[ELF] - Linkerscript: do not fail on additional semicolons in linkerscript.

[grimar]

Summary

Linux kernel linkerscript contains additional semicolon (last line):

.apicdrivers : AT(ADDR(.apicdrivers) - LOAD_OFFSET) {
  __apicdrivers = .;
  *(.apicdrivers);

I checked that both gold and bfd are able to parse something like:

.text : { ;;*(.text);;S = 0;; } }

Patch do the same.

[jhenderson]

ELF/LinkerScript.cpp

1497 ↗

(On Diff #86266)

Slight grammar issue here - I think this should say something like "Commands may be surrounded by unnecessary additional semicolons", or better yet something like "Unnecessary semicolons are permitted here, but should be ignored."

[grimar]

• Addressed review comments.

ELF/LinkerScript.cpp

1497 ↗

(On Diff #86266)

Fixed, thanks.

[ruiu]

lld/trunk/ELF/LinkerScript.cpp

1497–1498

This is indeed better, but I think a more natural description about the grammar is that empty commands are allowed.

lld/trunk/ELF/LinkerScript.cpp

ScriptParser::readOutputSectionDescription(StringRef OutSec) {
   if (consume("ONLY_IF_RO"))
     Cmd->Constraint = ConstraintKind::ReadOnly;
   if (consume("ONLY_IF_RW"))
     Cmd->Constraint = ConstraintKind::ReadWrite;
   expect("{");
 
   while (!Error && !consume("}")) {
     StringRef Tok = next();
-    if (SymbolAssignment *Assignment = readProvideOrAssignment(Tok)) {
+    if (Tok == ";") {
+      // Commands may contain excessive additional semicolons around.
+      // We should be able to parse it.
+    } else if (SymbolAssignment *Assignment = readProvideOrAssignment(Tok)) {
       Cmd->Commands.emplace_back(Assignment);
     } else if (BytesDataCommand *Data = readBytesDataCommand(Tok)) {
       Cmd->Commands.emplace_back(Data);
     } else if (Tok == "ASSERT") {
       Cmd->Commands.emplace_back(new AssertCommand(readAssert()));
       expect(";");
     } else if (Tok == "CONSTRUCTORS") {
       // CONSTRUCTORS is a keyword to make the linker recognize C++ ctors/dtors

lld/trunk/test/ELF/linkerscript/sections.s

 # SEC-MULTI: 1 .text         0000000e {{[0-9a-f]*}} TEXT DATA
 # SEC-MULTI: 2 .data         00000023 {{[0-9a-f]*}} DATA
 # SEC-MULTI: 3 .bss          00000002 {{[0-9a-f]*}} BSS
 # SEC-MULTI: 4 .comment      00000008 {{[0-9a-f]*}}
 # SEC-MULTI: 5 .symtab       00000030 {{[0-9a-f]*}}
 # SEC-MULTI: 6 .shstrtab     00000035 {{[0-9a-f]*}}
 # SEC-MULTI: 7 .strtab       00000008 {{[0-9a-f]*}}
 
+# Input section pattern contains additional semicolon.
+# Case found in linux kernel script. Check we are able to parse it.
+# RUN: echo "SECTIONS { .text : { ;;*(.text);;S = 0;; } }" > %t.script
+# RUN: ld.lld -o /dev/null --script %t.script %t
+
 .globl _start
 _start:
     mov $60, %rax
     mov $42, %rdi
 
 .section .data,"aw"
 .quad 10, 10, 20, 20
 .section other,"aw"
 .short 10
 .byte 20
 .section .bss,"",@nobits
 .short 0