X-Git-Url: https://git.openssl.org/?a=blobdiff_plain;f=crypto%2Fmodes%2Fgcm128.c;h=fda6f052d2023e80f67cf4afae32767392a9b7a3;hb=e4e5bc39f9481b9c0842ad5b6d0ac338a9742281;hp=e7d173678657f36939cbb8b8a9c15ba5c02526a2;hpb=96a4cf8c292df00f564afaf84256058486e8a1c4;p=openssl.git

diff --git a/crypto/modes/gcm128.c b/crypto/modes/gcm128.c
index e7d1736786..fda6f052d2 100644
--- a/crypto/modes/gcm128.c
+++ b/crypto/modes/gcm128.c
@@ -148,7 +148,6 @@ static void gcm_gmult_8bit(u64 Xi[2], const u128 Htable[256])
 	const u8 *xi = (const u8 *)Xi+15;
 	size_t rem, n = *xi;
 	const union { long one; char little; } is_endian = {1};
-	__fips_constseg
 	static const size_t rem_8bit[256] = {
 		PACK(0x0000), PACK(0x01C2), PACK(0x0384), PACK(0x0246),
 		PACK(0x0708), PACK(0x06CA), PACK(0x048C), PACK(0x054E),
@@ -327,7 +326,6 @@ static void gcm_init_4bit(u128 Htable[16], u64 H[2])
 }
 
 #ifndef GHASH_ASM
-__fips_constseg
 static const size_t rem_4bit[16] = {
 	PACK(0x0000), PACK(0x1C20), PACK(0x3840), PACK(0x2460),
 	PACK(0x7080), PACK(0x6CA0), PACK(0x48C0), PACK(0x54E0),
@@ -463,7 +461,6 @@ static void gcm_ghash_4bit(u64 Xi[2],const u128 Htable[16],
      */ 
     u128 Hshr4[16];	/* Htable shifted right by 4 bits */
     u8   Hshl4[16];	/* Htable shifted left  by 4 bits */
-    __fips_constseg
     static const unsigned short rem_8bit[256] = {
 	0x0000, 0x01C2, 0x0384, 0x0246, 0x0708, 0x06CA, 0x048C, 0x054E,
 	0x0E10, 0x0FD2, 0x0D94, 0x0C56, 0x0918, 0x08DA, 0x0A9C, 0x0B5E,
@@ -645,7 +642,7 @@ static void gcm_gmult_1bit(u64 Xi[2],const u64 H[2])
 
 #endif
 
-#if	TABLE_BITS==4 && defined(GHASH_ASM)
+#if	TABLE_BITS==4 && (defined(GHASH_ASM) || defined(OPENSSL_CPUID_OBJ))
 # if	!defined(I386_ONLY) && \
 	(defined(__i386)	|| defined(__i386__)	|| \
 	 defined(__x86_64)	|| defined(__x86_64__)	|| \
@@ -658,6 +655,16 @@ void gcm_init_clmul(u128 Htable[16],const u64 Xi[2]);
 void gcm_gmult_clmul(u64 Xi[2],const u128 Htable[16]);
 void gcm_ghash_clmul(u64 Xi[2],const u128 Htable[16],const u8 *inp,size_t len);
 
+#if defined(__i386) || defined(__i386__) || defined(_M_IX86)
+# define gcm_init_avx	gcm_init_clmul
+# define gcm_gmult_avx	gcm_gmult_clmul
+# define gcm_ghash_avx	gcm_ghash_clmul
+#else
+void gcm_init_avx(u128 Htable[16],const u64 Xi[2]);
+void gcm_gmult_avx(u64 Xi[2],const u128 Htable[16]);
+void gcm_ghash_avx(u64 Xi[2],const u128 Htable[16],const u8 *inp,size_t len);
+#endif
+
 #  if	defined(__i386) || defined(__i386__) || defined(_M_IX86)
 #   define GHASH_ASM_X86
 void gcm_gmult_4bit_mmx(u64 Xi[2],const u128 Htable[16]);
@@ -666,13 +673,21 @@ void gcm_ghash_4bit_mmx(u64 Xi[2],const u128 Htable[16],const u8 *inp,size_t len
 void gcm_gmult_4bit_x86(u64 Xi[2],const u128 Htable[16]);
 void gcm_ghash_4bit_x86(u64 Xi[2],const u128 Htable[16],const u8 *inp,size_t len);
 #  endif
-# elif defined(__arm__) || defined(__arm)
+# elif defined(__arm__) || defined(__arm) || defined(__aarch64__)
 #  include "arm_arch.h"
 #  if __ARM_ARCH__>=7
 #   define GHASH_ASM_ARM
 #   define GCM_FUNCREF_4BIT
+#   define PMULL_CAPABLE	(OPENSSL_armcap_P & ARMV8_PMULL)
+#   if defined(__arm__) || defined(__arm)
+#    define NEON_CAPABLE	(OPENSSL_armcap_P & ARMV7_NEON)
+#   endif
+void gcm_init_neon(u128 Htable[16],const u64 Xi[2]);
 void gcm_gmult_neon(u64 Xi[2],const u128 Htable[16]);
 void gcm_ghash_neon(u64 Xi[2],const u128 Htable[16],const u8 *inp,size_t len);
+void gcm_init_v8(u128 Htable[16],const u64 Xi[2]);
+void gcm_gmult_v8(u64 Xi[2],const u128 Htable[16]);
+void gcm_ghash_v8(u64 Xi[2],const u128 Htable[16],const u8 *inp,size_t len);
 #  endif
 # elif defined(__sparc__) || defined(__sparc)
 #  include "sparc_arch.h"
@@ -682,6 +697,13 @@ extern unsigned int OPENSSL_sparcv9cap_P[];
 void gcm_init_vis3(u128 Htable[16],const u64 Xi[2]);
 void gcm_gmult_vis3(u64 Xi[2],const u128 Htable[16]);
 void gcm_ghash_vis3(u64 Xi[2],const u128 Htable[16],const u8 *inp,size_t len);
+#elif defined(OPENSSL_CPUID_OBJ) && (defined(__powerpc__) || defined(__ppc__) || defined(_ARCH_PPC))
+#  include "ppc_arch.h"
+#  define GHASH_ASM_PPC
+#  define GCM_FUNCREF_4BIT
+void gcm_init_p8(u128 Htable[16],const u64 Xi[2]);
+void gcm_gmult_p8(u64 Xi[2],const u128 Htable[16]);
+void gcm_ghash_p8(u64 Xi[2],const u128 Htable[16],const u8 *inp,size_t len);
 # endif
 #endif
 
@@ -726,9 +748,15 @@ void CRYPTO_gcm128_init(GCM128_CONTEXT *ctx,void *key,block128_f block)
 #  if	!defined(GHASH_ASM_X86) || defined(OPENSSL_IA32_SSE2)
 	if (OPENSSL_ia32cap_P[0]&(1<<24) &&	/* check FXSR bit */
 	    OPENSSL_ia32cap_P[1]&(1<<1) ) {	/* check PCLMULQDQ bit */
-		gcm_init_clmul(ctx->Htable,ctx->H.u);
-		ctx->gmult = gcm_gmult_clmul;
-		ctx->ghash = gcm_ghash_clmul;
+		if (((OPENSSL_ia32cap_P[1]>>22)&0x41)==0x41) {	/* AVX+MOVBE */
+			gcm_init_avx(ctx->Htable,ctx->H.u);
+			ctx->gmult = gcm_gmult_avx;
+			ctx->ghash = gcm_ghash_avx;
+		} else {
+			gcm_init_clmul(ctx->Htable,ctx->H.u);
+			ctx->gmult = gcm_gmult_clmul;
+			ctx->ghash = gcm_ghash_clmul;
+		}
 		return;
 	}
 #  endif
@@ -750,10 +778,21 @@ void CRYPTO_gcm128_init(GCM128_CONTEXT *ctx,void *key,block128_f block)
 	ctx->ghash = gcm_ghash_4bit;
 #  endif
 # elif	defined(GHASH_ASM_ARM)
-	if (OPENSSL_armcap_P & ARMV7_NEON) {
+#  ifdef PMULL_CAPABLE
+	if (PMULL_CAPABLE) {
+		gcm_init_v8(ctx->Htable,ctx->H.u);
+		ctx->gmult = gcm_gmult_v8;
+		ctx->ghash = gcm_ghash_v8;
+	} else
+#  endif
+#  ifdef NEON_CAPABLE
+	if (NEON_CAPABLE) {
+		gcm_init_neon(ctx->Htable,ctx->H.u);
 		ctx->gmult = gcm_gmult_neon;
 		ctx->ghash = gcm_ghash_neon;
-	} else {
+	} else
+#  endif
+	{
 		gcm_init_4bit(ctx->Htable,ctx->H.u);
 		ctx->gmult = gcm_gmult_4bit;
 		ctx->ghash = gcm_ghash_4bit;
@@ -768,6 +807,16 @@ void CRYPTO_gcm128_init(GCM128_CONTEXT *ctx,void *key,block128_f block)
 		ctx->gmult = gcm_gmult_4bit;
 		ctx->ghash = gcm_ghash_4bit;
 	}
+# elif	defined(GHASH_ASM_PPC)
+	if (OPENSSL_ppccap_P & PPC_CRYPTO207) {
+		gcm_init_p8(ctx->Htable,ctx->H.u);
+		ctx->gmult = gcm_gmult_p8;
+		ctx->ghash = gcm_ghash_p8;
+	} else {
+		gcm_init_4bit(ctx->Htable,ctx->H.u);
+		ctx->gmult = gcm_gmult_4bit;
+		ctx->ghash = gcm_ghash_4bit;
+	}
 # else
 	gcm_init_4bit(ctx->Htable,ctx->H.u);
 # endif
@@ -831,7 +880,11 @@ void CRYPTO_gcm128_setiv(GCM128_CONTEXT *ctx,const unsigned char *iv,size_t len)
 		GCM_MUL(ctx,Yi);
 
 		if (is_endian.little)
+#ifdef BSWAP4
+			ctr = BSWAP4(ctx->Yi.d[3]);
+#else
 			ctr = GETU32(ctx->Yi.c+12);
+#endif
 		else
 			ctr = ctx->Yi.d[3];
 	}
@@ -839,7 +892,11 @@ void CRYPTO_gcm128_setiv(GCM128_CONTEXT *ctx,const unsigned char *iv,size_t len)
 	(*ctx->block)(ctx->Yi.c,ctx->EK0.c,ctx->key);
 	++ctr;
 	if (is_endian.little)
+#ifdef BSWAP4
+		ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 		PUTU32(ctx->Yi.c+12,ctr);
+#endif
 	else
 		ctx->Yi.d[3] = ctr;
 }
@@ -934,7 +991,11 @@ int CRYPTO_gcm128_encrypt(GCM128_CONTEXT *ctx,
 	}
 
 	if (is_endian.little)
+#ifdef BSWAP4
+		ctr = BSWAP4(ctx->Yi.d[3]);
+#else
 		ctr = GETU32(ctx->Yi.c+12);
+#endif
 	else
 		ctr = ctx->Yi.d[3];
 
@@ -968,7 +1029,11 @@ int CRYPTO_gcm128_encrypt(GCM128_CONTEXT *ctx,
 			(*block)(ctx->Yi.c,ctx->EKi.c,key);
 			++ctr;
 			if (is_endian.little)
+#ifdef BSWAP4
+				ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 				PUTU32(ctx->Yi.c+12,ctr);
+#endif
 			else
 				ctx->Yi.d[3] = ctr;
 			for (i=0; i<16/sizeof(size_t); ++i)
@@ -990,7 +1055,11 @@ int CRYPTO_gcm128_encrypt(GCM128_CONTEXT *ctx,
 			(*block)(ctx->Yi.c,ctx->EKi.c,key);
 			++ctr;
 			if (is_endian.little)
+#ifdef BSWAP4
+				ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 				PUTU32(ctx->Yi.c+12,ctr);
+#endif
 			else
 				ctx->Yi.d[3] = ctr;
 			for (i=0; i<16/sizeof(size_t); ++i)
@@ -1009,7 +1078,11 @@ int CRYPTO_gcm128_encrypt(GCM128_CONTEXT *ctx,
 			(*block)(ctx->Yi.c,ctx->EKi.c,key);
 			++ctr;
 			if (is_endian.little)
+#ifdef BSWAP4
+				ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 				PUTU32(ctx->Yi.c+12,ctr);
+#endif
 			else
 				ctx->Yi.d[3] = ctr;
 			for (i=0; i<16/sizeof(size_t); ++i)
@@ -1025,7 +1098,11 @@ int CRYPTO_gcm128_encrypt(GCM128_CONTEXT *ctx,
 			(*block)(ctx->Yi.c,ctx->EKi.c,key);
 			++ctr;
 			if (is_endian.little)
+#ifdef BSWAP4
+				ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 				PUTU32(ctx->Yi.c+12,ctr);
+#endif
 			else
 				ctx->Yi.d[3] = ctr;
 			while (len--) {
@@ -1043,7 +1120,11 @@ int CRYPTO_gcm128_encrypt(GCM128_CONTEXT *ctx,
 			(*block)(ctx->Yi.c,ctx->EKi.c,key);
 			++ctr;
 			if (is_endian.little)
+#ifdef BSWAP4
+				ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 				PUTU32(ctx->Yi.c+12,ctr);
+#endif
 			else
 				ctx->Yi.d[3] = ctr;
 		}
@@ -1087,7 +1168,11 @@ int CRYPTO_gcm128_decrypt(GCM128_CONTEXT *ctx,
 	}
 
 	if (is_endian.little)
+#ifdef BSWAP4
+		ctr = BSWAP4(ctx->Yi.d[3]);
+#else
 		ctr = GETU32(ctx->Yi.c+12);
+#endif
 	else
 		ctr = ctx->Yi.d[3];
 
@@ -1124,7 +1209,11 @@ int CRYPTO_gcm128_decrypt(GCM128_CONTEXT *ctx,
 			(*block)(ctx->Yi.c,ctx->EKi.c,key);
 			++ctr;
 			if (is_endian.little)
+#ifdef BSWAP4
+				ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 				PUTU32(ctx->Yi.c+12,ctr);
+#endif
 			else
 				ctx->Yi.d[3] = ctr;
 			for (i=0; i<16/sizeof(size_t); ++i)
@@ -1144,7 +1233,11 @@ int CRYPTO_gcm128_decrypt(GCM128_CONTEXT *ctx,
 			(*block)(ctx->Yi.c,ctx->EKi.c,key);
 			++ctr;
 			if (is_endian.little)
+#ifdef BSWAP4
+				ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 				PUTU32(ctx->Yi.c+12,ctr);
+#endif
 			else
 				ctx->Yi.d[3] = ctr;
 			for (i=0; i<16/sizeof(size_t); ++i)
@@ -1162,7 +1255,11 @@ int CRYPTO_gcm128_decrypt(GCM128_CONTEXT *ctx,
 			(*block)(ctx->Yi.c,ctx->EKi.c,key);
 			++ctr;
 			if (is_endian.little)
+#ifdef BSWAP4
+				ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 				PUTU32(ctx->Yi.c+12,ctr);
+#endif
 			else
 				ctx->Yi.d[3] = ctr;
 			for (i=0; i<16/sizeof(size_t); ++i) {
@@ -1180,7 +1277,11 @@ int CRYPTO_gcm128_decrypt(GCM128_CONTEXT *ctx,
 			(*block)(ctx->Yi.c,ctx->EKi.c,key);
 			++ctr;
 			if (is_endian.little)
+#ifdef BSWAP4
+				ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 				PUTU32(ctx->Yi.c+12,ctr);
+#endif
 			else
 				ctx->Yi.d[3] = ctr;
 			while (len--) {
@@ -1201,7 +1302,11 @@ int CRYPTO_gcm128_decrypt(GCM128_CONTEXT *ctx,
 			(*block)(ctx->Yi.c,ctx->EKi.c,key);
 			++ctr;
 			if (is_endian.little)
+#ifdef BSWAP4
+				ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 				PUTU32(ctx->Yi.c+12,ctr);
+#endif
 			else
 				ctx->Yi.d[3] = ctr;
 		}
@@ -1246,7 +1351,11 @@ int CRYPTO_gcm128_encrypt_ctr32(GCM128_CONTEXT *ctx,
 	}
 
 	if (is_endian.little)
+#ifdef BSWAP4
+		ctr = BSWAP4(ctx->Yi.d[3]);
+#else
 		ctr = GETU32(ctx->Yi.c+12);
+#endif
 	else
 		ctr = ctx->Yi.d[3];
 
@@ -1268,7 +1377,11 @@ int CRYPTO_gcm128_encrypt_ctr32(GCM128_CONTEXT *ctx,
 		(*stream)(in,out,GHASH_CHUNK/16,key,ctx->Yi.c);
 		ctr += GHASH_CHUNK/16;
 		if (is_endian.little)
+#ifdef BSWAP4
+			ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 			PUTU32(ctx->Yi.c+12,ctr);
+#endif
 		else
 			ctx->Yi.d[3] = ctr;
 		GHASH(ctx,out,GHASH_CHUNK);
@@ -1283,7 +1396,11 @@ int CRYPTO_gcm128_encrypt_ctr32(GCM128_CONTEXT *ctx,
 		(*stream)(in,out,j,key,ctx->Yi.c);
 		ctr += (unsigned int)j;
 		if (is_endian.little)
+#ifdef BSWAP4
+			ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 			PUTU32(ctx->Yi.c+12,ctr);
+#endif
 		else
 			ctx->Yi.d[3] = ctr;
 		in  += i;
@@ -1303,7 +1420,11 @@ int CRYPTO_gcm128_encrypt_ctr32(GCM128_CONTEXT *ctx,
 		(*ctx->block)(ctx->Yi.c,ctx->EKi.c,key);
 		++ctr;
 		if (is_endian.little)
+#ifdef BSWAP4
+			ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 			PUTU32(ctx->Yi.c+12,ctr);
+#endif
 		else
 			ctx->Yi.d[3] = ctr;
 		while (len--) {
@@ -1345,7 +1466,11 @@ int CRYPTO_gcm128_decrypt_ctr32(GCM128_CONTEXT *ctx,
 	}
 
 	if (is_endian.little)
+#ifdef BSWAP4
+		ctr = BSWAP4(ctx->Yi.d[3]);
+#else
 		ctr = GETU32(ctx->Yi.c+12);
+#endif
 	else
 		ctr = ctx->Yi.d[3];
 
@@ -1370,7 +1495,11 @@ int CRYPTO_gcm128_decrypt_ctr32(GCM128_CONTEXT *ctx,
 		(*stream)(in,out,GHASH_CHUNK/16,key,ctx->Yi.c);
 		ctr += GHASH_CHUNK/16;
 		if (is_endian.little)
+#ifdef BSWAP4
+			ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 			PUTU32(ctx->Yi.c+12,ctr);
+#endif
 		else
 			ctx->Yi.d[3] = ctr;
 		out += GHASH_CHUNK;
@@ -1396,7 +1525,11 @@ int CRYPTO_gcm128_decrypt_ctr32(GCM128_CONTEXT *ctx,
 		(*stream)(in,out,j,key,ctx->Yi.c);
 		ctr += (unsigned int)j;
 		if (is_endian.little)
+#ifdef BSWAP4
+			ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 			PUTU32(ctx->Yi.c+12,ctr);
+#endif
 		else
 			ctx->Yi.d[3] = ctr;
 		out += i;
@@ -1407,7 +1540,11 @@ int CRYPTO_gcm128_decrypt_ctr32(GCM128_CONTEXT *ctx,
 		(*ctx->block)(ctx->Yi.c,ctx->EKi.c,key);
 		++ctr;
 		if (is_endian.little)
+#ifdef BSWAP4
+			ctx->Yi.d[3] = BSWAP4(ctr);
+#else
 			PUTU32(ctx->Yi.c+12,ctr);
+#endif
 		else
 			ctx->Yi.d[3] = ctr;
 		while (len--) {
@@ -1703,6 +1840,46 @@ static const u8	IV18[]={0x93,0x13,0x22,0x5d,0xf8,0x84,0x06,0xe5,0x55,0x90,0x9c,0
 			0xa2,0x41,0x89,0x97,0x20,0x0e,0xf8,0x2e,0x44,0xae,0x7e,0x3f},
 		T18[]= {0xa4,0x4a,0x82,0x66,0xee,0x1c,0x8e,0xb0,0xc8,0xb5,0xd4,0xcf,0x5a,0xe9,0xf1,0x9a};
 
+/* Test Case 19 */
+#define K19 K1
+#define P19 P1
+#define IV19 IV1
+#define C19 C1
+static const u8 A19[]= {0xd9,0x31,0x32,0x25,0xf8,0x84,0x06,0xe5,0xa5,0x59,0x09,0xc5,0xaf,0xf5,0x26,0x9a,
+			0x86,0xa7,0xa9,0x53,0x15,0x34,0xf7,0xda,0x2e,0x4c,0x30,0x3d,0x8a,0x31,0x8a,0x72,
+			0x1c,0x3c,0x0c,0x95,0x95,0x68,0x09,0x53,0x2f,0xcf,0x0e,0x24,0x49,0xa6,0xb5,0x25,
+			0xb1,0x6a,0xed,0xf5,0xaa,0x0d,0xe6,0x57,0xba,0x63,0x7b,0x39,0x1a,0xaf,0xd2,0x55,
+			0x52,0x2d,0xc1,0xf0,0x99,0x56,0x7d,0x07,0xf4,0x7f,0x37,0xa3,0x2a,0x84,0x42,0x7d,
+			0x64,0x3a,0x8c,0xdc,0xbf,0xe5,0xc0,0xc9,0x75,0x98,0xa2,0xbd,0x25,0x55,0xd1,0xaa,
+			0x8c,0xb0,0x8e,0x48,0x59,0x0d,0xbb,0x3d,0xa7,0xb0,0x8b,0x10,0x56,0x82,0x88,0x38,
+			0xc5,0xf6,0x1e,0x63,0x93,0xba,0x7a,0x0a,0xbc,0xc9,0xf6,0x62,0x89,0x80,0x15,0xad},
+		T19[]= {0x5f,0xea,0x79,0x3a,0x2d,0x6f,0x97,0x4d,0x37,0xe6,0x8e,0x0c,0xb8,0xff,0x94,0x92};
+
+/* Test Case 20 */
+#define K20 K1
+#define A20 A1
+static const u8 IV20[64]={0xff,0xff,0xff,0xff},	/* this results in 0xff in counter LSB */
+		P20[288],
+		C20[]= {0x56,0xb3,0x37,0x3c,0xa9,0xef,0x6e,0x4a,0x2b,0x64,0xfe,0x1e,0x9a,0x17,0xb6,0x14,
+			0x25,0xf1,0x0d,0x47,0xa7,0x5a,0x5f,0xce,0x13,0xef,0xc6,0xbc,0x78,0x4a,0xf2,0x4f,
+			0x41,0x41,0xbd,0xd4,0x8c,0xf7,0xc7,0x70,0x88,0x7a,0xfd,0x57,0x3c,0xca,0x54,0x18,
+			0xa9,0xae,0xff,0xcd,0x7c,0x5c,0xed,0xdf,0xc6,0xa7,0x83,0x97,0xb9,0xa8,0x5b,0x49,
+			0x9d,0xa5,0x58,0x25,0x72,0x67,0xca,0xab,0x2a,0xd0,0xb2,0x3c,0xa4,0x76,0xa5,0x3c,
+			0xb1,0x7f,0xb4,0x1c,0x4b,0x8b,0x47,0x5c,0xb4,0xf3,0xf7,0x16,0x50,0x94,0xc2,0x29,
+			0xc9,0xe8,0xc4,0xdc,0x0a,0x2a,0x5f,0xf1,0x90,0x3e,0x50,0x15,0x11,0x22,0x13,0x76,
+			0xa1,0xcd,0xb8,0x36,0x4c,0x50,0x61,0xa2,0x0c,0xae,0x74,0xbc,0x4a,0xcd,0x76,0xce,
+			0xb0,0xab,0xc9,0xfd,0x32,0x17,0xef,0x9f,0x8c,0x90,0xbe,0x40,0x2d,0xdf,0x6d,0x86,
+			0x97,0xf4,0xf8,0x80,0xdf,0xf1,0x5b,0xfb,0x7a,0x6b,0x28,0x24,0x1e,0xc8,0xfe,0x18,
+			0x3c,0x2d,0x59,0xe3,0xf9,0xdf,0xff,0x65,0x3c,0x71,0x26,0xf0,0xac,0xb9,0xe6,0x42,
+			0x11,0xf4,0x2b,0xae,0x12,0xaf,0x46,0x2b,0x10,0x70,0xbe,0xf1,0xab,0x5e,0x36,0x06,
+			0x87,0x2c,0xa1,0x0d,0xee,0x15,0xb3,0x24,0x9b,0x1a,0x1b,0x95,0x8f,0x23,0x13,0x4c,
+			0x4b,0xcc,0xb7,0xd0,0x32,0x00,0xbc,0xe4,0x20,0xa2,0xf8,0xeb,0x66,0xdc,0xf3,0x64,
+			0x4d,0x14,0x23,0xc1,0xb5,0x69,0x90,0x03,0xc1,0x3e,0xce,0xf4,0xbf,0x38,0xa3,0xb6,
+			0x0e,0xed,0xc3,0x40,0x33,0xba,0xc1,0x90,0x27,0x83,0xdc,0x6d,0x89,0xe2,0xe7,0x74,
+			0x18,0x8a,0x43,0x9c,0x7e,0xbc,0xc0,0x67,0x2d,0xbd,0xa4,0xdd,0xcf,0xb2,0x79,0x46,
+			0x13,0xb0,0xbe,0x41,0x31,0x5e,0xf7,0x78,0x70,0x8a,0x70,0xee,0x7d,0x75,0x16,0x5c},
+		T20[]= {0x8b,0x30,0x7f,0x6b,0x33,0x28,0x6d,0x0a,0xb0,0x26,0xa9,0xed,0x3f,0xe1,0xe8,0x5f};
+
 #define TEST_CASE(n)	do {					\
 	u8 out[sizeof(P##n)];					\
 	AES_set_encrypt_key(K##n,sizeof(K##n)*8,&key);		\
@@ -1747,6 +1924,8 @@ int main()
 	TEST_CASE(16);
 	TEST_CASE(17);
 	TEST_CASE(18);
+	TEST_CASE(19);
+	TEST_CASE(20);
 
 #ifdef OPENSSL_CPUID_OBJ
 	{