[TextAPI] Fix off-by-one error in the bit index extraction loop

[mstojanovic]

Summary

The loop in findNextSetBit() runs one pass more than it should.

On 64-bit architectures this doesn't cause a problem, but 32-bit architectures mask the shift count to 5 bits which limits the count range to 0 to 31. Shifting by 32 has the same effect as shifting by 0, so if the first bit in the set is 1 the function will return with Index different from EndIndexVal. Because of that, the ranged for loops iterating thorough architectures will continue until hitting a 0 in the set, resulting in n additional iterations, where n is equal to the number of consecutive 1 bits at the start the set.

Ultimately TBDv1.WriteFile and TBDv2.WriteFile will output additional architectures causing a failure in the unit tests.

[ributzka]

Thanks for catching this. LGTM.

llvm/trunk/include/llvm/TextAPI/MachO/ArchitectureSet.h

class arch_iterator
       : public std::iterator<std::forward_iterator_tag, Architecture, size_t> {
   private:
     ArchSetType Index;
     Ty *ArchSet;
 
     void findNextSetBit() {
       if (Index == EndIndexVal)
         return;
-
-      do {
-        if (*ArchSet & (1UL << ++Index))
+      while (++Index < sizeof(Ty) * 8) {
+        if (*ArchSet & (1UL << Index))
           return;
-      } while (Index < sizeof(Ty) * 8);
+      }
 
       Index = EndIndexVal;
     }
 
   public:
     arch_iterator(Ty *ArchSet, ArchSetType Index = 0)
         : Index(Index), ArchSet(ArchSet) {
       if (Index != EndIndexVal && !(*ArchSet & (1UL << Index)))