eventfd: remove fput() call from possible IRQ context

[profile/ivi/kernel-x86-ivi.git] / fs / aio.c

diff --git a/fs/aio.c b/fs/aio.c
index d6f89d3..4a9d4d6 100644 (file)
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -443,7 +443,7 @@ static struct kiocb *__aio_get_req(struct kioctx *ctx)
        req->private = NULL;
        req->ki_iovec = NULL;
        INIT_LIST_HEAD(&req->ki_run_list);
-       req->ki_eventfd = ERR_PTR(-EINVAL);
+       req->ki_eventfd = NULL;
 
        /* Check if the completion queue has enough free space to
         * accept an event from this io.
@@ -485,8 +485,6 @@ static inline void really_put_req(struct kioctx *ctx, struct kiocb *req)
 {
        assert_spin_locked(&ctx->ctx_lock);
 
-       if (!IS_ERR(req->ki_eventfd))
-               fput(req->ki_eventfd);
        if (req->ki_dtor)
                req->ki_dtor(req);
        if (req->ki_iovec != &req->ki_inline_vec)
@@ -508,8 +506,11 @@ static void aio_fput_routine(struct work_struct *data)
                list_del(&req->ki_list);
                spin_unlock_irq(&fput_lock);
 
-               /* Complete the fput */
-               __fput(req->ki_filp);
+               /* Complete the fput(s) */
+               if (req->ki_filp != NULL)
+                       __fput(req->ki_filp);
+               if (req->ki_eventfd != NULL)
+                       __fput(req->ki_eventfd);
 
                /* Link the iocb into the context's free list */
                spin_lock_irq(&ctx->ctx_lock);
@@ -527,12 +528,14 @@ static void aio_fput_routine(struct work_struct *data)
  */
 static int __aio_put_req(struct kioctx *ctx, struct kiocb *req)
 {
+       int schedule_putreq = 0;
+
        dprintk(KERN_DEBUG "aio_put(%p): f_count=%ld\n",
                req, atomic_long_read(&req->ki_filp->f_count));
 
        assert_spin_locked(&ctx->ctx_lock);
 
-       req->ki_users --;
+       req->ki_users--;
        BUG_ON(req->ki_users < 0);
        if (likely(req->ki_users))
                return 0;
@@ -540,10 +543,23 @@ static int __aio_put_req(struct kioctx *ctx, struct kiocb *req)
        req->ki_cancel = NULL;
        req->ki_retry = NULL;
 
-       /* Must be done under the lock to serialise against cancellation.
-        * Call this aio_fput as it duplicates fput via the fput_work.
+       /*
+        * Try to optimize the aio and eventfd file* puts, by avoiding to
+        * schedule work in case it is not __fput() time. In normal cases,
+        * we would not be holding the last reference to the file*, so
+        * this function will be executed w/out any aio kthread wakeup.
         */
-       if (unlikely(atomic_long_dec_and_test(&req->ki_filp->f_count))) {
+       if (unlikely(atomic_long_dec_and_test(&req->ki_filp->f_count)))
+               schedule_putreq++;
+       else
+               req->ki_filp = NULL;
+       if (req->ki_eventfd != NULL) {
+               if (unlikely(atomic_long_dec_and_test(&req->ki_eventfd->f_count)))
+                       schedule_putreq++;
+               else
+                       req->ki_eventfd = NULL;
+       }
+       if (unlikely(schedule_putreq)) {
                get_ioctx(ctx);
                spin_lock(&fput_lock);
                list_add(&req->ki_list, &fput_head);
@@ -1009,7 +1025,7 @@ int aio_complete(struct kiocb *iocb, long res, long res2)
         * eventfd. The eventfd_signal() function is safe to be called
         * from IRQ context.
         */
-       if (!IS_ERR(iocb->ki_eventfd))
+       if (iocb->ki_eventfd != NULL)
                eventfd_signal(iocb->ki_eventfd, 1);
 
 put_rq:
@@ -1270,7 +1286,7 @@ static void io_destroy(struct kioctx *ioctx)
  *     pointer is passed for ctxp.  Will fail with -ENOSYS if not
  *     implemented.
  */
-asmlinkage long sys_io_setup(unsigned nr_events, aio_context_t __user *ctxp)
+SYSCALL_DEFINE2(io_setup, unsigned, nr_events, aio_context_t __user *, ctxp)
 {
        struct kioctx *ioctx = NULL;
        unsigned long ctx;
@@ -1308,7 +1324,7 @@ out:
  *     implemented.  May fail with -EFAULT if the context pointed to
  *     is invalid.
  */
-asmlinkage long sys_io_destroy(aio_context_t ctx)
+SYSCALL_DEFINE1(io_destroy, aio_context_t, ctx)
 {
        struct kioctx *ioctx = lookup_ioctx(ctx);
        if (likely(NULL != ioctx)) {
@@ -1608,6 +1624,7 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
                req->ki_eventfd = eventfd_fget((int) iocb->aio_resfd);
                if (IS_ERR(req->ki_eventfd)) {
                        ret = PTR_ERR(req->ki_eventfd);
+                       req->ki_eventfd = NULL;
                        goto out_put_req;
                }
        }
@@ -1662,8 +1679,8 @@ out_put_req:
  *     are available to queue any iocbs.  Will return 0 if nr is 0.  Will
  *     fail with -ENOSYS if not implemented.
  */
-asmlinkage long sys_io_submit(aio_context_t ctx_id, long nr,
-                             struct iocb __user * __user *iocbpp)
+SYSCALL_DEFINE3(io_submit, aio_context_t, ctx_id, long, nr,
+               struct iocb __user * __user *, iocbpp)
 {
        struct kioctx *ctx;
        long ret = 0;
@@ -1737,8 +1754,8 @@ static struct kiocb *lookup_kiocb(struct kioctx *ctx, struct iocb __user *iocb,
  *     invalid.  May fail with -EAGAIN if the iocb specified was not
  *     cancelled.  Will fail with -ENOSYS if not implemented.
  */
-asmlinkage long sys_io_cancel(aio_context_t ctx_id, struct iocb __user *iocb,
-                             struct io_event __user *result)
+SYSCALL_DEFINE3(io_cancel, aio_context_t, ctx_id, struct iocb __user *, iocb,
+               struct io_event __user *, result)
 {
        int (*cancel)(struct kiocb *iocb, struct io_event *res);
        struct kioctx *ctx;
@@ -1799,11 +1816,11 @@ asmlinkage long sys_io_cancel(aio_context_t ctx_id, struct iocb __user *iocb,
  *     will be updated if not NULL and the operation blocks.  Will fail
  *     with -ENOSYS if not implemented.
  */
-asmlinkage long sys_io_getevents(aio_context_t ctx_id,
-                                long min_nr,
-                                long nr,
-                                struct io_event __user *events,
-                                struct timespec __user *timeout)
+SYSCALL_DEFINE5(io_getevents, aio_context_t, ctx_id,
+               long, min_nr,
+               long, nr,
+               struct io_event __user *, events,
+               struct timespec __user *, timeout)
 {
        struct kioctx *ioctx = lookup_ioctx(ctx_id);
        long ret = -EINVAL;