[lld] Implement --dynamic-list-data

[zatrazz]

Summary

This patch implements the --dynamic-list-data option, which adds in
dynamic symbol table all the global data symbols.

This patch depends on http://reviews.llvm.org/D18771 and
http://reviews.llvm.org/D18772

[ruiu]

ELF/Writer.cpp

848

You are not using ELFT in this function.

[zatrazz]

ELF/Writer.cpp

848

Indeed (this is an artifact from a previous version), I will remove it.

[• rafael]

What is using this feature?

test/ELF/dynamic-list-data.s

3

not sure I can parse "is build in liked"

6

Why do you need the soname?

[zatrazz]

test/ELF/dynamic-list-data.s

3

Yeah the wording is fuzzy, I will change that.

6

Just to force test to be dynamically linked without the need to pass either the libc or another system library.

[zatrazz]

In D18773#391822, @rafael wrote:

What is using this feature?

As for --dynamic-list-cpp-{new,typeinfo} it is an option that might be used with -Bsymbolic to speed up internal shared library all (to bind internally), however I could not find any project that actually uses it (it though libstdc++.so could be a candidate, but it is not). As for D18774 I can drop the patch is the usercase is not strong enough for inclusion.

[• zzzoom]

UCX requires --dynamic-list-data:
https://github.com/openucx/ucx/issues/4549
https://github.com/openucx/ucx/issues/5545

[MaskRay]

In D18773#3017438, @zzzoom wrote:

UCX requires --dynamic-list-data:
https://github.com/openucx/ucx/issues/4549
https://github.com/openucx/ucx/issues/5545

This is not sufficient justification. See https://github.com/openucx/ucx/issues/4549#issuecomment-926167627
I don't think --dynamic-list-data is useful.

ELF/Config.h

struct Configuration {
   bool Trace;
   bool Verbose;
   bool WarnCommon;
   bool ZExecStack;
   bool ZNodelete;
   bool ZNow;
   bool ZOrigin;
   bool ZRelro;
+  bool DynamicListData;
   ELFKind EKind = ELFNoneKind;
   uint16_t EMachine = llvm::ELF::EM_NONE;
   uint64_t EntryAddr = -1;
   unsigned LtoO;
   unsigned Optimize;
 };
 
 // The only instance of Configuration struct.
 extern Configuration *Config;
 
 } // namespace elf
 } // namespace lld
 
 #endif

ELF/Driver.cpp

void LinkerDriver::readConfigs(opt::InputArgList &Args) {
   Config->Fini = getString(Args, OPT_fini, "_fini");
   Config->Init = getString(Args, OPT_init, "_init");
   Config->OutputFile = getString(Args, OPT_o);
   Config->SoName = getString(Args, OPT_soname);
   Config->Sysroot = getString(Args, OPT_sysroot);
   StringRef DynamicListPath = getString(Args, OPT_dynamic_list);
   if (!DynamicListPath.empty())
     readDynamicList(DynamicListPath);
+  Config->DynamicListData = Args.hasArg(OPT_dynamic_list_data);
 
   Config->Optimize = getInteger(Args, OPT_O, 0);
   Config->LtoO = getInteger(Args, OPT_lto_O, 2);
   if (Config->LtoO > 3)
     error("invalid optimization level for LTO: " + getString(Args, OPT_lto_O));
 
   Config->ZExecStack = hasZOption(Args, "execstack");
   Config->ZNodelete = hasZOption(Args, "nodelete");

ELF/Options.td

def wrap : Separate<["--", "-"], "wrap">, MetaVarName<"<symbol>">,
   HelpText<"Use wrapper functions for symbol">;
 
 def z : JoinedOrSeparate<["-"], "z">, MetaVarName<"<option>">,
   HelpText<"Linker option extensions">;
 
 def dynamic_list : Separate<["--", "-"], "dynamic-list">,
   HelpText<"Dynamic list to use">;
 
+def dynamic_list_data : Flag<["--", "-"], "dynamic-list-data">,
+  HelpText<"Add all global data to the dynamic symbol list">;
+
 // Aliases
 def alias_Bdynamic_call_shared: Flag<["-"], "call_shared">, Alias<Bdynamic>;
 def alias_Bdynamic_dy: Flag<["-"], "dy">, Alias<Bdynamic>;
 def alias_Bstatic_dn: Flag<["-"], "dn">, Alias<Bstatic>;
 def alias_Bstatic_non_shared: Flag<["-"], "non_shared">, Alias<Bstatic>;
 def alias_Bstatic_static: Flag<["-"], "static">, Alias<Bstatic>;
 def alias_L__library_path : Joined<["--"], "library-path=">, Alias<L>;
 def alias_discard_all_x: Flag<["-"], "x">, Alias<discard_all>;

ELF/Writer.cpp

template <class ELFT> static bool includeInSymtab(const SymbolBody &B) {
   if (auto *D = dyn_cast<DefinedRegular<ELFT>>(&B)) {
     // Exclude symbols pointing to garbage-collected sections.
     if (D->Section && !D->Section->Live)
       return false;
   }
   return true;
 }
 
+template <class ELFT>
 static bool includeInDynsym(const SymbolBody &B) {
   uint8_t V = B.getVisibility();
   if (V != STV_DEFAULT && V != STV_PROTECTED)
     return false;
   if (Config->ExportDynamic || Config->Shared)
     return true;
   if (DynList->isInDynamicList(B.getName()))
     return true;
+  // Handle --dynamic-list-data.
+  if (Config->DynamicListData)
+    if (B.Type == STT_OBJECT)
+      return true;
   return B.MustBeInDynSym;
 }
 
 // This class knows how to create an output section for a given
 // input section. Output section type is determined by various
 // factors, including input section's sh_flags, sh_type and
 // linker scripts.
 namespace {
if (auto *SC = dyn_cast<SharedSymbol<ELFT>>(Body))
       if (SC->needsCopy())
         CopyRelSymbols.push_back(SC);
 
     if (!includeInSymtab<ELFT>(*Body))
       continue;
     if (Out<ELFT>::SymTab)
       Out<ELFT>::SymTab->addSymbol(Body);
 
-    if (isOutputDynamic() && includeInDynsym(*Body))
+    if (isOutputDynamic() && includeInDynsym<ELFT>(*Body))
       Out<ELFT>::DynSymTab->addSymbol(Body);
   }
 
   // Do not proceed if there was an undefined symbol.
   if (HasError)
     return;
 
   addCommonSymbols(CommonSymbols);

test/ELF/dynamic-list-data.s

+# REQUIRES: x86
+
+## This object is build in linked against to force dynamic loading without
+## the use of full paths in linking command.
+# RUN: llvm-mc -filetype=obj -triple=x86_64-unknown-linux %p/Inputs/shared.s -o %t2.o
+# RUN: ld.lld -shared %t2.o -soname shared -o %t2.so
+
+# RUN: llvm-mc -filetype=obj -triple=x86_64-unknown-linux %s -o %t
+# RUN: ld.lld --dynamic-list-data %t %t2.so -o %t.exe
+# RUN: llvm-readobj -dyn-symbols %t.exe | FileCheck %s
+
+# CHECK:      DynamicSymbols [
+# CHECK-NEXT:   Symbol {
+# CHECK-NEXT:     Name: @ (0)
+# CHECK-NEXT:     Value: 0x0
+# CHECK-NEXT:     Size: 0
+# CHECK-NEXT:     Binding: Local (0x0)
+# CHECK-NEXT:     Type: None (0x0)
+# CHECK-NEXT:     Other: 0
+# CHECK-NEXT:     Section: Undefined (0x0)
+# CHECK-NEXT:   }
+# CHECK-NEXT:   Symbol {
+# CHECK-NEXT:     Name: data1@ (1)
+# CHECK-NEXT:     Value: 0x13000
+# CHECK-NEXT:     Size: 4
+# CHECK-NEXT:     Binding: Global (0x1)
+# CHECK-NEXT:     Type: Object (0x1)
+# CHECK-NEXT:     Other: 0
+# CHECK-NEXT:     Section: .bss (0x6)
+# CHECK-NEXT:   }
+# CHECK-NEXT:   Symbol {
+# CHECK-NEXT:     Name: data2@ (7)
+# CHECK-NEXT:     Value: 0x11000
+# CHECK-NEXT:     Size: 4
+# CHECK-NEXT:     Binding: Global (0x1)
+# CHECK-NEXT:     Type: Object (0x1)
+# CHECK-NEXT:     Other: 0
+# CHECK-NEXT:     Section: .text (0x4)
+# CHECK-NEXT:   }
+# CHECK-NEXT: ] 
+
+## int data1;
+.comm data1,4,4
+
+## int data2 = 5;
+.globl  data2
+.type data2, @object
+.size data2, 4
+data2:
+.long 5
+
+.globl _start
+_start:
+  retq