[ELF] Fix incorrect inclusion of program headers to PT_LOAD segment

[evgeny777]

Summary

Linker shouldn't include program headers in PT_LOAD in the following case:

1. PHDRS directive exists
2. PT_LOAD segment doesn't contain PHDRS and/or FILEHDR attribute

Below is an example of such script:

PHDRS { 
  hdr PT_PHDR FILEHDR PHDRS;
  all PT_LOAD; 
}
SECTIONS { 
   . = 0x10000;
   .text : { *(.text) } : all
  /* other sections go here */
}

Current revision will detect that there is a place to put ELF and program headers and will do this even though 'all' segment doesn't
have PHDRS and FILEHDR attributes.

To make this patch as simple as possible, both FILEHDR and PHDRS attributes are required to put ELF and program headers to PT_LOAD.
GNU gold has the same requirement.

[evgeny777]

Addressed review comments

[• rafael]

LGTM

[evgeny777]

Wrong diff for this revision. This was actually for
https://reviews.llvm.org/D24795

[evgeny777]

New revision created for the issue:
https://reviews.llvm.org/D25774

ELF/LinkerScript.cpp

 
 // Creates program headers as instructed by PHDRS linker script command.
 template <class ELFT>
 std::vector<PhdrEntry<ELFT>> LinkerScript<ELFT>::createPhdrs() {
   std::vector<PhdrEntry<ELFT>> Ret;
 
   // Process PHDRS and FILEHDR keywords because they are not
   // real output sections and cannot be added in the following loop.
+  std::vector<size_t> DefPhdrIds;
   for (const PhdrsCommand &Cmd : Opt.PhdrsCommands) {
     Ret.emplace_back(Cmd.Type, Cmd.Flags == UINT_MAX ? PF_R : Cmd.Flags);
     PhdrEntry<ELFT> &Phdr = Ret.back();
 
     if (Cmd.HasFilehdr)
       Phdr.add(Out<ELFT>::ElfHeader);
     if (Cmd.HasPhdrs)
       Phdr.add(Out<ELFT>::ProgramHeaders);
 
     if (Cmd.LMAExpr) {
       Phdr.H.p_paddr = Cmd.LMAExpr(0);
       Phdr.HasLMA = true;
     }
+
+    // If output section command doesn't specify any segments,
+    // and we haven't previously assigned any section to segment,
+    // then we simply assign section to the very first load segment.
+    // Below is an example of such linker script:
+    // PHDRS { seg PT_LOAD; }
+    // SECTIONS { .aaa : { *(.aaa) } }
+    if (DefPhdrIds.empty() && Phdr.H.p_type == PT_LOAD)
+      DefPhdrIds.push_back(Ret.size() - 1);
   }
 
   // Add output sections to program headers.
-  PhdrEntry<ELFT> *Load = nullptr;
-  uintX_t Flags = PF_R;
   for (OutputSectionBase<ELFT> *Sec : *OutputSections) {
     if (!(Sec->getFlags() & SHF_ALLOC))
       break;
 
     std::vector<size_t> PhdrIds = getPhdrIndices(Sec->getName());
-    if (!PhdrIds.empty()) {
-      // Assign headers specified by linker script
-      for (size_t Id : PhdrIds) {
-        Ret[Id].add(Sec);
-        if (Opt.PhdrsCommands[Id].Flags == UINT_MAX)
-          Ret[Id].H.p_flags |= Sec->getPhdrFlags();
-      }
-    } else {
-      // If we have no load segment or flags've changed then we want new load
-      // segment.
-      uintX_t NewFlags = Sec->getPhdrFlags();
-      if (Load == nullptr || Flags != NewFlags) {
-        Load = &*Ret.emplace(Ret.end(), PT_LOAD, NewFlags);
-        Flags = NewFlags;
-      }
-      Load->add(Sec);
-    }
+    if (PhdrIds.empty())
+      PhdrIds = std::move(DefPhdrIds);
+
+    DefPhdrIds = std::move(PhdrIds);
   }
   return Ret;
 }
 
 template <class ELFT> bool LinkerScript<ELFT>::ignoreInterpSection() {
   // Ignore .interp section in case we have PHDRS specification
   // and PT_INTERP isn't listed.
   return !Opt.PhdrsCommands.empty() &&

test/ELF/linkerscript/openbsd-randomize.s

 # RUN: llvm-mc -filetype=obj -triple=i686-unknown-linux %s -o %t.o
 # RUN: echo "PHDRS { text PT_LOAD FILEHDR PHDRS; rand PT_OPENBSD_RANDOMIZE; } \
 # RUN:       SECTIONS { . = SIZEOF_HEADERS; \
+# RUN:         .text : { *(.text) } \
 # RUN:         .openbsd.randomdata : { *(.openbsd.randomdata) } : rand }" > %t.script
 # RUN: ld.lld --script %t.script %t.o -o %t
 # RUN: llvm-readobj --program-headers -s %t | FileCheck %s
 
 # CHECK:      ProgramHeader {
 # CHECK:        Type: PT_OPENBSD_RANDOMIZE (0x65A3DBE6)
-# CHECK-NEXT:   Offset: 0x94
-# CHECK-NEXT:   VirtualAddress: 0x94
-# CHECK-NEXT:   PhysicalAddress: 0x94
+# CHECK-NEXT:   Offset: 0x74
+# CHECK-NEXT:   VirtualAddress: 0x74
+# CHECK-NEXT:   PhysicalAddress: 0x74
 # CHECK-NEXT:   FileSize: 8
 # CHECK-NEXT:   MemSize: 8
 # CHECK-NEXT:   Flags [ (0x4)
 # CHECK-NEXT:     PF_R (0x4)
 # CHECK-NEXT:   ]
 # CHECK-NEXT:   Alignment: 1
 # CHECK-NEXT: }
 
 .section .openbsd.randomdata, "a"
 .quad 0

test/ELF/linkerscript/phdrs-flags.s

 # REQUIRES: x86
 # RUN: llvm-mc -filetype=obj -triple=x86_64-unknown-linux %s -o %t
 # RUN: echo "PHDRS {all PT_LOAD FILEHDR PHDRS FLAGS (1 | 1 + 0x1);} \
 # RUN:       SECTIONS { \
 # RUN:           . = 0x10000200; \
 # RUN:           .text : {*(.text*)} :all \
 # RUN:           .foo : {*(.foo.*)} :all \
 # RUN:           .data : {*(.data.*)} :all}" > %t.script
-
 # RUN: ld.lld -o %t1 --script %t.script %t
 # RUN: llvm-readobj -program-headers %t1 | FileCheck %s
+
+# RUN: echo "PHDRS {all PT_LOAD FILEHDR PHDRS FLAGS (0x1);} \
+# RUN:       SECTIONS { \
+# RUN:           . = 0x10000200; \
+# RUN:           .text : {*(.text*)} :all \
+# RUN:           .foo : {*(.foo.*)}  \
+# RUN:           .data : {*(.data.*)} }" > %t.script
+# RUN: ld.lld -o %t1 --script %t.script %t
+# RUN: llvm-readobj -program-headers %t1 | FileCheck --check-prefix=DEFHDR %s
+
 # CHECK:     ProgramHeaders [
 # CHECK-NEXT:  ProgramHeader {
 # CHECK-NEXT:    Type: PT_LOAD (0x1)
 # CHECK-NEXT:    Offset: 0x0
 # CHECK-NEXT:    VirtualAddress: 0x10000000
 # CHECK-NEXT:    PhysicalAddress: 0x10000000
 # CHECK-NEXT:    FileSize: 521
 # CHECK-NEXT:    MemSize: 521
 # CHECK-NEXT:    Flags [
 # CHECK-NEXT:      PF_W (0x2)
 # CHECK-NEXT:      PF_X (0x1)
 # CHECK-NEXT:    ]
 
+# DEFHDR:     ProgramHeaders [
+# DEFHDR-NEXT:  ProgramHeader {
+# DEFHDR-NEXT:    Type: PT_LOAD (0x1)
+# DEFHDR-NEXT:    Offset: 0x0
+# DEFHDR-NEXT:    VirtualAddress: 0x10000000
+# DEFHDR-NEXT:    PhysicalAddress: 0x10000000
+# DEFHDR-NEXT:    FileSize: 521
+# DEFHDR-NEXT:    MemSize: 521
+# DEFHDR-NEXT:    Flags [ (0x1)
+# DEFHDR-NEXT:      PF_X (0x1)
+# DEFHDR-NEXT:    ]
+# DEFHDR-NEXT:    Alignment: 4096
+# DEFHDR-NEXT:  }
+
 .global _start
 _start:
  nop
 
 .section .foo.1,"a"
 foo1:
  .long 0
 
 .section .foo.2,"aw"
 foo2:
  .long 0

test/ELF/linkerscript/phdrs.s

 # RUN:       SECTIONS { \
 # RUN:           . = 0x10000200; \
 # RUN:           .text : {*(.text*)} :all \
 # RUN:           .foo : {*(.foo.*)} :all \
 # RUN:           .data : {*(.data.*)} :all}" > %t.script
 # RUN: ld.lld -o %t1 --script %t.script %t
 # RUN: llvm-readobj -program-headers %t1 | FileCheck --check-prefix=AT %s
 
+# RUN: echo "PHDRS {all PT_LOAD FILEHDR PHDRS ;} \
+# RUN:       SECTIONS { \
+# RUN:           . = 0x10000200; \
+# RUN:           .text : {*(.text*)} :all \
+# RUN:           .foo : {*(.foo.*)}  \
+# RUN:           .data : {*(.data.*)} }" > %t.script
+# RUN: ld.lld -o %t1 --script %t.script %t
+# RUN: llvm-readobj -program-headers %t1 | FileCheck --check-prefix=DEFHDR %s
+
 # CHECK:     ProgramHeaders [
 # CHECK-NEXT:  ProgramHeader {
 # CHECK-NEXT:    Type: PT_LOAD (0x1)
 # CHECK-NEXT:    Offset: 0x0
 # CHECK-NEXT:    VirtualAddress: 0x10000000
 # CHECK-NEXT:    PhysicalAddress: 0x10000000
 # CHECK-NEXT:    FileSize: 521
 # CHECK-NEXT:    MemSize: 521

 # INT-PHDRS-NEXT:        PF_R
 # INT-PHDRS-NEXT:        PF_W
 # INT-PHDRS-NEXT:        PF_X
 # INT-PHDRS-NEXT:      ]
 # INT-PHDRS-NEXT:      Alignment: 4
 # INT-PHDRS-NEXT:    }
 # INT-PHDRS-NEXT:  ]
 
+# DEFHDR:     ProgramHeaders [
+# DEFHDR-NEXT:  ProgramHeader {
+# DEFHDR-NEXT:    Type: PT_LOAD (0x1)
+# DEFHDR-NEXT:    Offset: 0x0
+# DEFHDR-NEXT:    VirtualAddress: 0x10000000
+# DEFHDR-NEXT:    PhysicalAddress: 0x10000000
+# DEFHDR-NEXT:    FileSize: 521
+# DEFHDR-NEXT:    MemSize: 521
+# DEFHDR-NEXT:    Flags [ (0x7)
+# DEFHDR-NEXT:      PF_R (0x4)
+# DEFHDR-NEXT:      PF_W (0x2)
+# DEFHDR-NEXT:      PF_X (0x1)
+# DEFHDR-NEXT:    ]
+
 .global _start
 _start:
  nop
 
 .section .foo.1,"a"
 foo1:
  .long 0
 
 .section .foo.2,"aw"
 foo2:
  .long 0