linux-kernel - Re: [PATCH 5.4 117/203] rsi: fix potential null dereference in rsi

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: <20200204083332.GE26725@localhost>
Date:   Tue, 4 Feb 2020 09:33:32 +0100
From:   Johan Hovold <johan@...nel.org>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     linux-kernel@...r.kernel.org, stable@...r.kernel.org,
        Amitkumar Karwar <amitkarwar@...il.com>,
        Siva Rebbagondla <siva8118@...il.com>,
        Kalle Valo <kvalo@...eaurora.org>,
        Denis Efremov <efremov@...ux.com>
Subject: Re: [PATCH 5.4 117/203] rsi: fix potential null dereference in
 rsi_probe()

On Fri, Jan 17, 2020 at 12:17:14AM +0100, Greg Kroah-Hartman wrote:
> From: Denis Efremov <efremov@...ux.com>
> 
> commit f170d44bc4ec2feae5f6206980e7ae7fbf0432a0 upstream.
> 
> The id pointer can be NULL in rsi_probe(). It is checked everywhere except
> for the else branch in the idProduct condition. The patch adds NULL check
> before the id dereference in the rsi_dbg() call.
> 
> Fixes: 54fdb318c111 ("rsi: add new device model for 9116")
> Cc: Amitkumar Karwar <amitkarwar@...il.com>
> Cc: Siva Rebbagondla <siva8118@...il.com>
> Cc: Kalle Valo <kvalo@...eaurora.org>
> Signed-off-by: Denis Efremov <efremov@...ux.com>
> Signed-off-by: Kalle Valo <kvalo@...eaurora.org>
> Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>

This commit is bogus and was reverted shortly after it was applied in
order to prevent autosel from picking it up for stable (reverted by
c5dcf8f0e850 ("Revert "rsi: fix potential null dereference in
rsi_probe()"")).

The revert has now been picked up by Sasha, but shouldn't an
explicit revert in the same pull-request prevent a bad patch from being
backported in the first place? Seems like something that could be
scripted. But perhaps the net-stable oddities come into play here.

> ---
>  drivers/net/wireless/rsi/rsi_91x_usb.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> --- a/drivers/net/wireless/rsi/rsi_91x_usb.c
> +++ b/drivers/net/wireless/rsi/rsi_91x_usb.c
> @@ -793,7 +793,7 @@ static int rsi_probe(struct usb_interfac
>  		adapter->device_model = RSI_DEV_9116;
>  	} else {
>  		rsi_dbg(ERR_ZONE, "%s: Unsupported RSI device id 0x%x\n",
> -			__func__, id->idProduct);
> +			__func__, id ? id->idProduct : 0x0);
>  		goto err1;
>  	}

Johan