Description of symbols is avalable here:

https://docs.oracle.com/cd/E53394_01/html/E54766/u-etext-3c.html

It is said that:
_etext - The address of _etext is the first
location after the last read-only loadable segment.

_edata - The address of _edata is the first
location after the last read-write loadable segment.

_end - If the address of _edata is greater than the address
of _etext, the address of _end is same as the address of _edata.

In real life _end and _edata has different values for that case.
Both gold/bfd set _edata to the end of the last non SHT_NOBITS section.
This patch do the same for consistency.

It should fix the https://llvm.org/bugs/show_bug.cgi?id=26729.

Differential revision: http://reviews.llvm.org/D17601

llvm-svn: 262019

diff --git a/lld/ELF/Symbols.cpp b/lld/ELF/Symbols.cpp
index a974778..bd6f6b4 100644 (file)
--- a/lld/ELF/Symbols.cpp
+++ b/lld/ELF/Symbols.cpp
@@ -214,6 +214,8 @@ std::unique_ptr<InputFile> Lazy::getMember() {
 }
 
 template <class ELFT> static void doInitSymbols() {
+  ElfSym<ELFT>::Etext.setBinding(STB_GLOBAL);
+  ElfSym<ELFT>::Edata.setBinding(STB_GLOBAL);
   ElfSym<ELFT>::End.setBinding(STB_GLOBAL);
   ElfSym<ELFT>::Ignored.setBinding(STB_WEAK);
   ElfSym<ELFT>::Ignored.setVisibility(STV_HIDDEN);

diff --git a/lld/ELF/Symbols.h b/lld/ELF/Symbols.h
index 073dfeb..0c95475 100644 (file)
--- a/lld/ELF/Symbols.h
+++ b/lld/ELF/Symbols.h
@@ -346,6 +346,12 @@ template <class ELFT> struct ElfSym {
   // output file's symbol table. It has weak binding and can be substituted.
   static Elf_Sym Ignored;
 
+  // The content for _etext and etext symbols.
+  static Elf_Sym Etext;
+
+  // The content for _edata and edata symbols.
+  static Elf_Sym Edata;
+
   // The content for _end and end symbols.
   static Elf_Sym End;
 
@@ -359,6 +365,8 @@ template <class ELFT> struct ElfSym {
 };
 
 template <class ELFT> typename ElfSym<ELFT>::Elf_Sym ElfSym<ELFT>::Ignored;
+template <class ELFT> typename ElfSym<ELFT>::Elf_Sym ElfSym<ELFT>::Etext;
+template <class ELFT> typename ElfSym<ELFT>::Elf_Sym ElfSym<ELFT>::Edata;
 template <class ELFT> typename ElfSym<ELFT>::Elf_Sym ElfSym<ELFT>::End;
 template <class ELFT> typename ElfSym<ELFT>::Elf_Sym ElfSym<ELFT>::MipsGp;
 template <class ELFT>

diff --git a/lld/ELF/Writer.cpp b/lld/ELF/Writer.cpp
index b87291f..cbd7853 100644 (file)
--- a/lld/ELF/Writer.cpp
+++ b/lld/ELF/Writer.cpp
@@ -914,20 +914,30 @@ template <class ELFT> void Writer<ELFT>::addReservedSymbols() {
   if (!isOutputDynamic())
     Symtab.addIgnored("__tls_get_addr");
 
+  auto Define = [this](StringRef Name, StringRef Alias, Elf_Sym &Sym) {
+    if (Symtab.find(Name))
+      Symtab.addAbsolute(Name, Sym);
+    if (SymbolBody *B = Symtab.find(Alias))
+      if (B->isUndefined())
+        Symtab.addAbsolute(Alias, Sym);
+  };
+
   // If the "_end" symbol is referenced, it is expected to point to the address
   // right after the data segment. Usually, this symbol points to the end
   // of .bss section or to the end of .data section if .bss section is absent.
   // We don't know the final address of _end yet, so just add a symbol here,
   // and fix ElfSym<ELFT>::End.st_value later.
-  if (Symtab.find("_end"))
-    Symtab.addAbsolute("_end", ElfSym<ELFT>::End);
-
   // Define "end" as an alias to "_end" if it is used but not defined.
   // We don't want to define that unconditionally because we don't want to
   // break programs that uses "end" as a regular symbol.
-  if (SymbolBody *B = Symtab.find("end"))
-    if (B->isUndefined())
-      Symtab.addAbsolute("end", ElfSym<ELFT>::End);
+  // The similar history with _etext/etext and _edata/edata:
+  // Address of _etext is the first location after the last read-only loadable
+  // segment. Address of _edata points to the end of the last non SHT_NOBITS
+  // section. That is how gold/bfd do. We update the values for these symbols
+  // later, after assigning sections to segments.
+  Define("_end", "end", ElfSym<ELFT>::End);
+  Define("_etext", "etext", ElfSym<ELFT>::Etext);
+  Define("_edata", "edata", ElfSym<ELFT>::Edata);
 }
 
 // Sort input sections by section name suffixes for
@@ -1454,6 +1464,17 @@ template <class ELFT> void Writer<ELFT>::fixAbsoluteSymbols() {
   // Update MIPS _gp absolute symbol so that it points to the static data.
   if (Config->EMachine == EM_MIPS)
     ElfSym<ELFT>::MipsGp.st_value = getMipsGpAddr<ELFT>();
+
+  // _etext points to location after the last read-only loadable segment.
+  // _edata points to the end of the last non SHT_NOBITS section.
+  for (OutputSectionBase<ELFT> *Sec : OutputSections) {
+    if (!(Sec->getFlags() & SHF_ALLOC))
+      continue;
+    if (!(Sec->getFlags() & SHF_WRITE))
+      ElfSym<ELFT>::Etext.st_value = Sec->getVA() + Sec->getSize();
+    if (Sec->getType() != SHT_NOBITS)
+      ElfSym<ELFT>::Edata.st_value = Sec->getVA() + Sec->getSize();
+  }
 }
 
 template <class ELFT> void Writer<ELFT>::writeHeader() {

diff --git a/lld/test/ELF/edata-etext.s b/lld/test/ELF/edata-etext.s
new file mode 100644 (file)
index 0000000..e0538d6
--- /dev/null
+++ b/lld/test/ELF/edata-etext.s
@@ -0,0 +1,117 @@
+# REQUIRES: x86
+# RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux %s -o %t.o
+# RUN: ld.lld %t.o -o %t
+# RUN: llvm-readobj -sections -symbols %t | FileCheck %s
+
+## This checks that:
+## 1) Address of _etext is the first location after the last read-only loadable segment.
+## 2) Address of _edata points to the end of the last non SHT_NOBITS section.
+##    That is how gold/bfd do. At the same time specs says: "If the address of _edata is
+##    greater than the address of _etext, the address of _end is same as the address
+##    of _edata." (https://docs.oracle.com/cd/E53394_01/html/E54766/u-etext-3c.html).
+## 3) Address of _end is different from _edata because of 2.
+# CHECK:      Section {
+# CHECK:         Index: 1
+# CHECK:         Name: .text
+# CHECK-NEXT:    Type: SHT_PROGBITS
+# CHECK-NEXT:    Flags [
+# CHECK-NEXT:      SHF_ALLOC
+# CHECK-NEXT:      SHF_EXECINSTR
+# CHECK-NEXT:    ]
+# CHECK-NEXT:    Address: 0x11000
+# CHECK-NEXT:    Offset: 0x1000
+# CHECK-NEXT:    Size: 1
+# CHECK-NEXT:    Link:
+# CHECK-NEXT:    Info:
+# CHECK-NEXT:    AddressAlignment:
+# CHECK-NEXT:    EntrySize: 0
+# CHECK-NEXT:  }
+# CHECK-NEXT:  Section {
+# CHECK-NEXT:    Index: 2
+# CHECK-NEXT:    Name: .data
+# CHECK-NEXT:    Type: SHT_PROGBITS
+# CHECK-NEXT:    Flags [
+# CHECK-NEXT:      SHF_ALLOC
+# CHECK-NEXT:      SHF_WRITE
+# CHECK-NEXT:    ]
+# CHECK-NEXT:    Address: 0x12000
+# CHECK-NEXT:    Offset: 0x2000
+# CHECK-NEXT:    Size: 2
+# CHECK-NEXT:    Link:
+# CHECK-NEXT:    Info:
+# CHECK-NEXT:    AddressAlignment:
+# CHECK-NEXT:    EntrySize:
+# CHECK-NEXT:  }
+# CHECK-NEXT:  Section {
+# CHECK-NEXT:    Index: 3
+# CHECK-NEXT:    Name: .bss
+# CHECK-NEXT:    Type: SHT_NOBITS
+# CHECK-NEXT:    Flags [
+# CHECK-NEXT:      SHF_ALLOC
+# CHECK-NEXT:      SHF_WRITE
+# CHECK-NEXT:    ]
+# CHECK-NEXT:    Address: 0x12004
+# CHECK-NEXT:    Offset: 0x2002
+# CHECK-NEXT:    Size: 6
+# CHECK-NEXT:    Link:
+# CHECK-NEXT:    Info:
+# CHECK-NEXT:    AddressAlignment:
+# CHECK-NEXT:    EntrySize:
+# CHECK-NEXT:  }
+# CHECK:      Symbols [
+# CHECK-NEXT:  Symbol {
+# CHECK-NEXT:    Name:
+# CHECK-NEXT:    Value: 0x0
+# CHECK-NEXT:    Size: 0
+# CHECK-NEXT:    Binding: Local
+# CHECK-NEXT:    Type: None
+# CHECK-NEXT:    Other: 0
+# CHECK-NEXT:    Section: Undefined
+# CHECK-NEXT:  }
+# CHECK-NEXT:  Symbol {
+# CHECK-NEXT:    Name: _start
+# CHECK-NEXT:    Value: 0x11000
+# CHECK-NEXT:    Size: 0
+# CHECK-NEXT:    Binding: Global
+# CHECK-NEXT:    Type: None
+# CHECK-NEXT:    Other: 0
+# CHECK-NEXT:    Section: .text
+# CHECK-NEXT:  }
+# CHECK-NEXT:  Symbol {
+# CHECK-NEXT:    Name: _edata
+# CHECK-NEXT:    Value: 0x12002
+# CHECK-NEXT:    Size: 0
+# CHECK-NEXT:    Binding: Global
+# CHECK-NEXT:    Type: None
+# CHECK-NEXT:    Other: 0
+# CHECK-NEXT:    Section: Absolute
+# CHECK-NEXT:  }
+# CHECK-NEXT:  Symbol {
+# CHECK-NEXT:    Name: _end
+# CHECK-NEXT:    Value: 0x1200A
+# CHECK-NEXT:    Size: 0
+# CHECK-NEXT:    Binding: Global
+# CHECK-NEXT:    Type: None
+# CHECK-NEXT:    Other: 0
+# CHECK-NEXT:    Section: Absolute
+# CHECK-NEXT:  }
+# CHECK-NEXT:  Symbol {
+# CHECK-NEXT:    Name: _etext
+# CHECK-NEXT:    Value: 0x11001
+# CHECK-NEXT:    Size: 0
+# CHECK-NEXT:    Binding: Global
+# CHECK-NEXT:    Type: None
+# CHECK-NEXT:    Other: 0
+# CHECK-NEXT:    Section: Absolute
+# CHECK-NEXT:  }
+# CHECK-NEXT: ]
+
+.global _start,_end,_etext,_edata
+.text
+_start:
+  nop
+.data
+  .word 1
+.bss
+  .align 4
+  .space 6