!129 fix-CVE-2023-3817

Merge pull request !129 from code4lala/fix-CVE-2023-3817

CHANGES.md

@@ -28,7 +28,24 @@ breaking changes, and mappings for the large list of deprecated functions.
 
 [Migration guide]: https://github.com/openssl/openssl/tree/master/doc/man7/migration_guide.pod
 
- * Fix DH_check() excessive time with over sized modulus
+ * Fix excessive time spent checking DH q parameter value.
+
+   The function DH_check() performs various checks on DH parameters. After
+   fixing CVE-2023-3446 it was discovered that a large q parameter value can
+   also trigger an overly long computation during some of these checks.
+   A correct q value, if present, cannot be larger than the modulus p
+   parameter, thus it is unnecessary to perform these checks if q is larger
+   than p.
+
+   If DH_check() is called with such q parameter value,
+   DH_CHECK_INVALID_Q_VALUE return flag is set and the computationally
+   intensive checks are skipped.
+
+   ([CVE-2023-3817])
+
+   *Tomáš Mráz*
+
+ * Fix DH_check() excessive time with over sized modulus.
 
    The function DH_check() performs various checks on DH parameters. One of
    those checks confirms that the modulus ("p" parameter) is not too large.
@@ -65,7 +82,7 @@ breaking changes, and mappings for the large list of deprecated functions.
    has to skip calls to `EVP_DecryptUpdate()` for empty associated data
    entries.
 
-   *Tomas Mraz*
+   *Tomáš Mráz*
 
  * Mitigate for the time it takes for `OBJ_obj2txt` to translate gigantic
    OBJECT IDENTIFIER sub-identifiers to canonical numeric text form.
@@ -19514,6 +19531,7 @@ ndif
 
 <!-- Links -->
 
+[CVE-2023-3817]: https://www.openssl.org/news/vulnerabilities.html#CVE-2023-3817
 [CVE-2023-3446]: https://www.openssl.org/news/vulnerabilities.html#CVE-2023-3446
 [CVE-2023-2975]: https://www.openssl.org/news/vulnerabilities.html#CVE-2023-2975
 [CVE-2023-2650]: https://www.openssl.org/news/vulnerabilities.html#CVE-2023-2650

NEWS.md

@@ -17,6 +17,7 @@ OpenSSL Releases
 
 OpenSSL 3.0
 -----------
+  * Fix excessive time spent checking DH q parameter value ([CVE-2023-3817])
   * Fix DH_check() excessive time with over sized modulus ([CVE-2023-3446])
   * Do not ignore empty associated data entries with AES-SIV ([CVE-2023-2975])
   * Mitigate for very slow `OBJ_obj2txt()` performance with gigantic OBJECT
@@ -1426,6 +1427,7 @@ OpenSSL 0.9.x
 
 <!-- Links -->
 
+[CVE-2023-3817]: https://www.openssl.org/news/vulnerabilities.html#CVE-2023-3817
 [CVE-2023-3446]: https://www.openssl.org/news/vulnerabilities.html#CVE-2023-3446
 [CVE-2023-2975]: https://www.openssl.org/news/vulnerabilities.html#CVE-2023-2975
 [CVE-2023-2650]: https://www.openssl.org/news/vulnerabilities.html#CVE-2023-2650

crypto/dh/dh_check.c

@@ -143,7 +143,7 @@ int DH_check(const DH *dh, int *ret)
 #ifdef FIPS_MODULE
     return DH_check_params(dh, ret);
 #else
-    int ok = 0, r;
+    int ok = 0, r, q_good = 0;
     BN_CTX *ctx = NULL;
     BIGNUM *t1 = NULL, *t2 = NULL;
     int nid = DH_get_nid((DH *)dh);
@@ -171,6 +171,13 @@ int DH_check(const DH *dh, int *ret)
         goto err;
 
     if (dh->params.q != NULL) {
+        if (BN_ucmp(dh->params.p, dh->params.q) > 0)
+            q_good = 1;
+        else
+            *ret |= DH_CHECK_INVALID_Q_VALUE;
+    }
+
+    if (q_good) {
         if (BN_cmp(dh->params.g, BN_value_one()) <= 0)
             *ret |= DH_NOT_SUITABLE_GENERATOR;
         else if (BN_cmp(dh->params.g, dh->params.p) >= 0)