squeezing out a little more speed

diff --git a/aes_enc-asm.S b/aes_enc-asm.S
index 72b9e6ecbb5d3cbe0b5856436beae9e6015a8147..a2235cf2cef7f48f1a478646c4e4b76545258a68 100644 (file)
--- a/aes_enc-asm.S
+++ b/aes_enc-asm.S
@@ -207,7 +207,7 @@ aes_encrypt_core:
        eor T1, xREDUCER
 3:                     
        eor T1, r1
-       eor T1, ST01
+       eor ST01, T1
        
        mov T2, ST02
        eor T2, ST03
@@ -216,7 +216,7 @@ aes_encrypt_core:
        eor T2, xREDUCER
 3:
        eor T2, r1
-       eor T2, ST02
+       eor ST02, T2
        
        mov T3, ST03
        eor T3, ST00
@@ -225,13 +225,9 @@ aes_encrypt_core:
        eor T3, xREDUCER
 3:     
        eor T3, r1
-       eor T3, ST03
+       eor ST03, T3
                
        mov ST00, T0
-       mov ST01, T1
-       mov ST02, T2
-       mov ST03, T3
-
 
        mov r1, ST10
        eor r1, ST11
@@ -254,7 +250,7 @@ aes_encrypt_core:
        eor T1, xREDUCER
 3:                     
        eor T1, r1
-       eor T1, ST11
+       eor ST11, T1
        
        mov T2, ST12
        eor T2, ST13
@@ -263,7 +259,7 @@ aes_encrypt_core:
        eor T2, xREDUCER
 3:
        eor T2, r1
-       eor T2, ST12
+       eor ST12, T2
        
        mov T3, ST13
        eor T3, ST10
@@ -272,12 +268,9 @@ aes_encrypt_core:
        eor T3, xREDUCER
 3:     
        eor T3, r1
-       eor T3, ST13
+       eor ST13, T3
                
        mov ST10, T0
-       mov ST11, T1
-       mov ST12, T2
-       mov ST13, T3
 
        mov r1, ST20
        eor r1, ST21
@@ -300,7 +293,7 @@ aes_encrypt_core:
        eor T1, xREDUCER
 3:                     
        eor T1, r1
-       eor T1, ST21
+       eor ST21, T1
        
        mov T2, ST22
        eor T2, ST23
@@ -309,7 +302,7 @@ aes_encrypt_core:
        eor T2, xREDUCER
 3:
        eor T2, r1
-       eor T2, ST22
+       eor ST22, T2
        
        mov T3, ST23
        eor T3, ST20
@@ -318,12 +311,9 @@ aes_encrypt_core:
        eor T3, xREDUCER
 3:     
        eor T3, r1
-       eor T3, ST23
+       eor ST23, T3
                
        mov ST20, T0
-       mov ST21, T1
-       mov ST22, T2
-       mov ST23, T3
 
        mov r1, ST30
        eor r1, ST31
@@ -346,7 +336,7 @@ aes_encrypt_core:
        eor T1, xREDUCER
 3:                     
        eor T1, r1
-       eor T1, ST31
+       eor ST31, T1
        
        mov T2, ST32
        eor T2, ST33
@@ -355,7 +345,7 @@ aes_encrypt_core:
        eor T2, xREDUCER
 3:
        eor T2, r1
-       eor T2, ST32
+       eor ST32, T2
        
        mov T3, ST33
        eor T3, ST30
@@ -364,12 +354,9 @@ aes_encrypt_core:
        eor T3, xREDUCER
 3:     
        eor T3, r1
-       eor T3, ST33
+       eor ST33, T3
                
        mov ST30, T0
-       mov ST31, T1
-       mov ST32, T2
-       mov ST33, T3
 
        /* mix colums (rows) done */
 

diff --git a/host/gf256_table_gen.c b/host/gf256_table_gen.c
index 87887d64cd2bb7b80bfe73faf5ab2492458588a3..0d55db48f44acaa66954151ae676f1166129f436 100644 (file)
--- a/host/gf256_table_gen.c
+++ b/host/gf256_table_gen.c
@@ -43,6 +43,23 @@ char* fmt_strings[]={"%d", "%3d",
                      "0x%X", "0x%2.2X",
                      "0%o", "0%3o"};
                      
+void print_help(void){
+       puts(
+"gf256_table_gen [-cCbBdhoaApP] [-l min] [-u max]\n"
+"                [-f formatstring] [-o file] -r reducer value {values} \n"
+"\n"
+" -c/-C turn C-header on/off (default: on)\n"
+" -b/-B turn braces on/off (defaul: on)\n"
+" -d/-h/-8 print values in decimal/hexadecimal/octal (default: hexadecimal)\n"
+" -a/-a  turn alignment on/off (default: on)\n"
+" -p/-P  turn base prefix on/off (default: on)\n"
+" -r reducer reducer is the reduction polynomial represented as numeric value\n"
+" -l min starting value for multiplication table (default: 0)\n"
+" -u max upper limit value for multiplication table (default: 255)\n"          
+" -f formatstring string for formating the output of a singel multiplicxation\n"
+" -o file outputfile\n"
+}
+                     
 int main(int argc, char** argv){
        int i,j;
        int c;
@@ -65,6 +82,7 @@ int main(int argc, char** argv){
                
                /* These options don't set a flag.
                   We distinguish them by their indices. */
+               {"help",    no_argument      , 0, '?'},     
                {"reducer", required_argument, 0, 'r'},
                {"min",     required_argument, 0, 'l'},
                {"max",     required_argument, 0, 'u'},
@@ -76,7 +94,7 @@ int main(int argc, char** argv){
        
        unsigned long ul_a;
        int columns=8;
-       unsigned long ul_reducer=0x1b, max=0xff, min=0x00;
+       unsigned long ul_reducer=0x00, max=0xff, min=0x00;
        uint8_t reducer, a;
        char** eptr;
        FILE* of = stdout;
@@ -90,7 +108,7 @@ int main(int argc, char** argv){
                        case 'B': print_braces=0; break;
                        case 'h': print_base = 1; break;
                        case 'd': print_base = 0; break;
-                       case 'o': print_base = 2; break;
+                       case '8': print_base = 2; break;
                        case 'a': print_align =1; break;
                        case 'A': print_align =0; break;
                        case 'p': print_prefix=1; break;
@@ -117,10 +135,16 @@ int main(int argc, char** argv){
                                          }
                                          break;
                        case 'f': fmt = optarg; break;
-                       default: break;
+                       case '?': 
+                       default: 
+                               print_help();
+                               break;
                }
        }
-       reducer = ul_reducer&0xff;
+       if(!reducer = ul_reducer&0xff){
+                       fprintf(stderr, "You must specify a reductionpolynomial!\n", argv[i]);
+                       return -1;
+       }
        if(!fmt)
                fmt = fmt_strings[print_prefix*6+print_base*2+print_align];