Databases
StackOverflow2013
Postsdbo
POBIGNUM strange behavior in a calculation loop
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POBIGNUM strange behavior in a calculation loop
    primarykey
    data
    text
    <p>I'm trying to implement a basic routine to perform some calculation on BIGNUM(s) and I've found a strange behavior. The functions are as follows</p> <pre><code>unsigned char *char_array_as_hex(unsigned char *chr_a, int len) { unsigned char *chr_s = (unsigned char *)malloc(len * 2); char buffer[5]; for (int i = 0; i &lt; len; i++) { sprintf(buffer, "%02X", chr_a[i]); chr_s[(2 * i) + 0] = buffer[0]; chr_s[(2 * i) + 1] = buffer[1]; } return chr_s; } </code></pre> <p>and</p> <pre><code>char *big_number_as_decimal_from_hex_array(unsigned char *chr_a, int len, BN_CTX *bn_ctx) { unsigned char *hex_s = char_array_as_hex(chr_a, len); BIGNUM *big_number = BN_CTX_get(bn_ctx); BN_hex2bn(&amp;big_number, (char *)hex_s); char *big_number_as_decimal = BN_bn2dec(big_number); free(hex_s); BN_free(big_number); return big_number_as_decimal; } </code></pre> <p>and</p> <pre><code>void test_compute_prime256v1() { BN_CTX *bn_ctx = BN_CTX_new(); BN_CTX_start(bn_ctx); unsigned char seed_a[20] = { 0xC4,0x9D,0x36,0x08,0x86,0xE7,0x04,0x93,0x6A,0x66, /* seed */ 0x78,0xE1,0x13,0x9D,0x26,0xB7,0x81,0x9F,0x7E,0x90 }; printf("s = %s\n", big_number_as_decimal_from_hex_array(seed_a, 20, bn_ctx)); unsigned char p_a[32] = { 0xFF,0xFF,0xFF,0xFF,0x00,0x00,0x00,0x01,0x00,0x00, /* p */ 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, 0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF, 0xFF,0xFF }; printf("p = %s\n", big_number_as_decimal_from_hex_array(p_a, 32, bn_ctx)); BN_CTX_end(bn_ctx); BN_CTX_free(bn_ctx); } </code></pre> <p>then I call "test_compute_prime256v1" in an Objective-C method. If I call it once or multiple times with a reasonable delay between each call it produces correct result however, when I call that function in a loop it produces different incorrect values</p> <pre><code>- (IBAction)btnOK_Clicked:(id)sender { for (int i = 1; i &lt; 10; i++) { printf("i = %d\n", i); test_compute_prime256v1(); } } </code></pre> <p>and a sample output was</p> <pre><code>i = 1 s = 1122468115042657169822351801880191947498376363664 p = 115792089210356248762697446949407573530086143415290314195533631308867097853951 i = 2 s = 1122468115042657169822351801880191947498376363664 p = 966134380529368896499052403318808180610643774633026536153469502543482958881555881553276... i = 3 s = 1122468115042657169822351801880191947498376363664 p = 115792089210356248762697446949407573530086143415290314195533631308867097853951 </code></pre> <p>Note: some numbers are trimmed to fit in. I have followed the suggestion in <a href="http://www.openssl.org/docs/crypto/BN_CTX_start.html" rel="nofollow">here</a>. Am I missing something? Is there any mistake somewhere? Anyone can help?</p> <p>Thanks</p> <p>EDITED:</p> <p>I made some modification to code but the issue still exists. I changed <code>big_number_as_decimal_from_hex_array</code> as follows</p> <pre><code>char *big_number_as_decimal_from_hex_array_ex(unsigned char *chr_a, int len) { BN_CTX *bn_ctx = BN_CTX_new(); BN_CTX_start(bn_ctx); unsigned char *hex_s = char_array_as_hex(chr_a, len); BIGNUM *big_number = BN_CTX_get(bn_ctx); BN_hex2bn(&amp;big_number, (char *)hex_s); char *big_number_as_decimal = BN_bn2dec(big_number); free(hex_s); BN_free(big_number); BN_CTX_end(bn_ctx); BN_CTX_free(bn_ctx); return big_number_as_decimal; } </code></pre> <p>and also</p> <pre><code>char *big_number_as_decimal_from_hex_array_ex_2(unsigned char *chr_a, int len) { BN_CTX *bn_ctx = BN_CTX_new(); unsigned char *hex_s = char_array_as_hex(chr_a, len); BIGNUM *big_number = BN_CTX_get(bn_ctx); BN_hex2bn(&amp;big_number, (char *)hex_s); char *big_number_as_decimal = BN_bn2dec(big_number); free(hex_s); BN_free(big_number); BN_CTX_free(bn_ctx); return big_number_as_decimal; } </code></pre> <p>I modified the <code>test_compute_prime256v1</code> as </p> <pre><code>void test_compute_prime256v1_ex() { unsigned char seed_a[20] = {...}; printf("s = %s\n", big_number_as_decimal_from_hex_array_ex(seed_a, 20)); unsigned char p_a[32] = {...}; printf("p = %s\n", big_number_as_decimal_from_hex_array_ex(p_a, 32)); // or unsigned char seed_a[20] = {...}; printf("s = %s\n", big_number_as_decimal_from_hex_array_ex_2(seed_a, 20)); unsigned char p_a[32] = {...}; printf("p = %s\n", big_number_as_decimal_from_hex_array_ex_2(p_a, 32)); } </code></pre> <p>but the code produces the same incorrect result in a looped calculation</p>
    singulars
    1. PO
      singulars
      1. PTAnswer
    1. This table or related slice is empty.
    1. PTQuestion
    1. USanonim
    1. USanonim
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. POBIGNUM strange behavior in a calculation loop
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. POBIGNUM strange behavior in a calculation loop
      1. This table or related slice is empty.
      1. VTApproveEditSuggestion
    1. COi tested your code under linux gcc 64bits and didn't got any weird result. i didn't used any gui though so code is not called within a gui-driven callback.
      singulars
      1. POBIGNUM strange behavior in a calculation loop
      1. USphilippe lhardy
    2. COquick guess would be to reset context between s computation and p or to get two bignumbers straigh away since http://www.openssl.org/docs/crypto/BN_CTX_start.html claims 'All BN_CTX_get() calls must be made before calling any other functions that use the ctx as an argument.'
      singulars
      1. POBIGNUM strange behavior in a calculation loop
      1. USphilippe lhardy
    3. COI modified the code to reflect your suggestion but the result still the same. It produces incorrect result in a looped calc. I've tested it on an iOS 6.0 simulator (iPhone 5.0) under mac-os-x lion 64bit.
      singulars
      1. POBIGNUM strange behavior in a calculation loop
      1. USanonim
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload