[lld/mac] Handle symbols from -U in treatUndefinedSymbol()

[thakis]

Summary

In ld64, -U section$start$FOO$bar handles section$start$FOO$bar
as a regular section$start symbol, that is section$start processing
happens before -U processing.

Likely, nobody uses that in practice so it doesn't seem very important
to be compatible with this, but it also moves the -U handling code next
to the -undefined dynamic_lookup handling code, which is nice because
they do the same thing. And, in fact, this did identify a bug in a corner
case in the intersection of -undefined dynamic_lookup and dead-stripping
(fix for that in D106565).

Vaguely related to PR50760.

No interesting behavior change.

[int3]

hmm, looks like there are behavioral changes based on the test failures...

[int3]

ah I see, it's a diff dependency issue. lgtm

lld/MachO/Config.h

 //===- Config.h -------------------------------------------------*- C++ -*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 #ifndef LLD_MACHO_CONFIG_H
 #define LLD_MACHO_CONFIG_H
 
 #include "llvm/ADT/CachedHashString.h"
 #include "llvm/ADT/DenseMap.h"
 #include "llvm/ADT/DenseSet.h"
 #include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/StringSet.h"
 #include "llvm/BinaryFormat/MachO.h"
 #include "llvm/Support/CachePruning.h"
 #include "llvm/Support/GlobPattern.h"
 #include "llvm/Support/VersionTuple.h"
 #include "llvm/TextAPI/Architecture.h"
 #include "llvm/TextAPI/Platform.h"
 #include "llvm/TextAPI/Target.h"
 
struct Configuration {
   ICFLevel icfLevel = ICFLevel::none;
   llvm::MachO::HeaderFileType outputType;
   std::vector<llvm::StringRef> systemLibraryRoots;
   std::vector<llvm::StringRef> librarySearchPaths;
   std::vector<llvm::StringRef> frameworkSearchPaths;
   std::vector<llvm::StringRef> runtimePaths;
   std::vector<std::string> astPaths;
   std::vector<Symbol *> explicitUndefineds;
+  llvm::StringSet<> explicitDynamicLookups;
   // There are typically few custom sectionAlignments or segmentProtections,
   // so use a vector instead of a map.
   std::vector<SectionAlign> sectionAlignments;
   std::vector<SegmentProtection> segmentProtections;
 
   llvm::DenseMap<llvm::StringRef, SymbolPriorityEntry> priorities;
   SectionRenameMap sectionRenameMap;
   SegmentRenameMap segmentRenameMap;

lld/MachO/Driver.cpp

if (!get_threadpool_strategy(config->thinLTOJobs))
     error("--thinlto-jobs: invalid job count: " + config->thinLTOJobs);
 
   for (const Arg *arg : args.filtered(OPT_u)) {
     config->explicitUndefineds.push_back(symtab->addUndefined(
         arg->getValue(), /*file=*/nullptr, /*isWeakRef=*/false));
   }
 
   for (const Arg *arg : args.filtered(OPT_U))
-    symtab->addDynamicLookup(arg->getValue());
+    config->explicitDynamicLookups.insert(arg->getValue());
 
   config->mapFile = args.getLastArgValue(OPT_map);
   config->optimize = args::getInteger(args, OPT_O, 1);
   config->outputFile = args.getLastArgValue(OPT_o, "a.out");
   config->finalOutput =
       args.getLastArgValue(OPT_final_output, config->outputFile);
   config->astPaths = args.getAllArgValues(OPT_add_ast_path);
   config->headerPad = args::getHex(args, OPT_headerpad, /*Default=*/32);

lld/MachO/SymbolTable.cpp

Defined *s = addDefined(name, nullptr, isec, value, /*size=*/0,
                           /*isWeakDef=*/false, isPrivateExtern,
                           /*isThumb=*/false, referencedDynamically,
                           /*noDeadStrip=*/false);
   s->includeInSymtab = includeInSymtab;
   return s;
 }
 
 void lld::macho::treatUndefinedSymbol(const Undefined &sym, StringRef source) {
+  // Handle -U.
+  if (config->explicitDynamicLookups.count(sym.getName())) {
+    symtab->addDynamicLookup(sym.getName());
+    return;
+  }
+
+  // Handle -undefined.
   auto message = [source, &sym]() {
     std::string message = "undefined symbol";
     if (config->archMultiple)
       message += (" for arch " + getArchitectureName(config->arch())).str();
     message += ": " + toString(sym);
     if (!source.empty())
       message += "\n>>> referenced by " + source.str();
     else

lld/test/MachO/dead-strip.s

 # RUN: %lld -lSystem -dead_strip %t/strip-dylib-ref.o %t/dylib.dylib \
 # RUN:     -o %t/strip-dylib-ref -U _ref_undef_fun -U _unref_undef_fun
 # RUN: llvm-objdump --syms --bind --lazy-bind --weak-bind %t/strip-dylib-ref | \
 # RUN:     FileCheck --check-prefix=STRIPDYLIB --implicit-check-not _unref %s
 # STRIPDYLIB:      SYMBOL TABLE:
 # STRIPDYLIB-NEXT:  l {{.*}} __dyld_private
 # STRIPDYLIB-NEXT:  g {{.*}} _main
 # STRIPDYLIB-NEXT:  g {{.*}} __mh_execute_header
-# STRIPDYLIB-NEXT:  *UND* _ref_undef_fun
 # STRIPDYLIB-NEXT:  *UND* dyld_stub_binder
 # STRIPDYLIB-NEXT:  *UND* _ref_dylib_fun
+# STRIPDYLIB-NEXT:  *UND* _ref_undef_fun
 # STRIPDYLIB:      Bind table:
 # STRIPDYLIB:      Lazy bind table:
 # STRIPDYLIB:       __DATA   __la_symbol_ptr {{.*}} flat-namespace _ref_undef_fun
 # STRIPDYLIB:       __DATA   __la_symbol_ptr {{.*}} dylib _ref_dylib_fun
 # STRIPDYLIB:      Weak bind table:
 ## Stubs smoke check: There should be two stubs entries, not four, but we
 ## don't verify that they belong to _ref_undef_fun and _ref_dylib_fun.
 # RUN: llvm-objdump -d --section=__stubs --section=__stub_helper \

 # LIVESUPP-NEXT:   l {{.*}} _ref_ls_fun_bw
 # LIVESUPP-NEXT:   l {{.*}} _ref_ls_dylib_fun
 # LIVESUPP-NEXT:   l {{.*}} _ref_ls_undef_fun
 # LIVESUPP-NEXT:   l {{.*}} __dyld_private
 # LIVESUPP-NEXT:   g {{.*}} _main
 # LIVESUPP-NEXT:   g {{.*}} _bar
 # LIVESUPP-NEXT:   g {{.*}} _foo
 # LIVESUPP-NEXT:   g {{.*}} __mh_execute_header
-# LIVESUPP-NEXT:   *UND* _ref_undef_fun
 # LIVESUPP-NEXT:   *UND* dyld_stub_binder
 # LIVESUPP-NEXT:   *UND* _ref_dylib_fun
+# LIVESUPP-NEXT:   *UND* _ref_undef_fun
 
 # RUN: llvm-mc -filetype=obj -triple=x86_64-apple-macos \
 # RUN:     %t/live-support-iterations.s -o %t/live-support-iterations.o
 # RUN: %lld -lSystem -dead_strip %t/live-support-iterations.o \
 # RUN:     -o %t/live-support-iterations
 # RUN: llvm-objdump --syms %t/live-support-iterations | \
 # RUN:     FileCheck --check-prefix=LIVESUPP2 --implicit-check-not _unref %s
 # LIVESUPP2-LABEL: SYMBOL TABLE: