[lld] [lld][AArch64][ELF][PAC] Support AUTH relocations and AUTH ELF marking (PR #72714)

[Daniil Kovalev via llvm-commits]

================
@@ -1141,12 +1142,37 @@ void RelocationScanner::processAux(RelExpr expr, RelType type, uint64_t offset,
         (rel == target->symbolicRel && !sym.isPreemptible)) {
       addRelativeReloc<true>(*sec, offset, sym, addend, expr, type);
       return;
-    } else if (rel != 0) {
+    }
+    if (rel != 0) {
       if (config->emachine == EM_MIPS && rel == target->symbolicRel)
         rel = target->relativeRel;
       std::lock_guard<std::mutex> lock(relocMutex);
-      sec->getPartition().relaDyn->addSymbolReloc(rel, *sec, offset, sym,
-                                                  addend, type);
+      Partition &part = sec->getPartition();
+      if (config->emachine == EM_AARCH64 && type == R_AARCH64_AUTH_ABS64) {
+        // For a preemptible symbol, we can't use a relative relocation. For an
+        // undefined symbol, we can't compute offset at link-time and use a
+        // relative relocation. Use a symbolic relocation instead.
+        if (sym.isPreemptible) {
+          part.relaDyn->addSymbolReloc(type, *sec, offset, sym, addend, type);
+        } else if (part.relrAuthDyn && sec->addralign >= 2 && offset % 2 == 0 &&
+                   isInt<32>(sym.getVA(addend))) {
+          // Implicit addend is below 32-bits so we can use the compressed
----------------
kovdan01 wrote:

I deleted the check against 32 bits since we anyway need to re-check that in `finalizeAddressDependentContent`. See b215b0d60ac0c2ac29aeb318da7c95725be6ef86

Regarding this:

> It seems that we should just remove the test here and unconditionally add relocations to part.relrAuthDyn->relocs and do the fixup in `finalizeAddressDependentContent`

I don't think so. We anyway must keep the check against `part.relrAuthDyn` since it might be `nullptr`, so, even if checks `sec->addralign >= 2 && offset % 2 == 0` go to `finalizeAddressDependentContent`, both `else if` and `else` branches here will still be present and the `processAux` won't become significantly more simple. It looks like that we'd better leave `else if (part.relrAuthDyn && sec->addralign >= 2 && offset % 2 == 0)` here just like we have for regular relr relocs in `addRelativeReloc`. It would keep things consistent and we'll only have a single check against 32 bits in `finalizeAddressDependentContent`, while checks `sec->addralign >= 2 && offset % 2 == 0` suite better here in `processAux`. Please let me know if I miss something.



https://github.com/llvm/llvm-project/pull/72714