/[pcre]/code/trunk/doc/html/pcrestack.html
ViewVC logotype

Contents of /code/trunk/doc/html/pcrestack.html

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1194 - (show annotations) (download) (as text)
Wed Oct 31 17:42:29 2012 UTC (20 months, 3 weeks ago) by ph10
File MIME type: text/html
File size: 9594 byte(s)
More documentation updates

1 <html>
2 <head>
3 <title>pcrestack specification</title>
4 </head>
5 <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
6 <h1>pcrestack man page</h1>
7 <p>
8 Return to the <a href="index.html">PCRE index page</a>.
9 </p>
10 <p>
11 This page is part of the PCRE HTML documentation. It was generated automatically
12 from the original man page. If there is any nonsense in it, please consult the
13 man page, in case the conversion went wrong.
14 <br>
15 <br><b>
16 PCRE DISCUSSION OF STACK USAGE
17 </b><br>
18 <P>
19 When you call <b>pcre[16|32]_exec()</b>, it makes use of an internal function
20 called <b>match()</b>. This calls itself recursively at branch points in the
21 pattern, in order to remember the state of the match so that it can back up and
22 try a different alternative if the first one fails. As matching proceeds deeper
23 and deeper into the tree of possibilities, the recursion depth increases. The
24 <b>match()</b> function is also called in other circumstances, for example,
25 whenever a parenthesized sub-pattern is entered, and in certain cases of
26 repetition.
27 </P>
28 <P>
29 Not all calls of <b>match()</b> increase the recursion depth; for an item such
30 as a* it may be called several times at the same level, after matching
31 different numbers of a's. Furthermore, in a number of cases where the result of
32 the recursive call would immediately be passed back as the result of the
33 current call (a "tail recursion"), the function is just restarted instead.
34 </P>
35 <P>
36 The above comments apply when <b>pcre[16|32]_exec()</b> is run in its normal
37 interpretive manner. If the pattern was studied with the
38 PCRE_STUDY_JIT_COMPILE option, and just-in-time compiling was successful, and
39 the options passed to <b>pcre[16|32]_exec()</b> were not incompatible, the matching
40 process uses the JIT-compiled code instead of the <b>match()</b> function. In
41 this case, the memory requirements are handled entirely differently. See the
42 <a href="pcrejit.html"><b>pcrejit</b></a>
43 documentation for details.
44 </P>
45 <P>
46 The <b>pcre[16|32]_dfa_exec()</b> function operates in an entirely different way,
47 and uses recursion only when there is a regular expression recursion or
48 subroutine call in the pattern. This includes the processing of assertion and
49 "once-only" subpatterns, which are handled like subroutine calls. Normally,
50 these are never very deep, and the limit on the complexity of
51 <b>pcre[16|32]_dfa_exec()</b> is controlled by the amount of workspace it is given.
52 However, it is possible to write patterns with runaway infinite recursions;
53 such patterns will cause <b>pcre[16|32]_dfa_exec()</b> to run out of stack. At
54 present, there is no protection against this.
55 </P>
56 <P>
57 The comments that follow do NOT apply to <b>pcre[16|32]_dfa_exec()</b>; they are
58 relevant only for <b>pcre[16|32]_exec()</b> without the JIT optimization.
59 </P>
60 <br><b>
61 Reducing <b>pcre[16|32]_exec()</b>'s stack usage
62 </b><br>
63 <P>
64 Each time that <b>match()</b> is actually called recursively, it uses memory
65 from the process stack. For certain kinds of pattern and data, very large
66 amounts of stack may be needed, despite the recognition of "tail recursion".
67 You can often reduce the amount of recursion, and therefore the amount of stack
68 used, by modifying the pattern that is being matched. Consider, for example,
69 this pattern:
70 <pre>
71 ([^&#60;]|&#60;(?!inet))+
72 </pre>
73 It matches from wherever it starts until it encounters "&#60;inet" or the end of
74 the data, and is the kind of pattern that might be used when processing an XML
75 file. Each iteration of the outer parentheses matches either one character that
76 is not "&#60;" or a "&#60;" that is not followed by "inet". However, each time a
77 parenthesis is processed, a recursion occurs, so this formulation uses a stack
78 frame for each matched character. For a long string, a lot of stack is
79 required. Consider now this rewritten pattern, which matches exactly the same
80 strings:
81 <pre>
82 ([^&#60;]++|&#60;(?!inet))+
83 </pre>
84 This uses very much less stack, because runs of characters that do not contain
85 "&#60;" are "swallowed" in one item inside the parentheses. Recursion happens only
86 when a "&#60;" character that is not followed by "inet" is encountered (and we
87 assume this is relatively rare). A possessive quantifier is used to stop any
88 backtracking into the runs of non-"&#60;" characters, but that is not related to
89 stack usage.
90 </P>
91 <P>
92 This example shows that one way of avoiding stack problems when matching long
93 subject strings is to write repeated parenthesized subpatterns to match more
94 than one character whenever possible.
95 </P>
96 <br><b>
97 Compiling PCRE to use heap instead of stack for <b>pcre[16|32]_exec()</b>
98 </b><br>
99 <P>
100 In environments where stack memory is constrained, you might want to compile
101 PCRE to use heap memory instead of stack for remembering back-up points when
102 <b>pcre[16|32]_exec()</b> is running. This makes it run a lot more slowly, however.
103 Details of how to do this are given in the
104 <a href="pcrebuild.html"><b>pcrebuild</b></a>
105 documentation. When built in this way, instead of using the stack, PCRE obtains
106 and frees memory by calling the functions that are pointed to by the
107 <b>pcre[16|32]_stack_malloc</b> and <b>pcre[16|32]_stack_free</b> variables. By
108 default, these point to <b>malloc()</b> and <b>free()</b>, but you can replace
109 the pointers to cause PCRE to use your own functions. Since the block sizes are
110 always the same, and are always freed in reverse order, it may be possible to
111 implement customized memory handlers that are more efficient than the standard
112 functions.
113 </P>
114 <br><b>
115 Limiting <b>pcre[16|32]_exec()</b>'s stack usage
116 </b><br>
117 <P>
118 You can set limits on the number of times that <b>match()</b> is called, both in
119 total and recursively. If a limit is exceeded, <b>pcre[16|32]_exec()</b> returns an
120 error code. Setting suitable limits should prevent it from running out of
121 stack. The default values of the limits are very large, and unlikely ever to
122 operate. They can be changed when PCRE is built, and they can also be set when
123 <b>pcre[16|32]_exec()</b> is called. For details of these interfaces, see the
124 <a href="pcrebuild.html"><b>pcrebuild</b></a>
125 documentation and the
126 <a href="pcreapi.html#extradata">section on extra data for <b>pcre[16|32]_exec()</b></a>
127 in the
128 <a href="pcreapi.html"><b>pcreapi</b></a>
129 documentation.
130 </P>
131 <P>
132 As a very rough rule of thumb, you should reckon on about 500 bytes per
133 recursion. Thus, if you want to limit your stack usage to 8Mb, you should set
134 the limit at 16000 recursions. A 64Mb stack, on the other hand, can support
135 around 128000 recursions.
136 </P>
137 <P>
138 In Unix-like environments, the <b>pcretest</b> test program has a command line
139 option (<b>-S</b>) that can be used to increase the size of its stack. As long
140 as the stack is large enough, another option (<b>-M</b>) can be used to find the
141 smallest limits that allow a particular pattern to match a given subject
142 string. This is done by calling <b>pcre[16|32]_exec()</b> repeatedly with different
143 limits.
144 </P>
145 <br><b>
146 Obtaining an estimate of stack usage
147 </b><br>
148 <P>
149 The actual amount of stack used per recursion can vary quite a lot, depending
150 on the compiler that was used to build PCRE and the optimization or debugging
151 options that were set for it. The rule of thumb value of 500 bytes mentioned
152 above may be larger or smaller than what is actually needed. A better
153 approximation can be obtained by running this command:
154 <pre>
155 pcretest -m -C
156 </pre>
157 The <b>-C</b> option causes <b>pcretest</b> to output information about the
158 options with which PCRE was compiled. When <b>-m</b> is also given (before
159 <b>-C</b>), information about stack use is given in a line like this:
160 <pre>
161 Match recursion uses stack: approximate frame size = 640 bytes
162 </pre>
163 The value is approximate because some recursions need a bit more (up to perhaps
164 16 more bytes).
165 </P>
166 <P>
167 If the above command is given when PCRE is compiled to use the heap instead of
168 the stack for recursion, the value that is output is the size of each block
169 that is obtained from the heap.
170 </P>
171 <br><b>
172 Changing stack size in Unix-like systems
173 </b><br>
174 <P>
175 In Unix-like environments, there is not often a problem with the stack unless
176 very long strings are involved, though the default limit on stack size varies
177 from system to system. Values from 8Mb to 64Mb are common. You can find your
178 default limit by running the command:
179 <pre>
180 ulimit -s
181 </pre>
182 Unfortunately, the effect of running out of stack is often SIGSEGV, though
183 sometimes a more explicit error message is given. You can normally increase the
184 limit on stack size by code such as this:
185 <pre>
186 struct rlimit rlim;
187 getrlimit(RLIMIT_STACK, &rlim);
188 rlim.rlim_cur = 100*1024*1024;
189 setrlimit(RLIMIT_STACK, &rlim);
190 </pre>
191 This reads the current limits (soft and hard) using <b>getrlimit()</b>, then
192 attempts to increase the soft limit to 100Mb using <b>setrlimit()</b>. You must
193 do this before calling <b>pcre[16|32]_exec()</b>.
194 </P>
195 <br><b>
196 Changing stack size in Mac OS X
197 </b><br>
198 <P>
199 Using <b>setrlimit()</b>, as described above, should also work on Mac OS X. It
200 is also possible to set a stack size when linking a program. There is a
201 discussion about stack sizes in Mac OS X at this web site:
202 <a href="http://developer.apple.com/qa/qa2005/qa1419.html">http://developer.apple.com/qa/qa2005/qa1419.html.</a>
203 </P>
204 <br><b>
205 AUTHOR
206 </b><br>
207 <P>
208 Philip Hazel
209 <br>
210 University Computing Service
211 <br>
212 Cambridge CB2 3QH, England.
213 <br>
214 </P>
215 <br><b>
216 REVISION
217 </b><br>
218 <P>
219 Last updated: 24 June 2012
220 <br>
221 Copyright &copy; 1997-2012 University of Cambridge.
222 <br>
223 <p>
224 Return to the <a href="index.html">PCRE index page</a>.
225 </p>

Properties

Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12