[libcxx-commits] [compiler-rt] [libcxx] [libcxxabi] [libunwind] [runtimes][PAC] Harden unwinding when possible (PR #143230)

[Daniil Kovalev via libcxx-commits]

================
@@ -1047,18 +1051,26 @@ class UnwindCursor : public AbstractUnwindCursor{
   bool getInfoFromFdeCie(const typename CFI_Parser<A>::FDE_Info &fdeInfo,
                          const typename CFI_Parser<A>::CIE_Info &cieInfo,
                          pint_t pc, uintptr_t dso_base);
-  bool getInfoFromDwarfSection(pint_t pc, const UnwindInfoSections &sects,
-                                            uint32_t fdeSectionOffsetHint=0);
+  bool getInfoFromDwarfSection(const typename R::link_reg_t &pc,
+                               const UnwindInfoSections &sects,
+                               uint32_t fdeSectionOffsetHint = 0);
   int stepWithDwarfFDE(bool stage2) {
+#if __has_extension(ptrauth_calls)
----------------
kovdan01 wrote:

In this patch, both `__has_extension(ptrauth_calls)` and `__has_feature(ptrauth_calls)` are used. While both work, I suggest to stick with identical style for such checks (if there are no other reasons for using different 'has extension/feature' in particular places).

Probably `__has_extension(ptrauth_calls)` is more preferable since this way of doing such a check is already used in existing codebase.

Applies to this place and two other similar places in this file.

https://github.com/llvm/llvm-project/pull/143230