On Thu, Jan 17, 2008 at 07:07:48PM -0500, Dave Thompson wrote:
3. Also (in g only) if aes-*-ige is included in the test set,
as it (now) is by default, subsequent tests or shutdown may fail
in any of various ways, because that mode uses IV of 2block = 32B
but only 1AESblock = 16B is
I valgrind'ed OpenSSL as follows:
I compiled OpenSSL (0.9.8g) with my own random number engine - in order to
generate
pseudo random numbers that are not based on unitialized values (if you run
openssl
without doing this you get infinite warnings - of course).
The results are as follows
I must ask if you compiled OpenSSL with the 'PURIFY' option.
My build is substantially silenced: only two distinct valgrind 'uninitialiased
value' errors when using my custom rand replacement. This is throughout a full
regression test of my software.
I did not use the purify option - there
On Sat, Jan 19, 2008 at 03:40:12PM -0500, Brad House wrote:
I compiled OpenSSL (0.9.8g) with my own random number engine - in order to
generate
pseudo random numbers that are not based on unitialized values (if you run
openssl
without doing this you get infinite warnings - of course).
The
The reason he said he used his own pseudo
random number generator is
actually because of the PURIFY option
that he didn't turn on.
Actually the reason I use my own random number generator is because the openssl
generator is not thread safe in the following way: I should be able to create a
