MEDIUM: regex: Use PCRE JIT in acl

This is a patch for using PCRE JIT in acl.

I notice regex are used in other places, but they are more complicated
to modify to use PCRE APIs. So I focused to acl in the first try.

BTW, I made a simple benchmark program for PCRE JIT beforehand.
https://github.com/hnakamur/pcre-jit-benchmark

I read the manual for PCRE JIT
http://www.manpagez.com/man/3/pcrejit/

and wrote my benchmark program.
https://github.com/hnakamur/pcre-jit-benchmark/blob/master/test-pcre.c

diff --git a/Makefile b/Makefile
index a78e9d99ab9428bd530d5bd5fbd6b39077d878bf..daa64ee54d5c3dd1482d3fe5583acd41af29a1ad 100644 (file)
--- a/Makefile
+++ b/Makefile
@@ -14,6 +14,7 @@
 #   USE_MY_SPLICE        : redefine the splice syscall if build fails without.
 #   USE_NETFILTER        : enable netfilter on Linux. Automatic.
 #   USE_PCRE             : enable use of libpcre for regex. Recommended.
+#   USE_PCRE_JIT         : enable use of libpcre jit for regex. Recommended.
 #   USE_POLL             : enable poll(). Automatic.
 #   USE_PRIVATE_CACHE    : disable shared memory cache of ssl sessions.
 #   USE_REGPARM          : enable regparm optimization. Recommended on x86.
@@ -521,7 +522,7 @@ endif
 endif
 endif
 
-ifneq ($(USE_PCRE)$(USE_STATIC_PCRE),)
+ifneq ($(USE_PCRE)$(USE_STATIC_PCRE)$(USE_PCRE_JIT),)
 # PCREDIR is used to automatically construct the PCRE_INC and PCRE_LIB paths,
 # by appending /include and /lib respectively. If your system does not use the
 # same sub-directories, simply force these variables instead of PCREDIR. It is
@@ -546,6 +547,10 @@ OPTIONS_CFLAGS  += -DUSE_PCRE $(if $(PCRE_INC),-I$(PCRE_INC))
 OPTIONS_LDFLAGS += $(if $(PCRE_LIB),-L$(PCRE_LIB)) -Wl,-Bstatic -lpcreposix -lpcre -Wl,-Bdynamic
 BUILD_OPTIONS   += $(call ignore_implicit,USE_STATIC_PCRE)
 endif
+# JIT PCRE
+ifneq ($(USE_PCRE_JIT),)
+OPTIONS_CFLAGS  += -DUSE_PCRE_JIT
+endif
 endif
 
 # This one can be changed to look for ebtree files in an external directory

diff --git a/include/common/regex.h b/include/common/regex.h
index 60c7f422db5479029d58247dec6b1809beddc846..bab1a55cc91a2ae8dfeddf98072fb2cd552c835b 100644 (file)
--- a/include/common/regex.h
+++ b/include/common/regex.h
@@ -27,8 +27,20 @@
 #ifdef USE_PCRE
 #include <pcre.h>
 #include <pcreposix.h>
-#else
+
+#ifdef USE_PCRE_JIT
+struct jit_regex {
+    pcre *reg;
+    pcre_extra *extra;
+};
+typedef struct jit_regex regex;
+#else /* no PCRE_JIT */
+typedef regex_t regex;
+#endif
+
+#else /* no PCRE */
 #include <regex.h>
+typedef regex_t regex;
 #endif
 
 /* what to do when a header matches a regex */
@@ -55,6 +67,24 @@ const char *check_replace_string(const char *str);
 const char *chain_regex(struct hdr_exp **head, const regex_t *preg,
                        int action, const char *replace, void *cond);
 
+static inline int regex_exec(const regex *preg, const char *subject, int length) {
+#ifdef USE_PCRE_JIT
+       return pcre_exec(preg->reg, preg->extra, subject, length, 0, 0, NULL, 0);
+#else
+       return regexec(preg, subject, 0, NULL, 0);
+#endif
+}
+
+static inline void regex_free(regex *preg) {
+#ifdef USE_PCRE_JIT
+       pcre_free_study(preg->extra);
+       pcre_free(preg->reg);
+       free(preg);
+#else
+       regfree(preg);
+#endif
+}
+
 #endif /* _COMMON_REGEX_H */
 
 /*

diff --git a/include/types/acl.h b/include/types/acl.h
index bf5537f93fef6822a385ffaeadd5c1bbc419fd59..2b0de0c21c34487024f5828cae70adf7634745fb 100644 (file)
--- a/include/types/acl.h
+++ b/include/types/acl.h
@@ -213,7 +213,7 @@ struct acl_pattern {
        union {
                void *ptr;              /* any data */
                char *str;              /* any string  */
-               regex_t *reg;           /* a compiled regex */
+               regex *reg;             /* a compiled regex */
        } ptr;                          /* indirect values, allocated */
        void(*freeptrbuf)(void *ptr);   /* a destructor able to free objects from the ptr */
        int len;                        /* data length when required  */

diff --git a/src/acl.c b/src/acl.c
index 753f109c9822f2699a29c1ce904b08949aba16ee..a0b16b14e85235281b6140c19579c226e71c3618 100644 (file)
--- a/src/acl.c
+++ b/src/acl.c
@@ -533,7 +533,7 @@ int acl_match_reg(struct sample *smp, struct acl_pattern *pattern)
        old_char = smp->data.str.str[smp->data.str.len];
        smp->data.str.str[smp->data.str.len] = 0;
 
-       if (regexec(pattern->ptr.reg, smp->data.str.str, 0, NULL, 0) == 0)
+       if (regex_exec(pattern->ptr.reg, smp->data.str.str, smp->data.str.len) == 0)
                ret = ACL_PAT_PASS;
        else
                ret = ACL_PAT_FAIL;
@@ -900,28 +900,47 @@ acl_parse_strcat(const char **text, struct acl_pattern *pattern, int *opaque, ch
 /* Free data allocated by acl_parse_reg */
 static void acl_free_reg(void *ptr)
 {
-       regfree((regex_t *)ptr);
+       regex_free(ptr);
 }
 
 /* Parse a regex. It is allocated. */
 int acl_parse_reg(const char **text, struct acl_pattern *pattern, int *opaque, char **err)
 {
-       regex_t *preg;
+       regex *preg;
        int icase;
 
-       preg = calloc(1, sizeof(regex_t));
+       preg = calloc(1, sizeof(*preg));
 
        if (!preg) {
                memprintf(err, "out of memory while loading pattern");
                return 0;
        }
 
+#ifdef USE_PCRE_JIT
+       icase = (pattern->flags & ACL_PAT_F_IGNORE_CASE) ? PCRE_CASELESS : 0;
+       preg->reg = pcre_compile(*text, PCRE_NO_AUTO_CAPTURE | icase, NULL, NULL,
+               NULL);
+       if (!preg->reg) {
+               free(preg);
+               memprintf(err, "regex '%s' is invalid", *text);
+               return 0;
+       }
+
+       preg->extra = pcre_study(preg->reg, PCRE_STUDY_JIT_COMPILE, NULL);
+       if (!preg->extra) {
+               pcre_free(preg->reg);
+               free(preg);
+               memprintf(err, "failed to compile regex '%s'", *text);
+               return 0;
+       }
+#else
        icase = (pattern->flags & ACL_PAT_F_IGNORE_CASE) ? REG_ICASE : 0;
        if (regcomp(preg, *text, REG_EXTENDED | REG_NOSUB | icase) != 0) {
                free(preg);
                memprintf(err, "regex '%s' is invalid", *text);
                return 0;
        }
+#endif
 
        pattern->ptr.reg = preg;
        pattern->freeptrbuf = &acl_free_reg;