[lld-macho][NFC] Remove redundant StringRef construction

[BertalanD]

Summary

It's only used in one branch, so we were unnecessarily calculating the
length of many symbol names.

Tiny speedup when linking chromium_framework on my M1 Mac mini:

x before.txt
+ after.txt

    N           Min           Max        Median           Avg        Stddev
x  10     3.9917109        4.0418     4.0318099     4.0203902   0.021459873
+  10      3.944725      4.053988     3.9708955     3.9825602   0.037257609
Difference at 95.0% confidence
	-0.03783 +/- 0.0285663
	-0.940953% +/- 0.710536%
	(Student's t, pooled s = 0.0304028)

[int3]

Nice find!

[thakis]

Thanks :)

lld/MachO/InputFiles.cpp

void ObjFile::parseSymbols(ArrayRef<typename LP::section> sectionHeaders,
   for (uint32_t i = 0; i < nList.size(); ++i) {
     const NList &sym = nList[i];
 
     // Ignore debug symbols for now.
     // FIXME: may need special handling.
     if (sym.n_type & N_STAB)
       continue;
 
-    StringRef name = strtab + sym.n_strx;
     if ((sym.n_type & N_TYPE) == N_SECT) {
       Subsections &subsections = sections[sym.n_sect - 1]->subsections;
       // parseSections() may have chosen not to parse this section.
       if (subsections.empty())
         continue;
       symbolsBySection[sym.n_sect - 1].push_back(i);
     } else if (isUndef(sym)) {
       undefineds.push_back(i);
     } else {
-      symbols[i] = parseNonSectionSymbol(sym, name);
+      symbols[i] = parseNonSectionSymbol(sym, StringRef(strtab + sym.n_strx));
     }
   }
 
   for (size_t i = 0; i < sections.size(); ++i) {
     Subsections &subsections = sections[i]->subsections;
     if (subsections.empty())
       continue;
     std::vector<uint32_t> &symbolIndices = symbolsBySection[i];