Add ecdsa_{read,write}_signature()
diff --git a/include/polarssl/ecdsa.h b/include/polarssl/ecdsa.h
index bc1afbb..6dcb283 100644
--- a/include/polarssl/ecdsa.h
+++ b/include/polarssl/ecdsa.h
@@ -84,6 +84,48 @@
const ecp_point *Q, const mpi *r, const mpi *s);
/**
+ * \brief Compute ECDSA signature and write it to buffer
+ *
+ * \param ctx ECDSA context
+ * \param hash Message hash
+ * \param hlen Length of hash
+ * \param sig Buffer that will hold the signature
+ * \param slen Length of the signature written
+ * \param f_rng RNG function
+ * \param p_rng RNG parameter
+ *
+ * \note The "sig" buffer must be at least as large as twice the
+ * size of the curve used, plus 7 (eg. 71 bytes if a 256-bit
+ * curve is used).
+ *
+ * \return 0 if successful,
+ * or a POLARSSL_ERR_ECP, POLARSSL_ERR_MPI or
+ * POLARSSL_ERR_ASN1 error code
+ */
+int ecdsa_write_signature( ecdsa_context *ctx,
+ const unsigned char *hash, size_t hlen,
+ unsigned char *sig, size_t *slen,
+ int (*f_rng)(void *, unsigned char *, size_t),
+ void *p_rng );
+
+/**
+ * \brief Read and verify an ECDSA signature
+ *
+ * \param ctx ECDSA context
+ * \param hash Message hash
+ * \param hlen Size of hash
+ * \param sig Signature to read and verify
+ * \param slen Size of sig
+ *
+ * \return 0 if successful,
+ * POLARSSL_ERR_ECP_BAD_INPUT_DATA if signature is invalid
+ * or a POLARSSL_ERR_ECP or POLARSSL_ERR_MPI error code
+ */
+int ecdsa_read_signature( ecdsa_context *ctx,
+ const unsigned char *hash, size_t hlen,
+ const unsigned char *sig, size_t slen );
+
+/**
* \brief Initialize context
*
* \param ctx Context to initialize
diff --git a/include/polarssl/ecp.h b/include/polarssl/ecp.h
index 7bd9bd5..2c0009c 100644
--- a/include/polarssl/ecp.h
+++ b/include/polarssl/ecp.h
@@ -126,9 +126,10 @@
#define POLARSSL_ECP_DP_SECP521R1 25
/**
- * Maximum bit size of the groups (that is, of N)
+ * Maximum size of the groups (that is, of N and P)
*/
-#define POLARSSL_ECP_MAX_N_BITS 521
+#define POLARSSL_ECP_MAX_BITS 521
+#define POLARSSL_ECP_MAX_BYTES ( ( POLARSSL_ECP_MAX_BITS + 7 ) / 8 )
/*
* Maximum window size (actually, NAF width) used for point multipliation.