linux-kernel - Re: [PATCH 2/4] iio: core: Add support for IIO_AVAIL_LIST_WITH

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: <20220731182204.02459834@jic23-huawei>
Date:   Sun, 31 Jul 2022 18:22:04 +0100
From:   Jonathan Cameron <jic23@...nel.org>
To:     Paul Cercueil <paul@...pouillou.net>
Cc:     Peter Rosin <peda@...ntia.se>,
        Lars-Peter Clausen <lars@...afoo.de>, list@...ndingux.net,
        linux-iio@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/4] iio: core: Add support for IIO_AVAIL_LIST_WITH_TYPE

On Thu, 21 Jul 2022 20:15:24 +0100
Paul Cercueil <paul@...pouillou.net> wrote:

> The IIO_AVAIL_LIST_WITH_TYPE specifies that the array that corresponds
> to the available values is composed by cells of 3 integers, the first
> two representing the value itself (similar to what you would have with
> IIO_AVAIL_LIST), and the third integer representing the encoding type of
> the value.
> 
> This can be used for instance when a driver's .read_avail() callback
> returns values which cannot be represented with an unique encoding type.
> 
> Signed-off-by: Paul Cercueil <paul@...pouillou.net>

Hi Paul,

Generally looks good to me, but I'm not sure the overflow checks will work
as expected.

Jonathan

> ---
>  drivers/iio/industrialio-core.c | 25 +++++++++++++++++++++++++
>  drivers/iio/inkern.c            | 23 +++++++++++++++++++++++
>  include/linux/iio/consumer.h    |  6 ++++--
>  include/linux/iio/types.h       |  1 +
>  4 files changed, 53 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/iio/industrialio-core.c b/drivers/iio/industrialio-core.c
> index adf054c7a75e..99ced9eab069 100644
> --- a/drivers/iio/industrialio-core.c
> +++ b/drivers/iio/industrialio-core.c
> @@ -838,6 +838,29 @@ static ssize_t iio_format_avail_range(char *buf, const int *vals, int type)
>  	return iio_format_list(buf, vals, type, 3, "[", "]");
>  }
>  
> +static ssize_t iio_format_avail_list_with_type(char *buf, const int *vals,
> +					       int length)
> +{
> +	ssize_t len = 0;
> +	int i;
> +
> +	for (i = 0; i < length; i += 3) {
> +		if (i != 0) {
> +			len += sysfs_emit_at(buf, len, " ");
> +			if (len >= PAGE_SIZE)
I don't think this check will trigger if I read the sysfs_emit_at() docs right.

> +				return -EFBIG;
> +		}
> +
> +		len += __iio_format_value(buf, len, vals[i + 2], 2, &vals[i]);
> +		if (len >= PAGE_SIZE)

Can this trigger?  sysfs_emit_at() inside __iio_format_value() will only return
characters written which I think is none if there isn't space left in the page...

I think the normal thing to do is to just not handle the overflow if it occurs...



> +			return -EFBIG;
> +	}
> +
> +	len += sysfs_emit_at(buf, len, "\n");
> +
> +	return len;
> +}
> +
>  static ssize_t iio_read_channel_info_avail(struct device *dev,
>  					   struct device_attribute *attr,
>  					   char *buf)
> @@ -860,6 +883,8 @@ static ssize_t iio_read_channel_info_avail(struct device *dev,
>  		return iio_format_avail_list(buf, vals, type, length);
>  	case IIO_AVAIL_RANGE:
>  		return iio_format_avail_range(buf, vals, type);
> +	case IIO_AVAIL_LIST_WITH_TYPE:
> +		return iio_format_avail_list_with_type(buf, vals, length);
>  	default:
>  		return -EINVAL;
>  	}
> diff --git a/drivers/iio/inkern.c b/drivers/iio/inkern.c
> index e8a25852f0df..92d225f1ddd5 100644
> --- a/drivers/iio/inkern.c
> +++ b/drivers/iio/inkern.c
> @@ -855,6 +855,29 @@ static int iio_channel_read_max(struct iio_channel *chan,
>  		}
>  		return 0;
>  
> +	case IIO_AVAIL_LIST_WITH_TYPE:
> +		if (length <= 0 || length % 3 != 0)
> +			return -EINVAL;
> +
> +		if (vals[length - 1] != IIO_VAL_INT) {
> +			/* FIXME: learn about max for other iio values */
> +			return -EINVAL;
> +		}
> +
> +		*val = vals[length - 3];
> +		length -= 3;
> +
> +		for (; length; length -= 3) {
> +			if (vals[length - 1] != IIO_VAL_INT) {
> +				/* FIXME: learn about max for other iio values */
> +				return -EINVAL;
> +			}
> +
> +			if (vals[length - 3] > *val)
> +				*val = vals[length - 3];
> +		}
> +		return 0;
> +
>  	default:
>  		return ret;
>  	}
> diff --git a/include/linux/iio/consumer.h b/include/linux/iio/consumer.h
> index 5fa5957586cf..99dd12e10fb6 100644
> --- a/include/linux/iio/consumer.h
> +++ b/include/linux/iio/consumer.h
> @@ -309,7 +309,8 @@ int iio_read_max_channel_raw(struct iio_channel *chan, int *val);
>   * @vals:		Available values read back.
>   * @length:		Number of entries in vals.
>   *
> - * Returns an error code, IIO_AVAIL_RANGE or IIO_AVAIL_LIST.
> + * Returns an error code, IIO_AVAIL_RANGE, IIO_AVAIL_LIST or
> + * IIO_AVAIL_LIST_WITH_TYPE.
>   *
>   * For ranges, three vals are always returned; min, step and max.
>   * For lists, all the possible values are enumerated.
> @@ -328,7 +329,8 @@ int iio_read_avail_channel_raw(struct iio_channel *chan,
>   * @length:		Number of entries in vals.
>   * @attribute:		info attribute to be read back.
>   *
> - * Returns an error code, IIO_AVAIL_RANGE or IIO_AVAIL_LIST.
> + * Returns an error code, IIO_AVAIL_RANGE, IIO_AVAIL_LIST or
> + * IIO_AVAIL_LIST_WITH_TYPE.
>   */
>  int iio_read_avail_channel_attribute(struct iio_channel *chan,
>  				     const int **vals, int *type, int *length,
> diff --git a/include/linux/iio/types.h b/include/linux/iio/types.h
> index a7aa91f3a8dc..9777d1357080 100644
> --- a/include/linux/iio/types.h
> +++ b/include/linux/iio/types.h
> @@ -32,6 +32,7 @@ enum iio_event_info {
>  enum iio_available_type {
>  	IIO_AVAIL_LIST,
>  	IIO_AVAIL_RANGE,
> +	IIO_AVAIL_LIST_WITH_TYPE,
>  };
>  
>  enum iio_chan_info_enum {