LLD: Fix large integer implicitly truncated to unsigned type warning

[kzhuravl]

Summary

Patch by @b-sumner

[ruiu]

Both Clang and MSVC don't emit a warning on this code. What compiler are you using?

[b-sumner]

g++4.8.2 is complaining:

../../../InputSection.h: In constructor 'lld::elf::SectionPiece::SectionPiece(size_t, uint32_t, bool)':
../../../InputSection.h:219:41: error: large integer implicitly truncated to unsigned type [-Werror=overflow]

Live(Live || !Config->GcSections) {}
                                ^

[ruiu]

Thanks, I confirmed.

Since we have plenty bits for OutputOff, changing the type of OutputOff from uint64_t to int64_t seems better than using an obscure very large constant.

[kzhuravl]

Address review feedback.

[ruiu]

LGTM

Please mention that this fixes a gcc warning.

[kzhuravl]

rL316365.

ELF/InputSection.h

 // be found by looking at the next one) and put the hash in a side table.
 struct SectionPiece {
   SectionPiece(size_t Off, uint32_t Hash, bool Live)
       : InputOff(Off), Hash(Hash), OutputOff(-1),
         Live(Live || !Config->GcSections) {}
 
   uint32_t InputOff;
   uint32_t Hash;
-  uint64_t OutputOff : 63;
+  int64_t OutputOff : 63;
   uint64_t Live : 1;
 };
 
 static_assert(sizeof(SectionPiece) == 16, "SectionPiece is too big");
 
 // This corresponds to a SHF_MERGE section of an input file.
 class MergeInputSection : public InputSectionBase {
 public: