[RFC,2/3] media: i2c: adv7842: remove open coded version of SMBus block read

  The open coded version differs from the one in the core in one way: the
buffer will be always copied back, even when the transfer failed. It
looks like it is expected that the sanity check for a correct CRC and
header will bail out later.

Use the block read from the I2C core and propagate a potential errno
further to the sanity check.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
---

Note: we could now make the error checking even stronger by checking if
the number of received bytes is I2C_SMBUS_BLOCK_MAX. But to avoid
regressions, I kept the logic as is, i.e. only check for errno.

 drivers/media/i2c/adv7511-v4l2.c | 40 +++++++++++---------------------
 1 file changed, 14 insertions(+), 26 deletions(-)
  

On Tue, Jan 12, 2021 at 05:41:28PM +0100, Wolfram Sang wrote:
> The open coded version differs from the one in the core in one way: the
> buffer will be always copied back, even when the transfer failed. It
> looks like it is expected that the sanity check for a correct CRC and
> header will bail out later.
> 
> Use the block read from the I2C core and propagate a potential errno
> further to the sanity check.
> 
> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>

$subject should say "adv7511", sorry!
  

> +static int adv7511_edid_rd(struct v4l2_subdev *sd, uint16_t len, uint8_t *buf)
>  {
>  	struct adv7511_state *state = get_adv7511_state(sd);
> +	s32 len;

And 'len' here shadows the function argument :( Ok, I need to resend
this patch. Still, looking forward to a comment if an approach like this
is acceptable.
  

Hi Wolfram,

On 12/01/2021 17:41, Wolfram Sang wrote:
> The open coded version differs from the one in the core in one way: the
> buffer will be always copied back, even when the transfer failed. It
> looks like it is expected that the sanity check for a correct CRC and
> header will bail out later.

Nah, it's just a bug. It should have returned and checked the error code,
so your patch does the right thing.

Regards,

	Hans

> 
> Use the block read from the I2C core and propagate a potential errno
> further to the sanity check.
> 
> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
> ---
> 
> Note: we could now make the error checking even stronger by checking if
> the number of received bytes is I2C_SMBUS_BLOCK_MAX. But to avoid
> regressions, I kept the logic as is, i.e. only check for errno.
> 
>  drivers/media/i2c/adv7511-v4l2.c | 40 +++++++++++---------------------
>  1 file changed, 14 insertions(+), 26 deletions(-)
> 
> diff --git a/drivers/media/i2c/adv7511-v4l2.c b/drivers/media/i2c/adv7511-v4l2.c
> index a3161d709015..0150f76dc6a6 100644
> --- a/drivers/media/i2c/adv7511-v4l2.c
> +++ b/drivers/media/i2c/adv7511-v4l2.c
> @@ -214,36 +214,24 @@ static inline void adv7511_wr_and_or(struct v4l2_subdev *sd, u8 reg, u8 clr_mask
>  	adv7511_wr(sd, reg, (adv7511_rd(sd, reg) & clr_mask) | val_mask);
>  }
>  
> -static int adv_smbus_read_i2c_block_data(struct i2c_client *client,
> -					 u8 command, unsigned length, u8 *values)
> -{
> -	union i2c_smbus_data data;
> -	int ret;
> -
> -	if (length > I2C_SMBUS_BLOCK_MAX)
> -		length = I2C_SMBUS_BLOCK_MAX;
> -	data.block[0] = length;
> -
> -	ret = i2c_smbus_xfer(client->adapter, client->addr, client->flags,
> -			     I2C_SMBUS_READ, command,
> -			     I2C_SMBUS_I2C_BLOCK_DATA, &data);
> -	memcpy(values, data.block + 1, length);
> -	return ret;
> -}
> -
> -static void adv7511_edid_rd(struct v4l2_subdev *sd, uint16_t len, uint8_t *buf)
> +static int adv7511_edid_rd(struct v4l2_subdev *sd, uint16_t len, uint8_t *buf)
>  {
>  	struct adv7511_state *state = get_adv7511_state(sd);
> +	s32 len;
>  	int i;
> -	int err = 0;
>  
>  	v4l2_dbg(1, debug, sd, "%s:\n", __func__);
>  
> -	for (i = 0; !err && i < len; i += I2C_SMBUS_BLOCK_MAX)
> -		err = adv_smbus_read_i2c_block_data(state->i2c_edid, i,
> +	for (i = 0; i < len; i += I2C_SMBUS_BLOCK_MAX) {
> +		len = i2c_smbus_read_i2c_block_data(state->i2c_edid, i,
>  						    I2C_SMBUS_BLOCK_MAX, buf + i);
> -	if (err)
> -		v4l2_err(sd, "%s: i2c read error\n", __func__);
> +		if (len < 0) {
> +			v4l2_err(sd, "%s: i2c read error\n", __func__);
> +			return len;
> +		}
> +	}
> +
> +	return 0;
>  }
>  
>  static inline int adv7511_cec_read(struct v4l2_subdev *sd, u8 reg)
> @@ -1668,20 +1656,20 @@ static bool adv7511_check_edid_status(struct v4l2_subdev *sd)
>  	if (edidRdy & MASK_ADV7511_EDID_RDY) {
>  		int segment = adv7511_rd(sd, 0xc4);
>  		struct adv7511_edid_detect ed;
> +		int err;
>  
>  		if (segment >= EDID_MAX_SEGM) {
>  			v4l2_err(sd, "edid segment number too big\n");
>  			return false;
>  		}
>  		v4l2_dbg(1, debug, sd, "%s: got segment %d\n", __func__, segment);
> -		adv7511_edid_rd(sd, 256, &state->edid.data[segment * 256]);
> +		err = adv7511_edid_rd(sd, 256, &state->edid.data[segment * 256]);
>  		adv7511_dbg_dump_edid(2, debug, sd, segment, &state->edid.data[segment * 256]);
>  		if (segment == 0) {
>  			state->edid.blocks = state->edid.data[0x7e] + 1;
>  			v4l2_dbg(1, debug, sd, "%s: %d blocks in total\n", __func__, state->edid.blocks);
>  		}
> -		if (!edid_verify_crc(sd, segment) ||
> -		    !edid_verify_header(sd, segment)) {
> +		if (err < 0 || !edid_verify_crc(sd, segment) || !edid_verify_header(sd, segment)) {
>  			/* edid crc error, force reread of edid segment */
>  			v4l2_err(sd, "%s: edid crc or header error\n", __func__);
>  			state->have_monitor = false;
>
  

Hello Hans,

I hope you are well!

> > The open coded version differs from the one in the core in one way: the
> > buffer will be always copied back, even when the transfer failed. It
> > looks like it is expected that the sanity check for a correct CRC and
> > header will bail out later.
> 
> Nah, it's just a bug. It should have returned and checked the error code,
> so your patch does the right thing.

I see. So, I will only update the commit message.

Thanks for the reviews!

All the best,

   Wolfram