fix ubsan warning about shifting signed numbers

diff --git a/src/protocols/der/encode.c b/src/protocols/der/encode.c
index 9107d47624d847001aa6f0ecf55a482f89286c05..6e394a3e018e373af082b6b8eed2ffcb7df4607b 100644 (file)
--- a/src/protocols/der/encode.c
+++ b/src/protocols/der/encode.c
@@ -161,7 +161,7 @@ static ssize_t fr_der_encode_integer(fr_dbuff_t *dbuff, fr_dcursor_t *cursor, UN
 {
        fr_dbuff_t       our_dbuff = FR_DBUFF(dbuff);
        fr_pair_t const *vp;
-       int64_t          value;
+       uint64_t         value;
        uint8_t          first_octet = 0;
        size_t           i, len;
 
@@ -183,7 +183,13 @@ static ssize_t fr_der_encode_integer(fr_dbuff_t *dbuff, fr_dcursor_t *cursor, UN
         *            second octet, followed by bits 8 to 1 of each octet in turn up to and including the last octet of
         *            the contents octets.
         */
-       value = vp->vp_int64;
+
+       /*
+        *      Yes, the type is FR_TYPE_INT64.  But we encode the
+        *      data as-is, without caring about things like signed
+        *      math.
+        */
+       value = vp->vp_uint64;
 
        for (i = 0, len = 0; i < sizeof(value); i++) {
                uint8_t byte = (value >> 56) & 0xff;