On Fri, Feb 26, 2021 at 01:59:14PM +0100, Maciej Kwapulinski wrote:

Greg Kroah-Hartman [off-list ref] writes:

quoted

On Tue, Feb 16, 2021 at 05:05:14PM +0100, Maciej Kwapulinski wrote:

....

quoted

quoted

--- /dev/null
+++ b/drivers/misc/gna/gna_driver.h

@@ -0,0 +1,41 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright(c) 2017-2021 Intel Corporation */
+
+#ifndef __GNA_DRIVER_H__
+#define __GNA_DRIVER_H__
+
+#include <linux/kernel.h>
+#include <linux/mutex.h>
+#include <linux/types.h>
+
+#define GNA_DRV_NAME	"gna"

Way too generic, no one knows what "gna" is.

"intel gna" is much more verbose in search engines.
As we do not (plan to) have more "gna" drivers, is the following ok?:

intel-gna

the change would imply the following:

prompt$ lspci -s 00:00.3 -vvvv
00:00.3 System peripheral: Intel Corporation Device 3190 (rev 03)
	Subsystem: Intel Corporation Device 2072
  ....
	Kernel driver in use: intel-gna
	Kernel modules: gna

is it ok?

Why not intel-gna as the kernel module as well?

also, how about the interface to library (it's part of one of next patches)?:
prompt$ file /dev/gna0
/dev/gna0: character special (235/0)

can "gna" stay intact here?

Again, I have no idea what "gna" is, so you might want to pick something
more descriptive?

I'm pointing this out, because gna exists on the market for a while and
changing the above may have some impact we'd like to avoid.

If it exists but Linux does not support it, how would anyone know about
it?  :)

Please use real terms where possible.

thanks,

greg k-h