linux-kernel - Re: [PATCH V2] tty: hvc: hvc_opal: eliminate uses of of_node

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <2024050323-rhyme-upchuck-4c4c@gregkh>
Date: Fri, 3 May 2024 15:58:13 +0200
From: Greg KH <gregkh@...uxfoundation.org>
To: Lu Dai <dai.lu@...rdes.com>
Cc: npiggin@...il.com, christophe.leroy@...roup.eu,
	naveen.n.rao@...ux.ibm.com, mpe@...erman.id.au,
	jirislaby@...nel.org, linuxppc-dev@...ts.ozlabs.org,
	linux-kernel@...r.kernel.org, linux-serial@...r.kernel.org,
	javier.carrasco.cruz@...il.com, shuah@...nel.org,
	julia.lawall@...ia.fr
Subject: Re: [PATCH V2] tty: hvc: hvc_opal: eliminate uses of of_node_put()

On Fri, May 03, 2024 at 04:52:15PM +0300, Lu Dai wrote:
> Make use of the __free() cleanup handler to automatically free nodes
> when they get out of scope.
> 
> Remove the need for a 'goto' as an effect.
> 
> Signed-off-by: Lu Dai <dai.lu@...rdes.com>
> ---
> Changes since v1:
> Move the assignment of 'opal' to its declaration
> Seperate the declaration of 'np'
> 
>  drivers/tty/hvc/hvc_opal.c | 13 +++++--------
>  1 file changed, 5 insertions(+), 8 deletions(-)
> 
> diff --git a/drivers/tty/hvc/hvc_opal.c b/drivers/tty/hvc/hvc_opal.c
> index 095c33ad10f8..c17e8343ea60 100644
> --- a/drivers/tty/hvc/hvc_opal.c
> +++ b/drivers/tty/hvc/hvc_opal.c
> @@ -327,19 +327,18 @@ static void udbg_init_opal_common(void)
>  
>  void __init hvc_opal_init_early(void)
>  {
> -	struct device_node *stdout_node = of_node_get(of_stdout);
> +	struct device_node *stdout_node __free(device_node) = of_node_get(of_stdout);
>  	const __be32 *termno;
>  	const struct hv_ops *ops;
>  	u32 index;
>  
>  	/* If the console wasn't in /chosen, try /ibm,opal */
>  	if (!stdout_node) {
> -		struct device_node *opal, *np;
> -
>  		/* Current OPAL takeover doesn't provide the stdout
>  		 * path, so we hard wire it
>  		 */
> -		opal = of_find_node_by_path("/ibm,opal/consoles");
> +		struct device_node *opal __free(device_node) =
> +			of_find_node_by_path("/ibm,opal/consoles");
>  		if (opal) {

No blank line?

>  			pr_devel("hvc_opal: Found consoles in new location\n");
>  		} else {
> @@ -350,13 +349,13 @@ void __init hvc_opal_init_early(void)
>  		}
>  		if (!opal)
>  			return;
> +		struct device_node *np;
>  		for_each_child_of_node(opal, np) {

Ick, no, don't do that please.  Take some time and become more familiar
with kernel coding style and issues, perhaps work in drivers/staging/
first, before attempting to do stuff like this that is not correct.

thanks,

greg k-h