Re: [PATCH v9 3/3] i2c: npcm7xx: Add support for slave mode for Nuvoton

From: Andy Shevchenko
Date: Mon Mar 23 2020 - 12:02:02 EST


On Mon, Mar 23, 2020 at 03:44:37PM +0200, Tali Perry wrote:
> Add support for slave mode for Nuvoton
> NPCM BMC I2C controller driver.

Same comments apply as per previous patch.

...

> +#if IS_ENABLED(CONFIG_I2C_SLAVE)

Perhaps create a separate module and compile it when user selects?
See I²C DesignWare split.

...

> +#if IS_ENABLED(CONFIG_I2C_SLAVE)
> + if (bus->slave) {
> + npcm_i2c_slave_enable_l(bus, I2C_SLAVE_ADDR1, bus->slave->addr,
> + true);
> + }

Put it on one line and drop {} -- it will be much better to read.

> +#endif

...

> +static void npcm_i2c_write_to_fifo_slave(struct npcm_i2c *bus,
> + u16 max_bytes_to_send)
> +{
> + // Fill the FIFO, while the FIFO is not full and there are more bytes to
> + // write
> + npcm_i2c_clear_fifo_int(bus);
> + npcm_i2c_clear_tx_fifo(bus);
> + iowrite8(0, bus->reg + NPCM_I2CTXF_CTL);
> +
> + while ((max_bytes_to_send--) && (I2C_HW_FIFO_SIZE -
> + npcm_i2c_get_fifo_fullness(bus))) {
> + if (bus->slv_wr_size > 0) {
> + npcm_i2c_wr_byte(bus,
> + bus->slv_wr_buf[bus->slv_wr_ind %
> + I2C_HW_FIFO_SIZE]);
> + bus->slv_wr_ind = (bus->slv_wr_ind + 1) %
> + I2C_HW_FIFO_SIZE;
> + bus->slv_wr_size--; // size indicates the # of bytes in
> + // the SW FIFO, not HW.
> + } else {
> + break;
> + }

This looks ugly. Can you redo the style.

> + }
> +}

...

> +#if IS_ENABLED(CONFIG_I2C_SLAVE)
> + if (bus->slave) {
> + bus->slv_rd_buf[bus->slv_rd_ind %
> + I2C_HW_FIFO_SIZE] = data;
> + bus->slv_rd_ind++;
> + if (bus->slv_rd_ind == 1 && bus->read_block_use)
> + // 1st byte is length in block protocol
> + bus->slv_rd_size = data +
> + (u8)bus->PEC_use +
> + (u8)bus->read_block_use;
> + }
> +#endif

Ditto.

...

> I2C_FUNC_SMBUS_EMUL |
> I2C_FUNC_SMBUS_BLOCK_DATA |
> I2C_FUNC_SMBUS_PEC
> +#if IS_ENABLED(CONFIG_I2C_SLAVE)
> + | I2C_FUNC_SLAVE
> +#endif

Completely inconsistent style.

> ;

--
With Best Regards,
Andy Shevchenko