| Message ID | 20240221-iio-use-cleanup-magic-v1-1-f9c292666f26@analog.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | iio: move IIO to the cleanup.h magic | expand |
On Wed, 21 Feb 2024 14:26:52 +0100 Nuno Sa via B4 Relay <devnull+nuno.sa.analog.com@kernel.org> wrote: > From: Nuno Sa <nuno.sa@analog.com> > > Use the new cleanup magic for handling mutexes in IIO. This allows us to > greatly simplify some code paths. > > Note that we keep the plain mutex calls in the > iio_device_release|acquire() APIs since in there the macros would likely > not help much (as we want to keep the lock acquired when he leave the > APIs). > > Signed-off-by: Nuno Sa <nuno.sa@analog.com> > --- > drivers/iio/industrialio-core.c | 52 +++++++++++++++-------------------------- > 1 file changed, 19 insertions(+), 33 deletions(-) > > diff --git a/drivers/iio/industrialio-core.c b/drivers/iio/industrialio-core.c > index 9b2877fe8689..7e6497828364 100644 > --- a/drivers/iio/industrialio-core.c > +++ b/drivers/iio/industrialio-core.c > @@ -11,6 +11,7 @@ > @@ -1806,31 +1803,22 @@ static long iio_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) > struct iio_dev *indio_dev = ib->indio_dev; > struct iio_dev_opaque *iio_dev_opaque = to_iio_dev_opaque(indio_dev); > struct iio_ioctl_handler *h; > - int ret = -ENODEV; > - > - mutex_lock(&iio_dev_opaque->info_exist_lock); > > + guard(mutex)(&iio_dev_opaque->info_exist_lock); > /* > * The NULL check here is required to prevent crashing when a device > * is being removed while userspace would still have open file handles > * to try to access this device. > */ > if (!indio_dev->info) > - goto out_unlock; > + return -ENODEV; > > list_for_each_entry(h, &iio_dev_opaque->ioctl_handlers, entry) { > - ret = h->ioctl(indio_dev, filp, cmd, arg); > - if (ret != IIO_IOCTL_UNHANDLED) > - break; > + if (h->ioctl(indio_dev, filp, cmd, arg) != IIO_IOCTL_UNHANDLED) > + return 0; Changes the return value if ret returns something other than IIO_IOCTL_UNHANDLED which it can I think... > } > > - if (ret == IIO_IOCTL_UNHANDLED) > - ret = -ENODEV; > - > -out_unlock: > - mutex_unlock(&iio_dev_opaque->info_exist_lock); > - > - return ret; > + return -ENODEV; > }
On Thu, 2024-02-22 at 19:22 +0000, Jonathan Cameron wrote: > On Wed, 21 Feb 2024 14:26:52 +0100 > Nuno Sa via B4 Relay <devnull+nuno.sa.analog.com@kernel.org> wrote: > > > From: Nuno Sa <nuno.sa@analog.com> > > > > Use the new cleanup magic for handling mutexes in IIO. This allows us to > > greatly simplify some code paths. > > > > Note that we keep the plain mutex calls in the > > iio_device_release|acquire() APIs since in there the macros would likely > > not help much (as we want to keep the lock acquired when he leave the > > APIs). > > > > Signed-off-by: Nuno Sa <nuno.sa@analog.com> > > --- > > drivers/iio/industrialio-core.c | 52 +++++++++++++++----------------------- > > --- > > 1 file changed, 19 insertions(+), 33 deletions(-) > > > > diff --git a/drivers/iio/industrialio-core.c b/drivers/iio/industrialio- > > core.c > > index 9b2877fe8689..7e6497828364 100644 > > --- a/drivers/iio/industrialio-core.c > > +++ b/drivers/iio/industrialio-core.c > > @@ -11,6 +11,7 @@ > > > > > @@ -1806,31 +1803,22 @@ static long iio_ioctl(struct file *filp, unsigned > > int cmd, unsigned long arg) > > struct iio_dev *indio_dev = ib->indio_dev; > > struct iio_dev_opaque *iio_dev_opaque = > > to_iio_dev_opaque(indio_dev); > > struct iio_ioctl_handler *h; > > - int ret = -ENODEV; > > - > > - mutex_lock(&iio_dev_opaque->info_exist_lock); > > > > + guard(mutex)(&iio_dev_opaque->info_exist_lock); > > /* > > * The NULL check here is required to prevent crashing when a > > device > > * is being removed while userspace would still have open file > > handles > > * to try to access this device. > > */ > > if (!indio_dev->info) > > - goto out_unlock; > > + return -ENODEV; > > > > list_for_each_entry(h, &iio_dev_opaque->ioctl_handlers, entry) { > > - ret = h->ioctl(indio_dev, filp, cmd, arg); > > - if (ret != IIO_IOCTL_UNHANDLED) > > - break; > > + if (h->ioctl(indio_dev, filp, cmd, arg) != > > IIO_IOCTL_UNHANDLED) > > + return 0; > > Changes the return value if ret returns something other than > IIO_IOCTL_UNHANDLED > which it can I think... Yeah, I think for the multi buffer support we can actually return the anon fd... Pffff, one of those changes I made right before sending out the series thinking I could remove one LOC without properly looking at the code :/ Will send a v2. - Nuno Sá
diff --git a/drivers/iio/industrialio-core.c b/drivers/iio/industrialio-core.c index 9b2877fe8689..7e6497828364 100644 --- a/drivers/iio/industrialio-core.c +++ b/drivers/iio/industrialio-core.c @@ -11,6 +11,7 @@ #include <linux/anon_inodes.h> #include <linux/cdev.h> +#include <linux/cleanup.h> #include <linux/debugfs.h> #include <linux/device.h> #include <linux/err.h> @@ -268,20 +269,16 @@ EXPORT_SYMBOL(iio_read_const_attr); */ int iio_device_set_clock(struct iio_dev *indio_dev, clockid_t clock_id) { - int ret; struct iio_dev_opaque *iio_dev_opaque = to_iio_dev_opaque(indio_dev); const struct iio_event_interface *ev_int = iio_dev_opaque->event_interface; - ret = mutex_lock_interruptible(&iio_dev_opaque->mlock); - if (ret) - return ret; - if ((ev_int && iio_event_enabled(ev_int)) || - iio_buffer_enabled(indio_dev)) { - mutex_unlock(&iio_dev_opaque->mlock); - return -EBUSY; + scoped_cond_guard(mutex_intr, return -EINTR, &iio_dev_opaque->mlock) { + if ((ev_int && iio_event_enabled(ev_int)) || + iio_buffer_enabled(indio_dev)) + return -EBUSY; + + iio_dev_opaque->clock_id = clock_id; } - iio_dev_opaque->clock_id = clock_id; - mutex_unlock(&iio_dev_opaque->mlock); return 0; } @@ -1806,31 +1803,22 @@ static long iio_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) struct iio_dev *indio_dev = ib->indio_dev; struct iio_dev_opaque *iio_dev_opaque = to_iio_dev_opaque(indio_dev); struct iio_ioctl_handler *h; - int ret = -ENODEV; - - mutex_lock(&iio_dev_opaque->info_exist_lock); + guard(mutex)(&iio_dev_opaque->info_exist_lock); /* * The NULL check here is required to prevent crashing when a device * is being removed while userspace would still have open file handles * to try to access this device. */ if (!indio_dev->info) - goto out_unlock; + return -ENODEV; list_for_each_entry(h, &iio_dev_opaque->ioctl_handlers, entry) { - ret = h->ioctl(indio_dev, filp, cmd, arg); - if (ret != IIO_IOCTL_UNHANDLED) - break; + if (h->ioctl(indio_dev, filp, cmd, arg) != IIO_IOCTL_UNHANDLED) + return 0; } - if (ret == IIO_IOCTL_UNHANDLED) - ret = -ENODEV; - -out_unlock: - mutex_unlock(&iio_dev_opaque->info_exist_lock); - - return ret; + return -ENODEV; } static const struct file_operations iio_buffer_fileops = { @@ -2058,18 +2046,16 @@ void iio_device_unregister(struct iio_dev *indio_dev) cdev_device_del(&iio_dev_opaque->chrdev, &indio_dev->dev); - mutex_lock(&iio_dev_opaque->info_exist_lock); + scoped_guard(mutex, &iio_dev_opaque->info_exist_lock) { + iio_device_unregister_debugfs(indio_dev); - iio_device_unregister_debugfs(indio_dev); + iio_disable_all_buffers(indio_dev); - iio_disable_all_buffers(indio_dev); + indio_dev->info = NULL; - indio_dev->info = NULL; - - iio_device_wakeup_eventset(indio_dev); - iio_buffer_wakeup_poll(indio_dev); - - mutex_unlock(&iio_dev_opaque->info_exist_lock); + iio_device_wakeup_eventset(indio_dev); + iio_buffer_wakeup_poll(indio_dev); + } iio_buffers_free_sysfs_and_mask(indio_dev); }