That is a gnutls problem.

I do not understand what is going on. rshift does not look different than add1.
Weird. Comparing with the non-asm versions might give further insight.

Here are the nm's .

If some extra output from libgcrypt-config is helpful for you, just let me know
and it will be added. But please use the mailing list for such a discussion.

We have never used pkg-config. That there has never been a build dependecy on it.

D54: 165_libgcrypt-mpi-symbol-alpha.diff

Ah well, this is an alpha. As a quick workaround you can use

--disable-padlock disables the asm code and thus you should not get the same
compile error. Please check that there is no

Assuming you're talking about an extra dependency for building gcrypt, it would
be easy to build and install the *.pc without using the autoconf macros that
come with pkg-config. That said, everyone has pkg-config these days--apparently
gcrypt had a build-dependency on it and nobody even noticed (IIRC it was being
build, just not installed)? That's a decent indicator that nobody would mind, IMHO.

It was accidently left in the SVN. I remove it.

D55: 164_gcrypt-pkgconfig.diff

Here you go. Just let me know if you need anything else.

Can you please send me the file config.log by private mail (wk@gnupg.org).

Thanks for the quick response. I'm using version 1.4.0 and the only
modification I made was to cipher/random.h. I added #include
"gpg-errors.h" to get rid of an error I was receiving concerning
gpg_error_t. Other than that, to configure I just took the defaults by
running ./configure. Everything appeared to compile and install without
error. However, when running make check I see the following:

I just added #include "gpg-error.h" to random.h and all compiled OK.

I'm pretty sure I have it installed correctly. There's really not that much to
it. All of the gpg-error files are in /usr/local and I do see the gpg-error.h
file with the typedef for gpg_error_t. Is it possible that there is a problem
with gcc-3.1?

Did you install libgpg-error properly? The type gpg_error_t is defined in
gpg-error.h and included by gcrypt.h.

I made the following changes in random.h that at least allowed the program to
finish making:

Given that libgcrypt builds fine on non-GNU platforms, I suggest to take this to
the GNU gold developers.

Thanks; I tried that with exactly the same result.

use

Fixed in my working copy. Thanks.

Okay, fixed in the trunk (r1276). Thanks.

You are obviously right.

Implemented in Libgcrypt trunk.

Partly done in 1.3. This will be improved over time. Using the visibility
attribute has higher priority, though.

I have commited a similar fix to the SVN (-r1214).

Moved to libgcrypt until we have implemented this feature.

Scheduled for 1.3

Fixed in 1.2.6 (to be released today) and in the trunk's SVN.

Fixed in in version 1.5. Released today.

Checking the current libgcrypt11 package from Sid, shows that tuis pathc is not
anymore used. Thus I cose this bug.

I think I addressed this with revision 1180. It is easy to add more fallback
types, although I stuck to socklen_t only for now. I will check if I can reach
some of the original submitters for testing.

I think it is sufficient to do it for trunk (1.3).

Yeah, I agree that we need to generate it with configure.

Marcus, any opinion on this?

Duplicate of T236

Resolved a long time ago.

We switched the thread support model in the meantime. You should not have this
problem anymore.

Fixed. Thanks.

Fixed in 1.2.1, releases earlier.