hostport.c:238:34: error: 'last_ai' may be used uninitialized in this function [-Werror=maybe-uninitialized]

[robert-scheck]

Using Fedora 33 on s390x, building of radsecproxy 1.9.0 fails like this (and only there, other hardware architectures succeed):

[…]
gcc -DPACKAGE_NAME=\"radsecproxy\" -DPACKAGE_TARNAME=\"radsecproxy\" -DPACKAGE_VERSION=\"1.9.0\" -DPACKAGE_STRING=\"radsecproxy\ 1.9.0\" -DPACKAGE_BUGREPORT=\"https://radsecproxy.github.io\" -DPACKAGE_URL=\"\" -DPACKAGE=\"radsecproxy\" -DVERSION=\"1.9.0\" -DHAVE_MALLOPT=1 -DHAVE_LIBNETTLE=1 -DUSE_OPENSSL=1 -I.  -DSYSCONFDIR=\"/etc\"  -g -Wall -Werror -fno-strict-aliasing -I/usr/include -I/usr/include/openssl -Wall -pedantic -Wno-long-long -pthread -DRADPROT_UDP -DRADPROT_TCP -DRADPROT_TLS -DRADPROT_DTLS -O2 -flto=auto -ffat-lto-objects -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -fstack-protector-strong -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1  -m64 -march=zEC12 -mtune=z13 -fasynchronous-unwind-tables -fstack-clash-protection -c -o hostport.o hostport.c
make[1]: Leaving directory '/builddir/build/BUILD/radsecproxy-1.9.0'
hostport.c: In function 'resolvepassiveaddrinfo':
hostport.c:238:34: error: 'last_ai' may be used uninitialized in this function [-Werror=maybe-uninitialized]
  238 |                 last_ai->ai_next = hp->addrinfo;
      |                 ~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~
cc1: all warnings being treated as errors
make[1]: *** [Makefile:643: hostport.o] Error 1

Full logs:

• https://kojipkgs.fedoraproject.org//packages/radsecproxy/1.9.0/1.fc33/data/logs/s390x/build.log
• https://koji.fedoraproject.org/koji/buildinfo?buildID=1757103

[robert-scheck]

As partially discussed separately via e-mail:

• Build failure happens on an actual IBM mainframe
  • Specifically: Fedora 33 in a chroot environment (via Fedora buildsys mock) on Fedora 34 in KVM guest on LPAR
• Initializing *last_ai in hostport.c:225 to NULL solves the build failure

[robert-scheck]

Shouldn't this commit also go to master branch?

[fmauchle]

Yes, it will. I'm typically fixing stuff that should get released in the next patch in the maint-* branch, and periodically merge these changes into master.