[ELF] - Linkerscript: implemented output section [address] attribute.

[grimar]

Summary

Output section description in SECTIONS looks like that:

section [address] [(type)] :
...
{
...
}

Patch implements support of address atribute.

[davide]

Do you have examples of linker scripts using this feature? I wasn't able to find any.

[emaste]

@davide we have them in FreeBSD's kernel linker script (for debug data) but they're all addr 0, e.g.:

/* DWARF debug sections.
   Symbols in the DWARF debugging sections are relative to the beginning
   of the section so we begin them at 0.  */
/* DWARF 1 */
.debug          0 : { *(.debug) }
.line           0 : { *(.line) }

[davide]

In D22689#493261, @emaste wrote:

@davide we have them in FreeBSD's kernel linker script (for debug data) but they're all addr 0, e.g.:

/* DWARF debug sections.
   Symbols in the DWARF debugging sections are relative to the beginning
   of the section so we begin them at 0.  */
/* DWARF 1 */
.debug          0 : { *(.debug) }
.line           0 : { *(.line) }

OK, fine.

[grimar]

In D22689#493163, @davide wrote:

Do you have examples of linker scripts using this feature? I wasn't able to find any.

.lrodata   ALIGN(CONSTANT (MAXPAGESIZE)) + (. & (CONSTANT (MAXPAGESIZE) - 1)) :
{
...
}

https://svnweb.freebsd.org/base/head/sys/conf/ldscript.amd64?revision=284870&view=markup#l176

[grimar]

I mean that "ALIGN(CONSTANT (MAXPAGESIZE)) + (. & (CONSTANT (MAXPAGESIZE) - 1))" is [address] here.

[ruiu]

LGTM

ELF/LinkerScript.cpp

830 ↗

(On Diff #65100)

Please add a comment.

// Read an address expression.
// https://sourceware.org/binutils/docs/ld/Output-Section-Address.html#Output-Section-Address

lld/trunk/ELF/LinkerScript.h

 // and ONLY_IF_RW respectively.
 enum ConstraintKind { NoConstraint, ReadOnly, ReadWrite };
 
 struct OutputSectionCommand : BaseCommand {
   OutputSectionCommand(StringRef Name)
       : BaseCommand(OutputSectionKind), Name(Name) {}
   static bool classof(const BaseCommand *C);
   StringRef Name;
+  Expr AddrExpr;
   std::vector<std::unique_ptr<BaseCommand>> Commands;
   std::vector<StringRef> Phdrs;
   std::vector<uint8_t> Filler;
   ConstraintKind Constraint = NoConstraint;
 };
 
 struct InputSectionDescription : BaseCommand {
   InputSectionDescription() : BaseCommand(InputSectionKind) {}

lld/trunk/ELF/LinkerScript.cpp

for (const std::unique_ptr<BaseCommand> &Base : Opt.Commands) {
     // Find all the sections with required name. There can be more than
     // one section with such name, if the alignment, flags or type
     // attribute differs.
     auto *Cmd = cast<OutputSectionCommand>(Base.get());
     for (OutputSectionBase<ELFT> *Sec : Sections) {
       if (Sec->getName() != Cmd->Name)
         continue;
 
+      if (Cmd->AddrExpr)
+        Dot = Cmd->AddrExpr(Dot);
+
       if ((Sec->getFlags() & SHF_TLS) && Sec->getType() == SHT_NOBITS) {
         uintX_t TVA = Dot + ThreadBssOffset;
         TVA = alignTo(TVA, Sec->getAlignment());
         Sec->setVA(TVA);
         ThreadBssOffset = TVA - Dot + Sec->getSize();
         continue;
       }
 
return StringSwitch<int>(Op)
       .Case("!=", 2)
       .Case("&", 1)
       .Default(-1);
 }
 
 void ScriptParser::readOutputSectionDescription(StringRef OutSec) {
   OutputSectionCommand *Cmd = new OutputSectionCommand(OutSec);
   Opt.Commands.emplace_back(Cmd);
+
+  // Read an address expression.
+  // https://sourceware.org/binutils/docs/ld/Output-Section-Address.html#Output-Section-Address
+  if (peek() != ":")
+    Cmd->AddrExpr = readExpr();
+
   expect(":");
 
   // Parse constraints.
   if (skip("ONLY_IF_RO"))
     Cmd->Constraint = ReadOnly;
   if (skip("ONLY_IF_RW"))
     Cmd->Constraint = ReadWrite;
   expect("{");

lld/trunk/test/ELF/linkerscript-outsections-addr.s

+# REQUIRES: x86
+# RUN: llvm-mc -filetype=obj -triple=x86_64-unknown-linux %s -o %t
+# RUN: echo "SECTIONS {                 \
+# RUN:   .aaa 0x2000 :                  \
+# RUN:   {                              \
+# RUN:    *(.aaa)                       \
+# RUN:   }                              \
+# RUN:   .bbb 0x1 ? 0x3000 : 0x4000 :   \
+# RUN:   {                              \
+# RUN:    *(.bbb)                       \
+# RUN:   }                              \
+# RUN:   .ccc ALIGN(CONSTANT (MAXPAGESIZE)) + (. & (CONSTANT (MAXPAGESIZE) - 1)) : \
+# RUN:   {                              \
+# RUN:    *(.ccc)                       \
+# RUN:   }                              \
+# RUN:   .ddd 0x5001 :                  \
+# RUN:   {                              \
+# RUN:    *(.ddd)                       \
+# RUN:   }                              \
+# RUN: }" > %t.script
+# RUN: ld.lld %t --script %t.script -o %tout
+# RUN: llvm-readobj -s %tout | FileCheck %s
+
+## Check:
+## 1) Simple constant as address.
+## 2) That something that contains ":" character, like ternary
+##    operator works as expression.
+## 3) That complex expressions work.
+## 4) That section alignment still applied to explicitly specified address.
+
+#CHECK:Sections [
+#CHECK:  Section {
+#CHECK:    Index: 0
+#CHECK:    Name:
+#CHECK:    Type: SHT_NULL
+#CHECK:    Flags [
+#CHECK:    ]
+#CHECK:    Address: 0x0
+#CHECK:    Offset: 0x0
+#CHECK:    Size: 0
+#CHECK:    Link: 0
+#CHECK:    Info: 0
+#CHECK:    AddressAlignment: 0
+#CHECK:    EntrySize: 0
+#CHECK:  }
+#CHECK:  Section {
+#CHECK:    Index: 1
+#CHECK:    Name: .aaa
+#CHECK:    Type: SHT_PROGBITS
+#CHECK:    Flags [
+#CHECK:      SHF_ALLOC
+#CHECK:    ]
+#CHECK:    Address: 0x2000
+#CHECK:    Offset: 0x1000
+#CHECK:    Size: 8
+#CHECK:    Link: 0
+#CHECK:    Info: 0
+#CHECK:    AddressAlignment: 1
+#CHECK:    EntrySize: 0
+#CHECK:  }
+#CHECK:  Section {
+#CHECK:    Index: 2
+#CHECK:    Name: .bbb
+#CHECK:    Type: SHT_PROGBITS
+#CHECK:    Flags [
+#CHECK:      SHF_ALLOC
+#CHECK:    ]
+#CHECK:    Address: 0x3000
+#CHECK:    Offset: 0x2000
+#CHECK:    Size: 8
+#CHECK:    Link: 0
+#CHECK:    Info: 0
+#CHECK:    AddressAlignment: 1
+#CHECK:    EntrySize: 0
+#CHECK:  }
+#CHECK:  Section {
+#CHECK:    Index: 3
+#CHECK:    Name: .ccc
+#CHECK:    Type: SHT_PROGBITS
+#CHECK:    Flags [
+#CHECK:      SHF_ALLOC
+#CHECK:    ]
+#CHECK:    Address: 0x4008
+#CHECK:    Offset: 0x2008
+#CHECK:    Size: 8
+#CHECK:    Link: 0
+#CHECK:    Info: 0
+#CHECK:    AddressAlignment: 1
+#CHECK:    EntrySize: 0
+#CHECK:  }
+#CHECK:  Section {
+#CHECK:    Index: 4
+#CHECK:    Name: .ddd
+#CHECK:    Type: SHT_PROGBITS
+#CHECK:    Flags [
+#CHECK:      SHF_ALLOC
+#CHECK:    ]
+#CHECK:    Address: 0x5010
+#CHECK:    Offset: 0x2010
+#CHECK:    Size: 8
+#CHECK:    Link: 0
+#CHECK:    Info: 0
+#CHECK:    AddressAlignment: 16
+#CHECK:    EntrySize: 0
+#CHECK:  }
+  
+.globl _start
+_start:
+nop
+
+.section .aaa, "a"
+.quad 0
+
+.section .bbb, "a"
+.quad 0
+
+.section .ccc, "a"
+.quad 0
+
+.section .ddd, "a"
+.align 16
+.quad 0

lld/trunk/test/ELF/linkerscript/linkerscript-diagnostic.s

 
 ## One more check that text of lines and pointer to 'bad' token are working ok.
 # RUN: echo "SECTIONS {" > %t.script
 # RUN: echo ".text : { *(.text) }" >> %t.script
 # RUN: echo ".keep : { *(.keep) }" >> %t.script
 # RUN: echo "boom .temp : { *(.temp) } }" >> %t.script
 # RUN: not ld.lld -shared %t -o %t1 --script %t.script > %t.log 2>&1
 # RUN: FileCheck -check-prefix=ERR7 %s < %t.log
-# ERR7:      line 4: : expected, but got .temp
+# ERR7:      line 4: malformed number: .temp
 # ERR7-NEXT: boom .temp : { *(.temp) } }
 # RUN: grep '^     ^' %t.log