linux-kernel - Re: [PATCH] wifi: cisco: do not assign -1 to unsigned char

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: <Y1fSuq4I9BBQ4o+J@zx2c4.com>
Date:   Tue, 25 Oct 2022 14:12:42 +0200
From:   "Jason A. Donenfeld" <Jason@...c4.com>
To:     David Laight <David.Laight@...LAB.COM>
Cc:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Kalle Valo <kvalo@...nel.org>,
        "linux-wireless@...r.kernel.org" <linux-wireless@...r.kernel.org>
Subject: Re: [PATCH] wifi: cisco: do not assign -1 to unsigned char

On Tue, Oct 25, 2022 at 10:11:44AM +0000, David Laight wrote:
> From: Jason A. Donenfeld
> > Sent: 24 October 2022 17:29
> > 
> > With char becoming unsigned by default, and with `char` alone being
> > ambiguous and based on architecture, we get a warning when assigning the
> > unchecked output of hex_to_bin() to that unsigned char. Mark `key` as a
> > `u8`, which matches the struct's type, and then check each call to
> > hex_to_bin() before casting.
> > 
> > Cc: Kalle Valo <kvalo@...nel.org>
> > Cc: linux-wireless@...r.kernel.org
> > Signed-off-by: Jason A. Donenfeld <Jason@...c4.com>
> > ---
> >  drivers/net/wireless/cisco/airo.c | 18 ++++++++++++++----
> >  1 file changed, 14 insertions(+), 4 deletions(-)
> > 
> > diff --git a/drivers/net/wireless/cisco/airo.c b/drivers/net/wireless/cisco/airo.c
> > index 10daef81c355..fb2c35bd73bb 100644
> > --- a/drivers/net/wireless/cisco/airo.c
> > +++ b/drivers/net/wireless/cisco/airo.c
> > @@ -5232,7 +5232,7 @@ static int get_wep_tx_idx(struct airo_info *ai)
> >  	return -1;
> >  }
> > 
> > -static int set_wep_key(struct airo_info *ai, u16 index, const char *key,
> > +static int set_wep_key(struct airo_info *ai, u16 index, const u8 *key,
> >  		       u16 keylen, int perm, int lock)
> >  {
> >  	static const unsigned char macaddr[ETH_ALEN] = { 0x01, 0, 0, 0, 0, 0 };
> > @@ -5283,7 +5283,7 @@ static void proc_wepkey_on_close(struct inode *inode, struct file *file)
> >  	struct net_device *dev = pde_data(inode);
> >  	struct airo_info *ai = dev->ml_priv;
> >  	int i, rc;
> > -	char key[16];
> > +	u8 key[16];
> >  	u16 index = 0;
> >  	int j = 0;
> > 
> > @@ -5311,12 +5311,22 @@ static void proc_wepkey_on_close(struct inode *inode, struct file *file)
> >  	}
> > 
> >  	for (i = 0; i < 16*3 && data->wbuffer[i+j]; i++) {
> > +		int val;
> > +
> > +		if (i % 3 == 2)
> > +			continue;
> > +
> > +		val = hex_to_bin(data->wbuffer[i+j]);
> > +		if (val < 0) {
> > +			airo_print_err(ai->dev->name, "WebKey passed invalid key hex");
> > +			return;
> > +		}
> >  		switch(i%3) {
> >  		case 0:
> > -			key[i/3] = hex_to_bin(data->wbuffer[i+j])<<4;
> > +			key[i/3] = (u8)val << 4;
> >  			break;
> >  		case 1:
> > -			key[i/3] |= hex_to_bin(data->wbuffer[i+j]);
> > +			key[i/3] |= (u8)val;
> >  			break;
> >  		}
> >  	}
> 
> That is about the crappiest loop I've seen.
> I was just going to point out that the (u8) casts aren't needed.
> Something like:
> 	for (i = 0, buf = data->wbuffer + j; i < 16; i++, buf += 3) {
> 		int val;
> 		if (!buf[0] || !buf[1])
> 			break;
> 		val = hex_to_bin(buf[0]) | hex_to_bin(buf[1]) << 8;
> 		if (val < 0) {
> 			airo_print_err(ai->dev->name, "WebKey passed invalid key hex");
> 			return;
> 		}
> 		key[i] = val;
> 		if (!buf[2])
> 			break;
> 	}
> 
> Although there should be a check for buf[2] being valid.
> Any I worry about exactly what happens if there aren't 16 full bytes.

buf[2] isn't checked. Presumably it's a space or something. Your <<8
also isn't right; this is a hex char. Anyway, I think I'd rather
minimize this delta and leave this patch as-is.

Jason