Re: [PATCH v6 1/6] printk: Save console options for add_preferred_console_match()

[Andy Shevchenko]

On Tue, Feb 13, 2024 at 10:45:08AM +0200, Tony Lindgren wrote:
> Driver subsystems may need to translate the preferred console name to the
> character device name used. We already do some of this in console_setup()
> with a few hardcoded names, but that does not scale well.
> 
> The console options are parsed early in console_setup(), and the consoles
> are added with __add_preferred_console(). At this point we don't know much
> about the character device names and device drivers getting probed.
> 
> To allow driver subsystems to set up a preferred console, let's save the
> kernel command line console options. To add a preferred console from a
> driver subsystem with optional character device name translation, let's
> add a new function add_preferred_console_match().
> 
> This allows the serial core layer to support console=DEVNAME:0.0 style
> hardware based addressing in addition to the current console=ttyS0 style
> naming. And we can start moving console_setup() character device parsing
> to the driver subsystem specific code.
> 
> We use a separate array from the console_cmdline array as the character
> device name and index may be unknown at the console_setup() time. And
> eventually there's no need to call __add_preferred_console() until the
> subsystem is ready to handle the console.
> 
> Adding the console name in addition to the character device name, and a
> flag for an added console, could be added to the struct console_cmdline.
> And the console_cmdline array handling could be modified accordingly. But
> that complicates things compared saving the console options, and then
> adding the consoles when the subsystems handling the consoles are ready.

..

> +int __init console_opt_save(const char *str, const char *brl_opt)
> +{
> +	struct console_option *con;
> +	const char *opt = NULL;
> +	size_t namelen, optlen;
> +	int i;

> +	namelen = strcspn(str, ",");
> +	if (!namelen)
> +		return -EINVAL;
> +
> +	optlen = strlen(str) - namelen;
> +	if (optlen > 1)
> +		opt = str + namelen + 1;
> +
> +	if (namelen >= CONSOLE_NAME_MAX || optlen >= CONSOLE_OPT_MAX)
> +		return -EINVAL;
> +
> +	for (i = 0; i < MAX_CMDLINECONSOLES; i++) {
> +		con = &conopt[i];
> +
> +		if (con->name[0]) {
> +			if (!strncmp(str, con->name, namelen))
> +				return 0;
> +			continue;
> +		}

> +		strscpy(con->name, str, namelen + 1);
> +		if (opt)
> +			strscpy(con->opt, opt, optlen + 1);

> +		/* See _braille_console_setup(), both empty and NULL are valid */
> +		if (brl_opt) {
> +			strscpy(con->brl_opt, brl_opt, CONSOLE_BRL_OPT_MAX);
> +			con->has_brl_opt = 1;
> +		}
> +
> +		return 0;
> +	}
> +
> +	return -ENOMEM;
> +}

With fresh look at the above, can we amend it like below?
(dropped NULL assignment, optimized strlen(), split checks, dropped unneeded +1 in strscpy() calls)

int __init console_opt_save(const char *str, const char *brl_opt)
{
	struct console_option *con;
	size_t namelen, optlen;
	const char *opt;
	int i;

	namelen = strcspn(str, ",");
	if (namelen == 0 || namelen >= CONSOLE_NAME_MAX)
		return -EINVAL;

	opt = str + namelen;
	if (*opt == ',')
		opt++;

	optlen = strlen(opt);
	if (optlen >= CONSOLE_OPT_MAX)
		return -EINVAL;

	for (i = 0; i < MAX_CMDLINECONSOLES; i++) {
		con = &conopt[i];

		if (con->name[0]) {
			if (!strncmp(str, con->name, namelen))
				return 0;
			continue;
		}

		strscpy(con->name, str, namelen);
		strscpy(con->opt, opt, optlen); // not sure if emptying opt is okay

		/* See _braille_console_setup(), both empty and NULL are valid */
		if (brl_opt) {
			strscpy(con->brl_opt, brl_opt, CONSOLE_BRL_OPT_MAX);
			con->has_brl_opt = 1;
		}

		return 0;
	}

	return -ENOMEM;
}

-- 
With Best Regards,
Andy Shevchenko