Squashed commit upgrading to mbedtls-2.16.5

Squash merging branch import/mbedtls-2.16.5

058aefb2bfa4 ("core: mbedtls: use SHA-256 crypto accelerated routines")
bcef9baed8f1 ("core: mbedtls: use SHA-1 crypto accelerated routines")
c9359f31db12 ("core: mbedtls: use AES crypto accelerated routines")
0e6c1e2642c7 ("core: merge tee_*_get_digest_size() into a single function")
0cb3c28a2f4d ("libmbedtls: mbedtls_mpi_exp_mod(): optimize mempool usage")
5abf0e6ab72e ("libmbedtls: mbedtls_mpi_exp_mod(): reduce stack usage")
2ccc08ac7fef ("libmbedtls: preserve mempool usage on reinit")
cd2a24648569 ("libmbedtls: mbedtls_mpi_exp_mod() initialize W")
7727182ecb56 ("libmbedtls: fix no CRT issue")
120737075dcf ("libmbedtls: add interfaces in mbedtls for context memory operation")
1126250b3af8 ("libmbedtls: add missing source file chachapoly.c")
23972e9f1c98 ("libmedtls: mpi_miller_rabin: increase count limit")
1fcbc05b3cd2 ("libmbedtls: add mbedtls_mpi_init_mempool()")
66e03f068078 ("libmbedtls: make mbedtls_mpi_mont*() available")
d07e0ce56236 ("libmbedtls: refine mbedtls license header")
491ee2cd0ff4 ("mbedtls: configure mbedtls to reach for config")
9b6cee685d9a ("mbedtls: remove default include/mbedtls/config.h")
84f7467a0a91 ("Import mbedtls-2.16.5")

Signed-off-by: Jerome Forissier <jerome@forissier.org>
Acked-by: Jens Wiklander <jens.wiklander@linaro.org>
diff --git a/lib/libmbedtls/mbedtls/library/ripemd160.c b/lib/libmbedtls/mbedtls/library/ripemd160.c
index 3a208c9..887ef9f 100644
--- a/lib/libmbedtls/mbedtls/library/ripemd160.c
+++ b/lib/libmbedtls/mbedtls/library/ripemd160.c
@@ -147,22 +147,29 @@
     D = Dp = ctx->state[3];
     E = Ep = ctx->state[4];
 
-#define F1( x, y, z )   ( x ^ y ^ z )
-#define F2( x, y, z )   ( ( x & y ) | ( ~x & z ) )
-#define F3( x, y, z )   ( ( x | ~y ) ^ z )
-#define F4( x, y, z )   ( ( x & z ) | ( y & ~z ) )
-#define F5( x, y, z )   ( x ^ ( y | ~z ) )
+#define F1( x, y, z )   ( (x) ^ (y) ^ (z) )
+#define F2( x, y, z )   ( ( (x) & (y) ) | ( ~(x) & (z) ) )
+#define F3( x, y, z )   ( ( (x) | ~(y) ) ^ (z) )
+#define F4( x, y, z )   ( ( (x) & (z) ) | ( (y) & ~(z) ) )
+#define F5( x, y, z )   ( (x) ^ ( (y) | ~(z) ) )
 
-#define S( x, n ) ( ( x << n ) | ( x >> (32 - n) ) )
+#define S( x, n ) ( ( (x) << (n) ) | ( (x) >> (32 - (n)) ) )
 
-#define P( a, b, c, d, e, r, s, f, k )      \
-    a += f( b, c, d ) + X[r] + k;           \
-    a = S( a, s ) + e;                      \
-    c = S( c, 10 );
+#define P( a, b, c, d, e, r, s, f, k )                \
+    do                                                \
+    {                                                 \
+        (a) += f( (b), (c), (d) ) + X[r] + (k);       \
+        (a) = S( (a), (s) ) + (e);                    \
+        (c) = S( (c), 10 );                           \
+    } while( 0 )
 
-#define P2( a, b, c, d, e, r, s, rp, sp )   \
-    P( a, b, c, d, e, r, s, F, K );         \
-    P( a ## p, b ## p, c ## p, d ## p, e ## p, rp, sp, Fp, Kp );
+#define P2( a, b, c, d, e, r, s, rp, sp )                               \
+    do                                                                  \
+    {                                                                   \
+        P( (a), (b), (c), (d), (e), (r), (s), F, K );                   \
+        P( a ## p, b ## p, c ## p, d ## p, e ## p,                      \
+           (rp), (sp), Fp, Kp );                                        \
+    } while( 0 )
 
 #define F   F1
 #define K   0x00000000