Author: Dann Corbit
Date: 16:46:39 08/05/99
Go up one level in this thread
On August 05, 1999 at 19:41:58, Eugene Nalimov wrote:
>I beleive the memory will be released by operating system.
>
>Eugene
>
>On August 05, 1999 at 19:23:16, Gerrit Reubold wrote:
>
>>Hi Eugene, hi all,
>>
>>I have successfully included Nalimov's tablebases in my program, they are
>>working great, thanks again for supplying them.
>>
>>However, I have a small problem with my implementation of Nalimov's probing
>>code, it works fine, but I get memory leaks when exiting the program. The memory
>>was allocated by "IInitializeTb(cPath)", is there a complementary function which
>>I forgot to call on exit? I am using MSVC 6.0, if it helps, I can mail or post
>>(parts of) the debug output.
>>This is not a big problem, because it happens when exiting the program, but it
>>would be nice to exit "clean".
>>
>>Has somebody the same problem? Is there a remedy?
Once you have entered the EGTB initialization functions and successfully
allocated memory, register a function that frees the memory with atexit().
Also, if you are using crafty, then register this funtion with atexit() if you
want purify/electric fence/boundschecker to pipe down:
void release (void)
{
int i;
free (local[0]);
for (i = 0; i < 32; i++)
free (args[i]);
free (EGTB_cache);
free (trans_ref_a_orig);
free (trans_ref_b_orig);
free (pawn_hash_table_orig);
}
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.