[wasm-ld] Add __global_base symbol to mark the value of --global-base

[quantum]

Summary

This is needed for address sanitizer on Emscripten. As everything in
memory starts at the value passed to --global-base, everything before
that can be used as shadow memory.

This symbol is added so that the library for the ASan runtime can know
where the shadow memory ends and real memory begins.

This is split from D63742.

[sbc100]

As I said in D63742, I don't really like that fact that __global_base and __data_end are use for the start and end of the static data but use completely different naming conventions, but since we already expose the --global-base command line arg I guess its hard to make the align without changing the command line arg and/or change the name of __data_end. So I think I'm ok with this for now, but I'd like to find a way to make these consistent in the future.

lld/test/wasm/global-base.ll

3

You could avoid creating a .ll file here maybe and simply make this a .test file (pure text) that compiles that simple wasm/Inputs/start.ll file.. then you could also drop the --no-entry

5

Do you need --no-gc-sections here?

lld/wasm/Writer.cpp

223

Looks like we need to figure what to do with GlobalBase in the case of --stack-first. But that is somewhat independent of this change.

[quantum]

Fix test as suggested in review.

[quantum]

lld/test/wasm/global-base.ll

3

Done.

5

Removed.

lld/wasm/Writer.cpp

223

I agree.

[sbc100]

lld/test/wasm/global-base.test

1 ↗

(On Diff #206726)

Can can remove all the leaving ; not too.

Still lgtm

[quantum]

Remove leading ;

lld/test/wasm/global-base.ll

+; RUN: llc -filetype=obj %s -o %t.o
+
+target triple = "wasm32-unknown-unknown"
+
+; RUN: wasm-ld -no-gc-sections --export=__global_base --export=__data_end --allow-undefined --no-entry -o %t.wasm %t.o
+; RUN: obj2yaml %t.wasm | FileCheck %s  -check-prefix=CHECK-1024
+; CHECK-1024:       - Type:            GLOBAL
+; CHECK-1024-NEXT:    Globals:
+; CHECK-1024-NEXT:      - Index:           0
+; CHECK-1024-NEXT:        Type:            I32
+; CHECK-1024-NEXT:        Mutable:         true
+; CHECK-1024-NEXT:        InitExpr:
+; CHECK-1024-NEXT:          Opcode:          I32_CONST
+; CHECK-1024-NEXT:          Value:           66560
+; CHECK-1024-NEXT:      - Index:           1
+; CHECK-1024-NEXT:        Type:            I32
+; CHECK-1024-NEXT:        Mutable:         false
+; CHECK-1024-NEXT:        InitExpr:
+; CHECK-1024-NEXT:          Opcode:          I32_CONST
+; CHECK-1024-NEXT:          Value:           1024
+; CHECK-1024-NEXT:      - Index:           2
+; CHECK-1024-NEXT:        Type:            I32
+; CHECK-1024-NEXT:        Mutable:         false
+; CHECK-1024-NEXT:        InitExpr:
+; CHECK-1024-NEXT:          Opcode:          I32_CONST
+; CHECK-1024-NEXT:          Value:           1024
+
+; CHECK-1024:       - Type:            EXPORT
+; CHECK-1024-NEXT:    Exports:
+; CHECK-1024-NEXT:      - Name:            memory
+; CHECK-1024-NEXT:        Kind:            MEMORY
+; CHECK-1024-NEXT:        Index:           0
+; CHECK-1024-NEXT:      - Name:            __data_end
+; CHECK-1024-NEXT:        Kind:            GLOBAL
+; CHECK-1024-NEXT:        Index:           1
+; CHECK-1024-NEXT:      - Name:            __global_base
+; CHECK-1024-NEXT:        Kind:            GLOBAL
+; CHECK-1024-NEXT:        Index:           2
+
+; RUN: wasm-ld -no-gc-sections --export=__global_base --export=__data_end --allow-undefined --global-base=16777216 --no-entry -o %t.wasm %t.o
+; RUN: obj2yaml %t.wasm | FileCheck %s  -check-prefix=CHECK-16777216
+; CHECK-16777216:       - Type:            GLOBAL
+; CHECK-16777216-NEXT:    Globals:
+; CHECK-16777216-NEXT:      - Index:           0
+; CHECK-16777216-NEXT:        Type:            I32
+; CHECK-16777216-NEXT:        Mutable:         true
+; CHECK-16777216-NEXT:        InitExpr:
+; CHECK-16777216-NEXT:          Opcode:          I32_CONST
+; CHECK-16777216-NEXT:          Value:           16842752
+; CHECK-16777216-NEXT:      - Index:           1
+; CHECK-16777216-NEXT:        Type:            I32
+; CHECK-16777216-NEXT:        Mutable:         false
+; CHECK-16777216-NEXT:        InitExpr:
+; CHECK-16777216-NEXT:          Opcode:          I32_CONST
+; CHECK-16777216-NEXT:          Value:           16777216
+; CHECK-16777216-NEXT:      - Index:           2
+; CHECK-16777216-NEXT:        Type:            I32
+; CHECK-16777216-NEXT:        Mutable:         false
+; CHECK-16777216-NEXT:        InitExpr:
+; CHECK-16777216-NEXT:          Opcode:          I32_CONST
+; CHECK-16777216-NEXT:          Value:           16777216
+
+; CHECK-16777216:       - Type:            EXPORT
+; CHECK-16777216-NEXT:    Exports:
+; CHECK-16777216-NEXT:      - Name:            memory
+; CHECK-16777216-NEXT:        Kind:            MEMORY
+; CHECK-16777216-NEXT:        Index:           0
+; CHECK-16777216-NEXT:      - Name:            __data_end
+; CHECK-16777216-NEXT:        Kind:            GLOBAL
+; CHECK-16777216-NEXT:        Index:           1
+; CHECK-16777216-NEXT:      - Name:            __global_base
+; CHECK-16777216-NEXT:        Kind:            GLOBAL
+; CHECK-16777216-NEXT:        Index:           2

lld/wasm/Driver.cpp

if (Config->Shared) {
     StackPointer->Live = true;
     // For non-PIC code
     // TODO(sbc): Remove WASM_SYMBOL_VISIBILITY_HIDDEN when the mutable global
     // spec proposal is implemented in all major browsers.
     // See: https://github.com/WebAssembly/mutable-global
     WasmSym::StackPointer = Symtab->addSyntheticGlobal(
         "__stack_pointer", WASM_SYMBOL_VISIBILITY_HIDDEN, StackPointer);
     WasmSym::DataEnd = Symtab->addOptionalDataSymbol("__data_end");
+    WasmSym::GlobalBase = Symtab->addOptionalDataSymbol("__global_base");
     WasmSym::HeapBase = Symtab->addOptionalDataSymbol("__heap_base");
   }
 
   if (Config->Pic) {
     // For PIC code, we import two global variables (__memory_base and
     // __table_base) from the environment and use these as the offset at
     // which to load our static data and function table.
     // See:

lld/wasm/Symbols.h

public:
   const WasmSignature *Signature = nullptr;
 
 private:
   llvm::object::Archive::Symbol ArchiveSymbol;
 };
 
 // linker-generated symbols
 struct WasmSym {
+  // __global_base
+  // Symbol marking the start of the global section.
+  static DefinedData *GlobalBase;
+
   // __stack_pointer
   // Global that holds the address of the top of the explicit value stack in
   // linear memory.
   static GlobalSymbol *StackPointer;
 
   // __data_end
   // Symbol marking the end of the data and bss.
   static DefinedData *DataEnd;

lld/wasm/Symbols.cpp

 using namespace llvm::wasm;
 using namespace lld;
 using namespace lld::wasm;
 
 DefinedFunction *WasmSym::CallCtors;
 DefinedFunction *WasmSym::ApplyRelocs;
 DefinedData *WasmSym::DsoHandle;
 DefinedData *WasmSym::DataEnd;
+DefinedData *WasmSym::GlobalBase;
 DefinedData *WasmSym::HeapBase;
 GlobalSymbol *WasmSym::StackPointer;
 UndefinedGlobal *WasmSym::TableBase;
 UndefinedGlobal *WasmSym::MemoryBase;
 
 WasmSymbolType Symbol::getWasmType() const {
   if (isa<FunctionSymbol>(this))
     return WASM_SYMBOL_TYPE_FUNCTION;

lld/wasm/Writer.cpp

auto PlaceStack = [&]() {
     auto *SP = cast<DefinedGlobal>(WasmSym::StackPointer);
     SP->Global->Global.InitExpr.Value.Int32 = MemoryPtr;
     log("mem: stack top   = " + Twine(MemoryPtr));
   };
 
   if (Config->StackFirst) {
     PlaceStack();
   } else {
     MemoryPtr = Config->GlobalBase;
     log("mem: global base = " + Twine(Config->GlobalBase));
   }
 
+  if (WasmSym::GlobalBase)
+    WasmSym::GlobalBase->setVirtualAddress(Config->GlobalBase);
+
   uint32_t DataStart = MemoryPtr;
 
   // Arbitrarily set __dso_handle handle to point to the start of the data
   // segments.
   if (WasmSym::DsoHandle)
     WasmSym::DsoHandle->setVirtualAddress(DataStart);
 
   Out.DylinkSec->MemAlign = 0;