gcc/g++ 4.2.1 -r334125 is getting devd related build failures on ci.freebsd.org

[Mark Millard]

[This submittal is in part an experiment with how Eitan's
MUA classifies the Email using an alternate yahoo.com
address. Eitan: the subject and this note are all unique
to this message.]

I've reported to Eitan that the gcc/g++ 4.2.1 architectures
are failing for devd build failures.

Examples include:

--- all_subdir_sbin ---
cc1plus: warnings being treated as errors
/usr/obj/usr/src/sparc64.sparc64/tmp/usr/include/c++/4.2/bits/basic_string.h: In member function 'std::basic_string<_CharT, _Traits, _Alloc>::_Rep* std::basic_string<_CharT, _Traits, _Alloc>::_M_rep() const [with _CharT = char, _Traits = std::char_traits<char>, _Alloc = std::allocator<char>]':
/usr/obj/usr/src/sparc64.sparc64/tmp/usr/include/c++/4.2/bits/basic_string.h:493:   instantiated from 'std::basic_string<_CharT, _Traits, _Alloc>::~basic_string() [with _CharT = char, _Traits = std::char_traits<char>, _Alloc = std::allocator<char>]'
/usr/src/sbin/devd/devd.hh:149:   instantiated from here
/usr/obj/usr/src/sparc64.sparc64/tmp/usr/include/c++/4.2/bits/basic_string.h:288: warning: cast from 'char*' to 'std::basic_string<char, std::char_traits<char>, std::allocator<char> >::_Rep*' increases required alignment of target type

mips64 and mips and sparc64 agree, for example.

powerpc and powerpc64 report:

--- all_subdir_sbin ---
cc1plus: warnings being treated as errors
/usr/src/sbin/devd/devd.hh: In function 'void __tcf_0(void*)':
/usr/src/sbin/devd/devd.hh:150: warning: inlining failed in call to 'virtual config::~config()': --param max-inline-insns-single limit reached
/usr/src/sbin/devd/devd.cc:175: warning: called from here

but I expect the two sets of types of error message are related
somehow and the "fix" fixes both types.


===
Mark Millard
marklmi at yahoo.com
( dsl-only.net went
away in early 2018-Mar)