Hi Rayagonda,
On 7/24/2019 1:28 AM, Rayagonda Kokatanur wrote:
> Use SMBUS_MASTER_DATA_READ.MASTER_RD_STATUS bit to check for RX
> FIFO empty condition because SMBUS_MASTER_FIFO_CONTROL.MASTER_RX_PKT_COUNT
> is not updated for read >= 64 bytes. This fixes the issue when trying to
> read from the I2C slave more than 63 bytes.
>
> Fixes: c24b8d574b7c ("i2c: iproc: Extend I2C read up to 255 bytes")
>
> Signed-off-by: Rayagonda Kokatanur <rayagonda.kokata...@broadcom.com>
> ---
> drivers/i2c/busses/i2c-bcm-iproc.c | 10 ++++++----
> 1 file changed, 6 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/i2c/busses/i2c-bcm-iproc.c
> b/drivers/i2c/busses/i2c-bcm-iproc.c
> index 2c7f145..d7fd76b 100644
> --- a/drivers/i2c/busses/i2c-bcm-iproc.c
> +++ b/drivers/i2c/busses/i2c-bcm-iproc.c
> @@ -392,16 +392,18 @@ static bool bcm_iproc_i2c_slave_isr(struct
> bcm_iproc_i2c_dev *iproc_i2c,
> static void bcm_iproc_i2c_read_valid_bytes(struct bcm_iproc_i2c_dev
> *iproc_i2c)
> {
> struct i2c_msg *msg = iproc_i2c->msg;
> + uint32_t val;
>
> /* Read valid data from RX FIFO */
> while (iproc_i2c->rx_bytes < msg->len) {
> - if (!((iproc_i2c_rd_reg(iproc_i2c, M_FIFO_CTRL_OFFSET) >>
> M_FIFO_RX_CNT_SHIFT)
> - & M_FIFO_RX_CNT_MASK))
> + val = iproc_i2c_rd_reg(iproc_i2c, M_RX_OFFSET);
> +
> + /* rx fifo empty */
> + if (!((val >> M_RX_STATUS_SHIFT) & M_RX_STATUS_MASK))
> break;
>
> msg->buf[iproc_i2c->rx_bytes] =
> - (iproc_i2c_rd_reg(iproc_i2c, M_RX_OFFSET) >>
> - M_RX_DATA_SHIFT) & M_RX_DATA_MASK;
> + (val >> M_RX_DATA_SHIFT) & M_RX_DATA_MASK;
> iproc_i2c->rx_bytes++;
> }
> }
>
Thanks for the fix. This fix looks good to me!
Reviewed-by: Ray Jui <ray....@broadcom.com>
