diff --git a/components/libc/posix/io/epoll/epoll.c b/components/libc/posix/io/epoll/epoll.c
index 1b63f79929..f5d4cbd938 100644
--- a/components/libc/posix/io/epoll/epoll.c
+++ b/components/libc/posix/io/epoll/epoll.c
@@ -164,7 +164,7 @@ static int epoll_poll(struct dfs_file *file, struct rt_pollreq *req)
                 fdlist = fdlist->next;
                 mask = epoll_get_event(fdlist, &fdlist->req);
 
-                if (mask & fdlist->epev.events)
+                if (mask & fdlist->revents)
                 {
                     events |= mask | POLLIN | EPOLLRDNORM;
                     break;
@@ -221,7 +221,7 @@ static int epoll_rdlist_add(struct rt_fd_list *fdl, rt_uint32_t revents)
         if (rdlist != RT_NULL)
         {
             rdlist->rdl_event = fdl;
-            rdlist->rdl_event->epev.events = fdl->epev.events & revents;
+            rdlist->rdl_event->epev.events = fdl->revents & revents;
             rdlist->next = ep->rdlist->next;
             rdlist->exclusive = 0;
             ep->rdlist->next = rdlist;
@@ -250,7 +250,7 @@ static int epoll_wqueue_callback(struct rt_wqueue_node *wait, void *key)
 
     fdlist = rt_container_of(wait, struct rt_fd_list, wqn);
 
-    if (fdlist->epev.events)
+    if (fdlist->revents)
     {
         epoll_rdlist_add(fdlist, (rt_ubase_t)key);
     }
@@ -279,10 +279,10 @@ static void epoll_ctl_install(struct rt_fd_list *fdlist, struct rt_eventpoll *ep
 {
     rt_uint32_t mask = 0;
 
-    fdlist->req._key = fdlist->epev.events;
+    fdlist->req._key = fdlist->revents;
 
     mask = epoll_get_event(fdlist, &fdlist->req);
-    if (mask & fdlist->epev.events)
+    if (mask & fdlist->revents)
     {
         epoll_rdlist_add(fdlist, mask);
     }
@@ -317,6 +317,10 @@ static int epoll_epf_init(int fd)
 
             rt_mutex_init(&ep->lock, EPOLL_MUTEX_NAME, RT_IPC_FLAG_FIFO);
 
+            #ifdef RT_USING_DFS_V2
+            df->fops = &epoll_fops;
+            #endif
+
             df->vnode = (struct dfs_vnode *)rt_malloc(sizeof(struct dfs_vnode));
             if (df->vnode)
             {
@@ -353,13 +357,13 @@ static int epoll_epf_init(int fd)
 
 static int epoll_do_create(int size)
 {
-    rt_err_t ret = 0;
+    rt_err_t ret = -1;
     int status;
     int fd;
 
     if (size < 0)
     {
-        ret = -EINVAL;
+        rt_set_errno(EINVAL);
     }
     else
     {
@@ -371,12 +375,12 @@ static int epoll_do_create(int size)
             if (status < 0)
             {
                 fd_release(fd);
-                ret = status;
+                rt_set_errno(-status);
             }
         }
         else
         {
-            ret = fd;
+            rt_set_errno(-fd);
         }
     }
 
@@ -518,23 +522,31 @@ static int epoll_do_ctl(int epfd, int op, int fd, struct epoll_event *event)
 {
     struct dfs_file *epdf;
     struct rt_eventpoll *ep;
-    rt_err_t ret = -EINVAL;
+    rt_err_t ret = 0;
 
     if (op & ~EFD_SHARED_EPOLL_TYPE)
-        return -EINVAL;
+    {
+        rt_set_errno(EINVAL);
+        return -1;
+    }
 
     if ((epfd == fd) || (epfd < 0) || (fd < 0) || (event->data.fd != fd))
-        return -EINVAL;
+    {
+        rt_set_errno(EINVAL);
+        return -1;
+    }
 
     if (!(event->events & EPOLLEXCLUSIVE_BITS))
-        return -EINVAL;
+    {
+        rt_set_errno(EINVAL);
+        return -1;
+    }
 
     epdf = fd_get(epfd);
 
     if (epdf->vnode->data)
     {
         ep = epdf->vnode->data;
-        ret = 0;
 
         rt_mutex_take(&ep->lock, RT_WAITING_FOREVER);
 
@@ -550,12 +562,18 @@ static int epoll_do_ctl(int epfd, int op, int fd, struct epoll_event *event)
             ret = epoll_ctl_mod(epdf, event);
             break;
         default:
-            ret = -EINVAL;
+            rt_set_errno(EINVAL);
             break;
         }
-    }
 
-    rt_mutex_release(&ep->lock);
+        if (ret < 0)
+        {
+            rt_set_errno(-ret);
+            ret = -1;
+        }
+
+        rt_mutex_release(&ep->lock);
+    }
 
     return ret;
 }
@@ -613,13 +631,13 @@ static int epoll_get_event(struct rt_fd_list *fl, rt_pollreq_t *req)
         {
             if (df->vnode->fops->poll)
             {
-                req->_key = fl->epev.events | POLLERR | POLLHUP;
+                req->_key = fl->revents | POLLERR | POLLHUP;
                 mask = df->vnode->fops->poll(df, req);
                 if (mask < 0)
                     return mask;
             }
 
-            mask &= fl->epev.events | EPOLLOUT | POLLERR;
+            mask &= fl->revents | EPOLLOUT | POLLERR;
         }
     }
 
@@ -659,7 +677,8 @@ static int epoll_do(struct rt_eventpoll *ep, struct epoll_event *events, int max
                         if (rdlist->rdl_event->revents & EPOLLET)
                         {
                             rt_wqueue_remove(&rdlist->rdl_event->wqn);
-                            epoll_get_event(rdlist->rdl_event, &rdlist->rdl_event->req);
+                            mask = epoll_get_event(rdlist->rdl_event, &rdlist->rdl_event->req);
+                            rdlist->rdl_event->epev.events = mask & rdlist->rdl_event->revents;
                             isfree = 1;
                         }
                         else
@@ -729,7 +748,7 @@ static int epoll_do_wait(int epfd, struct epoll_event *events, int maxevents, in
     struct rt_eventpoll *ep;
     struct dfs_file *df;
     lwp_sigset_t old_sig, new_sig;
-    rt_err_t ret = -EINVAL;
+    rt_err_t ret = 0;
 
     if (ss)
     {
@@ -755,6 +774,12 @@ static int epoll_do_wait(int epfd, struct epoll_event *events, int maxevents, in
         lwp_thread_signal_mask(rt_thread_self(), LWP_SIG_MASK_CMD_SET_MASK, &old_sig, RT_NULL);
     }
 
+    if (ret < 0)
+    {
+        rt_set_errno(-ret);
+        ret = -1;
+    }
+
     return ret;
 }
 
diff --git a/components/libc/posix/io/eventfd/eventfd.c b/components/libc/posix/io/eventfd/eventfd.c
index 1d6c79b809..b4dddc00f8 100644
--- a/components/libc/posix/io/eventfd/eventfd.c
+++ b/components/libc/posix/io/eventfd/eventfd.c
@@ -40,13 +40,13 @@ struct eventfd_ctx
 #ifndef RT_USING_DFS_V2
 static int eventfd_close(struct dfs_file *file);
 static int eventfd_poll(struct dfs_file *file, struct rt_pollreq *req);
-static int eventfd_read(struct dfs_file *file, void *buf, size_t count);
-static int eventfd_write(struct dfs_file *file, const void *buf, size_t count);
+static ssize_t eventfd_read(struct dfs_file *file, void *buf, size_t count);
+static ssize_t eventfd_write(struct dfs_file *file, const void *buf, size_t count);
 #else
 static int eventfd_close(struct dfs_file *file);
 static int eventfd_poll(struct dfs_file *file, struct rt_pollreq *req);
-static int eventfd_read(struct dfs_file *file, void *buf, size_t count, off_t *pos);
-static int eventfd_write(struct dfs_file *file, const void *buf, size_t count, off_t *pos);
+static ssize_t eventfd_read(struct dfs_file *file, void *buf, size_t count, off_t *pos);
+static ssize_t eventfd_write(struct dfs_file *file, const void *buf, size_t count, off_t *pos);
 #endif
 
 static const struct dfs_file_ops eventfd_fops =
@@ -76,7 +76,6 @@ static int eventfd_poll(struct dfs_file *file, struct rt_pollreq *req)
     count = ctx->count;
 
     rt_poll_add(&ctx->reader_queue, req);
-    rt_poll_add(&ctx->writer_queue, req);
 
     if (count > 0)
         events |= POLLIN;
@@ -91,9 +90,9 @@ static int eventfd_poll(struct dfs_file *file, struct rt_pollreq *req)
 }
 
 #ifndef RT_USING_DFS_V2
-static int eventfd_read(struct dfs_file *file, void *buf, size_t count)
+static ssize_t eventfd_read(struct dfs_file *file, void *buf, size_t count)
 #else
-static int eventfd_read(struct dfs_file *file, void *buf, size_t count, off_t *pos)
+static ssize_t eventfd_read(struct dfs_file *file, void *buf, size_t count, off_t *pos)
 #endif
 {
     struct eventfd_ctx *ctx = (struct eventfd_ctx *)file->vnode->data;
@@ -107,7 +106,7 @@ static int eventfd_read(struct dfs_file *file, void *buf, size_t count, off_t *p
 
     rt_mutex_take(&ctx->lock, RT_WAITING_FOREVER);
 
-    if (ctx->count == 0)
+    if (ctx->count <= 0)
     {
         if (file->flags & O_NONBLOCK)
         {
@@ -135,7 +134,6 @@ static int eventfd_read(struct dfs_file *file, void *buf, size_t count, off_t *p
     }
 
     ctx->count -= counter_num;
-
     (*buffer) = counter_num;
 
     rt_mutex_release(&ctx->lock);
@@ -144,9 +142,9 @@ static int eventfd_read(struct dfs_file *file, void *buf, size_t count, off_t *p
 }
 
 #ifndef RT_USING_DFS_V2
-static int eventfd_write(struct dfs_file *file, const void *buf, size_t count)
+static ssize_t eventfd_write(struct dfs_file *file, const void *buf, size_t count)
 #else
-static int eventfd_write(struct dfs_file *file, const void *buf, size_t count, off_t *pos)
+static ssize_t eventfd_write(struct dfs_file *file, const void *buf, size_t count, off_t *pos)
 #endif
 {
     struct eventfd_ctx *ctx = (struct eventfd_ctx *)file->vnode->data;
@@ -182,7 +180,6 @@ static int eventfd_write(struct dfs_file *file, const void *buf, size_t count, o
             /* Release the mutex to avoid a deadlock */
             rt_mutex_release(&ctx->lock);
             rt_wqueue_wait(&ctx->writer_queue, 0, RT_WAITING_FOREVER);
-            rt_wqueue_wakeup(&ctx->reader_queue, (void *)POLLIN);
             rt_mutex_take(&ctx->lock, RT_WAITING_FOREVER);
         }
     }
@@ -232,6 +229,11 @@ static int rt_eventfd_create(struct dfs_file *df, unsigned int count, int flags)
             rt_free(ctx);
             ret = -ENOMEM;
         }
+
+        #ifdef RT_USING_DFS_V2
+        df->fops = &eventfd_fops;
+        #endif
+
     }
 
     return ret;
@@ -245,7 +247,10 @@ static int do_eventfd(unsigned int count, int flags)
     rt_ssize_t ret = 0;
 
     if (flags & ~EFD_FLAGS_SET)
-        return -RT_EINVAL;
+    {
+        rt_set_errno(EINVAL);
+        return -1;
+    }
 
     fd = fd_new();
     if (fd >= 0)
@@ -257,12 +262,14 @@ static int do_eventfd(unsigned int count, int flags)
         if (status < 0)
         {
             fd_release(fd);
-            ret = status;
+            rt_set_errno(-status);
+            ret = -1;
         }
     }
     else
     {
-        ret = fd;
+        rt_set_errno(-fd);
+        ret = -1;
     }
 
     return ret;