linux-kernel - Re: [RFCv2 PATCH 2/4] bus: add Wiegand bus driver

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

Message-ID: <8259cc0c-93e2-4e62-3ce5-9372e2b40397@linaro.org>
Date:   Thu, 6 Oct 2022 10:29:48 +0200
From:   Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
To:     Martin Zaťovič <m.zatovic1@...il.com>,
        robh+dt@...nel.org, krzysztof.kozlowski+dt@...aro.org,
        linus.walleij@...aro.org, brgl@...ev.pl,
        gregkh@...uxfoundation.org, jeffrey.l.hugo@...il.com,
        andersson@...nel.org, Michael.Srba@...nam.cz, saravanak@...gle.com,
        mani@...nel.org, hemantk@...eaurora.org,
        linux-kernel@...r.kernel.org, devicetree@...r.kernel.org,
        linux-gpio@...r.kernel.org
Subject: Re: [RFCv2 PATCH 2/4] bus: add Wiegand bus driver

On 05/10/2022 16:57, Martin Zaťovič wrote:
> The Wiegand bus driver spawns devices and matches them with
> drivers.
> 
> Signed-off-by: Martin Zaťovič <m.zatovic1@...il.com>
> ---
> The bus driver currently assumes that any new Wiegand driver will
> have a matching entry in the devicetree. It is currently sufficient
> as I will only be implementing the GPIO driver. If someone
> implements a Wiegand driver that will not use devicetree, he will
> also have to edit this bus driver, in order to match properly. Is
> that a correct approach?

(...)

> +static struct wiegand_device *
> +of_register_wiegand_device(struct wiegand *wiegand, struct device_node *nc)
> +{
> +	struct wiegand_device *dev;
> +	const char *val;
> +	int ret;
> +
> +	dev = wiegand_alloc_device(wiegand);
> +	if (!dev) {
> +		dev_err(wiegand->dev,
> +			"Wiegand device alloc error for %pOF\n", nc);
> +		return ERR_PTR(-ENOMEM);
> +	}
> +
> +	ret = of_property_read_string(nc, "compatible", &val);
> +	if (ret) {
> +		dev_err(wiegand->dev, "%pOF has no valid 'compatible' property (%d)\n",
> +			nc, ret);
> +		goto err_put;
> +	}
> +
> +	if (strcmp(val, "wiegand,wiegand-gpio") == 0) {

This does not look right. Bus can have any device attached, so limiting
some bus behavior to a specific device is not really scalable.

Anyway device and node matching should not be with strcmp but rather
of_*_match().

Best regards,
Krzysztof