diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..398baf2
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,17 @@
+# Binaries for programs and plugins
+*.exe
+*.exe~
+*.dll
+*.so
+*.dylib
+
+# Test binary, built with `go test -c`
+*.test
+
+# Output of the go coverage tool, specifically when used with LiteIDE
+*.out
+
+# Dependency directories (remove the comment below to include it)
+# vendor/
+
+.idea
diff --git a/ants.go b/ants.go
index 8187074..5dccc54 100644
--- a/ants.go
+++ b/ants.go
@@ -51,6 +51,9 @@ var (
 
 	// ErrPoolClosed will be returned when submitting task to a closed pool.
 	ErrPoolClosed = errors.New("this pool has been closed")
+
+	// ErrPoolOverload will be returned when the pool is full and no workers available.
+	ErrPoolOverload = errors.New("too many goroutines blocked on submit or Nonblocking is set")
 	//---------------------------------------------------------------------------
 
 	// workerChanCap determines whether the channel of a worker should be a buffered channel
diff --git a/ants_benchmark_test.go b/ants_benchmark_test.go
index 74e255b..fa9c423 100644
--- a/ants_benchmark_test.go
+++ b/ants_benchmark_test.go
@@ -23,6 +23,7 @@
 package ants_test
 
 import (
+	"runtime"
 	"sync"
 	"testing"
 	"time"
@@ -46,6 +47,22 @@ func demoPoolFunc(args interface{}) {
 	time.Sleep(time.Duration(n) * time.Millisecond)
 }
 
+func longRunningFunc() {
+	for {
+		runtime.Gosched()
+	}
+}
+
+func longRunningPoolFunc(arg interface{}) {
+	if ch, ok := arg.(chan struct{}); ok {
+		<-ch
+		return
+	}
+	for {
+		runtime.Gosched()
+	}
+}
+
 func BenchmarkGoroutineWithFunc(b *testing.B) {
 	var wg sync.WaitGroup
 	for i := 0; i < b.N; i++ {
diff --git a/ants_test.go b/ants_test.go
index 91ade65..9394492 100644
--- a/ants_test.go
+++ b/ants_test.go
@@ -416,6 +416,155 @@ func TestPurgePreMalloc(t *testing.T) {
 	}
 }
 
+func TestNonblockingSubmit(t *testing.T) {
+	poolSize := 10
+	p, err := ants.NewPool(poolSize)
+	if err != nil {
+		t.Fatalf("create TimingPool failed: %s", err.Error())
+	}
+	p.Nonblocking = true
+	defer p.Release()
+	for i := 0; i < poolSize-1; i++ {
+		if err := p.Submit(longRunningFunc); err != nil {
+			t.Fatalf("nonblocking submit when pool is not full shouldn't return error")
+		}
+	}
+	ch := make(chan struct{})
+	f := func() {
+		<-ch
+	}
+	// p is full now.
+	if err := p.Submit(f); err != nil {
+		t.Fatalf("nonblocking submit when pool is not full shouldn't return error")
+	}
+	if err := p.Submit(demoFunc); err == nil || err != ants.ErrPoolOverload {
+		t.Fatalf("nonblocking submit when pool is full should get an ErrPoolOverload")
+	}
+	// interrupt f to get an available worker
+	close(ch)
+	time.Sleep(1 * time.Second)
+	if err := p.Submit(demoFunc); err != nil {
+		t.Fatalf("nonblocking submit when pool is not full shouldn't return error")
+	}
+}
+
+func TestMaxBlockingSubmit(t *testing.T) {
+	poolSize := 10
+	p, err := ants.NewPool(poolSize)
+	if err != nil {
+		t.Fatalf("create TimingPool failed: %s", err.Error())
+	}
+	p.MaxBlockingTasks = 1
+	defer p.Release()
+	for i := 0; i < poolSize-1; i++ {
+		if err := p.Submit(longRunningFunc); err != nil {
+			t.Fatalf("submit when pool is not full shouldn't return error")
+		}
+	}
+	ch := make(chan struct{})
+	f := func() {
+		<-ch
+	}
+	// p is full now.
+	if err := p.Submit(f); err != nil {
+		t.Fatalf("submit when pool is not full shouldn't return error")
+	}
+	var wg sync.WaitGroup
+	wg.Add(1)
+	errCh := make(chan error, 1)
+	go func() {
+		// should be blocked. blocking num == 1
+		if err := p.Submit(demoFunc); err != nil {
+			errCh <- err
+		}
+		wg.Done()
+	}()
+	time.Sleep(1 * time.Second)
+	// already reached max blocking limit
+	if err := p.Submit(demoFunc); err != ants.ErrPoolOverload {
+		t.Fatalf("blocking submit when pool reach max blocking submit should return ErrPoolOverload")
+	}
+	// interrupt f to make blocking submit successful.
+	close(ch)
+	wg.Wait()
+	select {
+	case <-errCh:
+		t.Fatalf("blocking submit when pool is full should not return error")
+	default:
+	}
+}
+
+func TestNonblockingSubmitWithFunc(t *testing.T) {
+	poolSize := 10
+	p, err := ants.NewPoolWithFunc(poolSize, longRunningPoolFunc)
+	if err != nil {
+		t.Fatalf("create TimingPool failed: %s", err.Error())
+	}
+	p.Nonblocking = true
+	defer p.Release()
+	for i := 0; i < poolSize-1; i++ {
+		if err := p.Invoke(nil); err != nil {
+			t.Fatalf("nonblocking submit when pool is not full shouldn't return error")
+		}
+	}
+	ch := make(chan struct{})
+	// p is full now.
+	if err := p.Invoke(ch); err != nil {
+		t.Fatalf("nonblocking submit when pool is not full shouldn't return error")
+	}
+	if err := p.Invoke(nil); err == nil || err != ants.ErrPoolOverload {
+		t.Fatalf("nonblocking submit when pool is full should get an ErrPoolOverload")
+	}
+	// interrupt f to get an available worker
+	close(ch)
+	time.Sleep(1 * time.Second)
+	if err := p.Invoke(nil); err != nil {
+		t.Fatalf("nonblocking submit when pool is not full shouldn't return error")
+	}
+}
+
+func TestMaxBlockingSubmitWithFunc(t *testing.T) {
+	poolSize := 10
+	p, err := ants.NewPoolWithFunc(poolSize, longRunningPoolFunc)
+	if err != nil {
+		t.Fatalf("create TimingPool failed: %s", err.Error())
+	}
+	p.MaxBlockingTasks = 1
+	defer p.Release()
+	for i := 0; i < poolSize-1; i++ {
+		if err := p.Invoke(Param); err != nil {
+			t.Fatalf("submit when pool is not full shouldn't return error")
+		}
+	}
+	ch := make(chan struct{})
+	// p is full now.
+	if err := p.Invoke(ch); err != nil {
+		t.Fatalf("submit when pool is not full shouldn't return error")
+	}
+	var wg sync.WaitGroup
+	wg.Add(1)
+	errCh := make(chan error, 1)
+	go func() {
+		// should be blocked. blocking num == 1
+		if err := p.Invoke(Param); err != nil {
+			errCh <- err
+		}
+		wg.Done()
+	}()
+	time.Sleep(1 * time.Second)
+	// already reached max blocking limit
+	if err := p.Invoke(Param); err != ants.ErrPoolOverload {
+		t.Fatalf("blocking submit when pool reach max blocking submit should return ErrPoolOverload: %v", err)
+	}
+	// interrupt one func to make blocking submit successful.
+	close(ch)
+	wg.Wait()
+	select {
+	case <-errCh:
+		t.Fatalf("blocking submit when pool is full should not return error")
+	default:
+	}
+}
 func TestRestCodeCoverage(t *testing.T) {
 	_, err := ants.NewUltimatePool(-1, -1, false)
 	t.Log(err)
diff --git a/pool.go b/pool.go
index ef8e977..d2b817d 100644
--- a/pool.go
+++ b/pool.go
@@ -60,6 +60,19 @@ type Pool struct {
 	// PanicHandler is used to handle panics from each worker goroutine.
 	// if nil, panics will be thrown out again from worker goroutines.
 	PanicHandler func(interface{})
+
+	// Max number of goroutine blocking on pool.Submit.
+	// 0 (default value) means no such limit.
+	MaxBlockingTasks int32
+
+	// goroutine already been blocked on pool.Submit
+	// protected by pool.lock
+	blockingNum int32
+
+	// When Nonblocking is true, Pool.Submit will never be blocked.
+	// ErrPoolOverload will be returned when Pool.Submit cannot be done at once.
+	// When Nonblocking is true, MaxBlockingTasks is inoperative.
+	Nonblocking bool
 }
 
 // Clear expired workers periodically.
@@ -151,7 +164,11 @@ func (p *Pool) Submit(task func()) error {
 	if atomic.LoadInt32(&p.release) == CLOSED {
 		return ErrPoolClosed
 	}
-	p.retrieveWorker().task <- task
+	if w := p.retrieveWorker(); w == nil {
+		return ErrPoolOverload
+	} else {
+		w.task <- task
+	}
 	return nil
 }
 
@@ -237,8 +254,18 @@ func (p *Pool) retrieveWorker() *Worker {
 		p.lock.Unlock()
 		spawnWorker()
 	} else {
+		if p.Nonblocking {
+			p.lock.Unlock()
+			return nil
+		}
 	Reentry:
+		if p.MaxBlockingTasks != 0 && p.blockingNum >= p.MaxBlockingTasks {
+			p.lock.Unlock()
+			return nil
+		}
+		p.blockingNum++
 		p.cond.Wait()
+		p.blockingNum--
 		if p.Running() == 0 {
 			p.lock.Unlock()
 			spawnWorker()
diff --git a/pool_func.go b/pool_func.go
index 3f7e202..e8a8f00 100644
--- a/pool_func.go
+++ b/pool_func.go
@@ -63,6 +63,19 @@ type PoolWithFunc struct {
 	// PanicHandler is used to handle panics from each worker goroutine.
 	// if nil, panics will be thrown out again from worker goroutines.
 	PanicHandler func(interface{})
+
+	// Max number of goroutine blocking on pool.Submit.
+	// 0 (default value) means no such limit.
+	MaxBlockingTasks int32
+
+	// goroutine already been blocked on pool.Submit
+	// protected by pool.lock
+	blockingNum int32
+
+	// When Nonblocking is true, Pool.Submit will never be blocked.
+	// ErrPoolOverload will be returned when Pool.Submit cannot be done at once.
+	// When Nonblocking is true, MaxBlockingTasks is inoperative.
+	Nonblocking bool
 }
 
 // Clear expired workers periodically.
@@ -156,7 +169,11 @@ func (p *PoolWithFunc) Invoke(args interface{}) error {
 	if atomic.LoadInt32(&p.release) == CLOSED {
 		return ErrPoolClosed
 	}
-	p.retrieveWorker().args <- args
+	if w := p.retrieveWorker(); w == nil {
+		return ErrPoolOverload
+	} else {
+		w.args <- args
+	}
 	return nil
 }
 
@@ -242,8 +259,18 @@ func (p *PoolWithFunc) retrieveWorker() *WorkerWithFunc {
 		p.lock.Unlock()
 		spawnWorker()
 	} else {
+		if p.Nonblocking {
+			p.lock.Unlock()
+			return nil
+		}
 	Reentry:
+		if p.MaxBlockingTasks != 0 && p.blockingNum >= p.MaxBlockingTasks {
+			p.lock.Unlock()
+			return nil
+		}
+		p.blockingNum++
 		p.cond.Wait()
+		p.blockingNum--
 		if p.Running() == 0 {
 			p.lock.Unlock()
 			spawnWorker()