Mark wrote:
SSL_MODE_ACCEPT_MOVING_WRITE_BUFFER is likely for those OSes (*glares
at Windows in particular*) which require locking allocated memory in
place, performing an operation on it, and then unlocking it (to allow
the OS to manage the placement of the memory block in physical
memory).
There are Windows functions (GlobalAlloc) that can allocate moveable
blocks of memory. Their use is discouraged and only really exist for
legacy code compatibility.
Okay, so how does either of those two interpretations of what it means
affect OpenSSL then ?
I read the first poster saying it is to do with pinning memory down, the
the second poster saying it is to do with movable blocks of memory. If
I am understanding; these are direct opposites. Maybe you are both
presuming a reason as opposed to knowing the historical background of
OpenSSL to know for sure why it existed ? I'm really after that
historical understanding.
The only use of the attribute in the OpenSSL source tree is in one place
for SSL3 protocol, one (equivalent) place for the SSL2 protocol, in a
bogus looking test that appears to have no useful purpose.
Should I seek to approach the development team to remove all traces of
the attribute if it serves no purpose. (Other than cause confuse and worry)
Darryl
______________________________________________________________________
OpenSSL Project http://www.openssl.org
User Support Mailing List openssl-users@openssl.org
Automated List Manager [EMAIL PROTECTED]
