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: <78670082-36a2-2520-ac7f-b71482540035@amd.com>
Date: Mon, 8 Sep 2025 11:10:11 -0500
From: Tom Lendacky <thomas.lendacky@....com>
To: Yunseong Kim <ysk@...lloc.com>, Herbert Xu <herbert@...dor.apana.org.au>,
 Mario Limonciello <mario.limonciello@....com>,
 John Allen <john.allen@....com>, "David S. Miller" <davem@...emloft.net>
Cc: linux-crypto@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] crypto: ccp: Fix incorrect return type for
 psp_get_capability()

On 9/3/25 08:55, Yunseong Kim wrote:
> psp_get_capability() is declared as returning an 'unsigned int'. However,
> it returns -ENODEV on failure when it cannot access the device registers
> (i.e., when ioread32 returns 0xffffffff).
> 
> Since -ENODEV is a negative value, returning it from a function declared as
> 'unsigned int' results in an implicit cast to a large positive integer.
> This prevents the caller psp_dev_init() from correctly detecting the
> error condition, leading to improper error handling.
> 
> Fix this by changing the return type of psp_get_capability() to 'int'.
> 
> Additionally, change the type of the local variable 'val' to 'u32', which

So you have a couple of changes going on here...  which would mean
separate patches. But since the raw field of psp_cap_register is an
unsigned int and ioread32 returns an unsigned int, this doesn't need to
be changed at all.

> is more appropriate for register access, and reformat the long dev_notice
> line to adhere to kernel coding style guidelines.
> 
> Signed-off-by: Yunseong Kim <ysk@...lloc.com>
> ---
>  drivers/crypto/ccp/psp-dev.c | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/crypto/ccp/psp-dev.c b/drivers/crypto/ccp/psp-dev.c
> index 1c5a7189631e..84dde53db25b 100644
> --- a/drivers/crypto/ccp/psp-dev.c
> +++ b/drivers/crypto/ccp/psp-dev.c
> @@ -140,9 +140,9 @@ static irqreturn_t psp_irq_handler(int irq, void *data)
>  	return IRQ_HANDLED;
>  }
>  
> -static unsigned int psp_get_capability(struct psp_device *psp)
> +static int psp_get_capability(struct psp_device *psp)

This change is ok, everything else below is not needed.

Thanks,
Tom

>  {
> -	unsigned int val = ioread32(psp->io_regs + psp->vdata->feature_reg);
> +	u32 val = ioread32(psp->io_regs + psp->vdata->feature_reg);
>  
>  	/*
>  	 * Check for a access to the registers.  If this read returns
> @@ -152,7 +152,8 @@ static unsigned int psp_get_capability(struct psp_device *psp)
>  	 * could get properly initialized).
>  	 */
>  	if (val == 0xffffffff) {
> -		dev_notice(psp->dev, "psp: unable to access the device: you might be running a broken BIOS.\n");
> +		dev_notice(psp->dev,
> +			"psp: unable to access the device: you might be running a broken BIOS.\n");
>  		return -ENODEV;
>  	}
>  	psp->capability.raw = val;

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ