8<link rel="up" href="Reporting-Bugs.html#Reporting-Bugs" title="Reporting Bugs"> 9<link rel="prev" href="Bug-Criteria.html#Bug-Criteria" title="Bug Criteria"> 10<link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage"> 11<!-- 12Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 132000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011 14Free Software Foundation, Inc. 15 16Permission is granted to copy, distribute and/or modify this document 17under the terms of the GNU Free Documentation License, Version 1.3 18or any later version published by the Free Software Foundation; 19with no Invariant Sections, with no Front-Cover Texts, and with no 20Back-Cover Texts. A copy of the license is included in the 21section entitled ``GNU Free Documentation License''. 22 23--> 24<meta http-equiv="Content-Style-Type" content="text/css"> 25<style type="text/css"><!-- 26 pre.display { font-family:inherit } 27 pre.format { font-family:inherit } 28 pre.smalldisplay { font-family:inherit; font-size:smaller } 29 pre.smallformat { font-family:inherit; font-size:smaller } 30 pre.smallexample { font-size:smaller } 31 pre.smalllisp { font-size:smaller } 32 span.sc { font-variant:small-caps } 33 span.roman { font-family:serif; font-weight:normal; } 34 span.sansserif { font-family:sans-serif; font-weight:normal; } 35--></style> 36<link rel="stylesheet" type="text/css" href="../cs.css"> 37</head> 38<body> 39<div class="node"> 40<a name="Bug-Reporting"></a> 41<p> 42Previous: <a rel="previous" accesskey="p" href="Bug-Criteria.html#Bug-Criteria">Bug Criteria</a>, 43Up: <a rel="up" accesskey="u" href="Reporting-Bugs.html#Reporting-Bugs">Reporting Bugs</a> 44<hr> 45</div> 46 47<h3 class="section">19.2 How to Report Bugs</h3> 48 49<p><a name="index-bug-reports-156"></a><a name="index-bugs_002c-reporting-157"></a> 50A number of companies and individuals offer support for <span class="sc">gnu</span> 51products. If you obtained the binary utilities from a support 52organization, we recommend you contact that organization first. 53 54 <p>You can find contact information for many support companies and 55individuals in the file <samp><span class="file">etc/SERVICE</span></samp> in the <span class="sc">gnu</span> Emacs 56distribution. 57 58 <p>In any event, we also recommend that you send bug reports for the binary 59utilities to <a href="https://support.codesourcery.com/GNUToolchain/">https://support.codesourcery.com/GNUToolchain/</a>. 60 61 <p>The fundamental principle of reporting bugs usefully is this: 62<strong>report all the facts</strong>. If you are not sure whether to state a 63fact or leave it out, state it! 64 65 <p>Often people omit facts because they think they know what causes the 66problem and assume that some details do not matter. Thus, you might 67assume that the name of a file you use in an example does not matter. 68Well, probably it does not, but one cannot be sure. Perhaps the bug is 69a stray memory reference which happens to fetch from the location where 70that pathname is stored in memory; perhaps, if the pathname were 71different, the contents of that location would fool the utility into 72doing the right thing despite the bug. Play it safe and give a 73specific, complete example. That is the easiest thing for you to do, 74and the most helpful. 75 76 <p>Keep in mind that the purpose of a bug report is to enable us to fix the bug if 77it is new to us. Therefore, always write your bug reports on the assumption 78that the bug has not been reported previously. 79 80 <p>Sometimes people give a few sketchy facts and ask, “Does this ring a 81bell?” This cannot help us fix a bug, so it is basically useless. We 82respond by asking for enough details to enable us to investigate. 83You might as well expedite matters by sending them to begin with. 84 85 <p>To enable us to fix the bug, you should include all these things: 86 87 <ul> 88<li>The version of the utility. Each utility announces it if you start it 89with the <samp><span class="option">--version</span></samp> argument. 90 91 <p>Without this, we will not know whether there is any point in looking for 92the bug in the current version of the binary utilities. 93 94 <li>Any patches you may have applied to the source, including any patches 95made to the <code>BFD</code> library. 96 97 <li>The type of machine you are using, and the operating system name and 98version number. 99 100 <li>What compiler (and its version) was used to compile the utilities—e.g. 101“<code>gcc-2.7</code>”. 102 103 <li>The command arguments you gave the utility to observe the bug. To 104guarantee you will not omit something important, list them all. A copy 105of the Makefile (or the output from make) is sufficient. 106 107 <p>If we were to try to guess the arguments, we would probably guess wrong 108and then we might not encounter the bug. 109 110 <li>A complete input file, or set of input files, that will reproduce the 111bug. If the utility is reading an object file or files, then it is 112generally most helpful to send the actual object files. 113 114 <p>If the source files were produced exclusively using <span class="sc">gnu</span> programs 115(e.g., <samp><span class="command">gcc</span></samp>, <samp><span class="command">gas</span></samp>, and/or the <span class="sc">gnu</span> <samp><span class="command">ld</span></samp>), then it 116may be OK to send the source files rather than the object files. In 117this case, be sure to say exactly what version of <samp><span class="command">gcc</span></samp>, or 118whatever, was used to produce the object files. Also say how 119<samp><span class="command">gcc</span></samp>, or whatever, was configured. 120 121 <li>A description of what behavior you observe that you believe is 122incorrect. For example, “It gets a fatal signal.” 123 124 <p>Of course, if the bug is that the utility gets a fatal signal, then we 125will certainly notice it. But if the bug is incorrect output, we might 126not notice unless it is glaringly wrong. You might as well not give us 127a chance to make a mistake. 128 129 <p>Even if the problem you experience is a fatal signal, you should still 130say so explicitly. Suppose something strange is going on, such as your 131copy of the utility is out of sync, or you have encountered a bug in 132the C library on your system. (This has happened!) Your copy might 133crash and ours would not. If you told us to expect a crash, then when 134ours fails to crash, we would know that the bug was not happening for 135us. If you had not told us to expect a crash, then we would not be able 136to draw any conclusion from our observations. 137 138 <li>If you wish to suggest changes to the source, send us context diffs, as 139generated by <samp><span class="command">diff</span></samp> with the <samp><span class="option">-u</span></samp>, <samp><span class="option">-c</span></samp>, or <samp><span class="option">-p</span></samp> 140option. Always send diffs from the old file to the new file. If you 141wish to discuss something in the <samp><span class="command">ld</span></samp> source, refer to it by 142context, not by line number. 143 144 <p>The line numbers in our development sources will not match those in your 145sources. Your line numbers would convey no useful information to us. 146</ul> 147 148 <p>Here are some things that are not necessary: 149 150 <ul> 151<li>A description of the envelope of the bug. 152 153 <p>Often people who encounter a bug spend a lot of time investigating 154which changes to the input file will make the bug go away and which 155changes will not affect it. 156 157 <p>This is often time consuming and not very useful, because the way we 158will find the bug is by running a single example under the debugger 159with breakpoints, not by pure deduction from a series of examples. 160We recommend that you save your time for something else. 161 162 <p>Of course, if you can find a simpler example to report <em>instead</em> 163of the original one, that is a convenience for us. Errors in the 164output will be easier to spot, running under the debugger will take 165less time, and so on. 166 167 <p>However, simplification is not vital; if you do not want to do this, 168report the bug anyway and send us the entire test case you used. 169 170 <li>A patch for the bug. 171 172 <p>A patch for the bug does help us if it is a good one. But do not omit 173the necessary information, such as the test case, on the assumption that 174a patch is all we need. We might see problems with your patch and decide 175to fix the problem another way, or we might not understand it at all. 176 177 <p>Sometimes with programs as complicated as the binary utilities it is 178very hard to construct an example that will make the program follow a 179certain path through the code. If you do not send us the example, we 180will not be able to construct one, so we will not be able to verify that 181the bug is fixed. 182 183 <p>And if we cannot understand what bug you are trying to fix, or why your 184patch should be an improvement, we will not install it. A test case will 185help us to understand. 186 187 <li>A guess about what the bug is or what it depends on. 188 189 <p>Such guesses are usually wrong. Even we cannot guess right about such 190things without first using the debugger to find the facts. 191</ul> 192 193 </body></html> 194 195