Re: [PATCH 2/2] platform/x86: Add Intel Software Defined Silicon driver

On Mon, Sep 27, 2021 at 10:53:09AM -0700, David E. Box wrote:

On Mon, 2021-09-27 at 06:03 +0200, Greg KH wrote:

quoted

On Sun, Sep 26, 2021 at 06:15:16PM -0700, David E. Box wrote:

quoted

quoted

quoted

+static int sdsi_remove(struct platform_device *pdev)
+{
+       struct sdsi_priv *priv = platform_get_drvdata(pdev);
+
+       priv->dev_present = false;
+       sysfs_remove_bin_file(&priv->pdev->dev.kobj, &priv->registers_bin_attr);
+       misc_deregister(&priv->miscdev);
+       kref_put(&priv->kref, sdsi_priv_release);

Why do you need a kref for a structure that already can be controlled by
a different lifetime rule?

Which rule am I missing? This kref allows the structure to remain in case the device is removed
while the file is open.

This device is on a hardware bus that allows removal?

Well the device can be unbound. A test case covers this.

Great, where are these tests?  Why not add them to the kernel tree
itself in the proper location?

And in the real-world, who would ever unbind this?

quoted

Anyway, you now are dealing with lifetime rules of 3 structures all at
once, and the interactions between them is not very obvious.  It would
probably be simpler just to stick with 2, right?  You really only care
about the misc structure here.

In the case that the device is unbound, both the pdev and miscdev go away. Something has to outlive
them in order to handle any open files still trying to use the ioctl.

I do not think that the miscdev goes away if the file handle is still
open, right?

thanks,

greg k-h