linux-kernel - Re: [PATCH 2/2] genirq: Use firmware identifier while adding domain

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: <57CD8194.2090306@arm.com>
Date:   Mon, 5 Sep 2016 15:30:44 +0100
From:   Marc Zyngier <marc.zyngier@....com>
To:     Punit Agrawal <punit.agrawal@....com>, tglx@...utronix.de,
        jiang.liu@...ux.intel.com
Cc:     linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/2] genirq: Use firmware identifier while adding domain

On 31/05/16 13:56, Punit Agrawal wrote:
> Use the firmware provided identifier for the domain name.
> 
> Signed-off-by: Punit Agrawal <punit.agrawal@....com>
> ---
>  kernel/irq/irqdomain.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c
> index 1fe2fea..3af09e1 100644
> --- a/kernel/irq/irqdomain.c
> +++ b/kernel/irq/irqdomain.c
> @@ -114,6 +114,10 @@ struct irq_domain *__irq_domain_add(struct fwnode_handle *fwnode, int size,
>  	domain->hwirq_max = hwirq_max;
>  	domain->revmap_size = size;
>  	domain->revmap_direct_max_irq = direct_max;
> +	if (is_fwnode_irqchip(fwnode))
> +		domain->name = container_of(fwnode, struct irqchip_fwid, fwnode)->name;

So this thing worries me to no end. Look at the way irqchip_fwid is
constructed:

	name = kasprintf(GFP_KERNEL, "irqchip@%p", data);

You'd end-up disclosing a kernel address, which is not a very good idea.

> +	else
> +		domain->name = of_node_full_name(of_node);

And what if the node gets pruned (as it can happen on OpenFirmware
implementations)?

>  	irq_domain_check_hierarchy(domain);
>  
>  	mutex_lock(&irq_domain_mutex);
> 

Thanks,

	M.
-- 
Jazz is not dead. It just smells funny...