build fails for hurd-i386 (Debian Bug 881179)

[dun]

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=881179

Builds of munge for hurd-i386 (admittedly not a release architecture) have been failing:

  configure: checking authentication support
  checking for getpeereid... no
  checking for getpeerucred... no
  checking ucred.h usability... no
  checking ucred.h presence... no
  checking for ucred.h... no
  checking for struct ucred... no
  checking for SO_PEERCRED sockopt... no
  checking for struct xucred... no
  checking for LOCAL_PEERCRED sockopt... no
  checking for struct strrecvfd... yes
  checking if file descriptors can be passed over a fifo... no
  checking for /dev/spx... no
  checking for authentication method... failed
  configure: error: cannot determine authentication method

The immediate cause of failure for the fifo-based test was the Hurd's famous lack of PATH_MAX:

  configure:14718: checking if file descriptors can be passed over a fifo
  configure:14773: gcc -o conftest -g -O2 -fdebug-prefix-map=/<<PKGBUILDDIR>>=. -specs=/usr/share/dpkg/pie-compile.specs -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -specs=/usr/share/dpkg/pie-link.specs -Wl,-z,relro -Wl,-z,now conftest.c  >&5
  conftest.c: In function 'main':
  conftest.c:91:11: error: 'PATH_MAX' undeclared (first use in this function)

When locally patched to avoid relying on PATH_MAX, the test compiled, but proceeded to fail at runtime with ENOTTY (Inappropriate ioctl for device). I suspect, but have NOT confirmed, that a variant of this approach that uses I_RECVFD over a Unix domain socket would fare better.

Could you please take a look?

Thanks!

--
Aaron M. Ucko, KB1CJC (amu at alum.mit.edu, ucko at debian.org)
http://www.mit.edu/~amu/ | http://stuff.mit.edu/cgi/finger/[email protected]

[dun]

Removing the bug label since this is a failure to build on an unsupported platform, but keeping the issue open until I have a chance to spin up a hurd vm to investigate.