Note that all *_it variables are suddenly non-existant according to
libeay.num.  This is a bug that will be corrected.  Please be patient.

to include (but the OpenSSL_0_9_6-stable branche still has
inconsistencies)

and make all files the depend on it include it without prefixing it
with openssl/.

This means that all Makefiles will have $(TOP) as one of the include
directories.

compared to it has the type size_t.  Included the needed headers to
make that happen.

get a chance to be defined.

missed any.

This compiles and runs on Linux, and external applications have no
problems with it.  The definite test will be to build this on VMS.

sure they are available in opensslconf.h, by giving them names starting
with "OPENSSL_" to avoid conflicts with other packages and by making
sure e_os2.h will cover all platform-specific cases together with
opensslconf.h.

I've checked fairly well that nothing breaks with this (apart from
external software that will adapt if they have used something like
NO_KRB5), but I can't guarantee it completely, so a review of this
change would be a good thing.

entropy devices or sockets.

during connect() and other calls. First seen on Unixware-7.

Unify access to EGD-socket for all RAND_egd_*() methods.

Problem reported by "Wolfgang Marczy" <WMarczy@topcall.co.at>
in a message to openssl-dev (19 Dec 2000 13:40:51 +0100).

OCSP basic response verify. Very incomplete
but will verify the signatures on a response
and locate the signers certifcate.

Still needs to implement a proper OCSP certificate
verify.

Fix warning in RAND_egd().

only queried when the /dev/[u]random devices did not return enough
entropy. Only the amount of entropy missing to reach the required minimum
is queried, as EGD may be drained.
Queried locations are: /etc/entropy, /var/run/egd-pool

correctly, but RAND_add(..,n) counts the increasing n several times.
Only RAND_add(..,n) once entropy collection is finished.

them for a short period of time (actually, poll them with select(),
then read() whatever is there), which is about 10ms (hard-coded value)
each.

Separate Windows and Unixly code, and start on a VMS variant that
currently just returns 0.

Spotted by "David Schwartz" <davids@webmaster.com>.

The randomness may not actually be very good (we don't know).

At the same time, add VMS support for Rijndael.

returns -1...

acceptable, since all that happens if it fails is a library with
an index, which makes linking slower, but still working correctly.

work better in Win32.  Verified by zhu qun-ying <qyzhu@krdl.org.sg>.

so disable that part.
Reported by Jeffrey Altman <jaltman@columbia.edu>

disagree on the proper syntax and type names.

Jeffrey Altman reminds us to initialize some variables and ercommends the use of LOadLibrary instead of GetModuleHandle

problems with GetCursorInfo, but also adds network statistics and
performance statistics where available.

doesn't quite work on WinNT 4 earlier than SP6.  It works fine on
Windows 98 and Windows 2000.

I'm disabling it for now.  What's really needed is some kind of check
to see if GetCursorInfo is safe to call, or alternatively, GetCursor
or GetCursorPos could be used, according to Jeffrey.

no way to make sure GetCursorInfo will give us a valid answer.

The copy at www.usenix.org is the old version.

- Make sure PCURSORINFO is defined even on systems that do not provide it.
- Change the reference to Peter Gutmann's paper.
- Make sure we don't walk the whole heap lists for performance reasons.
  Jeffrey Altman suggests following Peter Gutmann's advice to keep it
  to 50 heap entries per heap list.