I was checking out a small memory leak in one of my crypt32 tests, and I've finally found the culprit: a test that deliberately causes a page fault (to check that it's handled.) It's still handled, but the function that generates the page fault, CRYPT_AsnEncodeRdn, doesn't have its __TRY block executed. Instead, its caller, CRYPT_AsnEncodeName, does. Thus memory allocated in CRYPT_AsnEncodeRdn is never freed.
Is there anything I can do about that? --Juan
__________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com
Juan Lang wrote:
I was checking out a small memory leak in one of my crypt32 tests, and I've finally found the culprit: a test that deliberately causes a page fault (to check that it's handled.) It's still handled, but the function that generates the page fault, CRYPT_AsnEncodeRdn, doesn't have its __TRY block executed. Instead, its caller, CRYPT_AsnEncodeName, does. Thus memory allocated in CRYPT_AsnEncodeRdn is never freed.
Is there anything I can do about that?
Juan,
are you sure about that ? I just tested it here and it works fine.
A+
-
Eric Pouech -
Juan Lang