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 for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <58c1d1d4-d054-4aab-990b-e2083ceece4c@linux.dev>
Date: Mon, 28 Apr 2025 16:16:13 +0100
From: Vadim Fedorenko <vadim.fedorenko@...ux.dev>
To: Sagi Maimon <maimon.sagi@...il.com>, jonathan.lemon@...il.com,
 richardcochran@...il.com, andrew+netdev@...n.ch, davem@...emloft.net,
 edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com
Cc: linux-kernel@...r.kernel.org, netdev@...r.kernel.org,
 Sagi Maimon <sagi.maimon@...ran.com>
Subject: Re: [PATCH v1] ptp: ocp: Fix NULL dereference in Adva board SMA sysfs
 operations

On 28/04/2025 15:37, Sagi Maimon wrote:
> From: Sagi Maimon <sagi.maimon@...ran.com>
> 
> On Adva boards, SMA sysfs store/get operations can call
> __handle_signal_outputs() or __handle_signal_inputs() while the `irig`
> and `dcf` pointers are uninitialized, leading to a NULL pointer
> dereference in __handle_signal() and causing a kernel crash. Add
> Adva-specific callbacks ptp_ocp_sma_adva_set_outputs() and
> ptp_ocp_sma_adva_set_inputs() to the ptp_ocp driver, and include NULL
> checks for `irig` and `dcf` to prevent crashes.
> 
> Fixes: ef61f5528fca ("ptp: ocp: add Adva timecard support")
> Signed-off-by: Sagi Maimon <sagi.maimon@...ran.com>
> ---
>   drivers/ptp/ptp_ocp.c | 62 +++++++++++++++++++++++++++++++++++++++++--
>   1 file changed, 60 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/ptp/ptp_ocp.c b/drivers/ptp/ptp_ocp.c
> index faf6e027f89a..3eaa2005b3b2 100644
> --- a/drivers/ptp/ptp_ocp.c
> +++ b/drivers/ptp/ptp_ocp.c
> @@ -2578,12 +2578,70 @@ static const struct ocp_sma_op ocp_fb_sma_op = {
>   	.set_output	= ptp_ocp_sma_fb_set_output,
>   };
>   
> +static int
> +ptp_ocp_sma_adva_set_output(struct ptp_ocp *bp, int sma_nr, u32 val)
> +{
> +	u32 reg, mask, shift;
> +	unsigned long flags;
> +	u32 __iomem *gpio;
> +
> +	gpio = sma_nr > 2 ? &bp->sma_map1->gpio2 : &bp->sma_map2->gpio2;
> +	shift = sma_nr & 1 ? 0 : 16;
> +
> +	mask = 0xffff << (16 - shift);
> +
> +	spin_lock_irqsave(&bp->lock, flags);
> +
> +	reg = ioread32(gpio);
> +	reg = (reg & mask) | (val << shift);
> +
> +	if (bp->irig_out)
> +		ptp_ocp_irig_out(bp, reg & 0x00100010);
> +	if (bp->dcf_out)
> +		ptp_ocp_dcf_out(bp, reg & 0x00200020);

You never initialize neither irig_out nor dcf_out, both checks will
always fail. Looks like a dead code - please, remove it.

> +
> +	iowrite32(reg, gpio);
> +
> +	spin_unlock_irqrestore(&bp->lock, flags);
> +
> +	return 0;
> +}
> +
> +static int
> +ptp_ocp_sma_adva_set_inputs(struct ptp_ocp *bp, int sma_nr, u32 val)
> +{
> +	u32 reg, mask, shift;
> +	unsigned long flags;
> +	u32 __iomem *gpio;
> +
> +	gpio = sma_nr > 2 ? &bp->sma_map2->gpio1 : &bp->sma_map1->gpio1;
> +	shift = sma_nr & 1 ? 0 : 16;
> +
> +	mask = 0xffff << (16 - shift);
> +
> +	spin_lock_irqsave(&bp->lock, flags);
> +
> +	reg = ioread32(gpio);
> +	reg = (reg & mask) | (val << shift);
> +
> +	if (bp->irig_in)
> +		ptp_ocp_irig_in(bp, reg & 0x00100010);
> +	if (bp->dcf_in)
> +		ptp_ocp_dcf_in(bp, reg & 0x00200020);

The same goes here - neither irig_in, nor dcf_in will have real values
for your hardware, please remove this checks.

> +
> +	iowrite32(reg, gpio);
> +
> +	spin_unlock_irqrestore(&bp->lock, flags);
> +
> +	return 0;
> +}
> +
>   static const struct ocp_sma_op ocp_adva_sma_op = {
>   	.tbl		= { ptp_ocp_adva_sma_in, ptp_ocp_adva_sma_out },
>   	.init		= ptp_ocp_sma_fb_init,
>   	.get		= ptp_ocp_sma_fb_get,
> -	.set_inputs	= ptp_ocp_sma_fb_set_inputs,
> -	.set_output	= ptp_ocp_sma_fb_set_output,
> +	.set_inputs	= ptp_ocp_sma_adva_set_inputs,
> +	.set_output	= ptp_ocp_sma_adva_set_output,
>   };
>   
>   static int

pw-bot: cr

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ