diff --git a/mongoose.c b/mongoose.c
index 283e98f5e4..7adeffe419 100644
--- a/mongoose.c
+++ b/mongoose.c
@@ -6787,6 +6787,8 @@ struct connstate {
   uint64_t timer;                         // TCP timer (see 'ttype' below)
   uint32_t acked;                         // Last ACK-ed number
   size_t unacked;                         // Not acked bytes
+  uint32_t maxseq;                        // Max send seq (ack + window)
+  uint16_t win;                           // destination current window size
   uint16_t dmss;                          // destination MSS (from TCP opts)
   uint8_t mac[sizeof(struct mg_l2addr)];  // Peer hw address
   uint8_t ttype;                          // Timer type:
@@ -8032,6 +8034,7 @@ static struct mg_connection *accept_conn(struct mg_connection *lsn,
   s = (struct connstate *) (c + 1);
   s->dmss = mss;  // from options in client SYN
   s->seq = mg_ntohl(pkt->tcp->ack), s->ack = mg_ntohl(pkt->tcp->seq);
+  s->win = mg_ntohs(pkt->tcp->win), s->maxseq = (uint32_t)(s->seq + s->win);
 #if MG_ENABLE_IPV6
   if (lsn->loc.is_ip6) {
     c->rem.addr.ip6[0] = pkt->ip6->src[0],
@@ -8124,10 +8127,13 @@ long mg_io_send(struct mg_connection *c, const void *buf, size_t len) {
   len = trim_len(c, len);
   if (c->is_udp) {
     if (!udp_send(c, buf, len)) return MG_IO_WAIT;
-  } else {  // TCP, cap to peer's MSS
+  } else {  // TCP, cap to peer's MSS and check window
     struct mg_tcpip_if *ifp = c->mgr->ifp;
     size_t sent;
+    uint32_t room = s->maxseq - s->seq;
+    if (room == 0) return MG_IO_WAIT;
     if (len > s->dmss) len = s->dmss;  // RFC-6691: reduce if sending opts
+    if ((uint32_t) len > room) len = room;
     sent = tx_tcp(ifp, s->mac, &c->loc, &c->rem, TH_PUSH | TH_ACK,
                   mg_htonl(s->seq), mg_htonl(s->ack), buf, len);
     if (sent == 0) {
@@ -8162,6 +8168,12 @@ static void handle_tls_recv(struct mg_connection *c) {
   }
 }
 
+static void handle_ack(struct connstate *s, uint32_t ackno, uint16_t win) {
+  if (ackno < (s->seq - s->win) || ackno > s->seq) return;
+  s->maxseq = (uint32_t)(ackno + win);
+  s->win = win;
+}
+
 static void read_conn(struct mg_connection *c, struct pkt *pkt) {
   struct connstate *s = (struct connstate *) (c + 1);
   struct mg_iobuf *io = c->is_tls ? &c->rtls : &c->recv;
@@ -8201,6 +8213,8 @@ static void read_conn(struct mg_connection *c, struct pkt *pkt) {
            mg_htonl(s->ack), NULL, 0);
     return;                        // no data to process
   } else if (pkt->pay.len == 0) {  // this is an ACK
+    if (pkt->tcp->flags & TH_ACK)
+      handle_ack(s, mg_ntohl(pkt->tcp->ack), mg_ntohs(pkt->tcp->win));
     if (s->fin_rcvd && s->ttype == MIP_TTYPE_FIN) s->twclosure = true;
     return;  // no data to process
   } else if (seq != s->ack) {
@@ -8218,6 +8232,8 @@ static void read_conn(struct mg_connection *c, struct pkt *pkt) {
     mg_error(c, "oom");
     return;  // drop it
   }
+    if (pkt->tcp->flags & TH_ACK)
+      handle_ack(s, mg_ntohl(pkt->tcp->ack), mg_ntohs(pkt->tcp->win));
   // Copy TCP payload into the IO buffer. If the connection is plain text,
   // we copy to c->recv. If the connection is TLS, this data is encrypted,
   // therefore we copy that encrypted data to the c->rtls iobuffer instead,
@@ -8336,6 +8352,7 @@ static void rx_tcp(struct mg_tcpip_if *ifp, struct pkt *pkt) {
     if (!handle_opt(s, pkt->tcp, pkt->ip6 != NULL))
       return;  // process options (MSS)
     s->seq = mg_ntohl(pkt->tcp->ack), s->ack = mg_ntohl(pkt->tcp->seq) + 1;
+    s->win = mg_ntohs(pkt->tcp->win), s->maxseq = (uint32_t)(s->seq + s->win);
     tx_tcp_ctrlresp(ifp, pkt, TH_ACK, pkt->tcp->ack);
     c->is_connecting = 0;  // Client connected
     settmout(c, MIP_TTYPE_KEEPALIVE);
diff --git a/src/net_builtin.c b/src/net_builtin.c
index bea2bc7272..1598d247c4 100644
--- a/src/net_builtin.c
+++ b/src/net_builtin.c
@@ -23,6 +23,8 @@ struct connstate {
   uint64_t timer;                         // TCP timer (see 'ttype' below)
   uint32_t acked;                         // Last ACK-ed number
   size_t unacked;                         // Not acked bytes
+  uint32_t maxseq;                        // Max send seq (ack + window)
+  uint16_t win;                           // destination current window size
   uint16_t dmss;                          // destination MSS (from TCP opts)
   uint8_t mac[sizeof(struct mg_l2addr)];  // Peer hw address
   uint8_t ttype;                          // Timer type:
@@ -1268,6 +1270,7 @@ static struct mg_connection *accept_conn(struct mg_connection *lsn,
   s = (struct connstate *) (c + 1);
   s->dmss = mss;  // from options in client SYN
   s->seq = mg_ntohl(pkt->tcp->ack), s->ack = mg_ntohl(pkt->tcp->seq);
+  s->win = mg_ntohs(pkt->tcp->win), s->maxseq = (uint32_t)(s->seq + s->win);
 #if MG_ENABLE_IPV6
   if (lsn->loc.is_ip6) {
     c->rem.addr.ip6[0] = pkt->ip6->src[0],
@@ -1360,10 +1363,13 @@ long mg_io_send(struct mg_connection *c, const void *buf, size_t len) {
   len = trim_len(c, len);
   if (c->is_udp) {
     if (!udp_send(c, buf, len)) return MG_IO_WAIT;
-  } else {  // TCP, cap to peer's MSS
+  } else {  // TCP, cap to peer's MSS and check window
     struct mg_tcpip_if *ifp = c->mgr->ifp;
     size_t sent;
+    uint32_t room = s->maxseq - s->seq;
+    if (room == 0) return MG_IO_WAIT;
     if (len > s->dmss) len = s->dmss;  // RFC-6691: reduce if sending opts
+    if ((uint32_t) len > room) len = room;
     sent = tx_tcp(ifp, s->mac, &c->loc, &c->rem, TH_PUSH | TH_ACK,
                   mg_htonl(s->seq), mg_htonl(s->ack), buf, len);
     if (sent == 0) {
@@ -1398,6 +1404,12 @@ static void handle_tls_recv(struct mg_connection *c) {
   }
 }
 
+static void handle_ack(struct connstate *s, uint32_t ackno, uint16_t win) {
+  if (ackno < (s->seq - s->win) || ackno > s->seq) return;
+  s->maxseq = (uint32_t)(ackno + win);
+  s->win = win;
+}
+
 static void read_conn(struct mg_connection *c, struct pkt *pkt) {
   struct connstate *s = (struct connstate *) (c + 1);
   struct mg_iobuf *io = c->is_tls ? &c->rtls : &c->recv;
@@ -1437,6 +1449,8 @@ static void read_conn(struct mg_connection *c, struct pkt *pkt) {
            mg_htonl(s->ack), NULL, 0);
     return;                        // no data to process
   } else if (pkt->pay.len == 0) {  // this is an ACK
+    if (pkt->tcp->flags & TH_ACK)
+      handle_ack(s, mg_ntohl(pkt->tcp->ack), mg_ntohs(pkt->tcp->win));
     if (s->fin_rcvd && s->ttype == MIP_TTYPE_FIN) s->twclosure = true;
     return;  // no data to process
   } else if (seq != s->ack) {
@@ -1454,6 +1468,8 @@ static void read_conn(struct mg_connection *c, struct pkt *pkt) {
     mg_error(c, "oom");
     return;  // drop it
   }
+    if (pkt->tcp->flags & TH_ACK)
+      handle_ack(s, mg_ntohl(pkt->tcp->ack), mg_ntohs(pkt->tcp->win));
   // Copy TCP payload into the IO buffer. If the connection is plain text,
   // we copy to c->recv. If the connection is TLS, this data is encrypted,
   // therefore we copy that encrypted data to the c->rtls iobuffer instead,
@@ -1572,6 +1588,7 @@ static void rx_tcp(struct mg_tcpip_if *ifp, struct pkt *pkt) {
     if (!handle_opt(s, pkt->tcp, pkt->ip6 != NULL))
       return;  // process options (MSS)
     s->seq = mg_ntohl(pkt->tcp->ack), s->ack = mg_ntohl(pkt->tcp->seq) + 1;
+    s->win = mg_ntohs(pkt->tcp->win), s->maxseq = (uint32_t)(s->seq + s->win);
     tx_tcp_ctrlresp(ifp, pkt, TH_ACK, pkt->tcp->ack);
     c->is_connecting = 0;  // Client connected
     settmout(c, MIP_TTYPE_KEEPALIVE);
diff --git a/test/mip_test.c b/test/mip_test.c
index ca63981bec..e7fc338e20 100644
--- a/test/mip_test.c
+++ b/test/mip_test.c
@@ -5,6 +5,8 @@
 
 #include "driver_mock.c"
 
+#define TCP_TEST_WIN 3000 // arbitrary peer window size to test txwindow
+
 static int s_num_tests = 0;
 static bool s_error = false;
 static int s_sent_fragment = 0;
@@ -116,6 +118,17 @@ static void tcpclosure_fn(struct mg_connection *c, int ev, void *ev_data) {
   (void) c, (void) ev_data;
 }
 
+static void txwindow_fn(struct mg_connection *c, int ev, void *ev_data) {
+  if (ev == MG_EV_ACCEPT) {
+    char bigdata[2 * TCP_TEST_WIN + 256];
+    *(int *) c->fn_data = 0;
+    mg_send(c, bigdata, sizeof(bigdata));
+  } else if (ev == MG_EV_WRITE) {
+     ++(*(int *) c->fn_data);
+  }
+  (void) c, (void) ev_data;
+}
+
 static void client_fn(struct mg_connection *c, int ev, void *ev_data) {
   if (ev == MG_EV_ERROR || ev == MG_EV_CONNECT) (*(int *) c->fn_data) = ev;
   (void) c, (void) ev_data;
@@ -139,7 +152,7 @@ static void frag_send_fn(struct mg_connection *c, int ev, void *ev_data) {
   if (ev == MG_EV_POLL) {
     if (!s_sent) {
       struct connstate *s = (struct connstate *) (c + 1);
-      s->dmss = 1500;      // mock set some destination MSS way larger
+      s->dmss = 1500, s->maxseq = 1500; // mock set dest MSS and win way larger
       c->send.len = 1200;  // setting TCP payload size
       s_sent = true;
     }
@@ -214,6 +227,7 @@ static void create_tcp_seg(struct eth *e, struct ipp *ipp, uint32_t seq,
   t.ack = mg_htonl(ack);
   t.sport = mg_htons(sport);
   t.dport = mg_htons(dport);
+  t.win = mg_htons(TCP_TEST_WIN);
   t.off = (uint8_t) ((sizeof(t) / 4) << 4) + (uint8_t) ((opts_len / 4) << 4);
   memcpy(s_driver_data.buf, e, sizeof(*e));
 #if MG_ENABLE_IPV6
@@ -898,6 +912,54 @@ static void test_tcp_retransmit(void) {
   mg_mgr_free(&mgr);
 }
 
+
+static void test_tcp_txwindow(void) {
+  struct mg_mgr mgr;
+  struct eth e;
+  struct ip ip;
+  struct ipp ipp;
+  struct tcp *t = (struct tcp *) (s_driver_data.buf + sizeof(e) + sizeof(ip));
+  int count = 0, stallcount;
+  uint32_t seq;
+  //bool response_recv = true;
+  struct mg_tcpip_driver driver;
+  struct mg_tcpip_if mif;
+
+  ipp.ip4 = &ip;
+  ipp.ip6 = NULL;
+
+  init_tcp_tests(&mgr, &e, &ipp, &driver, &mif, txwindow_fn);
+  mgr.conns->fn_data = &count;
+  init_tcp_handshake(&e, &ipp, &mgr);  // starts with seq_no=1000, ackno=2
+  ASSERT((t->seq == mg_htonl(2)));
+  do {
+    while (!received_response(&s_driver_data)) mg_mgr_poll(&mgr, 0);
+    seq = (uint32_t)(mg_htonl(t->seq) + s_driver_data.len - (size_t)((char *)((uint32_t *)t + (t->off >> 4)) - s_driver_data.buf));
+  } while (seq < (TCP_TEST_WIN + 2));
+  stallcount = count;
+  mg_mgr_poll(&mgr, 0), s_driver_data.len = 0;
+  mg_mgr_poll(&mgr, 0), s_driver_data.len = 0;
+  ASSERT((stallcount == count));
+  s_driver_data.tx_ready = false;
+  create_tcp_simpleseg(&e, &ipp, 1001, seq - TCP_TEST_WIN/2, TH_ACK, 0); // send ACK for half window
+  while (!received_response(&s_driver_data)) mg_mgr_poll(&mgr, 0);
+  ASSERT((stallcount < count));
+  do {
+    while (!received_response(&s_driver_data)) mg_mgr_poll(&mgr, 0);
+    seq = (uint32_t)(mg_htonl(t->seq) + s_driver_data.len - (size_t)((char *)((uint32_t *)t + (t->off >> 4)) - s_driver_data.buf));
+  } while (seq < (TCP_TEST_WIN + TCP_TEST_WIN/2 + 2));
+  stallcount = count;
+  mg_mgr_poll(&mgr, 0), s_driver_data.len = 0;
+  mg_mgr_poll(&mgr, 0), s_driver_data.len = 0;
+  ASSERT((stallcount == count));
+  s_driver_data.tx_ready = false;
+  create_tcp_simpleseg(&e, &ipp, 1001, seq, TH_ACK, 0); // send ACK
+  while (!received_response(&s_driver_data)) mg_mgr_poll(&mgr, 0);
+  ASSERT((stallcount < count));
+  s_driver_data.len = 0;
+  mg_mgr_free(&mgr);
+}
+
 static void test_frag_recv_path(void) {
   struct mg_mgr mgr;
   struct eth e;
@@ -1058,6 +1120,7 @@ static void test_tcp(bool ipv6) {
   if (!ipv6) {
     test_tcp_backlog();
     test_tcp_retransmit();
+    test_tcp_txwindow();
   }
 }