linux-kernel - Re: [PATCH] rtc: abx80x: Fix return value of nvmem callback on read

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]

Date: Thu, 13 Jun 2024 09:18:41 +0300
From: Dan Carpenter <dan.carpenter@...aro.org>
To: Joy Chakraborty <joychakr@...gle.com>
Cc: Sean Anderson <sean.anderson@...o.com>,
	Alexandre Belloni <alexandre.belloni@...tlin.com>,
	Srinivas Kandagatla <srinivas.kandagatla@...aro.org>,
	linux-rtc@...r.kernel.org, linux-kernel@...r.kernel.org,
	stable@...r.kernel.org
Subject: Re: [PATCH] rtc: abx80x: Fix return value of nvmem callback on read

On Wed, Jun 12, 2024 at 06:05:54PM +0000, Joy Chakraborty wrote:
> Read callbacks registered with nvmem core expect 0 to be returned on
> success and a negative value to be returned on failure.
> 
> abx80x_nvmem_xfer() on read calls i2c_smbus_read_i2c_block_data() which
> returns the number of bytes read on success as per its api description,
> this return value is handled as an error and returned to nvmem even on
> success.
> 
> Fix to handle all possible values that would be returned by
> i2c_smbus_read_i2c_block_data().
> 
> Fixes: e90ff8ede777 ("rtc: abx80x: Add nvmem support")
> Cc: stable@...r.kernel.org
> Signed-off-by: Joy Chakraborty <joychakr@...gle.com>
> ---
>  drivers/rtc/rtc-abx80x.c | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/rtc/rtc-abx80x.c b/drivers/rtc/rtc-abx80x.c
> index fde2b8054c2e..0f5847d1ca2a 100644
> --- a/drivers/rtc/rtc-abx80x.c
> +++ b/drivers/rtc/rtc-abx80x.c
> @@ -711,9 +711,16 @@ static int abx80x_nvmem_xfer(struct abx80x_priv *priv, unsigned int offset,
>  		else
>  			ret = i2c_smbus_read_i2c_block_data(priv->client, reg,
>  							    len, val);
> -		if (ret)
> +		if (ret < 0)
>  			return ret;
>  
> +		if (!write) {
> +			if (ret)
> +				len = ret;
> +			else
> +				return -EIO;
> +		}

I guess this is the conservative approach.  Ie.  Don't break things
which aren't already broken.  But I suspect the correct approach is to
say:

	if (ret != len)
		return -EIO;

Ah well.  Being conservative is good.  It probably doesn't ever happen
in real life so it probably doesn't matter either way.

I don't really like the if (write) follow by and if (!write)...  It
would add more lines, but improve readability if we just duplicate the
code a big:

	if (write) {
		ret = write();
		if (ret)
			return ret;
	} else {
		ret = read();
		if (ret <= 0)
			return ret ?: -EIO;
		len = ret;
	}

regards,
dan carpenter