WebAssembly: Never write more than 32-bits for WebAssembly::OPERAND_OFFSET32

[sbc100]

Summary

A bug was found where an offset of -1 would generate an encoding
of max int64 which is invalid in the binary format.

[sbc100]

update test

[sbc100]

remove debugging

[sbc100]

• update test

[dschuff]

LGTM for the encoding... I hope the -1 offset wasn't being generated by the compiler though?

[sbc100]

In D45280#1057520, @dschuff wrote:

LGTM for the encoding... I hope the -1 offset wasn't being generated by the compiler though?

It does come from the compiler:

void _start(void) {
  char *oob = (char *) -1; *oob = 0;
}

lib/Target/WebAssembly/MCTargetDesc/WebAssemblyMCCodeEmitter.cpp

 #include "llvm/ADT/Statistic.h"
 #include "llvm/MC/MCCodeEmitter.h"
 #include "llvm/MC/MCFixup.h"
 #include "llvm/MC/MCInst.h"
 #include "llvm/MC/MCInstrInfo.h"
 #include "llvm/MC/MCRegisterInfo.h"
 #include "llvm/MC/MCSubtargetInfo.h"
 #include "llvm/MC/MCSymbol.h"
+#include "llvm/Support/Debug.h"
 #include "llvm/Support/EndianStream.h"
 #include "llvm/Support/LEB128.h"
 #include "llvm/Support/raw_ostream.h"
+
 using namespace llvm;
 
 #define DEBUG_TYPE "mccodeemitter"
 
 STATISTIC(MCNumEmitted, "Number of MC instructions emitted.");
 STATISTIC(MCNumFixups, "Number of MC fixups created.");
 
 namespace {
for (unsigned i = 0, e = MI.getNumOperands(); i < e; ++i) {
     const MCOperand &MO = MI.getOperand(i);
     if (MO.isReg()) {
       /* nothing to encode */
     } else if (MO.isImm()) {
       if (i < Desc.getNumOperands()) {
         assert(Desc.TSFlags == 0 &&
                "WebAssembly non-variable_ops don't use TSFlags");
         const MCOperandInfo &Info = Desc.OpInfo[i];
+        DEBUG(dbgs() << "Encoding immediate: type=" << int(Info.OperandType) << "\n");
         if (Info.OperandType == WebAssembly::OPERAND_I32IMM) {
           encodeSLEB128(int32_t(MO.getImm()), OS);
+        } else if (Info.OperandType == WebAssembly::OPERAND_OFFSET32) {
+          encodeULEB128(uint32_t(MO.getImm()), OS);
         } else if (Info.OperandType == WebAssembly::OPERAND_I64IMM) {
           encodeSLEB128(int64_t(MO.getImm()), OS);
         } else if (Info.OperandType == WebAssembly::OPERAND_GLOBAL) {
           llvm_unreachable("wasm globals should only be accessed symbolicly");
         } else if (Info.OperandType == WebAssembly::OPERAND_SIGNATURE) {
           OS << uint8_t(MO.getImm());
         } else {
           encodeULEB128(uint64_t(MO.getImm()), OS);

test/MC/WebAssembly/offset.ll

+; RUN: llc -filetype=obj %s -o - | obj2yaml | FileCheck %s
+
+target triple = "wasm32-unknown-unknown-wasm"
+
+
+; CHECK:        - Type:            CODE
+; CHECK-NEXT:     Functions:
+; CHECK-NEXT:       - Index:           0
+; CHECK-NEXT:         Locals:
+; CHECK-NEXT:         Body:            41002802FFFFFFFF0F0B
+define i32 @load_i32_from_negative_address() {
+  %s = inttoptr i32 -1 to i32*
+  %t = load i32, i32* %s
+  ret i32 %t
+}
+
+; CHECK-NEXT:       - Index:           1
+; CHECK-NEXT:         Locals:          
+; CHECK-NEXT:         Body:            41002802030B
+define i32 @load_i32_from_wrapped_address() {
+  %s = inttoptr i32 4294967299 to i32*
+  %t = load i32, i32* %s
+  ret i32 %t
+}
+