I needed a 64-bit compatible version of the framework but when I
created the 64-bit target the unit tests failed. The only thing that
looked suspicious was the SHA1 implementation used by hmac_sha1, so I
replaced the HMAC function in hmac_sha1.c with the Common Crypto
implementation, CCHmac, and the unit tests all passed.
> I needed a 64-bit compatible version of the framework but when I
> created the 64-bit target the unit tests failed. The only thing that
> looked suspicious was the SHA1 implementation used by hmac_sha1, so I
> replaced the HMAC function in hmac_sha1.c with the Common Crypto
> implementation, CCHmac, and the unit tests all passed.
On Mon, Nov 16, 2009 at 12:00 PM, Karl Adam <karla...@mac.com> wrote:
> No reason at all, and this is a good fix. Do you have a patch?
> _Karl
> On Nov 15, 2009, at 15:01:14, Doug Richardson wrote:
> > I needed a 64-bit compatible version of the framework but when I
> > created the 64-bit target the unit tests failed. The only thing that
> > looked suspicious was the SHA1 implementation used by hmac_sha1, so I
> > replaced the HMAC function in hmac_sha1.c with the Common Crypto
> > implementation, CCHmac, and the unit tests all passed.
> On Mon, Nov 16, 2009 at 12:00 PM, Karl Adam <karla...@mac.com> wrote:
> No reason at all, and this is a good fix. Do you have a patch?
> _Karl
> On Nov 15, 2009, at 15:01:14, Doug Richardson wrote:
> > I needed a 64-bit compatible version of the framework but when I
> > created the 64-bit target the unit tests failed. The only thing that
> > looked suspicious was the SHA1 implementation used by hmac_sha1,
> so I
> > replaced the HMAC function in hmac_sha1.c with the Common Crypto
> > implementation, CCHmac, and the unit tests all passed.