1PCRE2JIT(3) Library Functions Manual PCRE2JIT(3)
2
6 PCRE2 - Perl-compatible regular expressions (revised API)
7
9 Just-in-time compiling is a heavyweight optimization that can greatly
11 speed up pattern matching. However, it comes at the cost of extra pro‐
12 cessing before the match is performed, so it is of most benefit when
13 the same pattern is going to be matched many times. This does not nec‐
14 essarily mean many calls of a matching function; if the pattern is not
15 anchored, matching attempts may take place many times at various posi‐
16 tions in the subject, even for a single call. Therefore, if the subject
17 string is very long, it may still pay to use JIT even for one-off
18 matches. JIT support is available for all of the 8-bit, 16-bit and
19 32-bit PCRE2 libraries.
20 JIT support applies only to the traditional Perl-compatible matching
22 function. It does not apply when the DFA matching function is being
23 used. The code for this support was written by Zoltan Herczeg.
24
26 JIT support is an optional feature of PCRE2. The "configure" option
28 --enable-jit (or equivalent CMake option) must be set when PCRE2 is
29 built if you want to use JIT. The support is limited to the following
30 hardware platforms:
31 ARM 32-bit (v5, v7, and Thumb2)
33 ARM 64-bit
34 Intel x86 32-bit and 64-bit
35 MIPS 32-bit and 64-bit
36 Power PC 32-bit and 64-bit
37 SPARC 32-bit
38 If --enable-jit is set on an unsupported platform, compilation fails.
40 A program can tell if JIT support is available by calling pcre2_con‐
42 fig() with the PCRE2_CONFIG_JIT option. The result is 1 when JIT is
43 available, and 0 otherwise. However, a simple program does not need to
44 check this in order to use JIT. The API is implemented in a way that
45 falls back to the interpretive code if JIT is not available. For pro‐
46 grams that need the best possible performance, there is also a "fast
47 path" API that is JIT-specific.
48
50 To make use of the JIT support in the simplest way, all you have to do
52 is to call pcre2_jit_compile() after successfully compiling a pattern
53 with pcre2_compile(). This function has two arguments: the first is the
54 compiled pattern pointer that was returned by pcre2_compile(), and the
55 second is zero or more of the following option bits: PCRE2_JIT_COM‐
56 PLETE, PCRE2_JIT_PARTIAL_HARD, or PCRE2_JIT_PARTIAL_SOFT.
57 If JIT support is not available, a call to pcre2_jit_compile() does
59 nothing and returns PCRE2_ERROR_JIT_BADOPTION. Otherwise, the compiled
60 pattern is passed to the JIT compiler, which turns it into machine code
61 that executes much faster than the normal interpretive code, but yields
62 exactly the same results. The returned value from pcre2_jit_compile()
63 is zero on success, or a negative error code.
64 There is a limit to the size of pattern that JIT supports, imposed by
66 the size of machine stack that it uses. The exact rules are not docu‐
67 mented because they may change at any time, in particular, when new
68 optimizations are introduced. If a pattern is too big, a call to
69 pcre2_jit_compile() returns PCRE2_ERROR_NOMEMORY.
70 PCRE2_JIT_COMPLETE requests the JIT compiler to generate code for com‐
72 plete matches. If you want to run partial matches using the PCRE2_PAR‐
73 TIAL_HARD or PCRE2_PARTIAL_SOFT options of pcre2_match(), you should
74 set one or both of the other options as well as, or instead of
75 PCRE2_JIT_COMPLETE. The JIT compiler generates different optimized code
76 for each of the three modes (normal, soft partial, hard partial). When
77 pcre2_match() is called, the appropriate code is run if it is avail‐
78 able. Otherwise, the pattern is matched using interpretive code.
79 You can call pcre2_jit_compile() multiple times for the same compiled
81 pattern. It does nothing if it has previously compiled code for any of
82 the option bits. For example, you can call it once with PCRE2_JIT_COM‐
83 PLETE and (perhaps later, when you find you need partial matching)
84 again with PCRE2_JIT_COMPLETE and PCRE2_JIT_PARTIAL_HARD. This time it
85 will ignore PCRE2_JIT_COMPLETE and just compile code for partial match‐
86 ing. If pcre2_jit_compile() is called with no option bits set, it imme‐
87 diately returns zero. This is an alternative way of testing whether JIT
88 is available.
89 At present, it is not possible to free JIT compiled code except when
91 the entire compiled pattern is freed by calling pcre2_code_free().
92 In some circumstances you may need to call additional functions. These
94 are described in the section entitled "Controlling the JIT stack"
95 below.
96 There are some pcre2_match() options that are not supported by JIT, and
98 there are also some pattern items that JIT cannot handle. Details are
99 given below. In both cases, matching automatically falls back to the
100 interpretive code. If you want to know whether JIT was actually used
101 for a particular match, you should arrange for a JIT callback function
102 to be set up as described in the section entitled "Controlling the JIT
103 stack" below, even if you do not need to supply a non-default JIT
104 stack. Such a callback function is called whenever JIT code is about to
105 be obeyed. If the match-time options are not right for JIT execution,
106 the callback function is not obeyed.
107 If the JIT compiler finds an unsupported item, no JIT data is gener‐
109 ated. You can find out if JIT matching is available after compiling a
110 pattern by calling pcre2_pattern_info() with the PCRE2_INFO_JITSIZE
111 option. A non-zero result means that JIT compilation was successful. A
112 result of 0 means that JIT support is not available, or the pattern was
113 not processed by pcre2_jit_compile(), or the JIT compiler was not able
114 to handle the pattern.
115
117 When a pattern is compiled with the PCRE2_UTF option, the interpretive
119 matching function expects its subject string to be a valid sequence of
120 UTF code units. If it is not, the result is undefined. This is also
121 true by default of matching via JIT. However, if the option
122 PCRE2_JIT_INVALID_UTF is passed to pcre2_jit_compile(), code that can
123 process a subject containing invalid UTF is compiled.
124 In this mode, an invalid code unit sequence never matches any pattern
126 item. It does not match dot, it does not match \p{Any}, it does not
127 even match negative items such as [^X]. A lookbehind assertion fails if
128 it encounters an invalid sequence while moving the current point back‐
129 wards. In other words, an invalid UTF code unit sequence acts as a bar‐
130 rier which no match can cross. Reaching an invalid sequence causes an
131 immediate backtrack.
132 Using this option, an application can run matches in arbitrary data,
134 knowing that any matched strings that are returned will be valid UTF.
135 This can be useful when searching for text in executable or other
136 binary files.
137
139 The pcre2_match() options that are supported for JIT matching are
141 PCRE2_COPY_MATCHED_SUBJECT, PCRE2_NOTBOL, PCRE2_NOTEOL, PCRE2_NOTEMPTY,
142 PCRE2_NOTEMPTY_ATSTART, PCRE2_NO_UTF_CHECK, PCRE2_PARTIAL_HARD, and
143 PCRE2_PARTIAL_SOFT. The PCRE2_ANCHORED and PCRE2_ENDANCHORED options
144 are not supported at match time.
145 If the PCRE2_NO_JIT option is passed to pcre2_match() it disables the
147 use of JIT, forcing matching by the interpreter code.
148 The only unsupported pattern items are \C (match a single data unit)
150 when running in a UTF mode, and a callout immediately before an asser‐
151 tion condition in a conditional group.
152
154 When a pattern is matched using JIT matching, the return values are the
156 same as those given by the interpretive pcre2_match() code, with the
157 addition of one new error code: PCRE2_ERROR_JIT_STACKLIMIT. This means
158 that the memory used for the JIT stack was insufficient. See "Control‐
159 ling the JIT stack" below for a discussion of JIT stack usage.
160 The error code PCRE2_ERROR_MATCHLIMIT is returned by the JIT code if
162 searching a very large pattern tree goes on for too long, as it is in
163 the same circumstance when JIT is not used, but the details of exactly
164 what is counted are not the same. The PCRE2_ERROR_DEPTHLIMIT error code
165 is never returned when JIT matching is used.
166
168 When the compiled JIT code runs, it needs a block of memory to use as a
170 stack. By default, it uses 32KiB on the machine stack. However, some
171 large or complicated patterns need more than this. The error
172 PCRE2_ERROR_JIT_STACKLIMIT is given when there is not enough stack.
173 Three functions are provided for managing blocks of memory for use as
174 JIT stacks. There is further discussion about the use of JIT stacks in
175 the section entitled "JIT stack FAQ" below.
176 The pcre2_jit_stack_create() function creates a JIT stack. Its argu‐
178 ments are a starting size, a maximum size, and a general context (for
179 memory allocation functions, or NULL for standard memory allocation).
180 It returns a pointer to an opaque structure of type pcre2_jit_stack, or
181 NULL if there is an error. The pcre2_jit_stack_free() function is used
182 to free a stack that is no longer needed. If its argument is NULL, this
183 function returns immediately, without doing anything. (For the techni‐
184 cally minded: the address space is allocated by mmap or VirtualAlloc.)
185 A maximum stack size of 512KiB to 1MiB should be more than enough for
186 any pattern.
187 The pcre2_jit_stack_assign() function specifies which stack JIT code
189 should use. Its arguments are as follows:
190 pcre2_match_context *mcontext
192 pcre2_jit_callback callback
193 void *data
194 The first argument is a pointer to a match context. When this is subse‐
196 quently passed to a matching function, its information determines which
197 JIT stack is used. If this argument is NULL, the function returns imme‐
198 diately, without doing anything. There are three cases for the values
199 of the other two options:
200 (1) If callback is NULL and data is NULL, an internal 32KiB block
202 on the machine stack is used. This is the default when a match
203 context is created.
204 (2) If callback is NULL and data is not NULL, data must be
206 a pointer to a valid JIT stack, the result of calling
207 pcre2_jit_stack_create().
208 (3) If callback is not NULL, it must point to a function that is
210 called with data as an argument at the start of matching, in
211 order to set up a JIT stack. If the return from the callback
212 function is NULL, the internal 32KiB stack is used; otherwise the
213 return value must be a valid JIT stack, the result of calling
214 pcre2_jit_stack_create().
215 A callback function is obeyed whenever JIT code is about to be run; it
217 is not obeyed when pcre2_match() is called with options that are incom‐
218 patible for JIT matching. A callback function can therefore be used to
219 determine whether a match operation was executed by JIT or by the
220 interpreter.
221 You may safely use the same JIT stack for more than one pattern (either
223 by assigning directly or by callback), as long as the patterns are
224 matched sequentially in the same thread. Currently, the only way to set
225 up non-sequential matches in one thread is to use callouts: if a call‐
226 out function starts another match, that match must use a different JIT
227 stack to the one used for currently suspended match(es).
228 In a multithread application, if you do not specify a JIT stack, or if
230 you assign or pass back NULL from a callback, that is thread-safe,
231 because each thread has its own machine stack. However, if you assign
232 or pass back a non-NULL JIT stack, this must be a different stack for
233 each thread so that the application is thread-safe.
234 Strictly speaking, even more is allowed. You can assign the same non-
236 NULL stack to a match context that is used by any number of patterns,
237 as long as they are not used for matching by multiple threads at the
238 same time. For example, you could use the same stack in all compiled
239 patterns, with a global mutex in the callback to wait until the stack
240 is available for use. However, this is an inefficient solution, and not
241 recommended.
242 This is a suggestion for how a multithreaded program that needs to set
244 up non-default JIT stacks might operate:
245 During thread initalization
247 thread_local_var = pcre2_jit_stack_create(...)
248 During thread exit
250 pcre2_jit_stack_free(thread_local_var)
251 Use a one-line callback function
253 return thread_local_var
254 All the functions described in this section do nothing if JIT is not
256 available.
257
259 (1) Why do we need JIT stacks?
261 PCRE2 (and JIT) is a recursive, depth-first engine, so it needs a stack
263 where the local data of the current node is pushed before checking its
264 child nodes. Allocating real machine stack on some platforms is diffi‐
265 cult. For example, the stack chain needs to be updated every time if we
266 extend the stack on PowerPC. Although it is possible, its updating
267 time overhead decreases performance. So we do the recursion in memory.
268 (2) Why don't we simply allocate blocks of memory with malloc()?
270 Modern operating systems have a nice feature: they can reserve an
272 address space instead of allocating memory. We can safely allocate mem‐
273 ory pages inside this address space, so the stack could grow without
274 moving memory data (this is important because of pointers). Thus we can
275 allocate 1MiB address space, and use only a single memory page (usually
276 4KiB) if that is enough. However, we can still grow up to 1MiB anytime
277 if needed.
278 (3) Who "owns" a JIT stack?
280 The owner of the stack is the user program, not the JIT studied pattern
282 or anything else. The user program must ensure that if a stack is being
283 used by pcre2_match(), (that is, it is assigned to a match context that
284 is passed to the pattern currently running), that stack must not be
285 used by any other threads (to avoid overwriting the same memory area).
286 The best practice for multithreaded programs is to allocate a stack for
287 each thread, and return this stack through the JIT callback function.
288 (4) When should a JIT stack be freed?
290 You can free a JIT stack at any time, as long as it will not be used by
292 pcre2_match() again. When you assign the stack to a match context, only
293 a pointer is set. There is no reference counting or any other magic.
294 You can free compiled patterns, contexts, and stacks in any order, any‐
295 time. Just do not call pcre2_match() with a match context pointing to
296 an already freed stack, as that will cause SEGFAULT. (Also, do not free
297 a stack currently used by pcre2_match() in another thread). You can
298 also replace the stack in a context at any time when it is not in use.
299 You should free the previous stack before assigning a replacement.
300 (5) Should I allocate/free a stack every time before/after calling
302 pcre2_match()?
303 No, because this is too costly in terms of resources. However, you
305 could implement some clever idea which release the stack if it is not
306 used in let's say two minutes. The JIT callback can help to achieve
307 this without keeping a list of patterns.
308 (6) OK, the stack is for long term memory allocation. But what happens
310 if a pattern causes stack overflow with a stack of 1MiB? Is that 1MiB
311 kept until the stack is freed?
312 Especially on embedded sytems, it might be a good idea to release mem‐
314 ory sometimes without freeing the stack. There is no API for this at
315 the moment. Probably a function call which returns with the currently
316 allocated memory for any stack and another which allows releasing mem‐
317 ory (shrinking the stack) would be a good idea if someone needs this.
318 (7) This is too much of a headache. Isn't there any better solution for
320 JIT stack handling?
321 No, thanks to Windows. If POSIX threads were used everywhere, we could
323 throw out this complicated API.
324
326 void pcre2_jit_free_unused_memory(pcre2_general_context *gcontext);
328 The JIT executable allocator does not free all memory when it is possi‐
330 ble. It expects new allocations, and keeps some free memory around to
331 improve allocation speed. However, in low memory conditions, it might
332 be better to free all possible memory. You can cause this to happen by
333 calling pcre2_jit_free_unused_memory(). Its argument is a general con‐
334 text, for custom memory management, or NULL for standard memory manage‐
335 ment.
336
338 This is a single-threaded example that specifies a JIT stack without
340 using a callback. A real program should include error checking after
341 all the function calls.
342 int rc;
344 pcre2_code *re;
345 pcre2_match_data *match_data;
346 pcre2_match_context *mcontext;
347 pcre2_jit_stack *jit_stack;
348 re = pcre2_compile(pattern, PCRE2_ZERO_TERMINATED, 0,
350 &errornumber, &erroffset, NULL);
351 rc = pcre2_jit_compile(re, PCRE2_JIT_COMPLETE);
352 mcontext = pcre2_match_context_create(NULL);
353 jit_stack = pcre2_jit_stack_create(32*1024, 512*1024, NULL);
354 pcre2_jit_stack_assign(mcontext, NULL, jit_stack);
355 match_data = pcre2_match_data_create(re, 10);
356 rc = pcre2_match(re, subject, length, 0, 0, match_data, mcontext);
357 /* Process result */
358 pcre2_code_free(re);
360 pcre2_match_data_free(match_data);
361 pcre2_match_context_free(mcontext);
362 pcre2_jit_stack_free(jit_stack);
363
366 Because the API described above falls back to interpreted matching when
368 JIT is not available, it is convenient for programs that are written
369 for general use in many environments. However, calling JIT via
370 pcre2_match() does have a performance impact. Programs that are written
371 for use where JIT is known to be available, and which need the best
372 possible performance, can instead use a "fast path" API to call JIT
373 matching directly instead of calling pcre2_match() (obviously only for
374 patterns that have been successfully processed by pcre2_jit_compile()).
375 The fast path function is called pcre2_jit_match(), and it takes
377 exactly the same arguments as pcre2_match(). However, the subject
378 string must be specified with a length; PCRE2_ZERO_TERMINATED is not
379 supported. Unsupported option bits (for example, PCRE2_ANCHORED,
380 PCRE2_ENDANCHORED and PCRE2_COPY_MATCHED_SUBJECT) are ignored, as is
381 the PCRE2_NO_JIT option. The return values are also the same as for
382 pcre2_match(), plus PCRE2_ERROR_JIT_BADOPTION if a matching mode (par‐
383 tial or complete) is requested that was not compiled.
384 When you call pcre2_match(), as well as testing for invalid options, a
386 number of other sanity checks are performed on the arguments. For exam‐
387 ple, if the subject pointer is NULL, an immediate error is given. Also,
388 unless PCRE2_NO_UTF_CHECK is set, a UTF subject string is tested for
389 validity. In the interests of speed, these checks do not happen on the
390 JIT fast path, and if invalid data is passed, the result is undefined.
391 Bypassing the sanity checks and the pcre2_match() wrapping can give
393 speedups of more than 10%.
394
396 pcre2api(3)
398
400 Philip Hazel (FAQ by Zoltan Herczeg)
402 University Computing Service
403 Cambridge, England.
404
406 Last updated: 06 March 2019
408 Copyright (c) 1997-2019 University of Cambridge.
409PCRE2 10.33 06 March 2019 PCRE2JIT(3)