Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Sender: kvalo=codeaurora.org@mg.codeaurora.org
DMARC-Filter: OpenDMARC Filter v1.3.2 smtp.codeaurora.org 1B306C433CA
From:   Kalle Valo <kvalo@codeaurora.org>
To:     Stanislav Yakovlev <stas.yakovlev@gmail.com>
Cc:     Dan Carpenter <dan.carpenter@oracle.com>,
        Jeff Garzik <jgarzik@pobox.com>,
        James Ketrenos <jketreno@linux.intel.com>,
        wireless <linux-wireless@vger.kernel.org>,
        kernel-janitors@vger.kernel.org,
        Linus Torvalds <torvalds@linuxfoundation.org>
Subject: Re: [PATCH] ipw2x00: potential buffer overflow in libipw_wx_set_encodeext()
References: <YD4enB/Er/5PWgUz@mwanda>
        <CA++WF2PBo1Mok+bnnUCEZxbbEacX7FcU7PYAPTo=iqgOb2+f3Q@mail.gmail.com>
Date:   Sun, 11 Apr 2021 12:03:11 +0300
In-Reply-To: <CA++WF2PBo1Mok+bnnUCEZxbbEacX7FcU7PYAPTo=iqgOb2+f3Q@mail.gmail.com>
        (Stanislav Yakovlev's message of "Sat, 10 Apr 2021 10:59:49 +0400")
Message-ID: <87blalmbw0.fsf@codeaurora.org>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.5 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

Stanislav Yakovlev <stas.yakovlev@gmail.com> writes:

> On Tue, 2 Mar 2021 at 15:16, Dan Carpenter <dan.carpenter@oracle.com> wrote:
>>
>> The "ext->key_len" is a u16 that comes from the user.  If it's over
>> SCM_KEY_LEN (32) that could lead to memory corruption.
>>
>> Fixes: e0d369d1d969 ("[PATCH] ieee82011: Added WE-18 support to
>> default wireless extension handler")
>> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
>> ---
>>  drivers/net/wireless/intel/ipw2x00/libipw_wx.c | 6 ++++--
>>  1 file changed, 4 insertions(+), 2 deletions(-)

[...]

>> --- a/drivers/net/wireless/intel/ipw2x00/libipw_wx.c
>> +++ b/drivers/net/wireless/intel/ipw2x00/libipw_wx.c
>> @@ -633,8 +633,10 @@ int libipw_wx_set_encodeext(struct libipw_device *ieee,
>>         }
>>
>>         if (ext->alg != IW_ENCODE_ALG_NONE) {
>> -               memcpy(sec.keys[idx], ext->key, ext->key_len);
>> -               sec.key_sizes[idx] = ext->key_len;
>> +               int len = min_t(int, ext->key_len, SCM_KEY_LEN);
>> +
>> +               memcpy(sec.keys[idx], ext->key, len);
>> +               sec.key_sizes[idx] = len;
>>                 sec.flags |= (1 << idx);
>>                 if (ext->alg == IW_ENCODE_ALG_WEP) {
>>                         sec.encode_alg[idx] = SEC_ALG_WEP;

In another thread Linus gave a good tip about clamp_val(), I think it
should be used it here as well to make the check safer and more
readable. And also elsewhere in wireless code with similar limits.

-- 
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches