COFF: Add a flag for disabling string tail merging.

[pcc]

Summary

We discovered (crbug.com/838449#c24) that string tail merging can
negatively affect compressed binary size, so provide a flag to turn
it off for users who care more about compressed size than uncompressed
size.

[ruiu]

LGTM

lld/trunk/COFF/Config.h

struct Configuration {
   bool Verbose = false;
   WindowsSubsystem Subsystem = llvm::COFF::IMAGE_SUBSYSTEM_UNKNOWN;
   Symbol *Entry = nullptr;
   bool NoEntry = false;
   std::string OutputFile;
   std::string ImportName;
   bool DoGC = true;
   bool DoICF = true;
+  bool TailMerge;
   bool Relocatable = true;
   bool Force = false;
   bool Debug = false;
   bool DebugDwarf = false;
   bool DebugGHashes = false;
   bool ShowTiming = false;
   unsigned DebugTypes = static_cast<unsigned>(DebugType::None);
   std::vector<std::string> NatvisFiles;

lld/trunk/COFF/Driver.cpp

void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
   // Handle /implib
   if (auto *Arg = Args.getLastArg(OPT_implib))
     Config->Implib = Arg->getValue();
 
   // Handle /opt.
   bool DoGC = !Args.hasArg(OPT_debug) || Args.hasArg(OPT_profile);
   unsigned ICFLevel =
       Args.hasArg(OPT_profile) ? 0 : 1; // 0: off, 1: limited, 2: on
+  unsigned TailMerge = 1;
   for (auto *Arg : Args.filtered(OPT_opt)) {
     std::string Str = StringRef(Arg->getValue()).lower();
     SmallVector<StringRef, 1> Vec;
     StringRef(Str).split(Vec, ',');
     for (StringRef S : Vec) {
       if (S == "ref") {
         DoGC = true;
       } else if (S == "noref") {
         DoGC = false;
       } else if (S == "icf" || S.startswith("icf=")) {
         ICFLevel = 2;
       } else if (S == "noicf") {
         ICFLevel = 0;
+      } else if (S == "lldtailmerge") {
+        TailMerge = 2;
+      } else if (S == "nolldtailmerge") {
+        TailMerge = 0;
       } else if (S.startswith("lldlto=")) {
         StringRef OptLevel = S.substr(7);
         if (OptLevel.getAsInteger(10, Config->LTOOptLevel) ||
             Config->LTOOptLevel > 3)
           error("/opt:lldlto: invalid optimization level: " + OptLevel);
       } else if (S.startswith("lldltojobs=")) {
         StringRef Jobs = S.substr(11);
         if (Jobs.getAsInteger(10, Config->LTOJobs) || Config->LTOJobs == 0)
void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
   // explicitly.
   // FIXME: LLD only implements "limited" ICF, i.e. it only merges identical
   // code. If the user passes /OPT:ICF explicitly, LLD should merge identical
   // comdat readonly data.
   if (ICFLevel == 1 && !DoGC)
     ICFLevel = 0;
   Config->DoGC = DoGC;
   Config->DoICF = ICFLevel > 0;
+  Config->TailMerge = (TailMerge == 1 && Config->DoICF) || TailMerge == 2;
 
   // Handle /lldsavetemps
   if (Args.hasArg(OPT_lldsavetemps))
     Config->SaveTemps = true;
 
   // Handle /kill-at
   if (Args.hasArg(OPT_kill_at))
     Config->KillAt = true;

lld/trunk/COFF/InputFiles.cpp

SectionChunk *ObjFile::readSection(uint32_t SectionNumber,
   if (C->isCodeView())
     DebugChunks.push_back(C);
   else if (Config->GuardCF != GuardCFLevel::Off && Name == ".gfids$y")
     GuardFidChunks.push_back(C);
   else if (Config->GuardCF != GuardCFLevel::Off && Name == ".gljmp$y")
     GuardLJmpChunks.push_back(C);
   else if (Name == ".sxdata")
     SXDataChunks.push_back(C);
-  else if (Config->DoICF && Sec->NumberOfRelocations == 0 && Name == ".rdata" &&
-           LeaderName.startswith("??_C@"))
+  else if (Config->TailMerge && Sec->NumberOfRelocations == 0 &&
+           Name == ".rdata" && LeaderName.startswith("??_C@"))
     // COFF sections that look like string literal sections (i.e. no
     // relocations, in .rdata, leader symbol name matches the MSVC name mangling
     // for string literals) are subject to string tail merging.
     MergeChunk::addSection(C);
   else
     Chunks.push_back(C);
 
   return C;

lld/trunk/test/COFF/string-tail-merge.s

 # REQUIRES: x86
 # RUN: llvm-mc -triple=x86_64-windows-msvc -filetype=obj -o %t.obj %s
 # RUN: lld-link %t.obj /out:%t.exe /entry:main /subsystem:console
 # RUN: llvm-objdump -s %t.exe | FileCheck %s
+# RUN: lld-link %t.obj /out:%t.exe /entry:main /subsystem:console /opt:noicf /opt:lldtailmerge
+# RUN: llvm-objdump -s %t.exe | FileCheck %s
+# RUN: lld-link %t.obj /out:%t.exe /entry:main /subsystem:console /opt:noicf
+# RUN: llvm-objdump -s %t.exe | FileCheck --check-prefix=NOSTM %s
+# RUN: lld-link %t.obj /out:%t.exe /entry:main /subsystem:console /opt:nolldtailmerge
+# RUN: llvm-objdump -s %t.exe | FileCheck --check-prefix=NOSTM %s
 
 # CHECK: Contents of section .text:
+# NOSTM: Contents of section .text:
 .globl main
 main:
 # CHECK-NEXT: 140001000 11200040 01000000 17200040 01000000
+# NOSTM-NEXT: 140001000 00200040 01000000 0c200040 01000000
 .8byte "??_C@_0M@LACCCNMM@hello?5world?$AA@"
 .8byte "??_C@_05MCBCHHEJ@world?$AA@"
 # CHECK-NEXT: 140001010 2a200040 01000000 36200040 01000000
+# NOSTM-NEXT: 140001010 12200040 01000000 2a200040 01000000
 .8byte "??_C@_1BI@HHJHKLLN@?$AAh?$AAe?$AAl?$AAl?$AAo?$AA?5?$AAw?$AAo?$AAr?$AAl?$AAd?$AA?$AA@"
 .8byte "??_C@_1M@NBBDDHIO@?$AAw?$AAo?$AAr?$AAl?$AAd?$AA?$AA@"
 # CHECK-NEXT: 140001020 00200040 01000000 0c200040 01000000
+# NOSTM-NEXT: 140001020 36200040 01000000 42200040 01000000
 .8byte "??_D@not_a_string_literal"
 .8byte "??_C@string_literal_with_relocs"
 # CHECK-NEXT: 140001030 00300040 01000000 1e200040 01000000
+# NOSTM-NEXT: 140001030 00300040 01000000 48200040 01000000
 .8byte "??_C@string_literal_in_wrong_section"
 .8byte "??_C@overaligned_string_literal"
 
 # CHECK: Contents of section .rdata:
 # CHECK-NEXT:  140002000 68656c6c 6f20776f 726c6400 6f826ca4  hello world.o.l.
 # CHECK-NEXT:  140002010 0068656c 6c6f2077 6f726c64 00006865  .hello world..he
 # CHECK-NEXT:  140002020 6c6c6f20 776f726c 64006800 65006c00  llo world.h.e.l.
 # CHECK-NEXT:  140002030 6c006f00 20007700 6f007200 6c006400  l.o. .w.o.r.l.d.
 # CHECK-NEXT:  140002040 0000                                 ..
 
+# NOSTM: Contents of section .rdata:
+# NOSTM-NEXT:  140002000 68656c6c 6f20776f 726c6400 776f726c  hello world.worl
+# NOSTM-NEXT:  140002010 64006800 65006c00 6c006f00 20007700  d.h.e.l.l.o. .w.
+# NOSTM-NEXT:  140002020 6f007200 6c006400 00007700 6f007200  o.r.l.d...w.o.r.
+# NOSTM-NEXT:  140002030 6c006400 00006865 6c6c6f20 776f726c  l.d...hello worl
+# NOSTM-NEXT:  140002040 64006f82 6ca40000 68656c6c 6f20776f  d.o.l...hello wo
+# NOSTM-NEXT:  140002050 726c6400                             rld.
 
 .section .rdata,"dr",discard,"??_C@_0M@LACCCNMM@hello?5world?$AA@"
 .globl "??_C@_0M@LACCCNMM@hello?5world?$AA@"
 "??_C@_0M@LACCCNMM@hello?5world?$AA@":
 .asciz "hello world"
 
 .section .rdata,"dr",discard,"??_C@_05MCBCHHEJ@world?$AA@"
 .globl "??_C@_05MCBCHHEJ@world?$AA@"