I have my data field as follows DATA = 0x02 0x01 0x02 0x03 0x04 0x05 0x06 0x07 Now I want to concatenate this data as follows DATA = 0x01020304050607. How can I do it using C program. I found a program in C for concatenation of data in an array and the program is as follows:
#include<stdio.h>
int main(void)
{
int num[3]={1, 2, 3}, n1, n2, new_num;
n1 = num[0] * 100;
n2 = num[1] * 10;
new_num = n1 + n2 + num[2];
printf("%d \n", new_num);
return 0;
}
For the hexadecimal data in the array how can I manipulate the above program to concatenate the hexadecimal data?
You need a 64 bit variable num as result, instead of 10 as factor you need 16, and instead of 100 as factor, you need 256.
But if your data is provided as an array of bytes, then you can simply insert complete bytes, i.e. repeatedly shifting by 8 bits (meaning a factor of 256):
int main(void)
{
uint8_t data[8] = { 0x02, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07 };
unsigned long long num = 0;
for (int i=0; i<8; i++) {
num <<=8; // shift by a complete byte, equal to num *= 256
num |= data[i]; // write the respective byte
}
printf("num is %016llx\n",num);
return 0;
}
Output:
num is 0201020304050607
Lest say you have input like
int DATA[8] = {0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07};
If you want output like 0x0001020304050607, to store this resultant output you need one variable of unsigned long long type. For e.g
int main(void) {
int DATA[8] = {0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07};
int ele = sizeof(DATA)/sizeof(DATA[0]);
unsigned long long mask = 0x00;
for(int row = 0; row < ele; row++) {
mask = mask << 8;/* every time left shifted by 8(0x01-> 0000 0001) times */
mask = DATA[row] | mask; /* put at correct location */
}
printf("%016llx\n",mask);
return 0;
}
Here's some kind of hack that writes your data directly into an integer, without any bitwise operators:
#include <stdio.h>
#include <stdint.h>
#include <string.h>
uint64_t numberize(const uint8_t from[8]) {
uint64_t r = 0;
uint8_t *p = &r;
#if '01' == 0x4849 // big endian
memcpy(p, from, 8);
#else // little endian
for (int i=7; i >= 0; --i)
*p++ = from[i];
#endif
return r;
}
int main() {
const uint8_t data[8] = { 0x02, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07 };
printf("result is %016llx\n", numberize(data));
return 0;
}
This does work and outputs this independently of the endianness of your machine:
result is 0201020304050607
The compile-time endianness test was taken from this SO answer.
Related
I did test a crc-16/ibm implementation i found on the net. when I test it with hex byte array it works fine but if I include some 0x00 values, then it doesn't give the proper result.
here is its code
unsigned short ComputeCRC16(const unsigned char* buf, unsigned int len) {
unsigned short crc = 0;
for (unsigned int j = 0; j < len; j++)
{
unsigned char b = buf[j];
for (unsigned char i = 0; i < 8; i++)
{
crc = ((b ^ (unsigned char)crc) & 1) ? ((crc >> 1) ^ 0xA001) : (crc >> 1);
b >>= 1;
}
}
return crc;
}
I tested it with this code:
int main() {
//fe b5 5f f7
unsigned char buf1[4096] = { 0xfe, 0xb5, 0x5f, 0xf7 };
//fe b5 00 5f f7 00
unsigned char buf2[4096] = { 0xfe, 0xb5, 0x00, 0x5f, 0xf7, 0x00 };
int a = strlen(buf1);
unsigned short res = ComputeCRC16(buf1, a);
printf("res = %04x\n", res); //res : 7858, the result is correct
int b = strlen(buf2);
unsigned short res = ComputeCRC16(buf2, b);
printf("res = %04x\n", res); //res : d781, the result is not correct
return 0; //the correct result : 26EE
}
to verify the result I use this website:
https://www.lammertbies.nl/comm/info/crc-calculation
Your CRC routine gives correct results. It is your test that's wrong. strlen(p) returns how many bytes there are before the first zero byte at p. For buf2, that's four, not five as you intended. For buf1 it's not even defined, since there can be anything in memory after that array. You might be getting four, if the compiler happened to put zeros after the array.
For testing, you should simply provide len manually. (buf1, 4), (buf2, 5).
By the way, that code could be more efficient. It doesn't have to test with b every time. Just exclusive-oring with b to start has the same effect:
crc ^= buf[j];
for (unsigned char i = 0; i < 8; i++)
crc = crc & 1 ? (crc >> 1) ^ 0xa001 : crc >> 1;
void OPC_N2_data_read()
{
unsigned int32 iobyte[3][4] = {0x00, };
int32 PM_int[3] = {0x00, };
float PM[3] = {0x00, };
int8 i,j,k;
int8 mask = 0x80;
const int16 mask1 = 0x0001;
const int16 mask0 = 0x0000;
int8 trash_byte = 0x32;
output_bit(ss,1);
output_bit(PM_CLOCK_pin,0);
delay_us (1);
output_bit(ss,0);
delay_us (2);
for( i = 0 ; i < 3 ; i ++ )
{
for ( j = 0 ; j < 4 ; j ++ )
{
for (k = 0 ; k < 8 ; k ++ )
{
output_bit(PM_CLOCK_pin,1);
iobyte[i][j] = iobyte[i][j] << 1;
if ( input ( PM_MISO_pin))
{
iobyte[i][j] |= mask1;
}
else
{
iobyte[i][j] |= mask0;
}
if ((trash_byte & mask) >0)
{
output_high(PM_MOSI_pin);
}
else
{
output_bit(PM_MOSI_pin,0);
}
delay_us(1);
output_bit(PM_CLOCK_pin,0);
delay_us(5);
mask = mask >>1;
}
}
}
delay_us(3);
output_high(ss);
for(i = 0; i<3; i++)
{
PM_int[i] = ((iobyte[i][0]<<24)|(iobyte[i][1]<<16)|(iobyte[i][2]<<8)|(iobyte[i][3]));
PM[i] = *(float*)&PM_int;
}
printf ("%x%x%x%x\r\n",iobyte[0][0],iobyte[0][1],iobyte[0][2],iobyte[0][3]);
printf ("%x%x%x%x\r\n",iobyte[1][0],iobyte[1][1],iobyte[1][2],iobyte[1][3]);
printf ("%x%x%x%x\r\n",iobyte[2][0],iobyte[2][1],iobyte[2][2],iobyte[2][3]);
printf ("%lx,%lx,%lx\r\n", PM_int[0],PM_int[1],PM_int[2]);
printf ("%3.5f,%3.5f,%3.5f\r\n", PM[0],PM[1],PM[2]);
}
I receive data from a 4-byte array.
This data is a float value.
I check through the computer, I see the following.
e911bd41 d867e641 8084e941 e911bd41,d867e641,8084e941 0.00000,0.00000,0.00000
Making the four data into one INT value works fine. PM_int[0],PM_int[1],PM_int[2]
However, if you try to convert this to a float value, only 0.00000 is displayed.
I do not know where the problem is.
Looking on your source code shows that you are assuming that the received iobyte[i][0] to iobyte[i][3] shall be assumed as a float when ordering in big-endian format:
PM_int[i] = ((iobyte[i][0]<<24)|(iobyte[i][1]<<16)|(iobyte[i][2]<<8)|(iobyte[i][3]));
But if you want to convert correctly a 32bits floating-point value, you have to know if the source CPU and the destination CPU are both big-endian or little-endian. Otherwise, it is necessary to reverse the byte order before trying to convert values. So, if your target has a different endianness, replace the above code by the following:
PM_int[i] = ((iobyte[i][3]<<24)|(iobyte[i][2]<<16)|(iobyte[i][1]<<8)|(iobyte[i][0]));
In your source code, an error of the cast formatting between the float PM[i] and the 32bits integer PM_int[i] compute a bad result. The cast must be performed to the good element of the array PM_int[i].
PM[i] = *((float*)&(PM_int[i]));
Instead of:
PM[i] = *(float*)&PM_int;
With the 3 received float values {0xe9, 0x11, 0xbd, 0x41 }, {0xd8, 0x67, 0xe6, 0x41 }, {0x80, 0x84, 0xe9, 0x41 }, the output becomes:
e911bd41
d867e641
8084e941
41bd11e9,41e667d8,41e98480
23.63375,28.80070,29.18970
Let's say that I have an array of 16 uint8_t as follows:
uint8_t array[] = {0x13, 0x01, 0x4E, 0x52, 0x31, 0x4A, 0x35, 0x36, 0x4C, 0x11, 0x21, 0xC6, 0x3C, 0x73, 0xC2, 0x41};
This array stores the data contained in a 128 bits register of an external peripheral. Some of the information it represents are stored on 2, 3, 8, 12 bits ... and so on.
What is the best and elegant way to slice it up and bit mask the information I need? (The problem is that some things that I need overlaps the length of one cell of the array)
If that can help, this snippet I wrote converts the whole array into a char* string. But casting this into an int is not option because.. well 16 bytes.
int i;
char str[33];
for(i = 0; i < sizeof(array) / sizeof(*array) ; i++) {
sprintf(str+2*i,"%02hX",array[i]);
}
puts(str);
13014E52314A35364C1121C63C73C241
Actually such problem also occures when trying to parse all kind of bitstreams, like video or image files or compressed data by algorithms like LZ*. So the approach used there is to implement a bitstream reader.
But in your case the bit sequence is fixed length and quite short, so one way is to manually check the field values using bitwise operations.
Or you can use this function that I just wrote, which can extract arbitrary number of bits from a uint8 array, starting from desired bit position:
uint32_t extract_bits(uint8_t *arr, unsigned int bit_index, unsigned int bit_count)
{
/* Assert that we are not requested to extract more than 32 bits */
uint32_t result = 0;
assert(bit_count <= sizeof(result)*8 && arr != NULL);
/* You can additionally check if you are trying to extract bits exceeding the 16 byte range */
assert(bit_index + bit_count <= 16 * 8);
unsigned int arr_id = bit_index / 8;
unsigned int bit_offset = bit_index % 8;
if (bit_offset > 0) {
/* Extract first 'unaligned_bit_count' bits, which happen to be non-byte-aligned.
* When we do extract those bits, the remaining will be byte-aligned so
* we will thread them in different manner.
*/
unsigned int unaligned_bit_count = 8 - bit_offset;
/* Check if we need less than the remaining unaligned bits */
if (bit_count < unaligned_bit_count) {
result = (arr[arr_id] >> bit_offset) & ((1 << bit_count) - 1);
return result;
}
/* We need them all */
result = arr[arr_id] >> bit_offset;
bit_count -= unaligned_bit_count;
/* Move to next byte element */
arr_id++;
}
while (bit_count > 0) {
/* Try to extract up to 8 bits per iteration */
int bits_to_extract = bit_count > 8 ? 8 : bit_count;
if (bits_to_extract < 8) {
result = (result << bits_to_extract) | (arr[arr_id] & ((1 << bits_to_extract)-1));
}else {
result = (result << bits_to_extract) | arr[arr_id];
}
bit_count -= bits_to_extract;
arr_id++;
}
return result;
}
Here is example of how it is used.
uint32_t r;
/* Extracts bits [7..8] and places them as most significant bits of 'r' */
r = extract_bits(arr, 7, 2)
/* Extracts bits [4..35] and places them as most significant bits of 'r' */
r = extract_bits(arr, 4, 32);
/* Visualize */
printf("slice=%x\n", r);
And then the visualisation of r is up to you. They can either be represented as hex dwords, characters, or however you decide.
I have a variable unsigned __int64 text=0x0; and I need to and update each 4 bits with different values.
text = b64b63...b3b2b1b0
I have a loop and inside of the loop must be like this :
for(i=0;i<16;i++)
{
b3b2b1b0=a[0]
b7b6b5b4=a[1]
b11b10b9b8=a[2]
..
..
..
}
I tried this one but didn't work
unsigned __int64 temp=0x0;
unsigned __int64 index=0x0;
for(i=0;i<16;i++)
{
index = (text>>(i*4))&0x0F;
temp = a[index];
text = text | temp<<(i*4);
}
thank you so much
You can use bitshifting and bitwise-OR to do this. Assuming the a[n] values are all between 0 and 15 and also unsigned __int64s:
unsigned __int64 text=0x0;
text = a[0] |
a[1] << 4 |
a[2] << 8 |
etc....;
If you don't trust the a[n] values, use (a[n] & 0xF) to clear all but the four least significant bits before shifting. If a[n] is not an unsigned __int64, cast it before shifting.
You could use bit fields:
typedef struct {
int n0 : 4;
int n1 : 4;
int n2 : 4;
int n3 : 4;
int n4 : 4;
int n5 : 4;
int n6 : 4;
int n7 : 4;
int n8 : 4;
int n9 : 4;
int n10 : 4;
int n11 : 4;
int n12 : 4;
int n13 : 4;
int n14 : 4;
int n15 : 4;
} S64;
typedef union {
uint64_t i;
S64 b;
} U64;
Note: you may need to pay attention to to the endianness of your target platform(s), if you require a specific ordering of the 4 bit fields within the 64 bit value.
If you must use a loop, you can do something like
uint64_t text = 0;
for (int i = 15; i >= 0; --i)
{
text <<= 4;
text |= a[i] & 0x0f; // Masking in case a[i] have more than the lowest four bits set
}
Try the below code. I have tried to copy 0x01, 0x02....0x0f into 16 different nibbles. The code extracts the values and shifts them appropriately combined with the OR operation.
#define NIBBLES 16 /* 64-bit has 16 nibbles */
#define MASK 0x0FLLU
int main()
{
unsigned int val[NIBBLES] = {0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f};
long long unsigned int num = 0, tmp;
int i, shift = 0;
for (i = 0; i < NIBBLES; i++) {
tmp = (val[i] & MASK) << shift;
num = num | tmp;
shift = shift + 4;
}
printf ("0x%llx\n", num);
}
Output: 0xfedcba9876543210
I have a couple uint8_t arrays in my c code, and I'd like to compare an arbitrary sequence bits from one with another. So for example, I have bitarray_1 and bitarray_2, and I'd like to compare bits 13 - 47 from bitarray_1 with bits 5-39 of bitarray_2. What is the most efficient way to do this?
Currently it's a huge bottleneck in my program, since I just have a naive implementation that copies the bits into the beginning of a new temporary array, and then uses memcmp on them.
three words: shift, mask and xor.
shift to get the same memory alignment for both bitarray. If not you will have to shift one of the arrays before comparing them. Your exemple is probably misleading because bits 13-47 and 5-39 have the same memory alignment on 8 bits addresses. This wouldn't be true if you were comparing say bits 14-48 with bits 5-39.
Once everything is aligned and exceeding bits cleared for table boundaries a xor is enough to perform the comparison of all the bits at once. Basically you can manage to do it with just one memory read for each array, which should be pretty efficient.
If memory alignment is the same for both arrays as in your example memcmp and special case for upper and lower bound is probably yet faster.
Also accessing array by uint32_t (or uint64_t on 64 bits architectures) should also be more efficient than accessing by uint8_t.
The principle is simple but as Andrejs said the implementation is not painless...
Here is how it goes (similarities with #caf proposal is no coincidence):
/* compare_bit_sequence() */
int compare_bit_sequence(uint8_t s1[], unsigned s1_off, uint8_t s2[], unsigned s2_off,
unsigned length)
{
const uint8_t mask_lo_bits[] =
{ 0x00, 0x01, 0x03, 0x07, 0x0f, 0x1f, 0x3f, 0x7f, 0xff };
const uint8_t clear_lo_bits[] =
{ 0xff, 0xfe, 0xfc, 0xf8, 0xf0, 0xe0, 0xc0, 0x80, 0x00 };
uint8_t v1;
uint8_t * max_s1;
unsigned end;
uint8_t lsl;
uint8_t v1_mask;
int delta;
/* Makes sure the offsets are less than 8 bits */
s1 += s1_off >> 3;
s1_off &= 7;
s2 += s2_off >> 3;
s2_off &= 7;
/* Make sure s2 is the sequence with the shorter offset */
if (s2_off > s1_off){
uint8_t * tmp_s;
unsigned tmp_off;
tmp_s = s2; s2 = s1; s1 = tmp_s;
tmp_off = s2_off; s2_off = s1_off; s1_off = tmp_off;
}
delta = s1_off;
/* handle the beginning, s2 incomplete */
if (s2_off > 0){
delta = s1_off - s2_off;
v1 = delta
? (s1[0] >> delta | s1[1] << (8 - delta)) & clear_lo_bits[delta]
: s1[0];
if (length <= 8 - s2_off){
if ((v1 ^ *s2)
& clear_lo_bits[s2_off]
& mask_lo_bits[s2_off + length]){
return NOT_EQUAL;
}
else {
return EQUAL;
}
}
else{
if ((v1 ^ *s2) & clear_lo_bits[s2_off]){
return NOT_EQUAL;
}
length -= 8 - s2_off;
}
s1++;
s2++;
}
/* main loop, we test one group of 8 bits of v2 at each loop */
max_s1 = s1 + (length >> 3);
lsl = 8 - delta;
v1_mask = clear_lo_bits[delta];
while (s1 < max_s1)
{
if ((*s1 >> delta | (*++s1 << lsl & v1_mask)) ^ *s2++)
{
return NOT_EQUAL;
}
}
/* last group of bits v2 incomplete */
end = length & 7;
if (end && ((*s2 ^ *s1 >> delta) & mask_lo_bits[end]))
{
return NOT_EQUAL;
}
return EQUAL;
}
All possible optimisations are not yet used. One promising one would be to use larger chunks of data (64 bits or 32 bits at once instead of 8), you could also detect cases where offset are synchronised for both arrays and in such cases use a memcmp instead of the main loop, replace modulos % 8 by logical operators & 7, replace '/ 8' by '>> 3', etc., have to branches of code instead of swapping s1 and s2, etc, but the main purpose is achieved : only one memory read and not memory write for each array item hence most of the work can take place inside processor registers.
bits 13 - 47 of bitarray_1 are the same as bits 5 - 39 of bitarray_1 + 1.
Compare the first 3 bits (5 - 7) with a mask and the other bits (8 - 39) with memcmp().
Rather than shift and copy the bits, maybe representing them differently is faster. You have to measure.
/* code skeleton */
static char bitarray_1_bis[BIT_ARRAY_SIZE*8+1];
static char bitarray_2_bis[BIT_ARRAY_SIZE*8+1];
static const char *lookup_table[] = {
"00000000", "00000001", "00000010" /* ... */
/* 256 strings */
/* ... */ "11111111"
};
/* copy every bit of bitarray_1 to an element of bitarray_1_bis */
for (k = 0; k < BIT_ARRAY_SIZE; k++) {
strcpy(bitarray_1_bis + 8*k, lookup_table[bitarray_1[k]]);
strcpy(bitarray_2_bis + 8*k, lookup_table[bitarray_2[k]]);
}
memcmp(bitarray_1_bis + 13, bitarray_2_bis + 5, 47 - 13 + 1);
You can (and should) limit the copy to the minimum possible.
I have no idea if it's faster, but it wouldn't surprise me if it was. Again, you have to measure.
The easiest way to do this is to convert the more complex case into a simpler case, then solve the simpler case.
In the following code, do_compare() solves the simpler case (where the sequences are never offset by more than 7 bits, s1 is always offset as much or more than s2, and the length of the sequence is non-zero). The compare_bit_sequence() function then takes care of converting the harder case to the easier case, and calls do_compare() to do the work.
This just does a single-pass through the bit sequences, so hopefully that's an improvement on your copy-and-memcmp implementation.
#define NOT_EQUAL 0
#define EQUAL 1
/* do_compare()
*
* Does the actual comparison, but has some preconditions on parameters to
* simplify things:
*
* length > 0
* 8 > s1_off >= s2_off
*/
int do_compare(const uint8_t s1[], const unsigned s1_off, const uint8_t s2[],
const unsigned s2_off, const unsigned length)
{
const uint8_t mask_lo_bits[] =
{ 0xff, 0x01, 0x03, 0x07, 0x0f, 0x1f, 0x3f, 0x7f, 0xff };
const uint8_t mask_hi_bits[] =
{ 0x00, 0x80, 0xc0, 0xe0, 0xf0, 0xf8, 0xfc, 0xfe, 0xff };
const unsigned msb = (length + s1_off - 1) / 8;
const unsigned s2_shl = s1_off - s2_off;
const unsigned s2_shr = 8 - s2_shl;
unsigned n;
uint8_t s1s2_diff, lo_bits = 0;
for (n = 0; n <= msb; n++)
{
/* Shift s2 so it is aligned with s1, pulling in low bits from
* the high bits of the previous byte, and store in s1s2_diff */
s1s2_diff = lo_bits | (s2[n] << s2_shl);
/* Save the bits needed to fill in the low-order bits of the next
* byte. HERE BE DRAGONS - since s2_shr can be 8, this below line
* only works because uint8_t is promoted to int, and we know that
* the width of int is guaranteed to be >= 16. If you change this
* routine to work with a wider type than uint8_t, you will need
* to special-case this line so that if s2_shr is the width of the
* type, you get lo_bits = 0. Don't say you weren't warned. */
lo_bits = s2[n] >> s2_shr;
/* XOR with s1[n] to determine bits that differ between s1 and s2 */
s1s2_diff ^= s1[n];
/* Look only at differences in the high bits in the first byte */
if (n == 0)
s1s2_diff &= mask_hi_bits[8 - s1_off];
/* Look only at differences in the low bits of the last byte */
if (n == msb)
s1s2_diff &= mask_lo_bits[(length + s1_off) % 8];
if (s1s2_diff)
return NOT_EQUAL;
}
return EQUAL;
}
/* compare_bit_sequence()
*
* Adjusts the parameters to match the preconditions for do_compare(), then
* calls it to do the work.
*/
int compare_bit_sequence(const uint8_t s1[], unsigned s1_off,
const uint8_t s2[], unsigned s2_off, unsigned length)
{
/* Handle length zero */
if (length == 0)
return EQUAL;
/* Makes sure the offsets are less than 8 bits */
s1 += s1_off / 8;
s1_off %= 8;
s2 += s2_off / 8;
s2_off %= 8;
/* Make sure s2 is the sequence with the shorter offset */
if (s1_off >= s2_off)
return do_compare(s1, s1_off, s2, s2_off, length);
else
return do_compare(s2, s2_off, s1, s1_off, length);
}
To do the comparison in your example, you'd call:
compare_bit_sequence(bitarray_1, 13, bitarray_2, 5, 35)
(Note that I am numbering the bits from zero, and assuming that the bitarrays are laid out little-endian, so this will start the comparison from the sixth-least-significant bit in bitarray2[0], and the sixth-least-signifcant bit in bitarray1[1]).
What about writing the function that will calculate the offsets from both arrays, apply the mask, shift the bits and store the result to the int so you may compare them. If the bits count (34 in your example) exceeds the length of the int - recurse or loop.
Sorry, the example will be pain in the ass.
Here is my unoptimized bit sequence comparison function:
#include <stdio.h>
#include <stdint.h>
// 01234567 01234567
uint8_t bitsA[] = { 0b01000000, 0b00010000 };
uint8_t bitsB[] = { 0b10000000, 0b00100000 };
int bit( uint8_t *bits, size_t bitpoz, size_t len ){
return (bitpoz<len)? !!(bits[bitpoz/8]&(1<<(7-bitpoz%8))): 0;
}
int bitcmp( uint8_t *bitsA, size_t firstA, size_t lenA,
uint8_t *bitsB, size_t firstB, size_t lenB ){
int cmp;
for( size_t i=0; i<lenA || i<lenB; i++ ){
if( (cmp = bit(bitsA,firstA+i,firstA+lenA) -
bit(bitsB,firstB+i,firstB+lenB)) ) return cmp;
}
return 0;
}
int main(){
printf( "cmp: %i\n", bitcmp( bitsA,1,11, bitsB,0,11 ) );
}
EDIT: Here is my (untested) bitstring equality test function:
#include <stdlib.h>
#include <stdint.h>
#define load_64bit(bits,first) (*(uint64_t*)bits<<first | *(bits+8)>>(8-first))
#define load_32bit(bits,first) (*(uint32_t*)bits<<first | *(bits+4)>>(8-first))
#define load_16bit(bits,first) (*(uint16_t*)bits<<first | *(bits+2)>>(8-first))
#define load_8bit( bits,first) ( *bits<<first | *(bits+1)>>(8-first))
static inline uint8_t last_bits( uint8_t *bits, size_t first, size_t size ){
return (first+size>8?load_8bit(bits,first):*bits<<first)>>(8-size);
}
int biteq( uint8_t *bitsA, size_t firstA,
uint8_t *bitsB, size_t firstB, size_t size ){
if( !size ) return 1;
bitsA+=firstA/8; firstA%=8;
bitsB+=firstB/8; firstB%=8;
for(; size>64;size-=64,bitsA+=8,bitsB+=8)
if(load_64bit(bitsA,firstA)!=load_64bit(bitsB,firstB)) return 0;
for(; size>32;size-=32,bitsA+=4,bitsB+=4)
if(load_32bit(bitsA,firstA)!=load_32bit(bitsB,firstB)) return 0;
for(; size>16;size-=16,bitsA+=2,bitsB+=2)
if(load_16bit(bitsA,firstA)!=load_16bit(bitsB,firstB)) return 0;
for(; size> 8;size-= 8,bitsA++, bitsB++ )
if(load_8bit( bitsA,firstA)!=load_8bit( bitsB,firstB)) return 0;
return !size ||
last_bits(bitsA,firstA,size)==last_bits(bitsB,firstB,size);
}
I made a simple measurement tool to see how fast is it:
#include <unistd.h>
#include <stdio.h>
#include <signal.h>
#define SIZE 1000000
uint8_t bitsC[SIZE];
volatile int end_loop;
void sigalrm_hnd( int sig ){ (void)sig; end_loop=1; }
int main(){
uint64_t loop_count; int cmp;
signal(SIGALRM,sigalrm_hnd);
loop_count=0; end_loop=0; alarm(10);
while( !end_loop ){
for( int i=1; i<7; i++ ){
loop_count++;
cmp = biteq( bitsC,i, bitsC,7-i,(SIZE-1)*8 );
if( !cmp ){ printf( "cmp: %i (==0)\n", cmp ); return -1; }
}
}
printf( "biteq: %.2f round/sec\n", loop_count/10.0 );
}
Result:
bitcmp: 8.40 round/sec
biteq: 363.60 round/sec
EDIT2: last_bits() changed.