Re: [openssl.org #3089] Building OpenSSL 1.0.1e with FIPS on Win64A

2013-07-11 Thread Steve Marquess
On 07/10/2013 03:46 PM, Graeme Perrow via RT wrote:
 I am trying to build the FIPS Object Module for Windows on an AMD64
 machine. I started with the instructions in section 4.3 of the User Guide
 2.0, and was able to build the FIPS module itself, but the instructions for
 building a FIPS-capable OpenSSL are specific to 32-bit Windows. I adjusted
 the build procedure as follows:
 ...
 Also (and more importantly), if I have to modify the build procedure for
 the FIPS-capable OpenSSL but not for the FIPS Object Module itself, does
 that mean my Module is not FIPS 140-2 validated?

I think this is more of a user list question.

OpenSSL proper (as opposed to the OpenSSL FIPS Object Module) is out of
scope of the FIPS 140-2 validation procedure, so you can hack it to your
hearts content. You need to embed the HMAC-SHA1 integrity check
(incore) digest in the FIPS module embedded in the shared library
executable file, but you aren't constrained to a specific command or
process.

Also note that you must verify the SHA1 digest of the FIPS module files
(as is done automatically in the fipsld script). Sort of moot if you
just generated those files, but a technical requirement nonetheless.

-Steve M.

-- 
Steve Marquess
OpenSSL Software Foundation, Inc.
1829 Mount Ephraim Road
Adamstown, MD  21710
USA
+1 877 673 6775 s/b
+1 301 874 2571 direct
marqu...@opensslfoundation.com
marqu...@openssl.com
__
OpenSSL Project http://www.openssl.org
Development Mailing List   openssl-dev@openssl.org
Automated List Manager   majord...@openssl.org


Re: [openssl.org #3089] Building OpenSSL 1.0.1e with FIPS on Win64A

2013-07-11 Thread Steve Marquess via RT
On 07/10/2013 03:46 PM, Graeme Perrow via RT wrote:
 I am trying to build the FIPS Object Module for Windows on an AMD64
 machine. I started with the instructions in section 4.3 of the User Guide
 2.0, and was able to build the FIPS module itself, but the instructions for
 building a FIPS-capable OpenSSL are specific to 32-bit Windows. I adjusted
 the build procedure as follows:
 ...
 Also (and more importantly), if I have to modify the build procedure for
 the FIPS-capable OpenSSL but not for the FIPS Object Module itself, does
 that mean my Module is not FIPS 140-2 validated?

I think this is more of a user list question.

OpenSSL proper (as opposed to the OpenSSL FIPS Object Module) is out of
scope of the FIPS 140-2 validation procedure, so you can hack it to your
hearts content. You need to embed the HMAC-SHA1 integrity check
(incore) digest in the FIPS module embedded in the shared library
executable file, but you aren't constrained to a specific command or
process.

Also note that you must verify the SHA1 digest of the FIPS module files
(as is done automatically in the fipsld script). Sort of moot if you
just generated those files, but a technical requirement nonetheless.

-Steve M.

-- 
Steve Marquess
OpenSSL Software Foundation, Inc.
1829 Mount Ephraim Road
Adamstown, MD  21710
USA
+1 877 673 6775 s/b
+1 301 874 2571 direct
marqu...@opensslfoundation.com
marqu...@openssl.com


__
OpenSSL Project http://www.openssl.org
Development Mailing List   openssl-dev@openssl.org
Automated List Manager   majord...@openssl.org


Crash in RAND_poll for IOS

2013-07-11 Thread Shivanand
Hi,
I have cross compiled openssl for ios and using it for generating random
number and other crypto stuff.

When I try to generate random number, I get a crash in XCode from RAND_poll.

Does the RAND_poll conform with ios ? Is there any known issue with openssl
random generation and ios?

Please let me know.

Regards
Shiva


RE: [Patch] ALPN Implementation for OpenSSL

2013-07-11 Thread Jeff Mendoza (MS OPEN TECH)
 Understood, I'll start working on this behavior:
 
 The client can send ALPN, NPN, or both.
   If the client only sends one: negotiation will take place normally.
   If the client sends both: the server will prefer ALPN. If nothing
 matches with ALPN, it will fall back to NPN and send its list.
 
 Also, we have received some feedback off-list on the code we have already
 posted, and will be reposting with some updates soon.

Hi All,

I have posted our updated ALPN patch, 
http://rt.openssl.org/Ticket/Display.html?id=3073. I'm happy to address any 
feedback. Also, there have been some questions about testing. The new patch has 
added support to s_client and s_server which should enable testing ALPN without 
any dependencies.

Thanks,
Jeff Mendoza


__
OpenSSL Project http://www.openssl.org
Development Mailing List   openssl-dev@openssl.org
Automated List Manager   majord...@openssl.org