aboutsummaryrefslogtreecommitdiffstatshomepage
diff options
context:
space:
mode:
-rw-r--r--examples/aes128cbc_example.c25
-rw-r--r--examples/aes128ecb_example.c25
-rw-r--r--examples/aes192cbc_example.c25
-rw-r--r--examples/aes192ecb_example.c25
-rw-r--r--examples/aes256cbc_example.c25
-rw-r--r--examples/aes256ecb_example.c25
-rw-r--r--include/aesni/api.h94
-rw-r--r--include/aesni/data.h19
-rw-r--r--include/aesni/raw.h53
-rw-r--r--src/aes128.asm209
-rw-r--r--src/aes192.asm261
-rw-r--r--src/aes256.asm302
-rw-r--r--test/aes128cbc_decrypt_block.c6
-rw-r--r--test/aes128cbc_encrypt_block.c5
-rw-r--r--test/aes128ecb_decrypt_block.c6
-rw-r--r--test/aes128ecb_encrypt_block.c5
-rw-r--r--test/aes192cbc_decrypt_block.c6
-rw-r--r--test/aes192cbc_encrypt_block.c5
-rw-r--r--test/aes192ecb_decrypt_block.c6
-rw-r--r--test/aes192ecb_encrypt_block.c5
-rw-r--r--test/aes256cbc_decrypt_block.c6
-rw-r--r--test/aes256cbc_encrypt_block.c5
-rw-r--r--test/aes256ecb_decrypt_block.c6
-rw-r--r--test/aes256ecb_encrypt_block.c5
24 files changed, 679 insertions, 475 deletions
diff --git a/examples/aes128cbc_example.c b/examples/aes128cbc_example.c
index ce568ea..94fb0d4 100644
--- a/examples/aes128cbc_example.c
+++ b/examples/aes128cbc_example.c
@@ -13,6 +13,7 @@
int main()
{
__declspec(align(16)) AesBlock128 plain, key, cypher, decrypted, iv;
+ __declspec(align(16)) Aes128KeySchedule key_schedule, inverted_schedule;
plain = make_aes_block128(0xffeeddcc, 0xbbaa9988, 0x77665544, 0x33221100);
key = make_aes_block128(0x0f0e0d0c, 0x0b0a0908, 0x07060504, 0x03020100);
@@ -32,13 +33,33 @@ int main()
printf(" %s\n", format_aes_block128_fips_style(&iv).str);
print_aes_block128_fips_matrix_style(&iv);
- cypher = aes128cbc_encrypt(plain, key, &iv);
+ aes128_expand_key_schedule(key, &key_schedule);
+
+ printf("\n");
+ printf("Key schedule:\n");
+ for (int i = 0; i < 11; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&key_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&key_schedule.keys[i]).str);
+ }
+
+ cypher = aes128cbc_encrypt(plain, &key_schedule, &iv);
printf("\n");
printf("Cypher: %s\n", format_aes_block128(&cypher).str);
printf(" %s\n", format_aes_block128_fips_style(&cypher).str);
print_aes_block128_fips_matrix_style(&cypher);
- decrypted = aes128cbc_decrypt(cypher, key, &iv);
+ aes128_invert_key_schedule(&key_schedule, &inverted_schedule);
+
+ printf("\n");
+ printf("Inverted key schedule:\n");
+ for (int i = 0; i < 11; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&inverted_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&inverted_schedule.keys[i]).str);
+ }
+
+ decrypted = aes128cbc_decrypt(cypher, &inverted_schedule, &iv);
printf("\n");
printf("Decrypted: %s\n", format_aes_block128(&decrypted).str);
printf(" %s\n", format_aes_block128_fips_style(&decrypted).str);
diff --git a/examples/aes128ecb_example.c b/examples/aes128ecb_example.c
index afb8dd9..aff4182 100644
--- a/examples/aes128ecb_example.c
+++ b/examples/aes128ecb_example.c
@@ -13,6 +13,7 @@
int main()
{
__declspec(align(16)) AesBlock128 plain, key, cypher, decrypted;
+ __declspec(align(16)) Aes128KeySchedule key_schedule, inverted_schedule;
plain = make_aes_block128(0xffeeddcc, 0xbbaa9988, 0x77665544, 0x33221100);
key = make_aes_block128(0x0f0e0d0c, 0x0b0a0908, 0x07060504, 0x03020100);
@@ -26,13 +27,33 @@ int main()
printf(" %s\n", format_aes_block128_fips_style(&key).str);
print_aes_block128_fips_matrix_style(&key);
- cypher = aes128ecb_encrypt(plain, key);
+ aes128_expand_key_schedule(key, &key_schedule);
+
+ printf("\n");
+ printf("Key schedule:\n");
+ for (int i = 0; i < 11; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&key_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&key_schedule.keys[i]).str);
+ }
+
+ cypher = aes128ecb_encrypt(plain, &key_schedule);
printf("\n");
printf("Cypher: %s\n", format_aes_block128(&cypher).str);
printf(" %s\n", format_aes_block128_fips_style(&cypher).str);
print_aes_block128_fips_matrix_style(&cypher);
- decrypted = aes128ecb_decrypt(cypher, key);
+ aes128_invert_key_schedule(&key_schedule, &inverted_schedule);
+
+ printf("\n");
+ printf("Inverted key schedule:\n");
+ for (int i = 0; i < 11; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&inverted_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&inverted_schedule.keys[i]).str);
+ }
+
+ decrypted = aes128ecb_decrypt(cypher, &inverted_schedule);
printf("\n");
printf("Decrypted: %s\n", format_aes_block128(&decrypted).str);
printf(" %s\n", format_aes_block128_fips_style(&decrypted).str);
diff --git a/examples/aes192cbc_example.c b/examples/aes192cbc_example.c
index 25719c8..4b583b6 100644
--- a/examples/aes192cbc_example.c
+++ b/examples/aes192cbc_example.c
@@ -14,6 +14,7 @@ int main()
{
__declspec(align(16)) AesBlock128 plain, cypher, decrypted, iv;
__declspec(align(16)) AesBlock192 key;
+ __declspec(align(16)) Aes192KeySchedule key_schedule, inverted_schedule;
plain = make_aes_block128(0xffeeddcc, 0xbbaa9988, 0x77665544, 0x33221100);
key = make_aes_block192(0x17161514, 0x13121110, 0x0f0e0d0c, 0x0b0a0908, 0x07060504, 0x03020100);
@@ -33,13 +34,33 @@ int main()
printf(" %s\n", format_aes_block128_fips_style(&iv).str);
print_aes_block128_fips_matrix_style(&iv);
- cypher = aes192cbc_encrypt(plain, &key, &iv);
+ aes192_expand_key_schedule(&key, &key_schedule);
+
+ printf("\n");
+ printf("Key schedule:\n");
+ for (int i = 0; i < 13; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&key_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&key_schedule.keys[i]).str);
+ }
+
+ cypher = aes192cbc_encrypt(plain, &key_schedule, &iv);
printf("\n");
printf("Cypher: %s\n", format_aes_block128(&cypher).str);
printf(" %s\n", format_aes_block128_fips_style(&cypher).str);
print_aes_block128_fips_matrix_style(&cypher);
- decrypted = aes192cbc_decrypt(cypher, &key, &iv);
+ aes192_invert_key_schedule(&key_schedule, &inverted_schedule);
+
+ printf("\n");
+ printf("Inverted key schedule:\n");
+ for (int i = 0; i < 13; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&inverted_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&inverted_schedule.keys[i]).str);
+ }
+
+ decrypted = aes192cbc_decrypt(cypher, &inverted_schedule, &iv);
printf("\n");
printf("Decrypted: %s\n", format_aes_block128(&decrypted).str);
printf(" %s\n", format_aes_block128_fips_style(&decrypted).str);
diff --git a/examples/aes192ecb_example.c b/examples/aes192ecb_example.c
index eb9b79d..eb3b329 100644
--- a/examples/aes192ecb_example.c
+++ b/examples/aes192ecb_example.c
@@ -14,6 +14,7 @@ int main()
{
__declspec(align(16)) AesBlock128 plain, cypher, decrypted;
__declspec(align(16)) AesBlock192 key;
+ __declspec(align(16)) Aes192KeySchedule key_schedule, inverted_schedule;
plain = make_aes_block128(0xffeeddcc, 0xbbaa9988, 0x77665544, 0x33221100);
key = make_aes_block192(0x17161514, 0x13121110, 0x0f0e0d0c, 0x0b0a0908, 0x07060504, 0x03020100);
@@ -27,13 +28,33 @@ int main()
printf(" %s\n", format_aes_block192_fips_style(&key).str);
print_aes_block192_fips_matrix_style(&key);
- cypher = aes192ecb_encrypt(plain, &key);
+ aes192_expand_key_schedule(&key, &key_schedule);
+
+ printf("\n");
+ printf("Key schedule:\n");
+ for (int i = 0; i < 13; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&key_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&key_schedule.keys[i]).str);
+ }
+
+ cypher = aes192ecb_encrypt(plain, &key_schedule);
printf("\n");
printf("Cypher: %s\n", format_aes_block128(&cypher).str);
printf(" %s\n", format_aes_block128_fips_style(&cypher).str);
print_aes_block128_fips_matrix_style(&cypher);
- decrypted = aes192ecb_decrypt(cypher, &key);
+ aes192_invert_key_schedule(&key_schedule, &inverted_schedule);
+
+ printf("\n");
+ printf("Inverted key schedule:\n");
+ for (int i = 0; i < 13; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&inverted_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&inverted_schedule.keys[i]).str);
+ }
+
+ decrypted = aes192ecb_decrypt(cypher, &inverted_schedule);
printf("\n");
printf("Decrypted: %s\n", format_aes_block128(&decrypted).str);
printf(" %s\n", format_aes_block128_fips_style(&decrypted).str);
diff --git a/examples/aes256cbc_example.c b/examples/aes256cbc_example.c
index 275b947..816683b 100644
--- a/examples/aes256cbc_example.c
+++ b/examples/aes256cbc_example.c
@@ -14,6 +14,7 @@ int main()
{
__declspec(align(16)) AesBlock128 plain, cypher, decrypted, iv;
__declspec(align(16)) AesBlock256 key;
+ __declspec(align(16)) Aes256KeySchedule key_schedule, inverted_schedule;
plain = make_aes_block128(0xffeeddcc, 0xbbaa9988, 0x77665544, 0x33221100);
key = make_aes_block256(0x1f1e1d1c, 0x1b1a1918, 0x17161514, 0x13121110, 0x0f0e0d0c, 0x0b0a0908, 0x07060504, 0x03020100);
@@ -33,13 +34,33 @@ int main()
printf(" %s\n", format_aes_block128_fips_style(&iv).str);
print_aes_block128_fips_matrix_style(&iv);
- cypher = aes256cbc_encrypt(plain, &key, &iv);
+ aes256_expand_key_schedule(&key, &key_schedule);
+
+ printf("\n");
+ printf("Key schedule:\n");
+ for (int i = 0; i < 15; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&key_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&key_schedule.keys[i]).str);
+ }
+
+ cypher = aes256cbc_encrypt(plain, &key_schedule, &iv);
printf("\n");
printf("Cypher: %s\n", format_aes_block128(&cypher).str);
printf(" %s\n", format_aes_block128_fips_style(&cypher).str);
print_aes_block128_fips_matrix_style(&cypher);
- decrypted = aes256cbc_decrypt(cypher, &key, &iv);
+ aes256_invert_key_schedule(&key_schedule, &inverted_schedule);
+
+ printf("\n");
+ printf("Inverted key schedule:\n");
+ for (int i = 0; i < 15; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&inverted_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&inverted_schedule.keys[i]).str);
+ }
+
+ decrypted = aes256cbc_decrypt(cypher, &inverted_schedule, &iv);
printf("\n");
printf("Decrypted: %s\n", format_aes_block128(&decrypted).str);
printf(" %s\n", format_aes_block128_fips_style(&decrypted).str);
diff --git a/examples/aes256ecb_example.c b/examples/aes256ecb_example.c
index d7a82ee..cd36b50 100644
--- a/examples/aes256ecb_example.c
+++ b/examples/aes256ecb_example.c
@@ -14,6 +14,7 @@ int main()
{
__declspec(align(16)) AesBlock128 plain, cypher, decrypted;
__declspec(align(16)) AesBlock256 key;
+ __declspec(align(16)) Aes256KeySchedule key_schedule, inverted_schedule;
plain = make_aes_block128(0xffeeddcc, 0xbbaa9988, 0x77665544, 0x33221100);
key = make_aes_block256(0x1f1e1d1c, 0x1b1a1918, 0x17161514, 0x13121110, 0x0f0e0d0c, 0x0b0a0908, 0x07060504, 0x03020100);
@@ -27,13 +28,33 @@ int main()
printf(" %s\n", format_aes_block256_fips_style(&key).str);
print_aes_block256_fips_matrix_style(&key);
- cypher = aes256ecb_encrypt(plain, &key);
+ aes256_expand_key_schedule(&key, &key_schedule);
+
+ printf("\n");
+ printf("Key schedule:\n");
+ for (int i = 0; i < 15; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&key_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&key_schedule.keys[i]).str);
+ }
+
+ cypher = aes256ecb_encrypt(plain, &key_schedule);
printf("\n");
printf("Cypher: %s\n", format_aes_block128(&cypher).str);
printf(" %s\n", format_aes_block128_fips_style(&cypher).str);
print_aes_block128_fips_matrix_style(&cypher);
- decrypted = aes256ecb_decrypt(cypher, &key);
+ aes256_invert_key_schedule(&key_schedule, &inverted_schedule);
+
+ printf("\n");
+ printf("Inverted key schedule:\n");
+ for (int i = 0; i < 15; ++i)
+ {
+ printf("\t[%d]: %s\n", i, format_aes_block128(&inverted_schedule.keys[i]).str);
+ printf("\t[%d]: %s\n", i, format_aes_block128_fips_style(&inverted_schedule.keys[i]).str);
+ }
+
+ decrypted = aes256ecb_decrypt(cypher, &inverted_schedule);
printf("\n");
printf("Decrypted: %s\n", format_aes_block128(&decrypted).str);
printf(" %s\n", format_aes_block128_fips_style(&decrypted).str);
diff --git a/include/aesni/api.h b/include/aesni/api.h
index d64b225..efcecc4 100644
--- a/include/aesni/api.h
+++ b/include/aesni/api.h
@@ -11,92 +11,134 @@
#include "data.h"
#include "raw.h"
+static __inline void __fastcall aes128_expand_key_schedule(
+ AesBlock128 key,
+ Aes128KeySchedule* key_schedule)
+{
+ raw_aes128_expand_key_schedule(key, key_schedule);
+}
+
+static __inline void __fastcall aes128_invert_key_schedule(
+ Aes128KeySchedule* key_schedule,
+ Aes128KeySchedule* inverted_schedule)
+{
+ raw_aes128_invert_key_schedule(key_schedule, inverted_schedule);
+}
+
static __inline AesBlock128 __fastcall aes128ecb_encrypt(
AesBlock128 plain,
- AesBlock128 key)
+ Aes128KeySchedule* key_schedule)
{
- return raw_aes128ecb_encrypt(plain, key);
+ return raw_aes128ecb_encrypt(plain, key_schedule);
}
static __inline AesBlock128 __fastcall aes128ecb_decrypt(
AesBlock128 cypher,
- AesBlock128 key)
+ Aes128KeySchedule* inverted_schedule)
{
- return raw_aes128ecb_decrypt(cypher, key);
+ return raw_aes128ecb_decrypt(cypher, inverted_schedule);
}
static __inline AesBlock128 __fastcall aes128cbc_encrypt(
AesBlock128 plain,
- AesBlock128 key,
- AesBlock128* initialization_vector)
+ Aes128KeySchedule* key_schedule,
+ AesBlock128* init_vector)
{
- return raw_aes128cbc_encrypt(plain, key, initialization_vector);
+ return raw_aes128cbc_encrypt(plain, key_schedule, init_vector);
}
static __inline AesBlock128 __fastcall aes128cbc_decrypt(
AesBlock128 cypher,
- AesBlock128 key,
- AesBlock128* initialization_vector)
+ Aes128KeySchedule* inverted_schedule,
+ AesBlock128* init_vector)
+{
+ return raw_aes128cbc_decrypt(cypher, inverted_schedule, init_vector);
+}
+
+static __inline void __fastcall aes192_expand_key_schedule(
+ AesBlock192* key,
+ Aes192KeySchedule* key_schedule)
{
- return raw_aes128cbc_decrypt(cypher, key, initialization_vector);
+ raw_aes192_expand_key_schedule(key->lo, key->hi, key_schedule);
+}
+
+static __inline void __fastcall aes192_invert_key_schedule(
+ Aes192KeySchedule* key_schedule,
+ Aes192KeySchedule* inverted_schedule)
+{
+ raw_aes192_invert_key_schedule(key_schedule, inverted_schedule);
}
static __inline AesBlock128 __fastcall aes192ecb_encrypt(
AesBlock128 plain,
- AesBlock192* key)
+ Aes192KeySchedule* key_schedule)
{
- return raw_aes192ecb_encrypt(plain, key->lo, key->hi);
+ return raw_aes192ecb_encrypt(plain, key_schedule);
}
static __inline AesBlock128 __fastcall aes192ecb_decrypt(
AesBlock128 cypher,
- AesBlock192* key)
+ Aes192KeySchedule* inverted_schedule)
{
- return raw_aes192ecb_decrypt(cypher, key->lo, key->hi);
+ return raw_aes192ecb_decrypt(cypher, inverted_schedule);
}
static __inline AesBlock128 __fastcall aes192cbc_encrypt(
AesBlock128 plain,
- AesBlock192* key,
+ Aes192KeySchedule* key_schedule,
AesBlock128* initialization_vector)
{
- return raw_aes192cbc_encrypt(plain, key->lo, key->hi, initialization_vector);
+ return raw_aes192cbc_encrypt(plain, key_schedule, initialization_vector);
}
static __inline AesBlock128 __fastcall aes192cbc_decrypt(
AesBlock128 cypher,
- AesBlock192* key,
+ Aes192KeySchedule* inverted_schedule,
AesBlock128* initialization_vector)
{
- return raw_aes192cbc_decrypt(cypher, key->lo, key->hi, initialization_vector);
+ return raw_aes192cbc_decrypt(cypher, inverted_schedule, initialization_vector);
+}
+
+static __inline void __fastcall aes256_expand_key_schedule(
+ AesBlock256* key,
+ Aes256KeySchedule* key_schedule)
+{
+ raw_aes256_expand_key_schedule(key->lo, key->hi, key_schedule);
+}
+
+static __inline void __fastcall aes256_invert_key_schedule(
+ Aes256KeySchedule* key_schedule,
+ Aes256KeySchedule* inverted_schedule)
+{
+ raw_aes256_invert_key_schedule(key_schedule, inverted_schedule);
}
static __inline AesBlock128 __fastcall aes256ecb_encrypt(
AesBlock128 plain,
- AesBlock256* key)
+ Aes256KeySchedule* key_schedule)
{
- return raw_aes256ecb_encrypt(plain, key->lo, key->hi);
+ return raw_aes256ecb_encrypt(plain, key_schedule);
}
static __inline AesBlock128 __fastcall aes256ecb_decrypt(
AesBlock128 cypher,
- AesBlock256* key)
+ Aes256KeySchedule* inverted_schedule)
{
- return raw_aes256ecb_decrypt(cypher, key->lo, key->hi);
+ return raw_aes256ecb_decrypt(cypher, inverted_schedule);
}
static __inline AesBlock128 __fastcall aes256cbc_encrypt(
AesBlock128 plain,
- AesBlock256* key,
+ Aes256KeySchedule* key_schedule,
AesBlock128* initialization_vector)
{
- return raw_aes256cbc_encrypt(plain, key->lo, key->hi, initialization_vector);
+ return raw_aes256cbc_encrypt(plain, key_schedule, initialization_vector);
}
static __inline AesBlock128 __fastcall aes256cbc_decrypt(
AesBlock128 cypher,
- AesBlock256* key,
+ Aes256KeySchedule* inverted_schedule,
AesBlock128* initialization_vector)
{
- return raw_aes256cbc_decrypt(cypher, key->lo, key->hi, initialization_vector);
+ return raw_aes256cbc_decrypt(cypher, inverted_schedule, initialization_vector);
}
diff --git a/include/aesni/data.h b/include/aesni/data.h
index 49cf5e5..889a9ff 100644
--- a/include/aesni/data.h
+++ b/include/aesni/data.h
@@ -26,6 +26,25 @@ typedef struct
}
AesBlock256;
+typedef struct
+{
+ AesBlock128 keys[11];
+}
+Aes128KeySchedule;
+
+typedef struct
+{
+ AesBlock128 keys[13];
+ char fillers[8];
+}
+Aes192KeySchedule;
+
+typedef struct
+{
+ AesBlock128 keys[15];
+}
+Aes256KeySchedule;
+
AesBlock128 make_aes_block128(int hi3, int hi2, int lo1, int lo0);
AesBlock192 make_aes_block192(int hi5, int hi4, int lo3, int lo2, int lo1, int lo0);
AesBlock256 make_aes_block256(int hi7, int hi6, int hi5, int hi4, int lo3, int lo2, int lo1, int lo0);
diff --git a/include/aesni/raw.h b/include/aesni/raw.h
index ffda8d2..db342c7 100644
--- a/include/aesni/raw.h
+++ b/include/aesni/raw.h
@@ -10,58 +10,73 @@
#include "data.h"
+void __fastcall raw_aes128_expand_key_schedule(
+ AesBlock128 key,
+ Aes128KeySchedule* key_schedule);
+void __fastcall raw_aes128_invert_key_schedule(
+ Aes128KeySchedule* key_schedule,
+ Aes128KeySchedule* inverted_schedule);
+
AesBlock128 __fastcall raw_aes128ecb_encrypt(
AesBlock128 plain,
- AesBlock128 key);
+ Aes128KeySchedule* key_schedule);
AesBlock128 __fastcall raw_aes128ecb_decrypt(
AesBlock128 cypher,
- AesBlock128 key);
+ Aes128KeySchedule* inverted_schedule);
AesBlock128 __fastcall raw_aes128cbc_encrypt(
AesBlock128 plain,
- AesBlock128 key,
+ Aes128KeySchedule* key_schedule,
AesBlock128* iv);
AesBlock128 __fastcall raw_aes128cbc_decrypt(
AesBlock128 cypher,
- AesBlock128 key,
+ Aes128KeySchedule* inverted_schedule,
AesBlock128* iv);
+void __fastcall raw_aes192_expand_key_schedule(
+ AesBlock128 key_lo,
+ AesBlock128 key_hi,
+ Aes192KeySchedule* key_schedule);
+void __fastcall raw_aes192_invert_key_schedule(
+ Aes192KeySchedule* key_schedule,
+ Aes192KeySchedule* inverted_schedule);
+
AesBlock128 __fastcall raw_aes192ecb_encrypt(
AesBlock128 plain,
- AesBlock128 key_lo,
- AesBlock128 key_hi);
+ Aes192KeySchedule* key_schedule);
AesBlock128 __fastcall raw_aes192ecb_decrypt(
AesBlock128 cypher,
- AesBlock128 key_lo,
- AesBlock128 key_hi);
+ Aes192KeySchedule* inverted_schedule);
AesBlock128 __fastcall raw_aes192cbc_encrypt(
AesBlock128 plain,
- AesBlock128 key_lo,
- AesBlock128 key_hi,
+ Aes192KeySchedule* key_schedule,
AesBlock128 *iv);
AesBlock128 __fastcall raw_aes192cbc_decrypt(
AesBlock128 cypher,
+ Aes192KeySchedule* inverted_schedule,
+ AesBlock128 *iv);
+
+void __fastcall raw_aes256_expand_key_schedule(
AesBlock128 key_lo,
AesBlock128 key_hi,
- AesBlock128 *iv);
+ Aes256KeySchedule* key_schedule);
+void __fastcall raw_aes256_invert_key_schedule(
+ Aes256KeySchedule* key_schedule,
+ Aes256KeySchedule* inverted_schedule);
AesBlock128 __fastcall raw_aes256ecb_encrypt(
AesBlock128 plain,
- AesBlock128 key_lo,
- AesBlock128 key_hi);
+ Aes256KeySchedule* key_schedule);
AesBlock128 __fastcall raw_aes256ecb_decrypt(
AesBlock128 cypher,
- AesBlock128 key_lo,
- AesBlock128 key_hi);
+ Aes256KeySchedule* inverted_schedule);
AesBlock128 __fastcall raw_aes256cbc_encrypt(
AesBlock128 plain,
- AesBlock128 key_lo,
- AesBlock128 key_hi,
+ Aes256KeySchedule* key_schedule,
AesBlock128 *iv);
AesBlock128 __fastcall raw_aes256cbc_decrypt(
AesBlock128 cypher,
- AesBlock128 key_lo,
- AesBlock128 key_hi,
+ Aes256KeySchedule* inverted_schedule,
AesBlock128 *iv);
diff --git a/src/aes128.asm b/src/aes128.asm
index 5d62621..577e582 100644
--- a/src/aes128.asm
+++ b/src/aes128.asm
@@ -6,68 +6,55 @@
.xmm
.model flat
-.data
-
-align 10h
-key_schedule oword 11 dup(0)
-
-align 10h
-inverted_key_schedule oword 11 dup(0)
-
.code
-@raw_aes128ecb_encrypt@32 proc
- call expand_keys128
- pxor xmm0, [key_schedule]
- aesenc xmm0, [key_schedule + 10h]
- aesenc xmm0, [key_schedule + 20h]
- aesenc xmm0, [key_schedule + 30h]
- aesenc xmm0, [key_schedule + 40h]
- aesenc xmm0, [key_schedule + 50h]
- aesenc xmm0, [key_schedule + 60h]
- aesenc xmm0, [key_schedule + 70h]
- aesenc xmm0, [key_schedule + 80h]
- aesenc xmm0, [key_schedule + 90h]
- aesenclast xmm0, [key_schedule + 0A0h]
+@raw_aes128ecb_encrypt@20 proc
+ pxor xmm0, [ecx]
+ aesenc xmm0, [ecx + 10h]
+ aesenc xmm0, [ecx + 20h]
+ aesenc xmm0, [ecx + 30h]
+ aesenc xmm0, [ecx + 40h]
+ aesenc xmm0, [ecx + 50h]
+ aesenc xmm0, [ecx + 60h]
+ aesenc xmm0, [ecx + 70h]
+ aesenc xmm0, [ecx + 80h]
+ aesenc xmm0, [ecx + 90h]
+ aesenclast xmm0, [ecx + 0A0h]
ret
-@raw_aes128ecb_encrypt@32 endp
+@raw_aes128ecb_encrypt@20 endp
+
+@raw_aes128cbc_encrypt@24 proc
+ pxor xmm0, [edx]
+ jmp @raw_aes128ecb_encrypt@20
+@raw_aes128cbc_encrypt@24 endp
-@raw_aes128cbc_encrypt@36 proc
+@raw_aes128ecb_decrypt@20 proc
pxor xmm0, [ecx]
- jmp @raw_aes128ecb_encrypt@32
-@raw_aes128cbc_encrypt@36 endp
-
-@raw_aes128ecb_decrypt@32 proc
- call expand_keys128
- pxor xmm0, [inverted_key_schedule]
- aesdec xmm0, [inverted_key_schedule + 10h]
- aesdec xmm0, [inverted_key_schedule + 20h]
- aesdec xmm0, [inverted_key_schedule + 30h]
- aesdec xmm0, [inverted_key_schedule + 40h]
- aesdec xmm0, [inverted_key_schedule + 50h]
- aesdec xmm0, [inverted_key_schedule + 60h]
- aesdec xmm0, [inverted_key_schedule + 70h]
- aesdec xmm0, [inverted_key_schedule + 80h]
- aesdec xmm0, [inverted_key_schedule + 90h]
- aesdeclast xmm0, [inverted_key_schedule + 0A0h]
+ aesdec xmm0, [ecx + 10h]
+ aesdec xmm0, [ecx + 20h]
+ aesdec xmm0, [ecx + 30h]
+ aesdec xmm0, [ecx + 40h]
+ aesdec xmm0, [ecx + 50h]
+ aesdec xmm0, [ecx + 60h]
+ aesdec xmm0, [ecx + 70h]
+ aesdec xmm0, [ecx + 80h]
+ aesdec xmm0, [ecx + 90h]
+ aesdeclast xmm0, [ecx + 0A0h]
ret
-@raw_aes128ecb_decrypt@32 endp
+@raw_aes128ecb_decrypt@20 endp
-@raw_aes128cbc_decrypt@36 proc
- push ecx
- call @raw_aes128ecb_decrypt@32
- pop ecx
- pxor xmm0, [ecx]
+@raw_aes128cbc_decrypt@24 proc
+ call @raw_aes128ecb_decrypt@20
+ pxor xmm0, [edx]
ret
-@raw_aes128cbc_decrypt@36 endp
+@raw_aes128cbc_decrypt@24 endp
-expand_keys128 proc
+@raw_aes128_expand_key_schedule@20 proc
; A "word" (in terms of the FIPS 187 standard) is a 32-bit block.
; Words are denoted by `w[N]`.
;
; A key schedule is composed of 10 "regular" keys and a dumb key for
; the "whitening" step.
- ; It's stored in `key_schedule`.
;
; A key schedule is thus composed of 44 "words".
; The FIPS standard includes an algorithm to calculate these words via
@@ -104,52 +91,51 @@ expand_keys128 proc
; and RotWord using `aeskeygenassist`, which is used in this routine.
;
; Preconditions:
- ; * xmm1[127:96] == w[3],
- ; * xmm1[95:64] == w[2],
- ; * xmm1[63:32] == w[1],
- ; * xmm1[31:0] == w[0].
+ ; * xmm0[127:96] == w[3],
+ ; * xmm0[95:64] == w[2],
+ ; * xmm0[63:32] == w[1],
+ ; * xmm0[31:0] == w[0].
- movdqa [key_schedule], xmm1 ; sets w[0], w[1], w[2], w[3]
+ movdqa [ecx], xmm0 ; sets w[0], w[1], w[2], w[3]
+ add ecx, 10h ; ecx = &w[4]
- lea ecx, [key_schedule + 10h] ; ecx = &w[4]
- aeskeygenassist xmm7, xmm1, 01h ; xmm7[127:96] = RotWord(SubWord(w[3]))^Rcon
+ aeskeygenassist xmm7, xmm0, 01h ; xmm7[127:96] = RotWord(SubWord(w[3]))^Rcon
call gen_round_key ; sets w[4], w[5], w[6], w[7]
- aeskeygenassist xmm7, xmm1, 02h ; xmm7[127:96] = RotWord(SubWord(w[7]))^Rcon
+ aeskeygenassist xmm7, xmm0, 02h ; xmm7[127:96] = RotWord(SubWord(w[7]))^Rcon
call gen_round_key ; sets w[8], w[9], w[10], w[11]
- aeskeygenassist xmm7, xmm1, 04h ; xmm7[127:96] = RotWord(SubWord(w[11]))^Rcon
+ aeskeygenassist xmm7, xmm0, 04h ; xmm7[127:96] = RotWord(SubWord(w[11]))^Rcon
call gen_round_key ; sets w[12], w[13], w[14], w[15]
- aeskeygenassist xmm7, xmm1, 08h ; xmm7[127:96] = RotWord(SubWord(w[15]))^Rcon
+ aeskeygenassist xmm7, xmm0, 08h ; xmm7[127:96] = RotWord(SubWord(w[15]))^Rcon
call gen_round_key ; sets w[16], w[17], w[18], w[19]
- aeskeygenassist xmm7, xmm1, 10h ; xmm7[127:96] = RotWord(SubWord(w[19]))^Rcon
+ aeskeygenassist xmm7, xmm0, 10h ; xmm7[127:96] = RotWord(SubWord(w[19]))^Rcon
call gen_round_key ; sets w[20], w[21], w[22], w[23]
- aeskeygenassist xmm7, xmm1, 20h ; xmm7[127:96] = RotWord(SubWord(w[23]))^Rcon
+ aeskeygenassist xmm7, xmm0, 20h ; xmm7[127:96] = RotWord(SubWord(w[23]))^Rcon
call gen_round_key ; sets w[24], w[25], w[26], w[27]
- aeskeygenassist xmm7, xmm1, 40h ; xmm7[127:96] = RotWord(SubWord(w[27]))^Rcon
+ aeskeygenassist xmm7, xmm0, 40h ; xmm7[127:96] = RotWord(SubWord(w[27]))^Rcon
call gen_round_key ; sets w[28], w[29], w[30], w[31]
- aeskeygenassist xmm7, xmm1, 80h ; xmm7[127:96] = RotWord(SubWord(w[31]))^Rcon
+ aeskeygenassist xmm7, xmm0, 80h ; xmm7[127:96] = RotWord(SubWord(w[31]))^Rcon
call gen_round_key ; sets w[32], w[33], w[34], w[35]
- aeskeygenassist xmm7, xmm1, 1Bh ; xmm7[127:96] = RotWord(SubWord(w[35]))^Rcon
+ aeskeygenassist xmm7, xmm0, 1Bh ; xmm7[127:96] = RotWord(SubWord(w[35]))^Rcon
call gen_round_key ; sets w[36], w[37], w[38], w[39]
- aeskeygenassist xmm7, xmm1, 36h ; xmm7[127:96] = RotWord(SubWord(w[39]))^Rcon
+ aeskeygenassist xmm7, xmm0, 36h ; xmm7[127:96] = RotWord(SubWord(w[39]))^Rcon
call gen_round_key ; sets w[40], w[41], w[42], w[43]
- call invert_key_schedule
ret
gen_round_key:
; Preconditions:
- ; * xmm1[127:96] == w[i+3],
- ; * xmm1[95:64] == w[i+2],
- ; * xmm1[63:32] == w[i+1],
- ; * xmm1[31:0] == w[i],
+ ; * xmm0[127:96] == w[i+3],
+ ; * xmm0[95:64] == w[i+2],
+ ; * xmm0[63:32] == w[i+1],
+ ; * xmm0[31:0] == w[i],
; * xmm7[127:96] == RotWord(SubWord(w[i+3]))^Rcon,
; * ecx == &w[i+4].
;
; Postconditions:
- ; * xmm1[127:96] == w[i+7] == RotWord(SubWord(w[i+3]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i],
- ; * xmm1[95:64] == w[i+6] == RotWord(SubWord(w[i+3]))^Rcon^w[i+2]^w[i+1]^w[i],
- ; * xmm1[63:32] == w[i+5] == RotWord(SubWord(w[i+3]))^Rcon^w[i+1]^w[i],
- ; * xmm1[31:0] == w[i+4] == RotWord(SubWord(w[i+3]))^Rcon^w[i],
+ ; * xmm0[127:96] == w[i+7] == RotWord(SubWord(w[i+3]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i],
+ ; * xmm0[95:64] == w[i+6] == RotWord(SubWord(w[i+3]))^Rcon^w[i+2]^w[i+1]^w[i],
+ ; * xmm0[63:32] == w[i+5] == RotWord(SubWord(w[i+3]))^Rcon^w[i+1]^w[i],
+ ; * xmm0[31:0] == w[i+4] == RotWord(SubWord(w[i+3]))^Rcon^w[i],
; * ecx == &w[i+8],
; * the value in xmm6 is also modified.
@@ -158,17 +144,17 @@ gen_round_key:
; w[i+2]^w[i+1]^w[i],
; w[i+1]^w[i] and
; w[i].
- movdqa xmm6, xmm1 ; xmm6 = xmm1
+ movdqa xmm6, xmm0 ; xmm6 = xmm0
pslldq xmm6, 4 ; xmm6 <<= 32
- pxor xmm1, xmm6 ; xmm1 ^= xmm6
+ pxor xmm0, xmm6 ; xmm0 ^= xmm6
pslldq xmm6, 4 ; xmm6 <<= 32
- pxor xmm1, xmm6 ; xmm1 ^= xmm6
+ pxor xmm0, xmm6 ; xmm0 ^= xmm6
pslldq xmm6, 4 ; xmm6 <<= 32
- pxor xmm1, xmm6 ; xmm1 ^= xmm6
- ; xmm1[127:96] == w[i+3]^w[i+2]^w[i+1]^w[i]
- ; xmm1[95:64] == w[i+2]^w[i+1]^w[i]
- ; xmm1[63:32] == w[i+1]^w[i]
- ; xmm1[31:0] == w[i]
+ pxor xmm0, xmm6 ; xmm0 ^= xmm6
+ ; xmm0[127:96] == w[i+3]^w[i+2]^w[i+1]^w[i]
+ ; xmm0[95:64] == w[i+2]^w[i+1]^w[i]
+ ; xmm0[63:32] == w[i+1]^w[i]
+ ; xmm0[31:0] == w[i]
; Calculate
; w[i+7] == RotWord(SubWord(w[i+3]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i],
@@ -176,51 +162,52 @@ gen_round_key:
; w[i+5] == RotWord(SubWord(w[i+3]))^Rcon^w[i+1]^w[i] and
; w[i+4] == RotWord(SubWord(w[i+3]))^Rcon^w[i].
pshufd xmm6, xmm7, 0FFh ; xmm6[127:96] = xmm6[95:64] = xmm6[63:32] = xmm6[31:0] = xmm7[127:96]
- pxor xmm1, xmm6 ; xmm1 ^= xmm6
- ; xmm1[127:96] == w[i+7] == RotWord(SubWord(w[i+3]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i]
- ; xmm1[95:64] == w[i+6] == RotWord(SubWord(w[i+3]))^Rcon^w[i+2]^w[i+1]^w[i]
- ; xmm1[63:32] == w[i+5] == RotWord(SubWord(w[i+3]))^Rcon^w[i+1]^w[i]
- ; xmm1[31:0] == w[i+4] == RotWord(SubWord(w[i+3]))^Rcon^w[i]
+ pxor xmm0, xmm6 ; xmm0 ^= xmm6
+ ; xmm0[127:96] == w[i+7] == RotWord(SubWord(w[i+3]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i]
+ ; xmm0[95:64] == w[i+6] == RotWord(SubWord(w[i+3]))^Rcon^w[i+2]^w[i+1]^w[i]
+ ; xmm0[63:32] == w[i+5] == RotWord(SubWord(w[i+3]))^Rcon^w[i+1]^w[i]
+ ; xmm0[31:0] == w[i+4] == RotWord(SubWord(w[i+3]))^Rcon^w[i]
; Set w[i+4], w[i+5], w[i+6] and w[i+7].
- movdqa [ecx], xmm1 ; w[i+7] = RotWord(SubWord(w[i+3]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i]
+ movdqa [ecx], xmm0 ; w[i+7] = RotWord(SubWord(w[i+3]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i]
; w[i+6] = RotWord(SubWord(w[i+3]))^Rcon^w[i+2]^w[i+1]^w[i]
; w[i+5] = RotWord(SubWord(w[i+3]))^Rcon^w[i+1]^w[i]
; w[i+4] = RotWord(SubWord(w[i+3]))^Rcon^w[i]
add ecx, 10h ; ecx = &w[i+8]
ret
+@raw_aes128_expand_key_schedule@20 endp
-invert_key_schedule:
- movdqa xmm7, [key_schedule]
- movdqa xmm6, [key_schedule + 0A0h]
- movdqa [inverted_key_schedule], xmm6
- movdqa [inverted_key_schedule + 0A0h], xmm7
+@raw_aes128_invert_key_schedule@8 proc
+ movdqa xmm7, [ecx]
+ movdqa xmm6, [ecx + 0A0h]
+ movdqa [edx], xmm6
+ movdqa [edx + 0A0h], xmm7
- aesimc xmm7, [key_schedule + 10h]
- aesimc xmm6, [key_schedule + 90h]
- movdqa [inverted_key_schedule + 10h], xmm6
- movdqa [inverted_key_schedule + 90h], xmm7
+ aesimc xmm7, [ecx + 10h]
+ aesimc xmm6, [ecx + 90h]
+ movdqa [edx + 10h], xmm6
+ movdqa [edx + 90h], xmm7
- aesimc xmm7, [key_schedule + 20h]
- aesimc xmm6, [key_schedule + 80h]
- movdqa [inverted_key_schedule + 20h], xmm6
- movdqa [inverted_key_schedule + 80h], xmm7
+ aesimc xmm7, [ecx + 20h]
+ aesimc xmm6, [ecx + 80h]
+ movdqa [edx + 20h], xmm6
+ movdqa [edx + 80h], xmm7
- aesimc xmm7, [key_schedule + 30h]
- aesimc xmm6, [key_schedule + 70h]
- movdqa [inverted_key_schedule + 30h], xmm6
- movdqa [inverted_key_schedule + 70h], xmm7
+ aesimc xmm7, [ecx + 30h]
+ aesimc xmm6, [ecx + 70h]
+ movdqa [edx + 30h], xmm6
+ movdqa [edx + 70h], xmm7
- aesimc xmm7, [key_schedule + 40h]
- aesimc xmm6, [key_schedule + 60h]
- movdqa [inverted_key_schedule + 40h], xmm6
- movdqa [inverted_key_schedule + 60h], xmm7
+ aesimc xmm7, [ecx + 40h]
+ aesimc xmm6, [ecx + 60h]
+ movdqa [edx + 40h], xmm6
+ movdqa [edx + 60h], xmm7
- aesimc xmm7, [key_schedule + 50h]
- movdqa [inverted_key_schedule + 50h], xmm7
+ aesimc xmm7, [ecx + 50h]
+ movdqa [edx + 50h], xmm7
ret
-expand_keys128 endp
+@raw_aes128_invert_key_schedule@8 endp
end
diff --git a/src/aes192.asm b/src/aes192.asm
index 55fd540..21e0fa1 100644
--- a/src/aes192.asm
+++ b/src/aes192.asm
@@ -6,72 +6,59 @@
.xmm
.model flat
-.data
-
-align 10h
-key_schedule oword 13 dup(0)
-
-align 10h
-inverted_key_schedule oword 13 dup(0)
-
.code
-@raw_aes192ecb_encrypt@48 proc
- call expand_keys192
- pxor xmm0, [key_schedule]
- aesenc xmm0, [key_schedule + 10h]
- aesenc xmm0, [key_schedule + 20h]
- aesenc xmm0, [key_schedule + 30h]
- aesenc xmm0, [key_schedule + 40h]
- aesenc xmm0, [key_schedule + 50h]
- aesenc xmm0, [key_schedule + 60h]
- aesenc xmm0, [key_schedule + 70h]
- aesenc xmm0, [key_schedule + 80h]
- aesenc xmm0, [key_schedule + 90h]
- aesenc xmm0, [key_schedule + 0A0h]
- aesenc xmm0, [key_schedule + 0B0h]
- aesenclast xmm0, [key_schedule + 0C0h]
+@raw_aes192ecb_encrypt@20 proc
+ pxor xmm0, [ecx]
+ aesenc xmm0, [ecx + 10h]
+ aesenc xmm0, [ecx + 20h]
+ aesenc xmm0, [ecx + 30h]
+ aesenc xmm0, [ecx + 40h]
+ aesenc xmm0, [ecx + 50h]
+ aesenc xmm0, [ecx + 60h]
+ aesenc xmm0, [ecx + 70h]
+ aesenc xmm0, [ecx + 80h]
+ aesenc xmm0, [ecx + 90h]
+ aesenc xmm0, [ecx + 0A0h]
+ aesenc xmm0, [ecx + 0B0h]
+ aesenclast xmm0, [ecx + 0C0h]
ret
-@raw_aes192ecb_encrypt@48 endp
+@raw_aes192ecb_encrypt@20 endp
-@raw_aes192cbc_encrypt@52 proc
- pxor xmm0, [ecx]
- jmp @raw_aes192ecb_encrypt@48
-@raw_aes192cbc_encrypt@52 endp
+@raw_aes192cbc_encrypt@24 proc
+ pxor xmm0, [edx]
+ jmp @raw_aes192ecb_encrypt@20
+@raw_aes192cbc_encrypt@24 endp
-@raw_aes192ecb_decrypt@48 proc
- call expand_keys192
- pxor xmm0, [inverted_key_schedule]
- aesdec xmm0, [inverted_key_schedule + 10h]
- aesdec xmm0, [inverted_key_schedule + 20h]
- aesdec xmm0, [inverted_key_schedule + 30h]
- aesdec xmm0, [inverted_key_schedule + 40h]
- aesdec xmm0, [inverted_key_schedule + 50h]
- aesdec xmm0, [inverted_key_schedule + 60h]
- aesdec xmm0, [inverted_key_schedule + 70h]
- aesdec xmm0, [inverted_key_schedule + 80h]
- aesdec xmm0, [inverted_key_schedule + 90h]
- aesdec xmm0, [inverted_key_schedule + 0A0h]
- aesdec xmm0, [inverted_key_schedule + 0B0h]
- aesdeclast xmm0, [inverted_key_schedule + 0C0h]
+@raw_aes192ecb_decrypt@20 proc
+ pxor xmm0, [ecx]
+ aesdec xmm0, [ecx + 10h]
+ aesdec xmm0, [ecx + 20h]
+ aesdec xmm0, [ecx + 30h]
+ aesdec xmm0, [ecx + 40h]
+ aesdec xmm0, [ecx + 50h]
+ aesdec xmm0, [ecx + 60h]
+ aesdec xmm0, [ecx + 70h]
+ aesdec xmm0, [ecx + 80h]
+ aesdec xmm0, [ecx + 90h]
+ aesdec xmm0, [ecx + 0A0h]
+ aesdec xmm0, [ecx + 0B0h]
+ aesdeclast xmm0, [ecx + 0C0h]
ret
-@raw_aes192ecb_decrypt@48 endp
+@raw_aes192ecb_decrypt@20 endp
-@raw_aes192cbc_decrypt@52 proc
- push ecx
- call @raw_aes192ecb_decrypt@48
- pop ecx
- pxor xmm0, [ecx]
+@raw_aes192cbc_decrypt@24 proc
+ call @raw_aes192ecb_decrypt@20
+ pxor xmm0, [edx]
ret
-@raw_aes192cbc_decrypt@52 endp
+@raw_aes192cbc_decrypt@24 endp
-expand_keys192 proc
+@raw_aes192_expand_key_schedule@36 proc
; A "word" (in terms of the FIPS 187 standard) is a 32-bit block.
; Words are denoted by `w[N]`.
;
; A key schedule is composed of 12 "regular" keys and a dumb key for
; the "whitening" step.
- ; It's stored in `key_schedule`.
;
; A key schedule is thus composed of 52 "words".
; The FIPS standard includes an algorithm to calculate these words via
@@ -116,59 +103,58 @@ expand_keys192 proc
; and RotWord using `aeskeygenassist`, which is used in this routine.
;
; Preconditions:
- ; * xmm2[63:32] == w[5],
- ; * xmm2[31:0] == w[4],
- ; * xmm1[127:96] == w[3],
- ; * xmm1[95:64] == w[2],
- ; * xmm1[63:32] == w[1],
- ; * xmm1[31:0] == w[0].
+ ; * xmm1[63:32] == w[5],
+ ; * xmm1[31:0] == w[4],
+ ; * xmm0[127:96] == w[3],
+ ; * xmm0[95:64] == w[2],
+ ; * xmm0[63:32] == w[1],
+ ; * xmm0[31:0] == w[0].
- movdqa [key_schedule], xmm1 ; sets w[0], w[1], w[2], w[3]
- movdqa [key_schedule + 10h], xmm2 ; sets w[4], w[5]
+ movdqa [ecx], xmm0 ; sets w[0], w[1], w[2], w[3]
+ movdqa [ecx + 10h], xmm1 ; sets w[4], w[5]
+ add ecx, 18h ; ecx = &w[6]
- lea ecx, [key_schedule + 18h] ; ecx = &w[6]
- aeskeygenassist xmm7, xmm2, 1 ; xmm7[63:32] = RotWord(SubWord(w[5]))^Rcon,
+ aeskeygenassist xmm7, xmm1, 1 ; xmm7[63:32] = RotWord(SubWord(w[5]))^Rcon,
call gen_round_key ; sets w[6], w[7], w[8], w[9], w[10], w[11]
- aeskeygenassist xmm7, xmm2, 2 ; xmm7[63:32] = RotWord(SubWord(w[11]))^Rcon
+ aeskeygenassist xmm7, xmm1, 2 ; xmm7[63:32] = RotWord(SubWord(w[11]))^Rcon
call gen_round_key ; sets w[12], w[13], w[14], w[15], w[16], w[17]
- aeskeygenassist xmm7, xmm2, 4 ; xmm7[63:32] = RotWord(SubWord(w[17]))^Rcon
+ aeskeygenassist xmm7, xmm1, 4 ; xmm7[63:32] = RotWord(SubWord(w[17]))^Rcon
call gen_round_key ; sets w[18], w[19], w[20], w[21], w[22], w[23]
- aeskeygenassist xmm7, xmm2, 8 ; xmm7[63:32] = RotWord(SubWord(w[23]))^Rcon
+ aeskeygenassist xmm7, xmm1, 8 ; xmm7[63:32] = RotWord(SubWord(w[23]))^Rcon
call gen_round_key ; sets w[24], w[25], w[26], w[27], w[28], w[29]
- aeskeygenassist xmm7, xmm2, 10h ; xmm7[63:32] = RotWord(SubWord(w[29]))^Rcon
+ aeskeygenassist xmm7, xmm1, 10h ; xmm7[63:32] = RotWord(SubWord(w[29]))^Rcon
call gen_round_key ; sets w[30], w[31], w[32], w[33], w[34], w[35]
- aeskeygenassist xmm7, xmm2, 20h ; xmm7[63:32] = RotWord(SubWord(w[35]))^Rcon
+ aeskeygenassist xmm7, xmm1, 20h ; xmm7[63:32] = RotWord(SubWord(w[35]))^Rcon
call gen_round_key ; sets w[36], w[37], w[38], w[39], w[40], w[41]
- aeskeygenassist xmm7, xmm2, 40h ; xmm7[63:32] = RotWord(SubWord(w[41]))^Rcon
+ aeskeygenassist xmm7, xmm1, 40h ; xmm7[63:32] = RotWord(SubWord(w[41]))^Rcon
call gen_round_key ; sets w[42], w[43], w[44], w[45], w[46], w[47]
- aeskeygenassist xmm7, xmm2, 80h ; xmm7[63:32] = RotWord(SubWord(w[49]))^Rcon
+ aeskeygenassist xmm7, xmm1, 80h ; xmm7[63:32] = RotWord(SubWord(w[49]))^Rcon
call gen_round_key ; sets w[48], w[49], w[50], w[51], w[52], w[53] // FIXME
- call invert_key_schedule
ret
gen_round_key:
; Preconditions:
- ; * xmm2[127:96] == 0,
- ; * xmm2[95:64] == 0,
- ; * xmm2[63:32] == w[i+5],
- ; * xmm2[31:0] == w[i+4],
- ; * xmm1[127:96] == w[i+3],
- ; * xmm1[95:64] == w[i+2],
- ; * xmm1[63:32] == w[i+1],
- ; * xmm1[31:0] == w[i],
+ ; * xmm1[127:96] == 0,
+ ; * xmm1[95:64] == 0,
+ ; * xmm1[63:32] == w[i+5],
+ ; * xmm1[31:0] == w[i+4],
+ ; * xmm0[127:96] == w[i+3],
+ ; * xmm0[95:64] == w[i+2],
+ ; * xmm0[63:32] == w[i+1],
+ ; * xmm0[31:0] == w[i],
; * xmm7[63:32] == RotWord(SubWord(w[i+5]))^Rcon,
; * ecx == &w[i+6].
;
; Postconditions:
- ; * xmm2[127:96] == 0,
- ; * xmm2[95:64] == 0,
- ; * xmm2[63:32] == w[i+11] == RotWord(SubWord(w[i+5]))^Rcon^w[i+5]^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i],
- ; * xmm2[31:0] == w[i+10] == RotWord(SubWord(w[i+5]))^Rcon^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i],
- ; * xmm1[127:96] == w[i+9] == RotWord(SubWord(w[i+5]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i],
- ; * xmm1[95:64] == w[i+8] == RotWord(SubWord(w[i+5]))^Rcon^w[i+2]^w[i+1]^w[i],
- ; * xmm1[63:32] == w[i+7] == RotWord(SubWord(w[i+5]))^Rcon^w[i+1]^w[i],
- ; * xmm1[31:0] == w[i+6] == RotWord(SubWord(w[i+5]))^Rcon^w[i],
+ ; * xmm1[127:96] == 0,
+ ; * xmm1[95:64] == 0,
+ ; * xmm1[63:32] == w[i+11] == RotWord(SubWord(w[i+5]))^Rcon^w[i+5]^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i],
+ ; * xmm1[31:0] == w[i+10] == RotWord(SubWord(w[i+5]))^Rcon^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i],
+ ; * xmm0[127:96] == w[i+9] == RotWord(SubWord(w[i+5]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i],
+ ; * xmm0[95:64] == w[i+8] == RotWord(SubWord(w[i+5]))^Rcon^w[i+2]^w[i+1]^w[i],
+ ; * xmm0[63:32] == w[i+7] == RotWord(SubWord(w[i+5]))^Rcon^w[i+1]^w[i],
+ ; * xmm0[31:0] == w[i+6] == RotWord(SubWord(w[i+5]))^Rcon^w[i],
; * ecx == &w[i+12],
; * the value in xmm6 is also modified.
@@ -177,17 +163,17 @@ gen_round_key:
; w[i+2]^w[i+1]^w[i],
; w[i+1]^w[i] and
; w[i].
- movdqa xmm6, xmm1 ; xmm6 = xmm1
+ movdqa xmm6, xmm0 ; xmm6 = xmm0
pslldq xmm6, 4 ; xmm6 <<= 32
- pxor xmm1, xmm6 ; xmm1 ^= xmm6
+ pxor xmm0, xmm6 ; xmm0 ^= xmm6
pslldq xmm6, 4 ; xmm6 <<= 32
- pxor xmm1, xmm6 ; xmm1 ^= xmm6
+ pxor xmm0, xmm6 ; xmm0 ^= xmm6
pslldq xmm6, 4 ; xmm6 <<= 32
- pxor xmm1, xmm6 ; xmm1 ^= xmm6
- ; xmm1[127:96] == w[i+3]^w[i+2]^w[i+1]^w[i]
- ; xmm1[95:64] == w[i+2]^w[i+1]^w[i]
- ; xmm1[63:32] == w[i+1]^w[i]
- ; xmm1[31:0] == w[i]
+ pxor xmm0, xmm6 ; xmm0 ^= xmm6
+ ; xmm0[127:96] == w[i+3]^w[i+2]^w[i+1]^w[i]
+ ; xmm0[95:64] == w[i+2]^w[i+1]^w[i]
+ ; xmm0[63:32] == w[i+1]^w[i]
+ ; xmm0[31:0] == w[i]
; Calculate
; w[i+9] == RotWord(SubWord(w[i+5]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i],
@@ -195,14 +181,14 @@ gen_round_key:
; w[i+7] == RotWord(SubWord(w[i+5]))^Rcon^w[i+1]^w[i] and
; w[i+6] == RotWord(SubWord(w[i+5]))^Rcon^w[i].
pshufd xmm6, xmm7, 55h ; xmm6[127:96] = xmm6[95:64] = xmm6[63:32] = xmm6[31:0] = xmm7[63:32]
- pxor xmm1, xmm6 ; xmm1 ^= xmm6
- ; xmm1[127:96] == w[i+9] == RotWord(SubWord(w[i+5]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i]
- ; xmm1[95:64] == w[i+8] == RotWord(SubWord(w[i+5]))^Rcon^w[i+2]^w[i+1]^w[i]
- ; xmm1[63:32] == w[i+7] == RotWord(SubWord(w[i+5]))^Rcon^w[i+1]^w[i]
- ; xmm1[31:0] == w[i+6] == RotWord(SubWord(w[i+5]))^Rcon^w[i]
+ pxor xmm0, xmm6 ; xmm0 ^= xmm6
+ ; xmm0[127:96] == w[i+9] == RotWord(SubWord(w[i+5]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i]
+ ; xmm0[95:64] == w[i+8] == RotWord(SubWord(w[i+5]))^Rcon^w[i+2]^w[i+1]^w[i]
+ ; xmm0[63:32] == w[i+7] == RotWord(SubWord(w[i+5]))^Rcon^w[i+1]^w[i]
+ ; xmm0[31:0] == w[i+6] == RotWord(SubWord(w[i+5]))^Rcon^w[i]
; Set w[i+6], w[i+7], w[i+8] and w[i+9].
- movdqu [ecx], xmm1 ; w[i+6] = RotWord(SubWord(w[i+5]))^Rcon^w[i]
+ movdqu [ecx], xmm0 ; w[i+6] = RotWord(SubWord(w[i+5]))^Rcon^w[i]
; w[i+7] = RotWord(SubWord(w[i+5]))^Rcon^w[i+1]^w[i]
; w[i+8] = RotWord(SubWord(w[i+5]))^Rcon^w[i+2]^w[i+1]^w[i]
; w[i+9] = RotWord(SubWord(w[i+5]))^Rcon^w[i+3]^w[i+2]^w[i+1]^w[i]
@@ -211,62 +197,63 @@ gen_round_key:
; Calculate
; w[i+5]^w[i+4],
; w[i+4].
- pshufd xmm6, xmm2, 0F3h ; xmm6 = xmm2[31:0] << 32
- pxor xmm2, xmm6 ; xmm2 ^= xmm7
- ; xmm2[63:32] == w[i+5]^w[i+4]
- ; xmm2[31:0] == w[i+4]
+ pshufd xmm6, xmm1, 0F3h ; xmm6 = xmm1[31:0] << 32
+ pxor xmm1, xmm6 ; xmm1 ^= xmm7
+ ; xmm1[63:32] == w[i+5]^w[i+4]
+ ; xmm1[31:0] == w[i+4]
; Calculate
; w[i+10] == RotWord(SubWord(w[i+5]))^Rcon^w[i+5]^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i],
; w[i+11] == RotWord(SubWord(w[i+5]))^Rcon^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i].
- pshufd xmm6, xmm1, 0FFh ; xmm6[127:96] = xmm6[95:64] = xmm6[63:32] = xmm6[31:0] = xmm1[127:96]
+ pshufd xmm6, xmm0, 0FFh ; xmm6[127:96] = xmm6[95:64] = xmm6[63:32] = xmm6[31:0] = xmm0[127:96]
psrldq xmm6, 8 ; xmm6 >>= 64
- pxor xmm2, xmm6 ; xmm2 ^= xmm6
- ; xmm2[63:32] == w[i+11] == RotWord(SubWord(w[i+5]))^Rcon^w[i+5]^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i]
- ; xmm2[31:0] == w[i+10] == RotWord(SubWord(w[i+5]))^Rcon^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i]
+ pxor xmm1, xmm6 ; xmm1 ^= xmm6
+ ; xmm1[63:32] == w[i+11] == RotWord(SubWord(w[i+5]))^Rcon^w[i+5]^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i]
+ ; xmm1[31:0] == w[i+10] == RotWord(SubWord(w[i+5]))^Rcon^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i]
; Set w[i+10] and w[i+11].
- movq qword ptr [ecx], xmm2 ; w[i+10] = RotWord(SubWord(w[i+5]))^Rcon^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i]
+ movq qword ptr [ecx], xmm1 ; w[i+10] = RotWord(SubWord(w[i+5]))^Rcon^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i]
; w[i+11] = RotWord(SubWord(w[i+5]))^Rcon^w[i+5]^w[i+4]^w[i+3]^w[i+2]^w[i+1]^w[i]
add ecx, 8 ; ecx = &w[i+12]
ret
+@raw_aes192_expand_key_schedule@36 endp
-invert_key_schedule:
- movdqa xmm7, [key_schedule]
- movdqa xmm6, [key_schedule + 0C0h]
- movdqa [inverted_key_schedule], xmm6
- movdqa [inverted_key_schedule + 0C0h], xmm7
+@raw_aes192_invert_key_schedule@8 proc
+ movdqa xmm7, [ecx]
+ movdqa xmm6, [ecx + 0C0h]
+ movdqa [edx], xmm6
+ movdqa [edx + 0C0h], xmm7
- aesimc xmm7, [key_schedule + 10h]
- aesimc xmm6, [key_schedule + 0B0h]
- movdqa [inverted_key_schedule + 10h], xmm6
- movdqa [inverted_key_schedule + 0B0h], xmm7
+ aesimc xmm7, [ecx + 10h]
+ aesimc xmm6, [ecx + 0B0h]
+ movdqa [edx + 10h], xmm6
+ movdqa [edx + 0B0h], xmm7
- aesimc xmm7, [key_schedule + 20h]
- aesimc xmm6, [key_schedule + 0A0h]
- movdqa [inverted_key_schedule + 20h], xmm6
- movdqa [inverted_key_schedule + 0A0h], xmm7
+ aesimc xmm7, [ecx + 20h]
+ aesimc xmm6, [ecx + 0A0h]
+ movdqa [edx + 20h], xmm6
+ movdqa [edx + 0A0h], xmm7
- aesimc xmm7, [key_schedule + 30h]
- aesimc xmm6, [key_schedule + 90h]
- movdqa [inverted_key_schedule + 30h], xmm6
- movdqa [inverted_key_schedule + 90h], xmm7
+ aesimc xmm7, [ecx + 30h]
+ aesimc xmm6, [ecx + 90h]
+ movdqa [edx + 30h], xmm6
+ movdqa [edx + 90h], xmm7
- aesimc xmm7, [key_schedule + 40h]
- aesimc xmm6, [key_schedule + 80h]
- movdqa [inverted_key_schedule + 40h], xmm6
- movdqa [inverted_key_schedule + 80h], xmm7
+ aesimc xmm7, [ecx + 40h]
+ aesimc xmm6, [ecx + 80h]
+ movdqa [edx + 40h], xmm6
+ movdqa [edx + 80h], xmm7
- aesimc xmm7, [key_schedule + 50h]
- aesimc xmm6, [key_schedule + 70h]
- movdqa [inverted_key_schedule + 50h], xmm6
- movdqa [inverted_key_schedule + 70h], xmm7
+ aesimc xmm7, [ecx + 50h]
+ aesimc xmm6, [ecx + 70h]
+ movdqa [edx + 50h], xmm6
+ movdqa [edx + 70h], xmm7
- aesimc xmm7, [key_schedule + 60h]
- movdqa [inverted_key_schedule + 60h], xmm7
+ aesimc xmm7, [ecx + 60h]
+ movdqa [edx + 60h], xmm7
ret
-expand_keys192 endp
+@raw_aes192_invert_key_schedule@8 endp
end
diff --git a/src/aes256.asm b/src/aes256.asm
index acea8e3..e59a253 100644
--- a/src/aes256.asm
+++ b/src/aes256.asm
@@ -6,76 +6,63 @@
.xmm
.model flat
-.data
-
-align 10h
-key_schedule oword 15 dup(0)
-
-align 10h
-inverse_key_schedule oword 15 dup(0)
-
.code
-@raw_aes256ecb_encrypt@48 proc
- call expand_keys256
- pxor xmm0, [key_schedule]
- aesenc xmm0, [key_schedule + 10h]
- aesenc xmm0, [key_schedule + 20h]
- aesenc xmm0, [key_schedule + 30h]
- aesenc xmm0, [key_schedule + 40h]
- aesenc xmm0, [key_schedule + 50h]
- aesenc xmm0, [key_schedule + 60h]
- aesenc xmm0, [key_schedule + 70h]
- aesenc xmm0, [key_schedule + 80h]
- aesenc xmm0, [key_schedule + 90h]
- aesenc xmm0, [key_schedule + 0A0h]
- aesenc xmm0, [key_schedule + 0B0h]
- aesenc xmm0, [key_schedule + 0C0h]
- aesenc xmm0, [key_schedule + 0D0h]
- aesenclast xmm0, [key_schedule + 0E0h]
+@raw_aes256ecb_encrypt@20 proc
+ pxor xmm0, [ecx]
+ aesenc xmm0, [ecx + 10h]
+ aesenc xmm0, [ecx + 20h]
+ aesenc xmm0, [ecx + 30h]
+ aesenc xmm0, [ecx + 40h]
+ aesenc xmm0, [ecx + 50h]
+ aesenc xmm0, [ecx + 60h]
+ aesenc xmm0, [ecx + 70h]
+ aesenc xmm0, [ecx + 80h]
+ aesenc xmm0, [ecx + 90h]
+ aesenc xmm0, [ecx + 0A0h]
+ aesenc xmm0, [ecx + 0B0h]
+ aesenc xmm0, [ecx + 0C0h]
+ aesenc xmm0, [ecx + 0D0h]
+ aesenclast xmm0, [ecx + 0E0h]
ret
-@raw_aes256ecb_encrypt@48 endp
+@raw_aes256ecb_encrypt@20 endp
+
+@raw_aes256cbc_encrypt@24 proc
+ pxor xmm0, [edx]
+ jmp @raw_aes256ecb_encrypt@20
+@raw_aes256cbc_encrypt@24 endp
-@raw_aes256cbc_encrypt@52 proc
+@raw_aes256ecb_decrypt@20 proc
pxor xmm0, [ecx]
- jmp @raw_aes256ecb_encrypt@48
-@raw_aes256cbc_encrypt@52 endp
-
-@raw_aes256ecb_decrypt@48 proc
- call expand_keys256
- pxor xmm0, [inverse_key_schedule]
- aesdec xmm0, [inverse_key_schedule + 10h]
- aesdec xmm0, [inverse_key_schedule + 20h]
- aesdec xmm0, [inverse_key_schedule + 30h]
- aesdec xmm0, [inverse_key_schedule + 40h]
- aesdec xmm0, [inverse_key_schedule + 50h]
- aesdec xmm0, [inverse_key_schedule + 60h]
- aesdec xmm0, [inverse_key_schedule + 70h]
- aesdec xmm0, [inverse_key_schedule + 80h]
- aesdec xmm0, [inverse_key_schedule + 90h]
- aesdec xmm0, [inverse_key_schedule + 0A0h]
- aesdec xmm0, [inverse_key_schedule + 0B0h]
- aesdec xmm0, [inverse_key_schedule + 0C0h]
- aesdec xmm0, [inverse_key_schedule + 0D0h]
- aesdeclast xmm0, [inverse_key_schedule + 0E0h]
+ aesdec xmm0, [ecx + 10h]
+ aesdec xmm0, [ecx + 20h]
+ aesdec xmm0, [ecx + 30h]
+ aesdec xmm0, [ecx + 40h]
+ aesdec xmm0, [ecx + 50h]
+ aesdec xmm0, [ecx + 60h]
+ aesdec xmm0, [ecx + 70h]
+ aesdec xmm0, [ecx + 80h]
+ aesdec xmm0, [ecx + 90h]
+ aesdec xmm0, [ecx + 0A0h]
+ aesdec xmm0, [ecx + 0B0h]
+ aesdec xmm0, [ecx + 0C0h]
+ aesdec xmm0, [ecx + 0D0h]
+ aesdeclast xmm0, [ecx + 0E0h]
ret
-@raw_aes256ecb_decrypt@48 endp
+@raw_aes256ecb_decrypt@20 endp
-@raw_aes256cbc_decrypt@52 proc
- push ecx
- call @raw_aes256ecb_decrypt@48
- pop ecx
- pxor xmm0, [ecx]
+@raw_aes256cbc_decrypt@24 proc
+ call @raw_aes256ecb_decrypt@20
+ pxor xmm0, [edx]
ret
-@raw_aes256cbc_decrypt@52 endp
+@raw_aes256cbc_decrypt@24 endp
-expand_keys256 proc
+@raw_aes256_expand_key_schedule@36 proc
; A "word" (in terms of the FIPS 187 standard) is a 32-bit block.
; Words are denoted by `w[N]`.
;
; A key schedule is composed of 14 "regular" keys and a dumb key for
; the "whitening" step.
- ; It's stored in `key_schedule`.
;
; A key schedule is thus composed of 60 "words".
; The FIPS standard includes an algorithm to calculate these words via
@@ -128,99 +115,97 @@ expand_keys256 proc
; and RotWord using `aeskeygenassist`, which is used in this routine.
;
; Preconditions:
- ; * xmm2[127:96] == w[7],
- ; * xmm2[95:64] == w[6],
- ; * xmm2[63:32] == w[5],
- ; * xmm2[31:0] == w[4],
- ; * xmm1[127:96] == w[3],
- ; * xmm1[95:64] == w[2],
- ; * xmm1[63:32] == w[1],
- ; * xmm1[31:0] == w[0].
-
- movdqa [key_schedule], xmm1 ; sets w[0], w[1], w[2], w[3]
- movdqa [key_schedule + 10h], xmm2 ; sets w[4], w[5], w[6], w[7]
-
- lea ecx, [key_schedule + 20h] ; ecx = &w[8]
-
- aeskeygenassist xmm7, xmm2, 1h ; xmm7[127:96] = RotWord(SubWord(w[7]))^Rcon
+ ; * xmm1[127:96] == w[7],
+ ; * xmm1[95:64] == w[6],
+ ; * xmm1[63:32] == w[5],
+ ; * xmm1[31:0] == w[4],
+ ; * xmm0[127:96] == w[3],
+ ; * xmm0[95:64] == w[2],
+ ; * xmm0[63:32] == w[1],
+ ; * xmm0[31:0] == w[0].
+
+ movdqa [ecx], xmm0 ; sets w[0], w[1], w[2], w[3]
+ movdqa [ecx + 10h], xmm1 ; sets w[4], w[5], w[6], w[7]
+ lea ecx, [ecx + 20h] ; ecx = &w[8]
+
+ aeskeygenassist xmm7, xmm1, 1h ; xmm7[127:96] = RotWord(SubWord(w[7]))^Rcon
pshufd xmm7, xmm7, 0FFh ; xmm7[95:64] = xmm7[63:32] = xmm7[31:0] = xmm7[127:96]
call gen_round_key ; sets w[8], w[9], w[10], w[11]
- aeskeygenassist xmm7, xmm2, 0 ; xmm7[95:64] = SubWord(w[11])
+ aeskeygenassist xmm7, xmm1, 0 ; xmm7[95:64] = SubWord(w[11])
pshufd xmm7, xmm7, 0AAh ; xmm7[127:96] = xmm7[63:32] = xmm7[31:0] = xmm7[95:64]
call gen_round_key ; sets w[12], w[13], w[14], w[15]
- aeskeygenassist xmm7, xmm2, 2h ; xmm7[127:96] = RotWord(SubWord(w[15]))^Rcon
+ aeskeygenassist xmm7, xmm1, 2h ; xmm7[127:96] = RotWord(SubWord(w[15]))^Rcon
pshufd xmm7, xmm7, 0FFh ; xmm7[95:64] = xmm7[63:32] = xmm7[31:0] = xmm7[127:96]
call gen_round_key ; sets w[16], w[17], w[18], w[19]
- aeskeygenassist xmm7, xmm2, 0 ; xmm7[95:64] = SubWord(w[19])
+ aeskeygenassist xmm7, xmm1, 0 ; xmm7[95:64] = SubWord(w[19])
pshufd xmm7, xmm7, 0AAh ; xmm7[127:96] = xmm7[63:32] = xmm7[31:0] = xmm7[95:64]
call gen_round_key ; sets w[20], w[21], w[22], w[23]
- aeskeygenassist xmm7, xmm2, 4h ; xmm7[127:96] = RotWord(SubWord(w[23]))^Rcon
+ aeskeygenassist xmm7, xmm1, 4h ; xmm7[127:96] = RotWord(SubWord(w[23]))^Rcon
pshufd xmm7, xmm7, 0FFh ; xmm7[95:64] = xmm7[63:32] = xmm7[31:0] = xmm7[127:96]
call gen_round_key ; sets w[24], w[25], w[26], w[27]
- aeskeygenassist xmm7, xmm2, 0 ; xmm7[95:64] = SubWord(w[27])
+ aeskeygenassist xmm7, xmm1, 0 ; xmm7[95:64] = SubWord(w[27])
pshufd xmm7, xmm7, 0AAh ; xmm7[127:96] = xmm7[63:32] = xmm7[31:0] = xmm7[95:64]
call gen_round_key ; sets w[28], w[29], w[30], w[31]
- aeskeygenassist xmm7, xmm2, 8h ; xmm7[127:96] = RotWord(SubWord(w[31]))^Rcon
+ aeskeygenassist xmm7, xmm1, 8h ; xmm7[127:96] = RotWord(SubWord(w[31]))^Rcon
pshufd xmm7, xmm7, 0FFh ; xmm7[95:64] = xmm7[63:32] = xmm7[31:0] = xmm7[127:96]
call gen_round_key ; sets w[32], w[33], w[34], w[35]
- aeskeygenassist xmm7, xmm2, 0 ; xmm7[95:64] = SubWord(w[35])
+ aeskeygenassist xmm7, xmm1, 0 ; xmm7[95:64] = SubWord(w[35])
pshufd xmm7, xmm7, 0AAh ; xmm7[127:96] = xmm7[63:32] = xmm7[31:0] = xmm7[95:64]
call gen_round_key ; sets w[36], w[37], w[38], w[39]
- aeskeygenassist xmm7, xmm2, 10h ; xmm7[127:96] = RotWord(SubWord(w[39]))^Rcon
+ aeskeygenassist xmm7, xmm1, 10h ; xmm7[127:96] = RotWord(SubWord(w[39]))^Rcon
pshufd xmm7, xmm7, 0FFh ; xmm7[95:64] = xmm7[63:32] = xmm7[31:0] = xmm7[127:96]
call gen_round_key ; sets w[40], w[41], w[42], w[43]
- aeskeygenassist xmm7, xmm2, 0 ; xmm7[95:64] = SubWord(w[43])
+ aeskeygenassist xmm7, xmm1, 0 ; xmm7[95:64] = SubWord(w[43])
pshufd xmm7, xmm7, 0AAh ; xmm7[127:96] = xmm7[63:32] = xmm7[31:0] = xmm7[95:64]
call gen_round_key ; sets w[44], w[45], w[46], w[47]
- aeskeygenassist xmm7, xmm2, 20h ; xmm7[127:96] = RotWord(SubWord(w[47]))^Rcon
+ aeskeygenassist xmm7, xmm1, 20h ; xmm7[127:96] = RotWord(SubWord(w[47]))^Rcon
pshufd xmm7, xmm7, 0FFh ; xmm7[95:64] = xmm7[63:32] = xmm7[31:0] = xmm7[127:96]
call gen_round_key ; sets w[48], w[49], w[50], w[51]
- aeskeygenassist xmm7, xmm2, 0 ; xmm7[95:64] = SubWord(w[51])
+ aeskeygenassist xmm7, xmm1, 0 ; xmm7[95:64] = SubWord(w[51])
pshufd xmm7, xmm7, 0AAh ; xmm7[127:96] = xmm7[63:32] = xmm7[31:0] = xmm7[95:64]
call gen_round_key ; sets w[52], w[53], w[54], w[55]
- aeskeygenassist xmm7, xmm2, 40h ; xmm7[127:96] = RotWord(SubWord(w[55]))^Rcon
+ aeskeygenassist xmm7, xmm1, 40h ; xmm7[127:96] = RotWord(SubWord(w[55]))^Rcon
pshufd xmm7, xmm7, 0FFh ; xmm7[95:64] = xmm7[63:32] = xmm7[31:0] = xmm7[127:96]
call gen_round_key ; sets w[56], w[57], w[58], w[59]
- call invert_key_schedule
ret
gen_round_key:
; Preconditions:
- ; * xmm2[127:96] == w[i+7],
- ; * xmm2[95:64] == w[i+6],
- ; * xmm2[63:32] == w[i+5],
- ; * xmm2[31:0] == w[i+4],
- ; * xmm1[127:96] == w[i+3],
- ; * xmm1[95:64] == w[i+2],
- ; * xmm1[63:32] == w[i+1],
- ; * xmm1[31:0] == w[i],
+ ; * xmm1[127:96] == w[i+7],
+ ; * xmm1[95:64] == w[i+6],
+ ; * xmm1[63:32] == w[i+5],
+ ; * xmm1[31:0] == w[i+4],
+ ; * xmm0[127:96] == w[i+3],
+ ; * xmm0[95:64] == w[i+2],
+ ; * xmm0[63:32] == w[i+1],
+ ; * xmm0[31:0] == w[i],
; * xmm7[127:96] == xmm7[95:64] == xmm7[63:32] == xmm7[31:0] == HWGEN,
; where HWGEN is either RotWord(SubWord(w[i+7]))^Rcon or SubWord(w[i+7]),
; depending on the number of the round being processed,
; * ecx == &w[i+8].
;
; Postconditions:
- ; * xmm2[127:96] == w[i+11] == HWGEN^w[i+3]^w[i+2]^w[i+1]^w[i],
- ; * xmm2[95:64] == w[i+10] == HWGEN^w[i+2]^w[i+1]^w[i],
- ; * xmm2[63:32] == w[i+9] == HWGEN^w[i+1]^w[i],
- ; * xmm2[31:0] == w[i+8] == HWGEN^w[i],
- ; * xmm1[127:96] == w[i+7],
- ; * xmm1[95:64] == w[i+6],
- ; * xmm1[63:32] == w[i+5],
- ; * xmm1[31:0] == w[i+4],
+ ; * xmm1[127:96] == w[i+11] == HWGEN^w[i+3]^w[i+2]^w[i+1]^w[i],
+ ; * xmm1[95:64] == w[i+10] == HWGEN^w[i+2]^w[i+1]^w[i],
+ ; * xmm1[63:32] == w[i+9] == HWGEN^w[i+1]^w[i],
+ ; * xmm1[31:0] == w[i+8] == HWGEN^w[i],
+ ; * xmm0[127:96] == w[i+7],
+ ; * xmm0[95:64] == w[i+6],
+ ; * xmm0[63:32] == w[i+5],
+ ; * xmm0[31:0] == w[i+4],
; * ecx == &w[i+12],
; * the value in xmm6 is also modified.
@@ -229,83 +214,84 @@ gen_round_key:
; w[i+2]^w[i+1]^w[i],
; w[i+1]^w[i] and
; w[i].
- movdqa xmm6, xmm1 ; xmm6 = xmm1
+ movdqa xmm6, xmm0 ; xmm6 = xmm0
pslldq xmm6, 4 ; xmm6 <<= 32
- pxor xmm1, xmm6 ; xmm1 ^= xmm6
+ pxor xmm0, xmm6 ; xmm0 ^= xmm6
pslldq xmm6, 4 ; xmm6 <<= 32
- pxor xmm1, xmm6 ; xmm1 ^= xmm6
+ pxor xmm0, xmm6 ; xmm0 ^= xmm6
pslldq xmm6, 4 ; xmm6 <<= 32
- pxor xmm1, xmm6 ; xmm1 ^= xmm6
- ; xmm1[127:96] == w[i+3]^w[i+2]^w[i+1]^w[i]
- ; xmm1[95:64] == w[i+2]^w[i+1]^w[i]
- ; xmm1[63:32] == w[i+1]^w[i]
- ; xmm1[31:0] == w[i]
+ pxor xmm0, xmm6 ; xmm0 ^= xmm6
+ ; xmm0[127:96] == w[i+3]^w[i+2]^w[i+1]^w[i]
+ ; xmm0[95:64] == w[i+2]^w[i+1]^w[i]
+ ; xmm0[63:32] == w[i+1]^w[i]
+ ; xmm0[31:0] == w[i]
; Calculate
; HWGEN^w[i+3]^w[i+2]^w[i+1]^w[i],
; HWGEN^w[i+2]^w[i+1]^w[i],
; HWGEN^w[i+1]^w[i] and
; HWGEN^w[i].
- pxor xmm1, xmm7 ; xmm1 ^= xmm7
- ; xmm1[127:96] == w[i+11] == HWGEN^w[i+3]^w[i+2]^w[i+1]^w[i]
- ; xmm1[95:64] == w[i+10] == HWGEN^w[i+2]^w[i+1]^w[i]
- ; xmm1[63:32] == w[i+9] == HWGEN^w[i+1]^w[i]
- ; xmm1[31:0] == w[i+8] == HWGEN^w[i]
+ pxor xmm0, xmm7 ; xmm0 ^= xmm7
+ ; xmm0[127:96] == w[i+11] == HWGEN^w[i+3]^w[i+2]^w[i+1]^w[i]
+ ; xmm0[95:64] == w[i+10] == HWGEN^w[i+2]^w[i+1]^w[i]
+ ; xmm0[63:32] == w[i+9] == HWGEN^w[i+1]^w[i]
+ ; xmm0[31:0] == w[i+8] == HWGEN^w[i]
; Set w[i+8], w[i+9], w[i+10] and w[i+11].
- movdqa [ecx], xmm1 ; w[i+8] = HWGEN^w[i]
+ movdqa [ecx], xmm0 ; w[i+8] = HWGEN^w[i]
; w[i+9] = HWGEN^w[i+1]^w[i]
; w[i+10] = HWGEN^w[i+2]^w[i+1]^w[i]
; w[i+11] = HWGEN^w[i+3]^w[i+2]^w[i+1]^w[i]
add ecx, 10h ; ecx = &w[i+12]
- ; Swap the values in xmm1 and xmm2.
- pxor xmm1, xmm2
- pxor xmm2, xmm1
- pxor xmm1, xmm2
+ ; Swap the values in xmm0 and xmm1.
+ pxor xmm0, xmm1
+ pxor xmm1, xmm0
+ pxor xmm0, xmm1
ret
-
-invert_key_schedule:
- movdqa xmm7, [key_schedule ]
- movdqa xmm6, [key_schedule + 0E0h]
- movdqa [inverse_key_schedule ], xmm6
- movdqa [inverse_key_schedule + 0E0h], xmm7
-
- aesimc xmm7, [key_schedule + 10h]
- aesimc xmm6, [key_schedule + 0D0h]
- movdqa [inverse_key_schedule + 10h], xmm6
- movdqa [inverse_key_schedule + 0D0h], xmm7
-
- aesimc xmm7, [key_schedule + 20h]
- aesimc xmm6, [key_schedule + 0C0h]
- movdqa [inverse_key_schedule + 20h], xmm6
- movdqa [inverse_key_schedule + 0C0h], xmm7
-
- aesimc xmm7, [key_schedule + 30h]
- aesimc xmm6, [key_schedule + 0B0h]
- movdqa [inverse_key_schedule + 30h], xmm6
- movdqa [inverse_key_schedule + 0B0h], xmm7
-
- aesimc xmm7, [key_schedule + 40h]
- aesimc xmm6, [key_schedule + 0A0h]
- movdqa [inverse_key_schedule + 40h], xmm6
- movdqa [inverse_key_schedule + 0A0h], xmm7
-
- aesimc xmm7, [key_schedule + 50h]
- aesimc xmm6, [key_schedule + 90h]
- movdqa [inverse_key_schedule + 50h], xmm6
- movdqa [inverse_key_schedule + 90h], xmm7
-
- aesimc xmm7, [key_schedule + 60h]
- aesimc xmm6, [key_schedule + 80h]
- movdqa [inverse_key_schedule + 60h], xmm6
- movdqa [inverse_key_schedule + 80h], xmm7
-
- aesimc xmm7, [key_schedule + 70h]
- movdqa [inverse_key_schedule + 70h], xmm7
+@raw_aes256_expand_key_schedule@36 endp
+
+@raw_aes256_invert_key_schedule@8 proc
+ movdqa xmm7, [ecx]
+ movdqa xmm6, [ecx + 0E0h]
+ movdqa [edx], xmm6
+ movdqa [edx + 0E0h], xmm7
+
+ aesimc xmm7, [ecx + 10h]
+ aesimc xmm6, [ecx + 0D0h]
+ movdqa [edx + 10h], xmm6
+ movdqa [edx + 0D0h], xmm7
+
+ aesimc xmm7, [ecx + 20h]
+ aesimc xmm6, [ecx + 0C0h]
+ movdqa [edx + 20h], xmm6
+ movdqa [edx + 0C0h], xmm7
+
+ aesimc xmm7, [ecx + 30h]
+ aesimc xmm6, [ecx + 0B0h]
+ movdqa [edx + 30h], xmm6
+ movdqa [edx + 0B0h], xmm7
+
+ aesimc xmm7, [ecx + 40h]
+ aesimc xmm6, [ecx + 0A0h]
+ movdqa [edx + 40h], xmm6
+ movdqa [edx + 0A0h], xmm7
+
+ aesimc xmm7, [ecx + 50h]
+ aesimc xmm6, [ecx + 90h]
+ movdqa [edx + 50h], xmm6
+ movdqa [edx + 90h], xmm7
+
+ aesimc xmm7, [ecx + 60h]
+ aesimc xmm6, [ecx + 80h]
+ movdqa [edx + 60h], xmm6
+ movdqa [edx + 80h], xmm7
+
+ aesimc xmm7, [ecx + 70h]
+ movdqa [edx + 70h], xmm7
ret
-expand_keys256 endp
+@raw_aes256_invert_key_schedule@8 endp
end
diff --git a/test/aes128cbc_decrypt_block.c b/test/aes128cbc_decrypt_block.c
index 1a77cdf..3bd5d74 100644
--- a/test/aes128cbc_decrypt_block.c
+++ b/test/aes128cbc_decrypt_block.c
@@ -20,6 +20,7 @@ static void exit_with_usage(const char* argv0)
int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, key, cipher, iv;
+ __declspec(align(16)) Aes128KeySchedule key_schedule, inverted_schedule;
if (argc < 3)
exit_with_usage(argv[0]);
@@ -36,6 +37,9 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes128_expand_key_schedule(key, &key_schedule);
+ aes128_invert_key_schedule(&key_schedule, &inverted_schedule);
+
for (int i = 3; i < argc; ++i)
{
if (parse_aes_block128(&cipher, argv[i]) != 0)
@@ -43,7 +47,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- plain = aes128cbc_decrypt(cipher, key, &iv);
+ plain = aes128cbc_decrypt(cipher, &inverted_schedule, &iv);
print_aes_block128(&plain);
}
diff --git a/test/aes128cbc_encrypt_block.c b/test/aes128cbc_encrypt_block.c
index 31bb479..e35a1bf 100644
--- a/test/aes128cbc_encrypt_block.c
+++ b/test/aes128cbc_encrypt_block.c
@@ -20,6 +20,7 @@ static void exit_with_usage(const char* argv0)
int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, key, cipher, iv;
+ __declspec(align(16)) Aes128KeySchedule key_schedule;
if (argc < 3)
exit_with_usage(argv[0]);
@@ -36,6 +37,8 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes128_expand_key_schedule(key, &key_schedule);
+
for (int i = 3; i < argc; ++i)
{
if (parse_aes_block128(&plain, argv[i]) != 0)
@@ -43,7 +46,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- cipher = aes128cbc_encrypt(plain, key, &iv);
+ cipher = aes128cbc_encrypt(plain, &key_schedule, &iv);
print_aes_block128(&cipher);
}
diff --git a/test/aes128ecb_decrypt_block.c b/test/aes128ecb_decrypt_block.c
index 188524b..70c7285 100644
--- a/test/aes128ecb_decrypt_block.c
+++ b/test/aes128ecb_decrypt_block.c
@@ -20,6 +20,7 @@ static void exit_with_usage(const char* argv0)
int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, key, cipher;
+ __declspec(align(16)) Aes128KeySchedule key_schedule, inverted_schedule;
if (argc < 2)
exit_with_usage(argv[0]);
@@ -30,6 +31,9 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes128_expand_key_schedule(key, &key_schedule);
+ aes128_invert_key_schedule(&key_schedule, &inverted_schedule);
+
for (int i = 2; i < argc; ++i)
{
if (parse_aes_block128(&cipher, argv[i]) != 0)
@@ -37,7 +41,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- plain = aes128ecb_decrypt(cipher, key);
+ plain = aes128ecb_decrypt(cipher, &inverted_schedule);
print_aes_block128(&plain);
}
diff --git a/test/aes128ecb_encrypt_block.c b/test/aes128ecb_encrypt_block.c
index 27c7a0e..8f0bf0a 100644
--- a/test/aes128ecb_encrypt_block.c
+++ b/test/aes128ecb_encrypt_block.c
@@ -20,6 +20,7 @@ static void exit_with_usage(const char* argv0)
int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, key, cipher;
+ __declspec(align(16)) Aes128KeySchedule key_schedule;
if (argc < 2)
exit_with_usage(argv[0]);
@@ -30,6 +31,8 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes128_expand_key_schedule(key, &key_schedule);
+
for (int i = 2; i < argc; ++i)
{
if (parse_aes_block128(&plain, argv[i]) != 0)
@@ -37,7 +40,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- cipher = aes128ecb_encrypt(plain, key);
+ cipher = aes128ecb_encrypt(plain, &key_schedule);
print_aes_block128(&cipher);
}
diff --git a/test/aes192cbc_decrypt_block.c b/test/aes192cbc_decrypt_block.c
index 0fddf43..3d912ba 100644
--- a/test/aes192cbc_decrypt_block.c
+++ b/test/aes192cbc_decrypt_block.c
@@ -21,6 +21,7 @@ int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, cipher, iv;
__declspec(align(16)) AesBlock192 key;
+ __declspec(align(16)) Aes192KeySchedule key_schedule, inverted_schedule;
if (argc < 3)
exit_with_usage(argv[0]);
@@ -37,6 +38,9 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes192_expand_key_schedule(&key, &key_schedule);
+ aes192_invert_key_schedule(&key_schedule, &inverted_schedule);
+
for (int i = 3; i < argc; ++i)
{
if (parse_aes_block128(&cipher, argv[i]) != 0)
@@ -44,7 +48,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- plain = aes192cbc_decrypt(cipher, &key, &iv);
+ plain = aes192cbc_decrypt(cipher, &inverted_schedule, &iv);
print_aes_block128(&plain);
}
diff --git a/test/aes192cbc_encrypt_block.c b/test/aes192cbc_encrypt_block.c
index 069c408..21765f5 100644
--- a/test/aes192cbc_encrypt_block.c
+++ b/test/aes192cbc_encrypt_block.c
@@ -21,6 +21,7 @@ int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, cipher, iv;
__declspec(align(16)) AesBlock192 key;
+ __declspec(align(16)) Aes192KeySchedule key_schedule;
if (argc < 3)
exit_with_usage(argv[0]);
@@ -37,6 +38,8 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes192_expand_key_schedule(&key, &key_schedule);
+
for (int i = 3; i < argc; ++i)
{
if (parse_aes_block128(&plain, argv[i]) != 0)
@@ -44,7 +47,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- cipher = aes192cbc_encrypt(plain, &key, &iv);
+ cipher = aes192cbc_encrypt(plain, &key_schedule, &iv);
print_aes_block128(&cipher);
}
diff --git a/test/aes192ecb_decrypt_block.c b/test/aes192ecb_decrypt_block.c
index 73827a2..e6e749a 100644
--- a/test/aes192ecb_decrypt_block.c
+++ b/test/aes192ecb_decrypt_block.c
@@ -21,6 +21,7 @@ int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, cipher;
__declspec(align(16)) AesBlock192 key;
+ __declspec(align(16)) Aes192KeySchedule key_schedule, inverted_schedule;
if (argc < 2)
exit_with_usage(argv[0]);
@@ -31,6 +32,9 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes192_expand_key_schedule(&key, &key_schedule);
+ aes192_invert_key_schedule(&key_schedule, &inverted_schedule);
+
for (int i = 2; i < argc; ++i)
{
if (parse_aes_block128(&cipher, argv[i]) != 0)
@@ -38,7 +42,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- plain = aes192ecb_decrypt(cipher, &key);
+ plain = aes192ecb_decrypt(cipher, &inverted_schedule);
print_aes_block128(&plain);
}
diff --git a/test/aes192ecb_encrypt_block.c b/test/aes192ecb_encrypt_block.c
index 9f2db50..2b4f688 100644
--- a/test/aes192ecb_encrypt_block.c
+++ b/test/aes192ecb_encrypt_block.c
@@ -21,6 +21,7 @@ int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, cipher;
__declspec(align(16)) AesBlock192 key;
+ __declspec(align(16)) Aes192KeySchedule key_schedule;
if (argc < 2)
exit_with_usage(argv[0]);
@@ -31,6 +32,8 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes192_expand_key_schedule(&key, &key_schedule);
+
for (int i = 2; i < argc; ++i)
{
if (parse_aes_block128(&plain, argv[i]) != 0)
@@ -38,7 +41,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- cipher = aes192ecb_encrypt(plain, &key);
+ cipher = aes192ecb_encrypt(plain, &key_schedule);
print_aes_block128(&cipher);
}
diff --git a/test/aes256cbc_decrypt_block.c b/test/aes256cbc_decrypt_block.c
index bac98e3..bb082d5 100644
--- a/test/aes256cbc_decrypt_block.c
+++ b/test/aes256cbc_decrypt_block.c
@@ -21,6 +21,7 @@ int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, cipher, iv;
__declspec(align(16)) AesBlock256 key;
+ __declspec(align(16)) Aes256KeySchedule key_schedule, inverted_schedule;
if (argc < 3)
exit_with_usage(argv[0]);
@@ -37,6 +38,9 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes256_expand_key_schedule(&key, &key_schedule);
+ aes256_invert_key_schedule(&key_schedule, &inverted_schedule);
+
for (int i = 3; i < argc; ++i)
{
if (parse_aes_block128(&cipher, argv[i]) != 0)
@@ -44,7 +48,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- plain = aes256cbc_decrypt(cipher, &key, &iv);
+ plain = aes256cbc_decrypt(cipher, &inverted_schedule, &iv);
print_aes_block128(&plain);
}
diff --git a/test/aes256cbc_encrypt_block.c b/test/aes256cbc_encrypt_block.c
index 08ca11b..1436020 100644
--- a/test/aes256cbc_encrypt_block.c
+++ b/test/aes256cbc_encrypt_block.c
@@ -21,6 +21,7 @@ int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, cipher, iv;
__declspec(align(16)) AesBlock256 key;
+ __declspec(align(16)) Aes256KeySchedule key_schedule;
if (argc < 3)
exit_with_usage(argv[0]);
@@ -37,6 +38,8 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes256_expand_key_schedule(&key, &key_schedule);
+
for (int i = 3; i < argc; ++i)
{
if (parse_aes_block128(&plain, argv[i]) != 0)
@@ -44,7 +47,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- cipher = aes256cbc_encrypt(plain, &key, &iv);
+ cipher = aes256cbc_encrypt(plain, &key_schedule, &iv);
print_aes_block128(&cipher);
}
diff --git a/test/aes256ecb_decrypt_block.c b/test/aes256ecb_decrypt_block.c
index f3126fe..a290eed 100644
--- a/test/aes256ecb_decrypt_block.c
+++ b/test/aes256ecb_decrypt_block.c
@@ -21,6 +21,7 @@ int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, cipher;
__declspec(align(16)) AesBlock256 key;
+ __declspec(align(16)) Aes256KeySchedule key_schedule, inverted_schedule;
if (argc < 2)
exit_with_usage(argv[0]);
@@ -31,6 +32,9 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes256_expand_key_schedule(&key, &key_schedule);
+ aes256_invert_key_schedule(&key_schedule, &inverted_schedule);
+
for (int i = 2; i < argc; ++i)
{
if (parse_aes_block128(&cipher, argv[i]) != 0)
@@ -38,7 +42,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- plain = aes256ecb_decrypt(cipher, &key);
+ plain = aes256ecb_decrypt(cipher, &inverted_schedule);
print_aes_block128(&plain);
}
diff --git a/test/aes256ecb_encrypt_block.c b/test/aes256ecb_encrypt_block.c
index 3fb979d..2a4899e 100644
--- a/test/aes256ecb_encrypt_block.c
+++ b/test/aes256ecb_encrypt_block.c
@@ -21,6 +21,7 @@ int main(int argc, char** argv)
{
__declspec(align(16)) AesBlock128 plain, cipher;
__declspec(align(16)) AesBlock256 key;
+ __declspec(align(16)) Aes256KeySchedule key_schedule;
if (argc < 2)
exit_with_usage(argv[0]);
@@ -31,6 +32,8 @@ int main(int argc, char** argv)
exit_with_usage(argv[0]);
}
+ aes256_expand_key_schedule(&key, &key_schedule);
+
for (int i = 2; i < argc; ++i)
{
if (parse_aes_block128(&plain, argv[i]) != 0)
@@ -38,7 +41,7 @@ int main(int argc, char** argv)
fprintf(stderr, "Invalid 128-bit AES block '%s'\n", argv[i]);
continue;
}
- cipher = aes256ecb_encrypt(plain, &key);
+ cipher = aes256ecb_encrypt(plain, &key_schedule);
print_aes_block128(&cipher);
}