[lld-macho] Don't crash upon seeing uleb128 ptr encoding while parsing CIE.

[oontvoo]

Summary

Follow up to D130804

Fixes #56576.

[oontvoo]

formatting

[int3]

Why not parse the pointer instead of ignoring it?

(I assume you're putting this diff up because libffi also uses DW_EH_PE_uleb128?)

lld/MachO/InputFiles.cpp

struct CIE {
   bool fdesHaveAug = false;
   uint8_t lsdaPtrSize = 0; // 0 => no LSDA
   uint8_t funcPtrSize = 0;
 };
 
 static uint8_t pointerEncodingToSize(uint8_t enc) {
   switch (enc & 0xf) {
   case dwarf::DW_EH_PE_absptr:
+  case dwarf::DW_EH_PE_signed:
     return target->wordSize;
+  case dwarf::DW_EH_PE_udata2:
+  case dwarf::DW_EH_PE_sdata2:
+    return 2;
   case dwarf::DW_EH_PE_sdata4:
+  case dwarf::DW_EH_PE_udata4:
     return 4;
   case dwarf::DW_EH_PE_sdata8:
+  case dwarf::DW_EH_PE_udata8:
     // ld64 doesn't actually support sdata8, but this seems simple enough...
     return 8;
+  case dwarf::DW_EH_PE_uleb128:
+    return 128;
+
   default:
     return 0;
   };
 }
 
+static bool supportedEnc(uint8_t enc) {
+  switch (enc & 0xf) {
+  case dwarf::DW_EH_PE_uleb128:
+  case dwarf::DW_EH_PE_udata2:
+  case dwarf::DW_EH_PE_sleb128:
+  case dwarf::DW_EH_PE_sdata2:
+    return false;
+  }
+  return true;
+}
+
 static CIE parseCIE(const InputSection *isec, const EhReader &reader,
                     size_t off) {
   // Handling the full generality of possible DWARF encodings would be a major
   // pain. We instead take advantage of our knowledge of how llvm-mc encodes
   // DWARF and handle just that.
   constexpr uint8_t expectedPersonalityEnc =
       dwarf::DW_EH_PE_pcrel | dwarf::DW_EH_PE_indirect | dwarf::DW_EH_PE_sdata4;
 
case 'L': {
       if (cie.lsdaPtrSize == 0)
         reader.failOn(off, "unexpected LSDA encoding 0x" +
                                Twine::utohexstr(lsdaEnc));
       break;
     }
     case 'R': {
       uint8_t pointerEnc = reader.readByte(&off);
       cie.funcPtrSize = pointerEncodingToSize(pointerEnc);
-      if (cie.funcPtrSize == 0 || !(pointerEnc & dwarf::DW_EH_PE_pcrel))
+
+      // If the encoding is unsupported, we just ignore it (similar to LD64),
+      // rather than crashing here.
+
+      if (supportedEnc(pointerEnc) &&
+          (cie.funcPtrSize == 0 || !(pointerEnc & dwarf::DW_EH_PE_pcrel)))
         reader.failOn(off, "unexpected pointer encoding 0x" +
                                Twine::utohexstr(pointerEnc));
+
       break;
     }
     default:
       break;
     }
   }
   if (personalityAddrOff != 0) {
     auto personalityRelocIt =

lld/test/MachO/eh-frame-uleb128.s

+# REQUIRES: x86
+
+## Test that we correctly ignore unsupported uleb128 DWARF pointer encoding, rather than crashing.
+## Similarly to the 3h-frame-sdata4.s test, this test uses hand-roll EH frames.
+
+# RUN: llvm-mc -filetype=obj -triple=x86_64-apple-darwin %s -o %t.o
+# RUN: %lld %t.o -o %t.out
+# RUN: llvm-objdump --macho --syms --dwarf=frames
+
+## There should be no eh_frame contents.
+
+.globl  _main
+_main:
+  retq
+
+.section __TEXT,__eh_frame,coalesced,no_toc+strip_static_syms+live_support
+EHFrame0:
+ .balign 8
+LCIE:
+ .set Lset0,LECIE-LSCIE
+ .long Lset0
+LSCIE:
+ .long 0
+ .byte 1
+ .ascii "zR\0"
+ .byte 1
+ .byte 1
+ .byte 1
+ .byte 1
+ .byte 1
+ .byte 1
+ .byte 0x80+16, 1
+
+ .balign 8
+LECIE:
+
+ .subsections_via_symbols