__attribute__((noinline)) void Piccolo80vperm_key_schedule(const u8* masterKey, u8* roundKeys)
{
/* Note : master key in rdi and round keys in rsi */
/* __cdecl calling convention */
asm (".intel_syntax noprefix");
Push_All_Regs();
asm("lea "tostr(PiccoloTcon80_)", [rip + PiccoloTcon80]");
/* Isolate interleaved k2 and k3 */
format_input(rdi+4, rdi+14, xmm0, xmm13, xmm14, xmm15);
/* Shuffle the result */
asm("pshufb xmm0, [rip + PiccoloKSShufa]");
/* Isolate interleaved k0 and k1 */
format_input(rdi, rdi+10, xmm1, xmm13, xmm14, xmm15);
asm("movdqa xmm3, xmm1");
/* Shuffle the result */
asm("pshufb xmm1, [rip + PiccoloKSShufa]");
/* Get the pre-whitening keys */
asm("pshufb xmm3, [rip + PiccoloKSShufwa]");
/* Isolate interleaved k4 */
format_input(rdi+6, rdi+16, xmm2, xmm13, xmm14, xmm15);
asm("movdqa xmm4, xmm2");
/* Shuffle the result */
asm("pshufb xmm2, [rip + PiccoloKSShufb]");
/* Get the post-whitening keys */
asm("pshufb xmm4, [rip + PiccoloKSShufwb]");
/* Go for the transformed key schedule */
asm("xor rcx, rcx");
/* Store the pre-whitening keys */
asm("movdqa [rsi+rcx], xmm3");
asm("add rcx, 16");
KEYSCHED80();
KEYSCHED80();
KEYSCHED80();
KEYSCHED80();
KEYSCHED80();
/* Store the post-whitening keys */
asm("movdqa [rsi+rcx], xmm4");
Pop_All_Regs();
asm (".att_syntax noprefix");
return;
}
static int crypto_ccm_auth(struct aead_request *req, struct scatterlist *plain,
unsigned int cryptlen)
{
struct crypto_aead *aead = crypto_aead_reqtfm(req);
struct crypto_ccm_ctx *ctx = crypto_aead_ctx(aead);
struct crypto_ccm_req_priv_ctx *pctx = crypto_ccm_reqctx(req);
struct crypto_cipher *cipher = ctx->cipher;
unsigned int assoclen = req->assoclen;
u8 *odata = pctx->odata;
u8 *idata = pctx->idata;
int err;
/* format control data for input */
err = format_input(odata, req, cryptlen);
if (err)
goto out;
/* encrypt first block to use as start in computing mac */
crypto_cipher_encrypt_one(cipher, odata, odata);
/* format associated data and compute into mac */
if (assoclen) {
pctx->ilen = format_adata(idata, assoclen);
get_data_to_compute(cipher, pctx, req->assoc, req->assoclen);
} else {
pctx->ilen = 0;
}
/* compute plaintext into mac */
if (cryptlen)
get_data_to_compute(cipher, pctx, plain, cryptlen);
out:
return err;
}
void PRESENT80_InitKEY(const unsigned char* userkey){
ALIGN16(unsigned char tk[16]);
memset(tk, 0, sizeof(tk));
memcpy(tk, userkey, 2);
memcpy(tk+8, userkey, 2);
dqword lowkey = LOAD(tk);
memcpy(tk, userkey+2, 8);
memcpy(tk+8, userkey+2, 8);
dqword highkey = LOAD(tk);
rks[0] = highkey;
format_input(&rks[0]);
for(int i=0; i<31; i++){
key_schedule(&highkey, &lowkey, i);
rks[i+1] = highkey;
format_input(&rks[i+1]);
}
}
//////////////////////////////// MAIN LAND //////////////////////////////
int main(int argc, char **argv)
{
char archname[261];
char **argv_f;
errno = 0;
format_input(argc,argv,&argv_f);
cla_opt_handle(argc,argv_f,archname);
return 0;
}
void PRESENT80_enc(dqword *input, const unsigned char* userkey){
format_input(input);
for(int i=0; i<31; i++){
addRoundKey(input, &rks[i]);
sBoxLayer(input);
pLayer(input);
}
addRoundKey(input, &rks[31]);
format_output(input);
}
char *add_some_space(char *command)
{
char *new_command;
new_command = malloc(ft_strlen(command) + 2 * nb_tokens(command) + 1);
if (new_command == NULL)
return (NULL);
format_input(command, new_command);
free(command);
return (new_command);
}
TokenizerT *TKCreate( char *separators, char *ts ) {
// Allocate memory for the TokenizerT object
TokenizerT *tokenStructP = malloc( sizeof( TokenizerT ) );
tokenStructP->delimiter = malloc( sizeof(char) * (strlen( separators ) + 1) );
tokenStructP->inputString = malloc( sizeof(char) * (strlen( ts ) + 1 ) );
// Initialize Variables
copy_string(tokenStructP->delimiter, separators);
copy_string(tokenStructP->inputString, ts);
// Format escape commands into single byte characters
format_input(tokenStructP->delimiter);
format_input(tokenStructP->inputString);
if(&tokenStructP->inputString != &ts) {
return tokenStructP;
}
else {
return NULL;
}
}
int main(int argc, char *argv[]) {
uchar *input_seqfile;
uchar *temp;
uint n_seq, x;
int l;
FILE *indfile, *strfile, *binfile;
configure_logmsg(MSG_DEBUG1);
parse_arguments(&input_seqfile, argc, argv);
configure_logmsg(verbosity_level);
logmsg(MSG_INFO,"Output basename set to %s\n",output_basename);
l = strlen(output_basename);
MA(temp, (l+6)*sizeof(char));
strcpy(temp, output_basename);
strcat(temp, ".ind");
indfile = openfile(temp, "w", "sequence index file");
x = INDFILE_MAGIC;
fwrite(&x, sizeof(uint), 1, indfile);
strcpy(temp, output_basename);
strcat(temp, ".seq");
strfile = openfile(temp, "w", "sequence string file");
x = STRFILE_MAGIC;
fwrite(&x, sizeof(uint), 1, strfile);
strcpy(temp, output_basename);
strcat(temp, ".sbin");
binfile = openfile(temp, "w", "sequence binary file");
x = BINFILE_MAGIC;
fwrite(&x, sizeof(uint), 1, binfile);
n_seq = format_input(input_seqfile, indfile, strfile, binfile);
fclose(indfile);
fclose(strfile);
fclose(binfile);
logmsg(MSG_INFO,"%d sequences formatted\n",n_seq);
return 0;
}
/* main */
int main() {
bool exit_flag = false;
int oldest_history = HISTORY_SIZE - 1,
status;
char *username,
input[MAX_INPUT_SIZE],
history[HISTORY_SIZE][MAX_INPUT_SIZE] = {'\0'},
*tokens[CMD_MAX];
signal(SIGINT, handle_sigint);
//shell loop
while(!exit_flag) {
//print the prompt and get the input
printf("%s> " , get_username());
fgets(input, MAX_INPUT_SIZE, stdin);
format_input(input);
//built-in exit function
if(!strcmp(input, "exit")) exit_flag = true;
//built-in history function
else if(!strcmp(input, "history")) {
int i;
for(i = oldest_history+1; i!=oldest_history; i = (i+1)%HISTORY_SIZE) {
if(history[i][0] != '\0') printf("%s\n", history[i]);
}
}
//non-built-in function
else if(strcmp(input, "")) {
int token_index,
num_tokens;
pid_t pid;
char *tokens[CMD_MAX];
//get the token list and size
num_tokens = make_tokenlist(input, tokens);
//fork a child process
pid = fork();
//parent: wait for child to finish
if(pid > 0) {
pid = wait(&status);
}
//child: execute command
if(pid == 0) {
//redirect output
token_index = contains_token(tokens, num_tokens, ">");
if(token_index >= 0 && token_index+1 < num_tokens) {
int fd_o = open(tokens[token_index+1], O_CREAT|O_TRUNC|O_WRONLY, 0644);
if(dup2(fd_o, STDOUT_FILENO)<0) {
perror("dup2 error");
exit(EXIT_FAILURE);
}
close(fd_o);
//remove ">" and the output file from the list of arguments
tokens[token_index ] = '\0';
tokens[token_index+1] = '\0';
}
//redirect input
token_index = contains_token(tokens, num_tokens, "<");
if(token_index >= 0 && token_index+1 < num_tokens) {
int fd_i = open(tokens[token_index+1], O_RDONLY);
dup2(fd_i, STDIN_FILENO);
close(fd_i);
//remove "<" and the input file from the list of arguments
tokens[token_index ] = '\0';
tokens[token_index+1] = '\0';
}
//handle pipe
int num_pipes = count_pipes(tokens, num_tokens);
if(num_pipes > 0) {
//get the indices of the pipes to extract the commands
int pipe_indices[CMD_MAX];
get_pipe_indices(tokens, num_tokens, pipe_indices);
//extract all the commands
char *commands[CMD_MAX][CMD_MAX];
for(int i=0; i<=num_pipes; i++) {
extract_tokens(tokens, commands[i], pipe_indices[i]+1, pipe_indices[i+1]);
}
//connect and run all the pipe commands
int fds[2],
in = STDIN_FILENO;
for(int i=0; i<num_pipes; i++) {
pipe(fds);
execute_piped_process(in, fds[1], commands[i]);
close(fds[1]);
in = fds[0];
}
//execute the last command
if(in != STDIN_FILENO) dup2(in, STDIN_FILENO);
execvp(commands[num_pipes][0], commands[num_pipes]);
}
//execute command
execvp(tokens[0], tokens);
printf("%s: invalid command\n", tokens[0]);
exit(EXIT_FAILURE);
}
}
//update history
if(strcmp(input, "") && status==0) {
strcpy( history[oldest_history], input );
oldest_history--;
if(oldest_history < 0) oldest_history = HISTORY_SIZE - 1;
}
}
}
/* the gateway function */
void mexFunction(int nlhs, mxArray *plhs[], int nrhs, const mxArray *prhs[]) {
int i;
int out_samples;
mxArray *outcell;
if (nrhs != 2 && nrhs != 1) {
help();
return;
}
if (nrhs == 2) {
parse_parameters(CTRL);
}
if (initialize()<0) {
mexErrMsgTxt("Error in module initialize() function\n");
}
allocate_memory();
for (i=0;i<2;i++) {
array_dims[i]=nof_output_itf;
}
if (nof_input_itf > 0) {
fill_input_lengths(IN);
format_input(IN);
}
for (i=0;i<nof_input_itf;i++) {
input_ptr[i] = &input_buffer[i*input_max_samples*input_sample_sz];
}
for (i=0;i<nof_output_itf;i++) {
output_ptr[i] = &output_buffer[i*output_max_samples*output_sample_sz];
}
out_samples = work(input_ptr, output_ptr);
if (out_samples < 0) {
mexErrMsgTxt("Error in module work() function\n");
}
if (stop()<0) {
mexErrMsgTxt("Error in module stop() function\n");
}
for (i=0;i<nof_output_itf;i++) {
if (!output_len[i]) {
output_len[i] = out_samples;
}
}
if (nof_output_itf>1) {
OUT=mxCreateCellMatrix(nof_output_itf,1);
for (i=0;i<nof_output_itf;i++) {
save_output(&outcell,&output_buffer[i*output_max_samples*output_sample_sz],output_len[i]);
mxSetCell(OUT,i,outcell);
}
} else {
save_output(&OUT,&output_buffer[0],output_len[0]);
}
return;
}
/* LED main encryption block: it supposes that the scheduled
keys are in memory pointed by the second argument */
__attribute__((noinline)) void LED128vperm_core(const u8* message, const u8* subkeys, u8* ciphertext)
{
/* Note : message is in rdi, subkeys in rsi and ciphertext in rdx */
/* __cdecl calling convention */
asm (".intel_syntax noprefix");
\
Push_All_Regs();
/* Key Index */
asm("xor "tostr(RoundCounter_)", "tostr(RoundCounter_)"");
asm("lea "tostr(RCBase_)", [rip + RC128_LED]");
/* Load constants (TBoxes) */
asm("movdqa "tostr(T1a_LED_)", [rip + T1a_LED]");
asm("movdqa "tostr(T1b_LED_)", [rip + T1b_LED]");
asm("movdqa "tostr(T2a_LED_)", [rip + T2a_LED]");
asm("movdqa "tostr(T2b_LED_)", [rip + T2b_LED]");
asm("movdqa "tostr(T3a_LED_)", [rip + T3a_LED]");
asm("movdqa "tostr(T3b_LED_)", [rip + T3b_LED]");
asm("movdqa "tostr(T4a_LED_)", [rip + T4a_LED]");
asm("movdqa "tostr(T4b_LED_)", [rip + T4b_LED]");
/* Load the the And mask */
asm("movdqa "tostr(AndMask_LED_)", [rip + AndMask_LED]");
asm("movdqa "tostr(OrMask_LED_)", [rip + OrMask_LED]");
/* Load the message */
asm("movdqa "tostr(State_)", ["tostr(Plaintext_)"]");
/* Transform the message from line wise to column wise */
format_input(State_);
/* Transform the keys from line wise to column wise */
asm("movdqa "tostr(Tmp4_)", ["tostr(Keys_)"]");
format_input(Tmp4_);
asm("movdqa ["tostr(Keys_)"], "tostr(Tmp4_)"");
asm("movdqa "tostr(Tmp4_)", ["tostr(Keys_)"+16]");
format_input(Tmp4_);
asm("movdqa ["tostr(Keys_)"+16], "tostr(Tmp4_)"");
/* Interleave the half keys */
asm("mov rax, ["tostr(Keys_)"+8]");
asm("mov rbx, ["tostr(Keys_)"+16]");
asm("mov ["tostr(Keys_)"+16], rax");
asm("mov ["tostr(Keys_)"+8], rbx");
asm("pxor "tostr(State_)", ["tostr(Keys_)"]");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"+16]");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"]");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"+16]");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"]");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"+16]");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"]");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"+16]");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"]");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"+16]");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"]");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"]+16");
LEDROUND();
LEDROUND();
LEDROUND();
LEDROUND();
asm("pxor "tostr(State_)", ["tostr(Keys_)"]");
/* Transform the message back from column wise to line wise */
format_output(State_);
/* Move back the result in the input message */
asm("movdqa ["tostr(Ciphertext_)"], "tostr(State_)"");
Pop_All_Regs();
asm (".att_syntax noprefix");
return;
}
/* Piccolo main encryption block: it supposes that the scheduled
keys are in memory pointed by the second argument */
__attribute__((noinline)) void Piccolo80vperm_core(const u8* message, const u8* subkeys, u8* ciphertext)
{
/* Note : message is in rdi, subkeys in rsi and ciphertext in rdx */
/* __cdecl calling convention */
asm (".intel_syntax noprefix");
Push_All_Regs();
/* Key Index */
asm("xor "tostr(RoundCounter_)", "tostr(RoundCounter_)"");
/* Load constants (SBoxes, multiplications ...) */
asm("movdqa "tostr(PiccoloSBoxL_)", [rip + PiccoloSBoxL]");
asm("movdqa "tostr(PiccoloSBoxH_)", [rip + PiccoloSBoxH]");
asm("movdqa "tostr(TwoMulPiccoloSBoxL_)", [rip + TwoMulPiccoloSBoxL]");
asm("movdqa "tostr(ThreeMulPiccoloSBoxL_)", [rip + ThreeMulPiccoloSBoxL]");
asm("movdqa "tostr(PiccoloThreeShuf_)", [rip + PiccoloThreeShuf]");
asm("movdqa "tostr(PiccoloOneShufa_)", [rip + PiccoloOneShufa]");
asm("movdqa "tostr(PiccoloOneShufb_)", [rip + PiccoloOneShufb]");
/* Load the masks */
asm("movdqa "tostr(PiccoloAndMaskL_)", [rip + PiccoloAndMaskL]");
/* Load Piccolo's Round Permutation PiccoloRP */
asm("movdqa "tostr(PiccoloRP_)", [rip + PiccoloRP]");
/* Scheduled keys from [rsi] and above */
/* Load the messages and format them */
format_input(Plaintext_, Plaintext_+8, State_, Tmp1_, Tmp2_, Tmp3_);
/* Pre Whitening AddRoundKey */
asm("pxor "tostr(State_)", ["tostr(Keys_)"+"tostr(RoundCounter_)"]");
asm("add "tostr(RoundCounter_)", 16");
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloROUND();
PiccoloLASTROUND();
/* Post Whitening AddRoundKey */
asm("pxor "tostr(State_)", ["tostr(Keys_)"+"tostr(RoundCounter_)"]");
/* Move back the result in the input message, formatted */
format_output(Ciphertext_, State_, Tmp1_, Tmp2_, Tmp3_);
Pop_All_Regs();
asm (".att_syntax noprefix");
return;
}
__attribute__((noinline)) void Piccolo128vperm_key_schedule(const u8* masterKey, u8* roundKeys)
{
/* Note : master key in rdi and round keys in rsi */
/* __cdecl calling convention */
asm (".intel_syntax noprefix");
Push_All_Regs();
asm("lea "tostr(PiccoloTcon128_)", [rip + PiccoloTcon128]");
/* Interleave keys */
format_input(rdi, rdi+16, xmm11, xmm13, xmm14, xmm15);
format_input(rdi+8, rdi+24, xmm12, xmm13, xmm14, xmm15);
#ifdef AVX
/* Pre-Whitening Key */
asm("vpshufb xmm0, xmm11, [rip + PiccoloKSShufw0]");
asm("movdqa [rsi], xmm0");
/* Post-Whitening Key */
asm("vpshufb xmm1, xmm12, [rip + PiccoloKSShufw1]");
asm("movdqa [rsi+512], xmm1");
/* Even keys */
/* --------- */
/* Isolate interleaved k0 */
asm("vpshufb xmm0, xmm11, [rip + PiccoloKSShuf0]");
/* Isolate interleaved k2 */
asm("vpshufb xmm2, xmm11, [rip + PiccoloKSShuf2]");
/* Isolate interleaved k4 */
asm("vpshufb xmm4, xmm12, [rip + PiccoloKSShuf0]");
/* Isolate interleaved k6 */
asm("vpshufb xmm6, xmm12, [rip + PiccoloKSShuf2]");
/* Odd keys */
/* -------- */
/* Isolate interleaved k1 */
asm("vpshufb xmm1, xmm11, [rip + PiccoloKSShuf1]");
/* Isolate interleaved k3 */
asm("vpshufb xmm3, xmm11, [rip + PiccoloKSShuf3]");
/* Isolate interleaved k5 */
asm("vpshufb xmm5, xmm12, [rip + PiccoloKSShuf1]");
/* Isolate interleaved k7 */
asm("vpshufb xmm7, xmm12, [rip + PiccoloKSShuf3]");
/* Compute the pairs */
/* 2 - 3 */
asm("vpxor xmm13, xmm2, xmm3");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"]");
asm("movdqa [rsi+16], xmm13");
/* 4 - 5 */
asm("vpxor xmm13, xmm4, xmm5");
asm("mov rcx, 2*2");
asm("vpxor xmm8, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*7");
asm("vpxor xmm9, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
asm("mov rcx, 2*9");
asm("vpxor xmm10, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm10");
/* 6 - 7 */
asm("vpxor xmm13, xmm6, xmm7");
asm("mov rcx, 2*3");
asm("vpxor xmm8, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*5");
asm("vpxor xmm9, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
/* 2 - 1 */
asm("vpxor xmm13, xmm2, xmm1");
asm("mov rcx, 2*4");
asm("vpxor xmm8, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*20");
asm("vpxor xmm9, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
/* 0 - 3 */
asm("vpxor xmm13, xmm0, xmm3");
asm("mov rcx, 2*6");
asm("vpxor xmm8, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*11");
asm("vpxor xmm9, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
asm("mov rcx, 2*13");
asm("vpxor xmm10, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm10");
/* 6 - 1 */
asm("vpxor xmm13, xmm6, xmm1");
asm("mov rcx, 2*8");
asm("vpxor xmm8, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*24");
asm("vpxor xmm9, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
/* 2 - 7 */
asm("vpxor xmm13, xmm2, xmm7");
asm("mov rcx, 2*10");
asm("vpxor xmm8, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*15");
asm("vpxor xmm9, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
asm("mov rcx, 2*17");
asm("vpxor xmm10, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm10");
/* 4 - 1 */
asm("vpxor xmm13, xmm4, xmm1");
asm("mov rcx, 2*12");
asm("vpxor xmm8, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*28");
asm("vpxor xmm9, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
/* 6 - 5 */
asm("vpxor xmm13, xmm6, xmm5");
asm("mov rcx, 2*14");
asm("vpxor xmm8, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*19");
asm("vpxor xmm9, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
asm("mov rcx, 2*21");
asm("vpxor xmm10, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm10");
/* 0 - 1 */
asm("vpxor xmm13, xmm0, xmm1");
asm("mov rcx, 2*16");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
/* 4 - 3 */
asm("vpxor xmm13, xmm4, xmm3");
asm("mov rcx, 2*18");
asm("vpxor xmm8, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*23");
asm("vpxor xmm9, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
asm("mov rcx, 2*25");
asm("vpxor xmm10, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm10");
/* 0 - 7 */
asm("vpxor xmm13, xmm0, xmm7");
asm("mov rcx, 2*22");
asm("vpxor xmm8, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*27");
asm("vpxor xmm9, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
asm("mov rcx, 2*29");
asm("vpxor xmm10, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm10");
/* 2 - 5 */
asm("vpxor xmm13, xmm2, xmm5");
asm("mov rcx, 2*26");
asm("vpxor xmm8, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*31");
asm("vpxor xmm9, xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
/* 6 - 3 */
asm("vpxor xmm13, xmm6, xmm3");
asm("mov rcx, 2*30");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
#else
/* Pre-Whitening Key */
asm("movdqa xmm0, xmm11");
asm("pshufb xmm0, [rip + PiccoloKSShufw0]");
asm("movdqa [rsi], xmm0");
/* Post-Whitening Key */
asm("movdqa xmm1, xmm12");
asm("pshufb xmm1, [rip + PiccoloKSShufw1]");
asm("movdqa [rsi+512], xmm1");
/* Even keys */
/* --------- */
/* Isolate interleaved k0 */
asm("movdqa xmm0, xmm11");
asm("pshufb xmm0, [rip + PiccoloKSShuf0]");
/* Isolate interleaved k2 */
asm("movdqa xmm2, xmm11");
asm("pshufb xmm2, [rip + PiccoloKSShuf2]");
/* Isolate interleaved k4 */
asm("movdqa xmm4, xmm12");
asm("pshufb xmm4, [rip + PiccoloKSShuf0]");
/* Isolate interleaved k6 */
asm("movdqa xmm6, xmm12");
asm("pshufb xmm6, [rip + PiccoloKSShuf2]");
/* Odd keys */
/* -------- */
/* Isolate interleaved k1 */
asm("movdqa xmm1, xmm11");
asm("pshufb xmm1, [rip + PiccoloKSShuf1]");
/* Isolate interleaved k3 */
asm("movdqa xmm3, xmm11");
asm("pshufb xmm3, [rip + PiccoloKSShuf3]");
/* Isolate interleaved k5 */
asm("movdqa xmm5, xmm12");
asm("pshufb xmm5, [rip + PiccoloKSShuf1]");
/* Isolate interleaved k7 */
asm("movdqa xmm7, xmm12");
asm("pshufb xmm7, [rip + PiccoloKSShuf3]");
/* Compute the pairs */
/* 2 - 3 */
asm("movdqa xmm13, xmm2");
asm("movdqa xmm14, xmm3");
asm("pxor xmm13, xmm14");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"]");
asm("movdqa [rsi+16], xmm13");
/* 4 - 5 */
asm("movdqa xmm13, xmm4");
asm("movdqa xmm14, xmm5");
asm("pxor xmm13, xmm14");
asm("movdqa xmm8, xmm13");
asm("movdqa xmm9, xmm13");
asm("mov rcx, 2*2");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
asm("mov rcx, 2*7");
asm("pxor xmm8, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*9");
asm("pxor xmm9, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
/* 6 - 7 */
asm("movdqa xmm13, xmm6");
asm("movdqa xmm14, xmm7");
asm("pxor xmm13, xmm14");
asm("movdqa xmm8, xmm13");
asm("mov rcx, 2*3");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
asm("mov rcx, 2*5");
asm("pxor xmm8, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
/* 2 - 1 */
asm("movdqa xmm13, xmm2");
asm("movdqa xmm14, xmm1");
asm("pxor xmm13, xmm14");
asm("movdqa xmm8, xmm13");
asm("mov rcx, 2*4");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
asm("mov rcx, 2*20");
asm("pxor xmm8, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
/* 0 - 3 */
asm("movdqa xmm13, xmm0");
asm("movdqa xmm14, xmm3");
asm("pxor xmm13, xmm14");
asm("movdqa xmm8, xmm13");
asm("movdqa xmm9, xmm13");
asm("mov rcx, 2*6");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
asm("mov rcx, 2*11");
asm("pxor xmm8, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*13");
asm("pxor xmm9, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
/* 6 - 1 */
asm("movdqa xmm13, xmm6");
asm("movdqa xmm14, xmm1");
asm("pxor xmm13, xmm14");
asm("movdqa xmm8, xmm13");
asm("mov rcx, 2*8");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
asm("mov rcx, 2*24");
asm("pxor xmm8, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
/* 2 - 7 */
asm("movdqa xmm13, xmm2");
asm("movdqa xmm14, xmm7");
asm("pxor xmm13, xmm14");
asm("movdqa xmm8, xmm13");
asm("movdqa xmm9, xmm13");
asm("mov rcx, 2*10");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
asm("mov rcx, 2*15");
asm("pxor xmm8, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*17");
asm("pxor xmm9, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
/* 4 - 1 */
asm("movdqa xmm13, xmm4");
asm("movdqa xmm14, xmm1");
asm("pxor xmm13, xmm14");
asm("movdqa xmm8, xmm13");
asm("mov rcx, 2*12");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
asm("mov rcx, 2*28");
asm("pxor xmm8, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
/* 6 - 5 */
asm("movdqa xmm13, xmm6");
asm("movdqa xmm14, xmm5");
asm("pxor xmm13, xmm14");
asm("movdqa xmm8, xmm13");
asm("movdqa xmm9, xmm13");
asm("mov rcx, 2*14");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
asm("mov rcx, 2*19");
asm("pxor xmm8, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*21");
asm("pxor xmm9, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
/* 0 - 1 */
asm("movdqa xmm13, xmm0");
asm("movdqa xmm14, xmm1");
asm("pxor xmm13, xmm14");
asm("mov rcx, 2*16");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
/* 4 - 3 */
asm("movdqa xmm13, xmm4");
asm("movdqa xmm14, xmm3");
asm("pxor xmm13, xmm14");
asm("movdqa xmm8, xmm13");
asm("movdqa xmm9, xmm13");
asm("mov rcx, 2*18");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
asm("mov rcx, 2*23");
asm("pxor xmm8, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*25");
asm("pxor xmm9, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
/* 0 - 7 */
asm("movdqa xmm13, xmm0");
asm("movdqa xmm14, xmm7");
asm("pxor xmm13, xmm14");
asm("movdqa xmm8, xmm13");
asm("movdqa xmm9, xmm13");
asm("mov rcx, 2*22");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
asm("mov rcx, 2*27");
asm("pxor xmm8, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
asm("mov rcx, 2*29");
asm("pxor xmm9, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm9");
/* 2 - 5 */
asm("movdqa xmm13, xmm2");
asm("movdqa xmm14, xmm5");
asm("pxor xmm13, xmm14");
asm("movdqa xmm8, xmm13");
asm("mov rcx, 2*26");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
asm("mov rcx, 2*31");
asm("pxor xmm8, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm8");
/* 6 - 3 */
asm("movdqa xmm13, xmm6");
asm("movdqa xmm14, xmm3");
asm("pxor xmm13, xmm14");
asm("mov rcx, 2*30");
asm("pxor xmm13, ["tostr(PiccoloTcon128_)"+8*rcx-16]");
asm("movdqa [rsi+8*rcx], xmm13");
#endif
Pop_All_Regs();
asm (".att_syntax noprefix");
return;
}
