[LLD] Added STARTUP linker script directive

The STARTUP linker script directive works similar to the INPUT directive
with one file specified, but it also puts the specified file as the
first file to be parsed, similar to putting an object file to be the
first parameter on the command line.

Errors are emitted if two STARTUP directives were found (there can't be
two or more first files) and if archive files are passed as the
argument (not yet supported).

Author: AndrijaSyrmia

lld/ELF/Config.h

@@ -105,7 +105,7 @@ struct VersionDefinition {
 class LinkerDriver {
 public:
   void linkerMain(ArrayRef<const char *> args);
-  void addFile(StringRef path, bool withLOption);
+  void addFile(StringRef path, bool withLOption, bool toBeginning);
   void addLibrary(StringRef name);
 
 private:

lld/ELF/Driver.cpp

@@ -225,8 +225,18 @@ static bool isBitcode(MemoryBufferRef mb) {
   return identify_magic(mb.getBuffer()) == llvm::file_magic::bitcode;
 }
 
+// Adds a file to the end or beginning of the file vector
+inline static void addFileToVec(std::vector<InputFile *> &files,
+                                InputFile *file, bool toBeginning) {
+  if (LLVM_UNLIKELY(toBeginning)) {
+    files.insert(files.begin(), file);
+  } else {
+    files.push_back(file);
+  }
+}
+
 // Opens a file and create a file object. Path has to be resolved already.
-void LinkerDriver::addFile(StringRef path, bool withLOption) {
+void LinkerDriver::addFile(StringRef path, bool withLOption, bool toBeginning) {
   using namespace sys::fs;
 
   std::optional<MemoryBufferRef> buffer = readFile(path);
@@ -235,7 +245,7 @@ void LinkerDriver::addFile(StringRef path, bool withLOption) {
   MemoryBufferRef mbref = *buffer;
 
   if (config->formatBinary) {
-    files.push_back(make<BinaryFile>(mbref));
+    addFileToVec(files, make<BinaryFile>(mbref), toBeginning);
     return;
   }
 
@@ -244,6 +254,10 @@ void LinkerDriver::addFile(StringRef path, bool withLOption) {
     readLinkerScript(mbref);
     return;
   case file_magic::archive: {
+    if (LLVM_UNLIKELY(toBeginning))
+      error(
+          "Adding archive files with STARTUP directive not supported, file: " +
+          path);
     auto members = getArchiveMembers(mbref);
     if (inWholeArchive) {
       for (const std::pair<MemoryBufferRef, uint64_t> &p : members) {
@@ -300,14 +314,14 @@ void LinkerDriver::addFile(StringRef path, bool withLOption) {
     auto *f =
         make<SharedFile>(mbref, withLOption ? path::filename(path) : path);
     f->init();
-    files.push_back(f);
+    addFileToVec(files, f, toBeginning);
     return;
   }
   case file_magic::bitcode:
-    files.push_back(make<BitcodeFile>(mbref, "", 0, inLib));
+    addFileToVec(files, make<BitcodeFile>(mbref, "", 0, inLib), toBeginning);
     break;
   case file_magic::elf_relocatable:
-    files.push_back(createObjFile(mbref, "", inLib));
+    addFileToVec(files, createObjFile(mbref, "", inLib), toBeginning);
     break;
   default:
     error(path + ": unknown file type");
@@ -317,7 +331,7 @@ void LinkerDriver::addFile(StringRef path, bool withLOption) {
 // Add a given library by searching it from input search paths.
 void LinkerDriver::addLibrary(StringRef name) {
   if (std::optional<std::string> path = searchLibrary(name))
-    addFile(saver().save(*path), /*withLOption=*/true);
+    addFile(saver().save(*path), /*withLOption=*/true, /*toBeginning=*/false);
   else
     error("unable to find library -l" + name, ErrorTag::LibNotFound, {name});
 }
@@ -1624,7 +1638,7 @@ void LinkerDriver::createFiles(opt::InputArgList &args) {
       hasInput = true;
       break;
     case OPT_INPUT:
-      addFile(arg->getValue(), /*withLOption=*/false);
+      addFile(arg->getValue(), /*withLOption=*/false, /*toBeginning=*/false);
       hasInput = true;
       break;
     case OPT_defsym: {

lld/ELF/InputFiles.cpp

@@ -358,11 +358,13 @@ static void addDependentLibrary(StringRef specifier, const InputFile *f) {
   if (!config->dependentLibraries)
     return;
   if (std::optional<std::string> s = searchLibraryBaseName(specifier))
-    ctx.driver.addFile(saver().save(*s), /*withLOption=*/true);
+    ctx.driver.addFile(saver().save(*s), /*withLOption=*/true,
+                       /*toBeginning=*/false);
   else if (std::optional<std::string> s = findFromSearchPaths(specifier))
-    ctx.driver.addFile(saver().save(*s), /*withLOption=*/true);
+    ctx.driver.addFile(saver().save(*s), /*withLOption=*/true,
+                       /*toBeginning=*/false);
   else if (fs::exists(specifier))
-    ctx.driver.addFile(specifier, /*withLOption=*/false);
+    ctx.driver.addFile(specifier, /*withLOption=*/false, /*toBeginning=*/false);
   else
     error(toString(f) +
           ": unable to find library from dependent library specifier: " +

lld/ELF/ScriptParser.cpp

@@ -66,14 +66,15 @@ class ScriptParser final : ScriptLexer {
   void readDefsym(StringRef name);
 
 private:
-  void addFile(StringRef path);
+  void addFile(StringRef path, bool toBeginning);
 
   void readAsNeeded();
   void readEntry();
   void readExtern();
   void readGroup();
   void readInclude();
   void readInput();
+  void readStartupInput();
   void readMemory();
   void readOutput();
   void readOutputArch();
@@ -141,6 +142,9 @@ class ScriptParser final : ScriptLexer {
 
   // A set to detect an INCLUDE() cycle.
   StringSet<> seen;
+
+  // Two startup directives don't make sense, error if this happens
+  bool seenStartup = false;
 };
 } // namespace
 
@@ -267,6 +271,8 @@ void ScriptParser::readLinkerScript() {
       readSearchDir();
     } else if (tok == "SECTIONS") {
       readSections();
+    } else if (tok == "STARTUP") {
+      readStartupInput();
     } else if (tok == "TARGET") {
       readTarget();
     } else if (tok == "VERSION") {
@@ -289,27 +295,28 @@ void ScriptParser::readDefsym(StringRef name) {
   script->sectionCommands.push_back(cmd);
 }
 
-void ScriptParser::addFile(StringRef s) {
+void ScriptParser::addFile(StringRef s, bool toBeginning) {
   if (isUnderSysroot && s.startswith("/")) {
     SmallString<128> pathData;
     StringRef path = (config->sysroot + s).toStringRef(pathData);
     if (sys::fs::exists(path))
-      ctx.driver.addFile(saver().save(path), /*withLOption=*/false);
+      ctx.driver.addFile(saver().save(path), /*withLOption=*/false,
+                         toBeginning);
     else
       setError("cannot find " + s + " inside " + config->sysroot);
     return;
   }
 
   if (s.startswith("/")) {
     // Case 1: s is an absolute path. Just open it.
-    ctx.driver.addFile(s, /*withLOption=*/false);
+    ctx.driver.addFile(s, /*withLOption=*/false, toBeginning);
   } else if (s.startswith("=")) {
     // Case 2: relative to the sysroot.
     if (config->sysroot.empty())
-      ctx.driver.addFile(s.substr(1), /*withLOption=*/false);
+      ctx.driver.addFile(s.substr(1), /*withLOption=*/false, toBeginning);
     else
       ctx.driver.addFile(saver().save(config->sysroot + "/" + s.substr(1)),
-                         /*withLOption=*/false);
+                         /*withLOption=*/false, toBeginning);
   } else if (s.startswith("-l")) {
     // Case 3: search in the list of library paths.
     ctx.driver.addLibrary(s.substr(2));
@@ -321,17 +328,18 @@ void ScriptParser::addFile(StringRef s) {
       SmallString<0> path(directory);
       sys::path::append(path, s);
       if (sys::fs::exists(path)) {
-        ctx.driver.addFile(path, /*withLOption=*/false);
+        ctx.driver.addFile(path, /*withLOption=*/false, toBeginning);
         return;
       }
     }
     // Then search in the current working directory.
     if (sys::fs::exists(s)) {
-      ctx.driver.addFile(s, /*withLOption=*/false);
+      ctx.driver.addFile(s, /*withLOption=*/false, toBeginning);
     } else {
       // Finally, search in the list of library paths.
       if (std::optional<std::string> path = findFromSearchPaths(s))
-        ctx.driver.addFile(saver().save(*path), /*withLOption=*/true);
+        ctx.driver.addFile(saver().save(*path), /*withLOption=*/true,
+                           toBeginning);
       else
         setError("unable to find " + s);
     }
@@ -343,7 +351,7 @@ void ScriptParser::readAsNeeded() {
   bool orig = config->asNeeded;
   config->asNeeded = true;
   while (!errorCount() && !consume(")"))
-    addFile(unquote(next()));
+    addFile(unquote(next()), /*toBeginning=*/false);
   config->asNeeded = orig;
 }
 
@@ -393,10 +401,22 @@ void ScriptParser::readInput() {
     if (consume("AS_NEEDED"))
       readAsNeeded();
     else
-      addFile(unquote(next()));
+      addFile(unquote(next()), /*toBeginning=*/false);
   }
 }
 
+// Works similar to the INPUT directive, but the file specified
+// in STARTUP is put as the first file to be parsed. Doesn't work
+// for archive files (error is emitted then).
+void ScriptParser::readStartupInput() {
+  if (this->seenStartup)
+    setError("multiple STARTUP directives seen");
+  this->seenStartup = true;
+  expect("(");
+  addFile(unquote(next()), /*toBeginning=*/true);
+  expect(")");
+}
+
 void ScriptParser::readOutput() {
   // -o <file> takes predecence over OUTPUT(<file>).
   expect("(");

lld/test/ELF/linkerscript/startup.s

@@ -0,0 +1,35 @@
+# REQUIRES: x86
+
+## New linker script directive, STARTUP, takes a file an adds it to the files
+## that should be parsed list (like INPUT). It also specifies that that file
+## (object or bitcode) is treated first, as if it were specified as the first
+## file on the command line. This doesn't work for archives.
+
+# RUN: llvm-mc -filetype=obj -triple=x86_64 %s -o %t1.o
+# RUN: echo '.globl b; b:' | llvm-mc -filetype=obj -triple=x86_64 - -o %t2.o
+
+# RUN: echo 'STARTUP(%t1.o)' > %t.script
+# RUN: ld.lld %t2.o -T %t.script -o %t
+# RUN: llvm-objdump -t %t | FileCheck %s
+
+# RUN: echo 'STARTUP(%t2.o)' >> %t.script
+
+# RUN: not ld.lld -T %t.script -o /dev/null 2>&1 | FileCheck %s \
+# RUN: --check-prefix=MULTIPLE-STARTUPS
+
+# RUN: llvm-ar rc %t2.a %t2.o
+# RUN: echo 'STARTUP(%t2.a)' > %t.script
+
+# RUN: not ld.lld -T %t.script -o /dev/null %t1.o 2>&1 | FileCheck %s \
+# RUN: --check-prefix=ARCHIVE-ERROR
+
+# CHECK: _start
+# CHECK: b
+
+# MULTIPLE-STARTUPS: multiple STARTUP directives seen
+
+# ARCHIVE-ERROR: Adding archive files with STARTUP directive not supported
+
+.globl _start
+_start:
+  call b