Techwizz/kernel-fxtec-pro1x
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

staging:iio:events: Use non-atmoic bitops

Browse source 
We always hold the waitqueue lock when modifying the flags field. So it is safe
to use the non-atomic bitops here instead of the atomic versions.

The lock has to be held, because we need to clear the busy flag and flush the
event FIFO in one atomic operation when closing the event file descriptor.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit is contained in:
Lars-Peter Clausen 2012-01-03 14:59:42 +01:00 committed by Greg Kroah-Hartman
parent e18045ed75
commit a046c1e866
1 changed files with 3 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
drivers/staging/iio/industrialio-event.c
View file

@ -124,7 +124,7 @@ static int iio_event_chrdev_release(struct inode *inode, struct file *filep)
struct iio_event_interface *ev_int = filep->private_data;
spin_lock(&ev_int->wait.lock);
clear_bit(IIO_BUSY_BIT_POS, &ev_int->flags);
__clear_bit(IIO_BUSY_BIT_POS, &ev_int->flags);
/*
* In order to maintain a clean state for reopening,
* clear out any awaiting events. The mask will prevent
@ -153,7 +153,7 @@ int iio_event_getfd(struct iio_dev *indio_dev)
return -ENODEV;
spin_lock(&ev_int->wait.lock);
if (test_and_set_bit(IIO_BUSY_BIT_POS, &ev_int->flags)) {
if (__test_and_set_bit(IIO_BUSY_BIT_POS, &ev_int->flags)) {
spin_unlock(&ev_int->wait.lock);
return -EBUSY;
}
@ -162,7 +162,7 @@ int iio_event_getfd(struct iio_dev *indio_dev)
&iio_event_chrdev_fileops, ev_int, O_RDONLY);
if (fd < 0) {
spin_lock(&ev_int->wait.lock);
clear_bit(IIO_BUSY_BIT_POS, &ev_int->flags);
__clear_bit(IIO_BUSY_BIT_POS, &ev_int->flags);
spin_unlock(&ev_int->wait.lock);
}
return fd;