[lld/mac] Crash even less on undefined symbols with --icf=all

[thakis]

Summary

Follow-up to https://reviews.llvm.org/D112643. Even after that change, we were
still asserting if two separate functions that are eligible for ICF (same size,
same data, same number of relocs, same reloc types, ...) referred to
Undefineds. This fixes that oversight.

[int3]

lld/MachO/ICF.cpp

117

hmm having an Undefined here basically means that the link will error out, right? maybe add a comment to that effect

(tbh I'm not sure the assert is worth much any more, basically the only thing we are still excluding is LazySymbols...)

[thakis]

lld/MachO/ICF.cpp

117

Correct, it'll error out later. Added a comment here and in the other place.

Thanks!

lld/MachO/ICF.cpp

auto f = [](const Reloc &ra, const Reloc &rb) {
     uint64_t valueA = 0;
     uint64_t valueB = 0;
     if (ra.referent.is<Symbol *>()) {
       const auto *sa = ra.referent.get<Symbol *>();
       const auto *sb = rb.referent.get<Symbol *>();
       if (sa->kind() != sb->kind())
         return false;
       if (!isa<Defined>(sa)) {
-        assert(isa<DylibSymbol>(sa));
+        assert(isa<DylibSymbol>(sa) || isa<Undefined>(sa));
         return sa == sb;
       }
       const auto *da = cast<Defined>(sa);
       const auto *db = cast<Defined>(sb);
       if (!da->isec || !db->isec) {
         assert(da->isAbsolute() && db->isAbsolute());
         return da->value == db->value;
       }

lld/test/MachO/invalid/undefined-symbol.s

 #--- foo.s
 .globl _foo
 .text
 _foo:
   callq _bar
   retq
 
 #--- main.s
-.globl _main
 .text
+
+_anotherref:
+  callq _foo
+  movq $0, %rax
+  retq
+
+.globl _main
 _main:
   callq _foo
   movq $0, %rax
   retq
+
+.subsections_via_symbols