#define filter(x) (filterlut[(x) & 0xfffff])\r
#endif\r
\r
-\r
-\r
-typedef struct bucket {\r
- uint32_t *head;\r
- uint32_t *bp;\r
-} bucket_t;\r
-\r
-typedef bucket_t bucket_array_t[2][0x100];\r
-\r
-typedef struct bucket_info {\r
- struct {\r
- uint32_t *head, *tail;\r
- } bucket_info[2][0x100];\r
- uint32_t numbuckets;\r
- } bucket_info_t;\r
-\r
-\r
-static void bucket_sort_intersect(uint32_t* const estart, uint32_t* const estop,\r
- uint32_t* const ostart, uint32_t* const ostop,\r
- bucket_info_t *bucket_info, bucket_array_t bucket)\r
-{\r
- uint32_t *p1, *p2;\r
- uint32_t *start[2];\r
- uint32_t *stop[2];\r
-\r
- start[0] = estart;\r
- stop[0] = estop;\r
- start[1] = ostart;\r
- stop[1] = ostop;\r
-\r
- // init buckets to be empty\r
- for (uint32_t i = 0; i < 2; i++) {\r
- for (uint32_t j = 0x00; j <= 0xff; j++) {\r
- bucket[i][j].bp = bucket[i][j].head;\r
- }\r
- }\r
-\r
- // sort the lists into the buckets based on the MSB (contribution bits)\r
- for (uint32_t i = 0; i < 2; i++) {\r
- for (p1 = start[i]; p1 <= stop[i]; p1++) {\r
- uint32_t bucket_index = (*p1 & 0xff000000) >> 24;\r
- *(bucket[i][bucket_index].bp++) = *p1;\r
- }\r
- }\r
-\r
-\r
- // write back intersecting buckets as sorted list.\r
- // fill in bucket_info with head and tail of the bucket contents in the list and number of non-empty buckets.\r
- uint32_t nonempty_bucket;\r
- for (uint32_t i = 0; i < 2; i++) {\r
- p1 = start[i];\r
- nonempty_bucket = 0;\r
- for (uint32_t j = 0x00; j <= 0xff; j++) {\r
- if (bucket[0][j].bp != bucket[0][j].head && bucket[1][j].bp != bucket[1][j].head) { // non-empty intersecting buckets only\r
- bucket_info->bucket_info[i][nonempty_bucket].head = p1;\r
- for (p2 = bucket[i][j].head; p2 < bucket[i][j].bp; *p1++ = *p2++);\r
- bucket_info->bucket_info[i][nonempty_bucket].tail = p1 - 1;\r
- nonempty_bucket++;\r
- }\r
- }\r
- bucket_info->numbuckets = nonempty_bucket;\r
- }\r
-}\r
-\r
-/** binsearch\r
- * Binary search for the first occurence of *stop's MSB in sorted [start,stop]\r
- */\r
-static inline uint32_t* binsearch(uint32_t *start, uint32_t *stop)\r
-{\r
- uint32_t mid, val = *stop & 0xff000000;\r
- while(start != stop)\r
- if(start[mid = (stop - start) >> 1] > val)\r
- stop = &start[mid];\r
- else\r
- start += mid + 1;\r
-\r
- return start;\r
-}\r
-\r
/** update_contribution\r
* helper, calculates the partial linear feedback contributions and puts in MSB\r
*/\r
-static inline void\r
-update_contribution(uint32_t *item, const uint32_t mask1, const uint32_t mask2)\r
+static inline void update_contribution(uint32_t *item, const uint32_t mask1, const uint32_t mask2)\r
{\r
uint32_t p = *item >> 25;\r
\r
/** extend_table\r
* using a bit of the keystream extend the table of possible lfsr states\r
*/\r
-static inline void\r
-extend_table(uint32_t *tbl, uint32_t **end, int bit, int m1, int m2, uint32_t in)\r
+static inline void extend_table(uint32_t *tbl, uint32_t **end, int bit, int m1, int m2, uint32_t in)\r
{\r
in <<= 24;\r
for(*tbl <<= 1; tbl <= *end; *++tbl <<= 1)\r
*/\r
static inline void extend_table_simple(uint32_t *tbl, uint32_t **end, int bit)\r
{\r
- for(*tbl <<= 1; tbl <= *end; *++tbl <<= 1)\r
+ for(*tbl <<= 1; tbl <= *end; *++tbl <<= 1) {\r
if(filter(*tbl) ^ filter(*tbl | 1)) { // replace\r
*tbl |= filter(*tbl) ^ bit;\r
} else if(filter(*tbl) == bit) { // insert\r
*++*end = *++tbl;\r
*tbl = tbl[-1] | 1;\r
- } else // drop\r
+ } else { // drop\r
*tbl-- = *(*end)--;\r
+ }\r
+ }\r
}\r
/** recover\r
* recursively narrow down the search space, 4 bits of keystream at a time\r
oks >>= 1;\r
eks >>= 1;\r
in >>= 2;\r
- extend_table(o_head, &o_tail, oks & 1, LF_POLY_EVEN << 1 | 1,\r
- LF_POLY_ODD << 1, 0);\r
+ extend_table(o_head, &o_tail, oks & 1, LF_POLY_EVEN << 1 | 1, LF_POLY_ODD << 1, 0);\r
if(o_head > o_tail)\r
return sl;\r
\r
- extend_table(e_head, &e_tail, eks & 1, LF_POLY_ODD,\r
- LF_POLY_EVEN << 1 | 1, in & 3);\r
+ extend_table(e_head, &e_tail, eks & 1, LF_POLY_ODD, LF_POLY_EVEN << 1 | 1, in & 3);\r
if(e_head > e_tail)\r
return sl;\r
}\r
\r
// allocate memory for out of place bucket_sort\r
bucket_array_t bucket;\r
- for (uint32_t i = 0; i < 2; i++)\r
- for (uint32_t j = 0; j <= 0xff; j++) {\r
- bucket[i][j].head = malloc(sizeof(uint32_t)<<14);\r
- if (!bucket[i][j].head) {\r
- goto out;\r
- }\r
- }\r
-\r
+ \r
+ if ( !bucket_malloc(bucket) ) goto out;\r
\r
// initialize statelists: add all possible states which would result into the rightmost 2 bits of the keystream\r
for(i = 1 << 20; i >= 0; --i) {\r
// 22 bits to go to recover 32 bits in total. From now on, we need to take the "in"\r
// parameter into account.\r
in = (in >> 16 & 0xff) | (in << 16) | (in & 0xff00); // Byte swapping\r
- recover(odd_head, odd_tail, oks,\r
- even_head, even_tail, eks, 11, statelist, in << 1, bucket);\r
-\r
+ recover(odd_head, odd_tail, oks, even_head, even_tail, eks, 11, statelist, in << 1, bucket);\r
\r
out:\r
free(odd_head);\r
free(even_head);\r
- for (uint32_t i = 0; i < 2; i++)\r
- for (uint32_t j = 0; j <= 0xff; j++)\r
- free(bucket[i][j].head);\r
-\r
+ bucket_free(bucket);\r
return statelist;\r
}\r
\r
--- /dev/null
+#include "bucketsort.h"
+
+bool bucket_malloc(bucket_array_t bucket) {
+ for (uint32_t i = 0; i < 2; i++) {
+ for (uint32_t j = 0; j <= 0xff; j++) {
+ bucket[i][j].head = malloc(sizeof(uint32_t)<<14);
+ if (!bucket[i][j].head) {
+ return false;
+ }
+ }
+ }
+ return true;
+}
+
+void bucket_free(bucket_array_t bucket) {
+ for (uint8_t i = 0; i < 2; i++)
+ for (uint8_t j = 0; j <= 0xff; j++)
+ free(bucket[i][j].head);
+}
+
+void bucket_sort_intersect(uint32_t* const estart, uint32_t* const estop,
+ uint32_t* const ostart, uint32_t* const ostop,
+ bucket_info_t *bucket_info, bucket_array_t bucket)
+{
+ uint32_t *p1, *p2;
+ uint32_t *start[2];
+ uint32_t *stop[2];
+
+ start[0] = estart;
+ stop[0] = estop;
+ start[1] = ostart;
+ stop[1] = ostop;
+
+ // init buckets to be empty
+ for (uint32_t i = 0; i < 2; i++) {
+ for (uint32_t j = 0x00; j <= 0xff; j++) {
+ bucket[i][j].bp = bucket[i][j].head;
+ }
+ }
+
+ // sort the lists into the buckets based on the MSB (contribution bits)
+ for (uint32_t i = 0; i < 2; i++) {
+ for (p1 = start[i]; p1 <= stop[i]; p1++) {
+ uint32_t bucket_index = (*p1 & 0xff000000) >> 24;
+ *(bucket[i][bucket_index].bp++) = *p1;
+ }
+ }
+
+ // write back intersecting buckets as sorted list.
+ // fill in bucket_info with head and tail of the bucket contents in the list and number of non-empty buckets.
+ uint32_t nonempty_bucket;
+ for (uint32_t i = 0; i < 2; i++) {
+ p1 = start[i];
+ nonempty_bucket = 0;
+ for (uint32_t j = 0x00; j <= 0xff; j++) {
+ if (bucket[0][j].bp != bucket[0][j].head && bucket[1][j].bp != bucket[1][j].head) { // non-empty intersecting buckets only
+ bucket_info->bucket_info[i][nonempty_bucket].head = p1;
+ for (p2 = bucket[i][j].head; p2 < bucket[i][j].bp; *p1++ = *p2++);
+ bucket_info->bucket_info[i][nonempty_bucket].tail = p1 - 1;
+ nonempty_bucket++;
+ }
+ }
+ bucket_info->numbuckets = nonempty_bucket;
+ }
+}
--- /dev/null
+#include "radixsort.h"
+
+uint64_t * radixSort(uint64_t * array, uint32_t size) {
+ rscounts_t counts;
+ memset(&counts, 0, 256 * 8 * sizeof(uint32_t));
+ uint64_t * cpy = (uint64_t *)malloc(size * sizeof(uint64_t));
+ uint32_t o8=0, o7=0, o6=0, o5=0, o4=0, o3=0, o2=0, o1=0;
+ uint32_t t8, t7, t6, t5, t4, t3, t2, t1;
+ uint32_t x;
+ // calculate counts
+ for(x = 0; x < size; ++x) {
+ t8 = array[x] & 0xff;
+ t7 = (array[x] >> 8) & 0xff;
+ t6 = (array[x] >> 16) & 0xff;
+ t5 = (array[x] >> 24) & 0xff;
+ t4 = (array[x] >> 32) & 0xff;
+ t3 = (array[x] >> 40) & 0xff;
+ t2 = (array[x] >> 48) & 0xff;
+ t1 = (array[x] >> 56) & 0xff;
+ counts.c8[t8]++;
+ counts.c7[t7]++;
+ counts.c6[t6]++;
+ counts.c5[t5]++;
+ counts.c4[t4]++;
+ counts.c3[t3]++;
+ counts.c2[t2]++;
+ counts.c1[t1]++;
+ }
+ // convert counts to offsets
+ for(x = 0; x < 256; ++x) {
+ t8 = o8 + counts.c8[x];
+ t7 = o7 + counts.c7[x];
+ t6 = o6 + counts.c6[x];
+ t5 = o5 + counts.c5[x];
+ t4 = o4 + counts.c4[x];
+ t3 = o3 + counts.c3[x];
+ t2 = o2 + counts.c2[x];
+ t1 = o1 + counts.c1[x];
+ counts.c8[x] = o8;
+ counts.c7[x] = o7;
+ counts.c6[x] = o6;
+ counts.c5[x] = o5;
+ counts.c4[x] = o4;
+ counts.c3[x] = o3;
+ counts.c2[x] = o2;
+ counts.c1[x] = o1;
+ o8 = t8;
+ o7 = t7;
+ o6 = t6;
+ o5 = t5;
+ o4 = t4;
+ o3 = t3;
+ o2 = t2;
+ o1 = t1;
+ }
+ // radix
+ for(x = 0; x < size; ++x) {
+ t8 = array[x] & 0xff;
+ cpy[counts.c8[t8]] = array[x];
+ counts.c8[t8]++;
+ }
+ for(x = 0; x < size; ++x) {
+ t7 = (cpy[x] >> 8) & 0xff;
+ array[counts.c7[t7]] = cpy[x];
+ counts.c7[t7]++;
+ }
+ for(x = 0; x < size; ++x) {
+ t6 = (array[x] >> 16) & 0xff;
+ cpy[counts.c6[t6]] = array[x];
+ counts.c6[t6]++;
+ }
+ for(x = 0; x < size; ++x) {
+ t5 = (cpy[x] >> 24) & 0xff;
+ array[counts.c5[t5]] = cpy[x];
+ counts.c5[t5]++;
+ }
+ for(x = 0; x < size; ++x) {
+ t4 = (array[x] >> 32) & 0xff;
+ cpy[counts.c4[t4]] = array[x];
+ counts.c4[t4]++;
+ }
+ for(x = 0; x < size; ++x) {
+ t3 = (cpy[x] >> 40) & 0xff;
+ array[counts.c3[t3]] = cpy[x];
+ counts.c3[t3]++;
+ }
+ for(x = 0; x < size; ++x) {
+ t2 = (array[x] >> 48) & 0xff;
+ cpy[counts.c2[t2]] = array[x];
+ counts.c2[t2]++;
+ }
+ for(x = 0; x < size; ++x) {
+ t1 = (cpy[x] >> 56) & 0xff;
+ array[counts.c1[t1]] = cpy[x];
+ counts.c1[t1]++;
+ }
+ free(cpy);
+ return array;
+}
\ No newline at end of file