linux-kernel - Re: [PATCH v2 1/2] crypto: hisilicon

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]

Date:   Wed, 24 May 2023 12:45:04 +0200
From:   Philipp Zabel <p.zabel@...gutronix.de>
To:     David Yang <mmyangfl@...il.com>, linux-crypto@...r.kernel.org
Cc:     Weili Qian <qianweili@...wei.com>,
        Zhou Wang <wangzhou1@...ilicon.com>,
        Herbert Xu <herbert@...dor.apana.org.au>,
        "David S. Miller" <davem@...emloft.net>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 1/2] crypto: hisilicon - Add HiSilicon ADVCA Subsystem

Hi David,

On Sa, 2023-05-13 at 15:43 +0800, David Yang wrote:
> HiSilicon ADVCA Subsystem contains various cryptographic devices, including
> symmetric key ciphers, hash functions, RSA algorithms, as well as key
> ladder and OTP memory.
> 
> This patch adds symmetric key cipher driver.
> 
> Signed-off-by: David Yang <mmyangfl@...il.com>
> ---
>  drivers/crypto/hisilicon/Kconfig              |    8 +
>  drivers/crypto/hisilicon/Makefile             |    1 +
>  drivers/crypto/hisilicon/advca/Makefile       |    1 +
>  .../crypto/hisilicon/advca/hisi-advca-muc.c   | 1527 +++++++++++++++++
>  4 files changed, 1537 insertions(+)
>  create mode 100644 drivers/crypto/hisilicon/advca/Makefile
>  create mode 100644 drivers/crypto/hisilicon/advca/hisi-advca-muc.c
> 
[...]
> diff --git a/drivers/crypto/hisilicon/advca/hisi-advca-muc.c b/drivers/crypto/hisilicon/advca/hisi-advca-muc.c
> new file mode 100644
> index 000000000000..362596a91e19
> --- /dev/null
> +++ b/drivers/crypto/hisilicon/advca/hisi-advca-muc.c
> @@ -0,0 +1,1527 @@
[...]
> +static int hica_muc_probe(struct platform_device *pdev)
> +{
[...]
> +	priv->rst = devm_reset_control_get_optional(dev, NULL);

Please use devm_reset_control_get_optional_exclusive() directly.

Since priv->rst is only ever used in hica_muc_probe(), it could be
stored in a local variable instead.

> +	if (IS_ERR(priv->rst))
> +		return PTR_ERR(priv->rst);
> +
> +	ret = platform_get_irq(pdev, 0);
> +	if (ret < 0)
> +		return ret;
> +	priv->irqs[0] = ret;
> +	priv->irqs[1] = platform_get_irq_optional(pdev, 1);
> +
> +	priv->algs_num = ARRAY_SIZE(hica_muc_tmpls);
> +	priv->algs = devm_kmalloc_array(dev, priv->algs_num,
> +					sizeof(priv->algs[0]), GFP_KERNEL);
> +	if (!priv->algs)
> +		return -ENOMEM;
> +
> +	init_completion(&priv->cond);
> +
> +	priv->dev = dev;
> +	platform_set_drvdata(pdev, priv);
> +	dev_set_drvdata(dev, priv);
> +
> +	/* bring up device */
> +	ret = reset_control_assert(priv->rst) ?:
> +	      clk_bulk_prepare_enable(priv->clks_num, priv->clks) ?:
> +	      reset_control_deassert(priv->rst);

Using the ternary operator like this is a bit unconventional. Here, the
clocks are kept enabled if the reset_control_deassert() fails.

It would be good to add an error path that disables the clocks.

> +	if (ret)
> +		return ret;
> +
> +	if (!(readl(priv->base + MUC_RST_STATUS) & MUC_STATE_VALID)) {
> +		msleep(20);
> +		if (!(readl(priv->base + MUC_RST_STATUS) & MUC_STATE_VALID)) {
> +			dev_err(dev, "cannot bring up device\n");
> +			return -ENODEV;

This also leaves the clocks enabled. There are some more return
statements with the same issue below.

regards
Philipp