On Wed, Jan 23, 2019 at 02:58:22PM +0200, Gilad Ben-Yossef wrote:
> Hi,
>
> On Tue, Jan 22, 2019 at 5:14 PM Greg Kroah-Hartman
> <[email protected]> wrote:
> >
> > When calling debugfs functions, there is no need to ever check the
> > return value. The function can work or not, but the code logic should
> > never do something different based on this.
>
>
>
> I get the part about not failing loading the driver just because some
> debugs entry isn't available, but wont it be weird if
> debugfs_create_dir() fails but debugfs_create_regset32() succeeds and
> we suddenly have weird files in the debugfs root dir?
> Not the end of the world of course but maybe it's better to avoid
> trying to create the files if the directory is not available?
See this patch to handle that theoretical issue:
https://lore.kernel.org/lkml/[email protected]/T/#me91cc3d16185be13d64f85c8477c543cbda9baf6
thanks,
greg k-h
