I linked all related tasks for the release of 1.9.1, so that making release notes can be easier and recording information here.

Patch lets it build on xenial for me, thank you.

Patch for this bug is available here, "attachment-0001.bin": https://lists.gnupg.org/pipermail/gcrypt-devel/2021-January/005079.html

I committed the partial result docker container, so I can restart it for investigation. So:

I tested xenial with gcc-5.3 (xenial distro repo) and gcc-5.4 (xenial-updates distro repo) and libgcrypt 1.9.0 from git repo and from tarball. I did not get any errors.

See T5267: Ed25519 backward compatible private key support for preceding ZERO(s) for the fix.

In the next few days I'll be able to boot into Leopard, Mac OS X 10.5.8, to test this. Right now it seems that I need to fix a few problems with updated software sources…

Thank you for your testing.

In T5268#142155, @gniibe wrote:

To support old macOS (< 10.12), I think that code should be something like this:

After libgcrypt 1.9.0 first built on Tiger and 'make check' succeeded I ran 'make clean' and applied the changes to random/rndlinux.c. A simple make succeeded without problem. 'make check' then ran into problems, spitting out many times messages messages like these:

Push the change.
For older versions of MacOS X, I'll handle it later.

https://developer.apple.com/library/archive/documentation/System/Conceptual/ManPages_iPhoneOS/man2/posix_spawn.2.html dated August 9, 2007.
So, I guess that posix_spawn became available in MacOS X Leopard (10.5).

I wrote:

I don't know when the symbol of getentropy was available on macOS.

@ballapete Thank you for testing.

Also support older MacOS X, which has no posix_spawn.

In T5264#142204, @jukivili wrote:

I also needed this patch:

0001-global-fix-compile-error-at-pragma-GCC-diagnostic.patch2 KBDownload

I am going to check whether I need it on Tiger!

In T5264#142204, @jukivili wrote:

I tested building on Ubuntu 8.04 (gcc-4.2) and got same error about cipher_bulk_ops_t. Applying patch fixed that problem.

I see now my mistake: I also have to re-write line #132 which I did not see yet!

With your other patch I still get:

libtool: compile:  /opt/local/bin/gcc-apple-4.2 -DHAVE_CONFIG_H -I. -I.. -I../src -I../src -I../mpi -I../mpi -I/opt/local/include -I/opt/local/include -pipe -Os -std=gnu89 -arch ppc -fno-delete-null-pointer-checks -Wall -MT cipher.lo -MD -MP -MF .deps/cipher.Tpo -c cipher.c  -fno-common -DPIC -o .libs/cipher.o
In file included from cipher.c:31:
./cipher-internal.h:145: error: redefinition of typedef 'cipher_bulk_ops_t'
../src/cipher-proto.h:132: error: previous declaration of 'cipher_bulk_ops_t' was here

In T5264#142203, @jukivili wrote:

Thanks for testing. However, I do not believe patch has been correctly applied.

The "Portfile" that conducts the build process has the patch contained, I sent you the list of files being patched, and here is the end of my translated patch set:

I tested building on Ubuntu 8.04 (gcc-4.2) and got same error about cipher_bulk_ops_t. Applying patch fixed that problem.

Thanks for testing. However, I do not believe patch has been correctly applied.

In T5264#142094, @jukivili wrote:

Here's patch to try out:

0001-cipher-proto-remove-forward-typedef-of-cipher_bulk_o.patch6 KBDownload

In T5268#142155, @gniibe wrote:

To support old macOS (< 10.12), I think that code should be something like this:

diff --git a/random/rndlinux.c b/random/rndlinux.c
index 04e2a464..f378a549 100644

In T5264#142064, @jukivili wrote:

In "src/cipher-proto.h", try removing typedef and leaving just forward declaration of structure.

Performing only this change leads consistently to the well-known error of:

In T5264#142059, @werner wrote:

Do not use -fno-common

The translated bulk patch fails. Compilation still runs into this error:

In T5264#142094, @jukivili wrote:

Here's patch to try out:

0001-cipher-proto-remove-forward-typedef-of-cipher_bulk_o.patch6 KBDownload

Alright!

@ballapete when you have time, could you also test https://dev.gnupg.org/T5268#142155 on Tiger?

To support old macOS (< 10.12), I think that code should be something like this:

• For Ed25519 private key segfault: see T5267
• For macOS getentropy: see T5268
• For invm failure: see T5269

Right now I am compiling with my old PowerBook in Tiger. In case a compilation fails I'll try this patch – on Tiger (Mac OS X 10.4.11)! It's a bit more primitive than Leopard.