libgcrypt: disable tests w/musl, seems to hang

[dtzWill]

Maybe same problem as on Darwin, unsure.

From gdb:

Thread 1 (process 23820):
#0 0x00007ffff7dab684 in __syscall_cp_c () from target:/nix/store/66m5z7marjbs7pa3gv8sf5j1qrqjacfj-musl-1.1.19/lib/ld-musl-x86_64.so.1
#1 0x00007ffff7daac69 in __timedwait_cp () from target:/nix/store/66m5z7marjbs7pa3gv8sf5j1qrqjacfj-musl-1.1.19/lib/ld-musl-x86_64.so.1
#2 0x00007ffff7daad17 in __timedwait () from target:/nix/store/66m5z7marjbs7pa3gv8sf5j1qrqjacfj-musl-1.1.19/lib/ld-musl-x86_64.so.1
#3 0x00007ffff7dacaf4 in pthread_mutex_timedlock () from target:/nix/store/66m5z7marjbs7pa3gv8sf5j1qrqjacfj-musl-1.1.19/lib/ld-musl-x86_64.so.1
#4 0x00007ffff781e409 in _gpgrt_lock_lock () from target:/nix/store/f7qid95jabfr665qc1kbcl6adf48gq7w-libgpg-error-1.28/lib/libgpg-error.so.0
#5 0x00007ffff7b035d5 in lock_rng () at ./rndjent.c:212
#6 0x00007ffff7b036ab in _gcry_rndjent_poll (add=0x0, origin=RANDOM_ORIGIN_INIT, length=0) at ./rndjent.c:268
#7 0x00007ffff7b038cf in _gcry_rndjent_get_version (r_active=0x7fffffffc800) at ./rndjent.c:339
#8 0x00007ffff7a44f7f in print_config (fp=0x6026e0, what=0x0) at global.c:391
#9 _gcry_get_config (mode=mode@entry=0, what=, what@entry=0x0) at global.c:420
#10 0x00007ffff7a456a3 in _gcry_vcontrol (cmd=, arg_ptr=) at global.c:652
#11 0x00007ffff7a41689 in gcry_control (cmd=cmd@entry=GCRYCTL_PRINT_CONFIG) at visibility.c:79
#12 0x0000000000400ec3 in main (argc=, argv=) at version.c:160

Planning to cherry-pick this onto 18.03 as well
(like was done for the Darwin fix).

• Tested using sandboxing (nix.useSandbox on NixOS, or option sandbox in nix.conf on non-NixOS)
• Built on platform(s)
  • NixOS
  • macOS
  • other Linux distributions
• Tested via one or more NixOS test(s) if existing and applicable for the change (look inside nixos/tests)
• Tested compilation of all pkgs that depend on this change using nix-shell -p nox --run "nox-review wip"
• Tested execution of all binary files (usually in ./result/bin/)
• Fits CONTRIBUTING.md.

---

[GrahamcOfBorg]

Success on aarch64-linux (full log)

Attempted: libgcrypt

Partial log (click to expand)

/nix/store/q9j7ky21scd3lxxjj5yr5n3jpsyd2bj7-libgcrypt-1.8.3

[GrahamcOfBorg]

Success on x86_64-linux (full log)

Attempted: libgcrypt

Partial log (click to expand)

/nix/store/150r1ggzh95354vpdrl0b896hd5hwv7b-libgcrypt-1.8.3

[GrahamcOfBorg]

Success on x86_64-darwin (full log)

Attempted: libgcrypt

Partial log (click to expand)

these paths will be fetched (0.45 MiB download, 1.44 MiB unpacked):
  /nix/store/15djki9fy45azrpnf3cxyy2jmrjymr7p-libgcrypt-1.8.3
  /nix/store/jfrrwb33bs6il3xdm79w8jl2frqnnl8z-libgpg-error-1.28
copying path '/nix/store/jfrrwb33bs6il3xdm79w8jl2frqnnl8z-libgpg-error-1.28' from 'https://cache.nixos.org'...
copying path '/nix/store/15djki9fy45azrpnf3cxyy2jmrjymr7p-libgcrypt-1.8.3' from 'https://cache.nixos.org'...
/nix/store/15djki9fy45azrpnf3cxyy2jmrjymr7p-libgcrypt-1.8.3

[dtzWill]

Oh, nevermind they seem to be trying to obtain a lock they already have, but aren't using a recursive lock?

Will investigate...

[dtzWill]

Exploring locking badness fix in #42150 which replaces this if patch looks good. Closing for now.

(EDIT: oops wrong PR number)