Fix O(n^2) performance when parsing a big pile of extrainfos

We were doing an O(n) strlen in router_get_extrainfo_hash() for
every one we tried to parse.  Instead, have
router_get_extrainfo_hash() take the length of the extrainfo as an
argument, so that when it's called from
extrainfo_parse_from_string(), it doesn't do a strlen() over the
whole pile of extrainfos.

diff --git a/changes/bug5828 b/changes/bug5828
new file mode 100644 (file)
index 0000000..3a1734d
--- /dev/null
+++ b/changes/bug5828
@@ -0,0 +1,3 @@
+  o Minor bugfixes (performance):
+    - Avoid O(n^2) performance characteristics when parsing a large
+      extrainfo cache. Fixes bug 5828; bugfix on 0.2.0.1-alpha.

diff --git a/src/or/router.c b/src/or/router.c
index c51bb5dc4db2067ce8a93c5620e95a2b3ca0c8a0..35955f3ee02585b12d283dab55e06ab173df6c95 100644 (file)
--- a/src/or/router.c
+++ b/src/or/router.c
@@ -1655,6 +1655,7 @@ router_rebuild_descriptor(int force)
     ei->cache_info.signed_descriptor_len =
       strlen(ei->cache_info.signed_descriptor_body);
     router_get_extrainfo_hash(ei->cache_info.signed_descriptor_body,
+                              ei->cache_info.signed_descriptor_len,
                               ei->cache_info.signed_descriptor_digest);
   }
 
@@ -2359,7 +2360,7 @@ extrainfo_dump_to_string(char **s_out, extrainfo_t *extrainfo,
   }
 
   memset(sig, 0, sizeof(sig));
-  if (router_get_extrainfo_hash(s, digest) < 0 ||
+  if (router_get_extrainfo_hash(s, strlen(s), digest) < 0 ||
       router_append_dirobj_signature(sig, sizeof(sig), digest, DIGEST_LEN,
                                      ident_key) < 0) {
     log_warn(LD_BUG, "Could not append signature to extra-info "

diff --git a/src/or/routerparse.c b/src/or/routerparse.c
index c5ce174911ce8d30f158e81b229f3f4b91e7f52d..781c57897d7b5c18630b521a5d676503d064bdef 100644 (file)
--- a/src/or/routerparse.c
+++ b/src/or/routerparse.c
@@ -682,12 +682,12 @@ router_get_networkstatus_v3_hash(const char *s, char *digest,
                               ' ', alg);
 }
 
-/** Set <b>digest</b> to the SHA-1 digest of the hash of the extrainfo
- * string in <b>s</b>.  Return 0 on success, -1 on failure. */
+/** Set <b>digest</b> to the SHA-1 digest of the hash of the <b>s_len</b>-byte
+ * extrainfo string at <b>s</b>.  Return 0 on success, -1 on failure. */
 int
-router_get_extrainfo_hash(const char *s, char *digest)
+router_get_extrainfo_hash(const char *s, size_t s_len, char *digest)
 {
-  return router_get_hash_impl(s, strlen(s), digest, "extra-info",
+  return router_get_hash_impl(s, s_len, digest, "extra-info",
                               "\nrouter-signature",'\n', DIGEST_SHA1);
 }
 
@@ -1643,7 +1643,7 @@ extrainfo_parse_entry_from_string(const char *s, const char *end,
   while (end > s+2 && *(end-1) == '\n' && *(end-2) == '\n')
     --end;
 
-  if (router_get_extrainfo_hash(s, digest) < 0) {
+  if (router_get_extrainfo_hash(s, end-s, digest) < 0) {
     log_warn(LD_DIR, "Couldn't compute router hash.");
     goto err;
   }

diff --git a/src/or/routerparse.h b/src/or/routerparse.h
index b274d2196cb738bdd58e51089beee4abb596b712..6486a09c73adf7f07514d758ad8ae05e59a0148d 100644 (file)
--- a/src/or/routerparse.h
+++ b/src/or/routerparse.h
@@ -19,7 +19,7 @@ int router_get_networkstatus_v2_hash(const char *s, char *digest);
 int router_get_networkstatus_v3_hash(const char *s, char *digest,
                                      digest_algorithm_t algorithm);
 int router_get_networkstatus_v3_hashes(const char *s, digests_t *digests);
-int router_get_extrainfo_hash(const char *s, char *digest);
+int router_get_extrainfo_hash(const char *s, size_t s_len, char *digest);
 #define DIROBJ_MAX_SIG_LEN 256
 int router_append_dirobj_signature(char *buf, size_t buf_len,
                                    const char *digest,