linux-kernel - Re: [PATCH 1/2] USB: array-index-out-of-bounds in ehci_brcm_hub

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: <YYTwMTbYPpBE8817@kroah.com>
Date:   Fri, 5 Nov 2021 09:49:53 +0100
From:   Greg KH <gregkh@...uxfoundation.org>
To:     Haimin Zhang <tcs.kernel@...il.com>
Cc:     linux-kernel@...r.kernel.org, linux-usb@...r.kernel.org,
        Haimin Zhang <tcs_kernel@...cent.com>,
        TCS Robot <tcs_robot@...cent.com>
Subject: Re: [PATCH 1/2] USB: array-index-out-of-bounds in
 ehci_brcm_hub_control

On Fri, Nov 05, 2021 at 04:32:50PM +0800, Haimin Zhang wrote:
> There isn't enough check parameter `wIndex` in the function
> `ehci_brcm_hub_control`;due to the size of array `port_status`
> is 15, so it may lead to out of bounds.

Odd use of ';'

And have you seen this get out of bounds?  If so, how?


> 
> Signed-off-by: Haimin Zhang <tcs_kernel@...cent.com>
> Reported-by: TCS Robot <tcs_robot@...cent.com>

Signed-off-by goes below the reported-by lines.

And why are these 2 patches not "threaded" properly in email.  How did
you send them?

Also, your "From" line does not match your signed-off-by line, so I
could not take these at all even if this was ok.

> ---
>  drivers/usb/host/ehci-brcm.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/usb/host/ehci-brcm.c b/drivers/usb/host/ehci-brcm.c
> index d3626bfa966b..a1e3290e5459 100644
> --- a/drivers/usb/host/ehci-brcm.c
> +++ b/drivers/usb/host/ehci-brcm.c
> @@ -62,8 +62,11 @@ static int ehci_brcm_hub_control(
>  	u32 __iomem	*status_reg;
>  	unsigned long flags;
>  	int retval, irq_disabled = 0;
> +	u32 temp;
>  
> -	status_reg = &ehci->regs->port_status[(wIndex & 0xff) - 1];
> +	temp = (wIndex & 0xff) - 1;
> +	if (temp < ports)
> +		status_reg = &ehci->regs->port_status[temp];

What if the test fails?  Should you do something about that?  status_reg
is now uninitialized, doesn't the code fail now?

How did you test this?

thanks,

greg k-h