Re: [PATCH] media: uvcvideo: Log driver load in uvc_probe function

[Laurent Pinchart <laurent.pinchart@ideasonboard.com>]

Hi Benjamin,

(CC'ing Shuah)

On Thu, Aug 07, 2025 at 09:53:41AM +0200, Benjamin Mugnier wrote:
> On 8/1/25 17:04, Laurent Pinchart wrote:
> > On Fri, Aug 01, 2025 at 03:53:26PM +0100, Imene Jaziri wrote:
> >> Add a pr_info() in the uvc_probe function to trace when the
> >> uvcvideo driver is loaded. This is for learning purposes.
> > 
> > What part of the learning instructions you are following instructed you
> > to submit this patch to kernel mailing lists ? We are regularly spammed
> > by similar patches, which indicates the instructions are not clear
> > enough.
> > 
> 
> I got curious too. It comes from the Linux Foundation training LFD103
> [1]. Chapter 8 describes this patch pretty much, and chapter 9 describes
> how to send the patch, but with a warning not to do so :
> 
>   [...]
>   At this time, you can run:
> 
>   git format-patch -1 <commit ID> --to=maintainer1 --to=maintainer2
> --cc=maillist1 --cc=maillist2
> 
>   This will generate a patch.
> 
>   Important Note:
>   Please note that this is just an example. Don’t send this patch upstream.
> 
>   You can revert this commit now.
> 
>   Please refer to the Select the recipients for your patch section in
> the Submitting patches: the essential guide to getting your code into
> the kernel document.
> 
>   When you have your own patch ready for submittal, you can follow this
> example process to generate the patch and send it upstream using the
> following command:
> 
>   git send-email <patch_file>
>   [...]

Imene provided me with that information in private.

> Looking at it I guess it's pretty easy to miss the note.

Yes, reading the text, I really can't blame the people training the
course for missing it. It's a bit like if a mushroom foraging book said

- go pick those red mushrooms with white dots on the cap in the forest
- cut them in pieces, pan-fry them with butter
- add a couple of eggs, salt and pepper to make an omelette
- the dish is ready ! enjoy your meal !
- you won't be eating that, it's poisonous. throw it away

:-)

> Maybe
> requesting to add '--dry-run' to the git send-email command could be a
> simple fix to prevent from copy/pasting ?

I've contacted Shuah, who told me she updated the training material.
Hopefully we'll soon stop getting such patches.

> [1] https://training.linuxfoundation.org/training/a-beginners-guide-to-linux-kernel-development-lfd103/

I wonder why the training materials are not public, given that the
training is free of charge. Did I miss something, or does the LF
encourage people to develop free software, but won't eat their own
dogfood and publish their materials under open licenses ?

> >> Signed-off-by: Imene Jaziri <imenjazirii18@gmail.com>
> >> ---
> >>  drivers/media/usb/uvc/uvc_driver.c | 3 +--
> >>  1 file changed, 1 insertion(+), 2 deletions(-)
> >>
> >> diff --git a/drivers/media/usb/uvc/uvc_driver.c b/drivers/media/usb/uvc/uvc_driver.c
> >> index da24a655ab68..4e5d1d636640 100644
> >> --- a/drivers/media/usb/uvc/uvc_driver.c
> >> +++ b/drivers/media/usb/uvc/uvc_driver.c
> >> @@ -2170,7 +2170,6 @@ static int uvc_probe(struct usb_interface *intf,
> >>  		(const struct uvc_device_info *)id->driver_info;
> >>  	int function;
> >>  	int ret;
> >> -
> >>  	/* Allocate memory for the device and initialize it. */
> >>  	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
> >>  	if (dev == NULL)
> >> @@ -2188,7 +2187,7 @@ static int uvc_probe(struct usb_interface *intf,
> >>  	dev->info = info ? info : &uvc_quirk_none;
> >>  	dev->quirks = uvc_quirks_param == -1
> >>  		    ? dev->info->quirks : uvc_quirks_param;
> >> -
> >> +	pr_info("I changed uvcvideo driver in the Linux Kernel\n");
> >>  	if (id->idVendor && id->idProduct)
> >>  		uvc_dbg(dev, PROBE, "Probing known UVC device %s (%04x:%04x)\n",
> >>  			udev->devpath, id->idVendor, id->idProduct);

-- 
Regards,

Laurent Pinchart