interp: bail out of loops that iterate too many times (#5395)

* interp: bail out of loops that iterate too many times The existing loop guard (errLoopUnrolled) only fires when a loop body emits runtime instructions. Loops that are fully evaluable at compile time, such as inserting thousands of entries into a map, were not caught and could hang the compiler. Add a per-basic-block iteration counter that triggers a recoverable error (errLoopTooLong) when any block is entered more than 1000 times in a single function call. This defers the init function to runtime, which is the same behavior as other interp bailouts. Profiling showed that 83% of CPU time was spent in GC, caused by allocation pressure from the interp memory cloning on each map mutation. The iteration limit avoids this entirely by bailing out before the quadratic cost becomes significant. Performance on the reproducer from #2090 (map init with strconv.Itoa): entries before after 5,000 7.4s 2.1s 10,000 17.5s 2.1s 20,000 48.0s 2.8s 65,536 >180s (OOM) 3.2s * Add -interp-loop-limit
2026-07-26 06:38:42 +00:00 · 2026-05-17 11:42:27 -07:00
parent e80e7e5c10
commit 89d9e33bca
7 changed files with 138 additions and 109 deletions
@@ -489,7 +489,7 @@ func Build(pkgName, outpath, tmpdir string, config *compileopts.Config) (BuildRe
 				if pkgInit.IsNil() {
 					panic("init not found for " + pkg.Pkg.Path())
 				}
-				err := interp.RunFunc(pkgInit, config.Options.InterpTimeout, config.DumpSSA())
+				err := interp.RunFunc(pkgInit, config.Options.InterpTimeout, config.Options.InterpMaxLoopIterations, config.DumpSSA())
 				if err != nil {
 					return err
 				}
@@ -1208,7 +1208,7 @@ func createEmbedObjectFile(data, hexSum, sourceFile, sourceDir, tmpdir string, c
 // needed to convert a program to its final form. Some transformations are not
 // optional and must be run as the compiler expects them to run.
 func optimizeProgram(mod llvm.Module, config *compileopts.Config) error {
-	err := interp.Run(mod, config.Options.InterpTimeout, config.DumpSSA())
+	err := interp.Run(mod, config.Options.InterpTimeout, config.Options.InterpMaxLoopIterations, config.DumpSSA())
 	if err != nil {
 		return err
 	}
@@ -21,45 +21,46 @@ var (
 // usually passed from the command line, but can also be passed in environment
 // variables for example.
 type Options struct {
-	GOOS            string // environment variable
-	GOARCH          string // environment variable
-	GOARM           string // environment variable (only used with GOARCH=arm)
-	GOMIPS          string // environment variable (only used with GOARCH=mips and GOARCH=mipsle)
-	Directory       string // working dir, leave it unset to use the current working dir
-	Target          string
-	BuildMode       string // -buildmode flag
-	Opt             string
-	GC              string
-	PanicStrategy   string
-	Scheduler       string
-	StackSize       uint64 // goroutine stack size (if none could be automatically determined)
-	Serial          string
-	Work            bool // -work flag to print temporary build directory
-	InterpTimeout   time.Duration
-	PrintIR         bool
-	DumpSSA         bool
-	VerifyIR        bool
-	SkipDWARF       bool
-	PrintCommands   func(cmd string, args ...string) `json:"-"`
-	Semaphore       chan struct{}                    `json:"-"` // -p flag controls cap
-	Debug           bool
-	Nobounds        bool
-	PrintSizes      string
-	PrintAllocs     *regexp.Regexp // regexp string
-	PrintStacks     bool
-	Tags            []string
-	GlobalValues    map[string]map[string]string // map[pkgpath]map[varname]value
-	TestConfig      TestConfig
-	Programmer      string
-	OpenOCDCommands []string
-	LLVMFeatures    string
-	Monitor         bool
-	BaudRate        int
-	Timeout         time.Duration
-	WITPackage      string // pass through to wasm-tools component embed invocation
-	WITWorld        string // pass through to wasm-tools component embed -w option
-	ExtLDFlags      []string
-	GoCompatibility bool // enable to check for Go version compatibility
+	GOOS                    string // environment variable
+	GOARCH                  string // environment variable
+	GOARM                   string // environment variable (only used with GOARCH=arm)
+	GOMIPS                  string // environment variable (only used with GOARCH=mips and GOARCH=mipsle)
+	Directory               string // working dir, leave it unset to use the current working dir
+	Target                  string
+	BuildMode               string // -buildmode flag
+	Opt                     string
+	GC                      string
+	PanicStrategy           string
+	Scheduler               string
+	StackSize               uint64 // goroutine stack size (if none could be automatically determined)
+	Serial                  string
+	Work                    bool // -work flag to print temporary build directory
+	InterpTimeout           time.Duration
+	InterpMaxLoopIterations int
+	PrintIR                 bool
+	DumpSSA                 bool
+	VerifyIR                bool
+	SkipDWARF               bool
+	PrintCommands           func(cmd string, args ...string) `json:"-"`
+	Semaphore               chan struct{}                    `json:"-"` // -p flag controls cap
+	Debug                   bool
+	Nobounds                bool
+	PrintSizes              string
+	PrintAllocs             *regexp.Regexp // regexp string
+	PrintStacks             bool
+	Tags                    []string
+	GlobalValues            map[string]map[string]string // map[pkgpath]map[varname]value
+	TestConfig              TestConfig
+	Programmer              string
+	OpenOCDCommands         []string
+	LLVMFeatures            string
+	Monitor                 bool
+	BaudRate                int
+	Timeout                 time.Duration
+	WITPackage              string // pass through to wasm-tools component embed invocation
+	WITWorld                string // pass through to wasm-tools component embed -w option
+	ExtLDFlags              []string
+	GoCompatibility         bool // enable to check for Go version compatibility
 }

 // Verify performs a validation on the given options, raising an error if options are not valid.
@@ -19,6 +19,7 @@ var (
 	errUnsupportedRuntimeInst = errors.New("interp: unsupported instruction (to be emitted at runtime)")
 	errMapAlreadyCreated      = errors.New("interp: map already created")
 	errLoopUnrolled           = errors.New("interp: loop unrolled")
+	errLoopTooLong            = errors.New("interp: loop ran too many iterations")
 )

 // This is one of the errors that can be returned from toLLVMValue when the
@@ -29,7 +30,7 @@ var errInvalidPtrToIntSize = errors.New("interp: ptrtoint integer size does not
 func isRecoverableError(err error) bool {
 	return err == errIntegerAsPointer || err == errUnsupportedInst ||
 		err == errUnsupportedRuntimeInst || err == errMapAlreadyCreated ||
-		err == errLoopUnrolled || err == errInvalidPtrToIntSize
+		err == errLoopUnrolled || err == errLoopTooLong || err == errInvalidPtrToIntSize
 }

 // ErrorLine is one line in a traceback. The position may be missing.
@@ -19,35 +19,37 @@ const checks = true

 // runner contains all state related to one interp run.
 type runner struct {
-	mod           llvm.Module
-	targetData    llvm.TargetData
-	builder       llvm.Builder
-	pointerSize   uint32                   // cached pointer size from the TargetData
-	dataPtrType   llvm.Type                // often used type so created in advance
-	uintptrType   llvm.Type                // equivalent to uintptr in Go
-	maxAlign      int                      // maximum alignment of an object, alignment of runtime.alloc() result
-	byteOrder     binary.ByteOrder         // big-endian or little-endian
-	debug         bool                     // log debug messages
-	pkgName       string                   // package name of the currently executing package
-	functionCache map[llvm.Value]*function // cache of compiled functions
-	objects       []object                 // slice of objects in memory
-	globals       map[llvm.Value]int       // map from global to index in objects slice
-	start         time.Time
-	timeout       time.Duration
-	callsExecuted uint64
+	mod               llvm.Module
+	targetData        llvm.TargetData
+	builder           llvm.Builder
+	pointerSize       uint32                   // cached pointer size from the TargetData
+	dataPtrType       llvm.Type                // often used type so created in advance
+	uintptrType       llvm.Type                // equivalent to uintptr in Go
+	maxAlign          int                      // maximum alignment of an object, alignment of runtime.alloc() result
+	byteOrder         binary.ByteOrder         // big-endian or little-endian
+	debug             bool                     // log debug messages
+	pkgName           string                   // package name of the currently executing package
+	functionCache     map[llvm.Value]*function // cache of compiled functions
+	objects           []object                 // slice of objects in memory
+	globals           map[llvm.Value]int       // map from global to index in objects slice
+	start             time.Time
+	timeout           time.Duration
+	maxLoopIterations int
+	callsExecuted     uint64
 }

-func newRunner(mod llvm.Module, timeout time.Duration, debug bool) *runner {
+func newRunner(mod llvm.Module, timeout time.Duration, maxLoopIterations int, debug bool) *runner {
 	r := runner{
-		mod:           mod,
-		targetData:    llvm.NewTargetData(mod.DataLayout()),
-		byteOrder:     llvmutil.ByteOrder(mod.Target()),
-		debug:         debug,
-		functionCache: make(map[llvm.Value]*function),
-		objects:       []object{{}},
-		globals:       make(map[llvm.Value]int),
-		start:         time.Now(),
-		timeout:       timeout,
+		mod:               mod,
+		targetData:        llvm.NewTargetData(mod.DataLayout()),
+		byteOrder:         llvmutil.ByteOrder(mod.Target()),
+		debug:             debug,
+		functionCache:     make(map[llvm.Value]*function),
+		objects:           []object{{}},
+		globals:           make(map[llvm.Value]int),
+		start:             time.Now(),
+		timeout:           timeout,
+		maxLoopIterations: maxLoopIterations,
 	}
 	r.pointerSize = uint32(r.targetData.PointerSize())
 	r.dataPtrType = llvm.PointerType(mod.Context().Int8Type(), 0)
@@ -64,8 +66,8 @@ func (r *runner) dispose() {

 // Run evaluates runtime.initAll function as much as possible at compile time.
 // Set debug to true if it should print output while running.
-func Run(mod llvm.Module, timeout time.Duration, debug bool) error {
-	r := newRunner(mod, timeout, debug)
+func Run(mod llvm.Module, timeout time.Duration, maxLoopIterations int, debug bool) error {
+	r := newRunner(mod, timeout, maxLoopIterations, debug)
 	defer r.dispose()

 	initAll := mod.NamedFunction("runtime.initAll")
@@ -204,10 +206,10 @@ func Run(mod llvm.Module, timeout time.Duration, debug bool) error {

 // RunFunc evaluates a single package initializer at compile time.
 // Set debug to true if it should print output while running.
-func RunFunc(fn llvm.Value, timeout time.Duration, debug bool) error {
+func RunFunc(fn llvm.Value, timeout time.Duration, maxLoopIterations int, debug bool) error {
 	// Create and initialize *runner object.
 	mod := fn.GlobalParent()
-	r := newRunner(mod, timeout, debug)
+	r := newRunner(mod, timeout, maxLoopIterations, debug)
 	defer r.dispose()
 	initName := fn.Name()
 	if !strings.HasSuffix(initName, ".init") {
@@ -44,7 +44,7 @@ func runTest(t *testing.T, pathPrefix string) {
 	defer mod.Dispose()

 	// Perform the transform.
-	err = Run(mod, 10*time.Minute, false)
+	err = Run(mod, 10*time.Minute, DefaultMaxInterpBlockEntries, false)
 	if err != nil {
 		if err, match := err.(*Error); match {
 			println(err.Error())
@@ -12,6 +12,12 @@ import (
 	"tinygo.org/x/go-llvm"
 )

+// DefaultMaxInterpBlockEntries is the default maximum number of times a single
+// basic block may be entered during interpretation of one function call. This
+// limits how far the interpreter will unroll or evaluate loops before deferring
+// the init function to runtime.
+const DefaultMaxInterpBlockEntries = 1000
+
 func (r *runner) run(fn *function, params []value, parentMem *memoryView, indent string) (value, memoryView, *Error) {
 	mem := memoryView{r: r, parent: parentMem}
 	locals := make([]value, len(fn.locals))
@@ -26,6 +32,10 @@ func (r *runner) run(fn *function, params []value, parentMem *memoryView, indent
 	// This is used to prevent unrolling.
 	var runtimeBlocks map[int]struct{}

+	// Track how many times each basic block has been entered, to detect
+	// loops that are too expensive to evaluate at compile time.
+	var blockCounts map[int]int
+
 	// Start with the first basic block and the first instruction.
 	// Branch instructions may modify both bb and instIndex when branching.
 	bb := fn.blocks[0]
@@ -36,6 +46,18 @@ func (r *runner) run(fn *function, params []value, parentMem *memoryView, indent
 	for instIndex := 0; instIndex < len(bb.instructions); instIndex++ {
 		if instIndex == 0 {
 			// This is the start of a new basic block.
+
+			// Check whether this block has been entered too many times,
+			// which indicates an expensive loop that should be deferred
+			// to runtime.
+			if blockCounts == nil {
+				blockCounts = make(map[int]int)
+			}
+			blockCounts[currentBB]++
+			if r.maxLoopIterations > 0 && blockCounts[currentBB] > r.maxLoopIterations {
+				return nil, mem, r.errorAt(bb.instructions[0], errLoopTooLong)
+			}
+
 			if len(mem.instructions) != startRTInsts {
 				if _, ok := runtimeBlocks[lastBB]; ok {
 					// This loop has been unrolled.
@@ -30,6 +30,7 @@ import (
 	"github.com/tinygo-org/tinygo/compileopts"
 	"github.com/tinygo-org/tinygo/diagnostics"
 	"github.com/tinygo-org/tinygo/goenv"
+	"github.com/tinygo-org/tinygo/interp"
 	"github.com/tinygo-org/tinygo/loader"
 	"golang.org/x/tools/go/buildutil"
 	"tinygo.org/x/espflasher/pkg/espflasher"
@@ -1738,6 +1739,7 @@ func main() {
 	serial := flag.String("serial", "", "which serial output to use (none, uart, usb, rtt)")
 	work := flag.Bool("work", false, "print the name of the temporary build directory and do not delete this directory on exit")
 	interpTimeout := flag.Duration("interp-timeout", 180*time.Second, "interp optimization pass timeout")
+	interpLoopLimit := flag.Int("interp-loop-limit", interp.DefaultMaxInterpBlockEntries, "maximum loop iterations during interp (0 to disable)")
 	var tags buildutil.TagsFlag
 	flag.Var(&tags, "tags", "a space-separated list of extra build tags")
 	target := flag.String("target", "", "chip/board name or JSON target specification file")
@@ -1855,42 +1857,43 @@ func main() {
 	}

 	options := &compileopts.Options{
-		GOOS:            goenv.Get("GOOS"),
-		GOARCH:          goenv.Get("GOARCH"),
-		GOARM:           goenv.Get("GOARM"),
-		GOMIPS:          goenv.Get("GOMIPS"),
-		Target:          *target,
-		BuildMode:       *buildMode,
-		StackSize:       stackSize,
-		Opt:             *opt,
-		GC:              *gc,
-		PanicStrategy:   *panicStrategy,
-		Scheduler:       *scheduler,
-		Serial:          *serial,
-		Work:            *work,
-		InterpTimeout:   *interpTimeout,
-		PrintIR:         *printIR,
-		DumpSSA:         *dumpSSA,
-		VerifyIR:        *verifyIR,
-		SkipDWARF:       *skipDwarf,
-		Semaphore:       make(chan struct{}, *parallelism),
-		Debug:           !*nodebug,
-		Nobounds:        *nobounds,
-		PrintSizes:      *printSize,
-		PrintStacks:     *printStacks,
-		PrintAllocs:     printAllocs,
-		Tags:            []string(tags),
-		TestConfig:      testConfig,
-		GlobalValues:    globalVarValues,
-		Programmer:      *programmer,
-		OpenOCDCommands: ocdCommands,
-		LLVMFeatures:    *llvmFeatures,
-		Monitor:         *monitor,
-		BaudRate:        *baudrate,
-		Timeout:         *timeout,
-		WITPackage:      witPackage,
-		WITWorld:        witWorld,
-		GoCompatibility: *gocompatibility,
+		GOOS:                    goenv.Get("GOOS"),
+		GOARCH:                  goenv.Get("GOARCH"),
+		GOARM:                   goenv.Get("GOARM"),
+		GOMIPS:                  goenv.Get("GOMIPS"),
+		Target:                  *target,
+		BuildMode:               *buildMode,
+		StackSize:               stackSize,
+		Opt:                     *opt,
+		GC:                      *gc,
+		PanicStrategy:           *panicStrategy,
+		Scheduler:               *scheduler,
+		Serial:                  *serial,
+		Work:                    *work,
+		InterpTimeout:           *interpTimeout,
+		InterpMaxLoopIterations: *interpLoopLimit,
+		PrintIR:                 *printIR,
+		DumpSSA:                 *dumpSSA,
+		VerifyIR:                *verifyIR,
+		SkipDWARF:               *skipDwarf,
+		Semaphore:               make(chan struct{}, *parallelism),
+		Debug:                   !*nodebug,
+		Nobounds:                *nobounds,
+		PrintSizes:              *printSize,
+		PrintStacks:             *printStacks,
+		PrintAllocs:             printAllocs,
+		Tags:                    []string(tags),
+		TestConfig:              testConfig,
+		GlobalValues:            globalVarValues,
+		Programmer:              *programmer,
+		OpenOCDCommands:         ocdCommands,
+		LLVMFeatures:            *llvmFeatures,
+		Monitor:                 *monitor,
+		BaudRate:                *baudrate,
+		Timeout:                 *timeout,
+		WITPackage:              witPackage,
+		WITWorld:                witWorld,
+		GoCompatibility:         *gocompatibility,
 	}
 	if *printCommands {
 		options.PrintCommands = printCommand