I need more information:

• where is pkg-config path for <host_alias>? How is it determined?
  • 32-bit: /lib or /lib32?
  • 64-bit: /lib or /lib64?
  • something like x32: where???

I consider:

• Single gpgrt-config is better (and simpler)
• new option --for-host=<host_alias>? (--host is already used for query for host)
• update *.m4 using this new option to provide host information to determine the path

libassuan: Done
libksba: Done
npth: Done
ntbtls: Done
libgcrypt: Done

A bit tricky, but this would be good to use gpgrt-config by gpg-error.m4.
I say "tricky", because its name is gpg-error.m4 but it configure GPGRT_CONFIG to access to GPG_ERROR_CONFIG.

It might be good idea to provide libgcrypt.pc in libgcrypt 1.8.x for forward compatibility with libgpg-error 1.33.

Well, I changed my mind. Use of new gpgrt-config requires software update to introduce gpgrt.m4 and update of configure.ac to switch gpgrt from gpg-error, in standard way.
That's too much this time. It's good to defer this change.

Sorry, there is no good way, but only workaround in this case, because it is complicated and it is basically wrong thing to do it by composite device (in my opinion). I'll describe detail in this comment.
I'm not Windows user, so, I don't have an idea how to recover from such a situation.

OK, I'll change to use gpgrt-config, along with requiring newer version of libgpg-error.

This is the dependency graph:

While we have no way to express possible range of values in structure member, we can do something like this (which is not needed actually, if the gmtime function returns correct value).

Here is the warning:

../../libgpg-error/doc/yat2m.c: In function 'isodatestring':
../../libgpg-error/doc/yat2m.c:373:29: warning: '%02d' directive writing between 2 and 11 bytes into a region of size between 4 and 11 [-Wformat-overflow=]
       sprintf (buffer,"%04d-%02d-%02d",
                             ^~~~
../../libgpg-error/doc/yat2m.c:373:23: note: directive argument in the range [-2147483647, 2147483647]
       sprintf (buffer,"%04d-%02d-%02d",
                       ^~~~~~~~~~~~~~~~
../../libgpg-error/doc/yat2m.c:373:7: note: 'sprintf' output between 11 and 36 bytes into a destination of size 16
       sprintf (buffer,"%04d-%02d-%02d",
       ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
                1900+tp->tm_year, tp->tm_mon+1, tp->tm_mday );
                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Apparently, it is not the bug of gpg, but you just specified wrong line in your /etc/apt/souces.list.d/skypeforlinux.list, where filename extension .gpg is irrelevant.

Done for libgpg-error.
Will extend to other software.

Applied to master: rE91bcb2c7d824: gpg-error-config: Support "auto", architecture independent script.

there should be clearer labelling of smartcards so that users can tell them apart more easily

I did a small update to D467: Enable dynamically defining pkg_config_libdir for multiarch. The name is changed to "auto" (was: "unknown"). It now support other cases where CC is not a kind of gcc.

Support the case when CC is the one of clang.

@werner, I think that the scope is different. The bug reporters' claim were basically "GnuPG's cross building is different (for them), why?". They didn't claim GnuPG were unable to be cross-build.

This patch D467: Enable dynamically defining pkg_config_libdir for multiarch, enables multiarch dynamic PKG_CONFIG_LIBDIR support.