Oliver Neukum <oneukum@xxxxxxxx> wrote:
>
> Hi,
>
> On 16.09.24 14:44, Jeongjun Park wrote:
> > Oliver Neukum <oneukum@xxxxxxxx> wrote:
> >>
> >>
> >>
> >> On 16.09.24 06:15, Greg KH wrote:
> >>> On Mon, Sep 16, 2024 at 01:06:29PM +0900, Jeongjun Park wrote:
>
> >>> Please use the guard() form here, it makes the change much simpler and
> >>> easier to review and maintain.
> >>
> >> That would break the O_NONBLOCK case.
> >>
> >> Looking at the code it indeed looks like iowarrior_read() can race
> >> with itself. Strictly speaking it always could happen if a task used
> >> fork() after open(). The driver tries to restrict its usage to one
> >> thread, but I doubt that the logic is functional.
> >>
> >> It seems to me the correct fix is something like this:
> >
> > Well, I don't know why it's necessary to modify it like this.
> > I think it would be more appropriate to patch it to make it
> > more maintainable by using guard() as Greg suggested.
>
> Allow me to explain detail.
>
> guard() internally uses mutex_lock(). That means that
>
> a) it will block
> b) having blocked it will sleep in the state TASK_UNINTERRUPTIBLE
>
> The driver itself uses TASK_INTERRUPTIBLE in iowarrior_read(),
> when it waits for IO. That is entirely correct, as it waits for
> an external device doing an operation that may never occur. You
> must use TASK_INTERRUPTIBLE.
>
> Now, if you use mutex_lock() to wait for a task waiting for IO
> to occur in the state TASK_INTERRUPTIBLE, you are indirectlywaiting for
> an event that you must wait for in TASK_INTERRUPTIBLE in the state
> TASK_UNINTERRUPTIBLE.
> That is a bug. You have created a task that cannot be killed (uid may not match),
> but may have to be killed. Furthermore you block even in case the
> device has been opened with O_NONBLOCK, which is a second bug.
>
> These limitations are inherent in guard(). Therefore you cannot use
> guard here.
Okay. But O_NONBLOCK flag check already exists, and I don't know
if we need to branch separately to mutex_trylock just because O_NONBLOCK
flag exists. I think mutex_lock_interruptible is enough.
And the point of locking is too late. I think it would be more appropriate to
read file->private_data and then lock it right away.
I think this patch is a more appropriate patch:
---
drivers/usb/misc/iowarrior.c | 41 +++++++++++++++++++++++++++---------
1 file changed, 31 insertions(+), 10 deletions(-)
diff --git a/drivers/usb/misc/iowarrior.c b/drivers/usb/misc/iowarrior.c
index 6d28467ce352..6fb4ecebbc15 100644
--- a/drivers/usb/misc/iowarrior.c
+++ b/drivers/usb/misc/iowarrior.c
@@ -277,28 +277,40 @@ static ssize_t iowarrior_read(struct file *file,
char __user *buffer,
struct iowarrior *dev;
int read_idx;
int offset;
+ int retval = 0;
dev = file->private_data;
+ if (mutex_lock_interruptible(&dev->mutex)) {
+ retval = -EAGAIN;
+ goto exit;
+ }
+
/* verify that the device wasn't unplugged */
- if (!dev || !dev->present)
- return -ENODEV;
+ if (!dev->present) {
+ retval = -ENODEV;
+ goto unlock_exit;
+ }
dev_dbg(&dev->interface->dev, "minor %d, count = %zd\n",
dev->minor, count);
/* read count must be packet size (+ time stamp) */
if ((count != dev->report_size)
- && (count != (dev->report_size + 1)))
- return -EINVAL;
+ && (count != (dev->report_size + 1))) {
+ retval = -EINVAL;
+ goto unlock_exit;
+ }
/* repeat until no buffer overrun in callback handler occur */
do {
atomic_set(&dev->overflow_flag, 0);
if ((read_idx = read_index(dev)) == -1) {
/* queue empty */
- if (file->f_flags & O_NONBLOCK)
- return -EAGAIN;
+ if (file->f_flags & O_NONBLOCK) {
+ retval = -EAGAIN;
+ goto unlock_exit;
+ }
else {
//next line will return when there is either new data,
or the device is unplugged
int r = wait_event_interruptible(dev->read_wait,
@@ -309,28 +321,37 @@ static ssize_t iowarrior_read(struct file *file,
char __user *buffer,
-1));
if (r) {
//we were interrupted by a signal
- return -ERESTART;
+ retval = -ERESTART;
+ goto unlock_exit;
}
if (!dev->present) {
//The device was unplugged
- return -ENODEV;
+ retval = -ENODEV;
+ goto unlock_exit;
}
if (read_idx == -1) {
// Can this happen ???
- return 0;
+ goto unlock_exit;
}
}
}
offset = read_idx * (dev->report_size + 1);
if (copy_to_user(buffer, dev->read_queue + offset, count)) {
- return -EFAULT;
+ retval = -EFAULT;
+ goto unlock_exit;
}
} while (atomic_read(&dev->overflow_flag));
read_idx = ++read_idx == MAX_INTERRUPT_BUFFER ? 0 : read_idx;
atomic_set(&dev->read_idx, read_idx);
+ mutex_unlock(&dev->mutex);
return count;
+
+unlock_exit:
+ mutex_unlock(&dev->mutex);
+exit:
+ return retval;
}
/*
--
>
> Regards
> Oliver
