RE: [patch v3 1/1] i2c: add master driver for mellanox systems

[Vadim Pasternak]

Hi Vladimir,

Thank you very much for your reviews.

> -----Original Message-----
> From: Vladimir Zapolskiy [mailto:vz@xxxxxxxxx]
> Sent: Friday, November 04, 2016 12:34 AM
> To: Vadim Pasternak <vadimp@xxxxxxxxxxxx>; wsa@xxxxxxxxxxxxx
> Cc: linux-i2c@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; jiri@xxxxxxxxxxx;
> Michael Shych <michaelsh@xxxxxxxxxxxx>
> Subject: Re: [patch v3 1/1] i2c: add master driver for mellanox systems
> 
> Hi Vadim,
> 
> please find some more review comments below.
> 
> On 03.11.2016 20:50, vadimp@xxxxxxxxxxxx wrote:
> > From: Vadim Pasternak <vadimp@xxxxxxxxxxxx>
> >
> > Device driver for Mellanox I2C controller logic, implemented in
> > Lattice CPLD device.
> > Device supports:
> >  - Master mode
> >  - One physical bus
> >  - Polling mode
> >
> > The Kconfig currently controlling compilation of this code is:
> > drivers/i2c/busses/Kconfig:config I2C_MLXCPLD
> >
> > Signed-off-by: Michael Shych <michaelsh@xxxxxxxxxxxx>
> > Signed-off-by: Vadim Pasternak <vadimp@xxxxxxxxxxxx>
> > Reviewed-by: Jiri Pirko <jiri@xxxxxxxxxxxx>
> > ---
> 
> [snip]
> 
> > diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
> > index d252276..be885d2 100644
> > --- a/drivers/i2c/busses/Kconfig
> > +++ b/drivers/i2c/busses/Kconfig
> > @@ -1150,6 +1150,18 @@ config I2C_ELEKTOR
> >  	  This support is also available as a module.  If so, the module
> >  	  will be called i2c-elektor.
> >
> > +config I2C_MLXCPLD
> > +	tristate "Mellanox I2C driver"
> > +	depends on X86_64
> > +	default y
> 
> I believe the controller is not a widely used piece of hardware found on every
> second laptop.
> 
> Please remove 'default y' line from the Kconfig record.
> 
> > +	help
> > +	  This exposes the Mellanox platform I2C busses to the linux I2C layer
> > +	  for X86 based systems.
> > +	  Controller is implemented as CPLD logic.
> > +
> > +	  This driver can also be built as a module. If so, the module will be
> > +	  called as i2c-mlxcpld.
> > +
> >  config I2C_PCA_ISA
> >  	tristate "PCA9564/PCA9665 on an ISA bus"
> >  	depends on ISA
> 
> [snip]
> 
> > +
> > +struct mlxcpld_i2c_priv {
> > +	struct i2c_adapter adap;
> > +	u16 dev_id;
> > +	u16 base_addr;
> > +	struct mutex lock;
> > +	struct mlxcpld_i2c_curr_transf xfer;
> > +	struct device *dev;
> > +};
> > +struct platform_device *mlxcpld_i2c_plat_dev;
> 
> This global variable is unused.
> 
> And it emits several static check warnings:
> 
> checkpatch --strict:
> 
> CHECK: Please use a blank line after function/struct/union/enum declarations
> #324: FILE: drivers/i2c/busses/i2c-mlxcpld.c:110:
> +};
> +struct platform_device *mlxcpld_i2c_plat_dev;
> 
> smatch:
> 
>   i2c-mlxcpld.c:110:24: warning: symbol 'mlxcpld_i2c_plat_dev' was not
> declared. Should it be static?
> 
> Please remove it as unused.
> 
> [snip]
> 
> > +
> > +/* Check validity of current i2c message and all transfer.
> > + * Calculate also coomon length of all i2c messages in transfer.
> > + */
> > +static int mlxcpld_i2c_invalid_len(struct mlxcpld_i2c_priv *priv,
> > +				   const struct i2c_msg *msg, u8 *comm_len) {
> > +	u8 max_len = msg->flags == I2C_M_RD ? MLXCPLD_I2C_DATA_REG_SZ -
> > +		     MLXCPLD_I2C_MAX_ADDR_LEN :
> MLXCPLD_I2C_DATA_REG_SZ;
> > +
> > +	if (msg->len < 0 || msg->len > max_len)
> 
> Compiler W=1 level warning:
> 
> i2c-mlxcpld.c: In function 'mlxcpld_i2c_invalid_len':
> i2c-mlxcpld.c:201:15: warning: comparison is always false due to limited range
> of data type [-Wtype-limits]
>   if (msg->len < 0 || msg->len > max_len)
>                ^
> 
> msg->len is unsigned.
> 
> > +		return -EINVAL;
> > +
> > +	*comm_len += msg->len;
> > +	if (*comm_len > MLXCPLD_I2C_DATA_REG_SZ)
> > +		return -EINVAL;
> > +
> > +	return 0;
> > +}
> > +
> > +/* Check validity of received i2c messages parameters.
> > + *  Returns 0 if OK, other - in case of invalid parameters
> > + *  or common length of data that should be passed to CPLD
> 
> Sorry for nitpicking, you may consider to remove one of two spaces after
> asterisks above.
> 
> [snip]
> 
> > +
> > +static void mlxcpld_i2c_set_transf_data(struct mlxcpld_i2c_priv *priv,
> > +					struct i2c_msg *msgs, int num,
> > +					u8 comm_len)
> > +{
> > +	priv->xfer.msg = msgs;
> > +	priv->xfer.msg_num = num;
> > +
> > +	/*
> > +	 * All upper layers currently are never use transfer with more than
> > +	 * 2 messages. Actually, it's also not so relevant in Mellanox systems
> > +	 * because of HW limitation. Max size of transfer is o more than 20B
> > +	 * in current x86 LPCI2C bridge.
> > +	 */
> > +	priv->xfer.cmd = (msgs[num - 1].flags & I2C_M_RD);
> 
> Round brackets are not needed above.
> 
> > +
> > +	if (priv->xfer.cmd == I2C_M_RD && comm_len != msgs[0].len) {
> > +		priv->xfer.addr_width = msgs[0].len;
> > +		priv->xfer.data_len = comm_len - priv->xfer.addr_width;
> > +	} else {
> > +		priv->xfer.addr_width = 0;
> > +		priv->xfer.data_len = comm_len;
> > +	}
> > +}
> > +
> 
> [snip]
> 
> > +/*
> > + * Wait for master transfer to complete.
> > + * It puts current process to sleep until we get interrupt or timeout expires.
> > + * Returns the number of transferred or read bytes or error (<0).
> > + */
> > +static int mlxcpld_i2c_wait_for_tc(struct mlxcpld_i2c_priv *priv) {
> > +	int status, i = 1, timeout = 0;
> 
> Please remove the assignment of 'i' variable, after update it is not needed
> anymore.
> 
> > +	u8 datalen;
> > +
> > +	do {
> > +		usleep_range(MLXCPLD_I2C_POLL_TIME / 2,
> MLXCPLD_I2C_POLL_TIME);
> > +		if (!mlxcpld_i2c_check_status(priv, &status))
> > +			break;
> > +		timeout += MLXCPLD_I2C_POLL_TIME;
> > +	} while (status == 0 && timeout < MLXCPLD_I2C_XFER_TO);
> > +
> > +	switch (status) {
> > +	case MLXCPLD_LPCI2C_NO_IND:
> > +		return -ETIMEDOUT;
> > +
> > +	case MLXCPLD_LPCI2C_ACK_IND:
> > +		if (priv->xfer.cmd != I2C_M_RD)
> > +			return (priv->xfer.addr_width + priv->xfer.data_len);
> > +
> > +		if (priv->xfer.msg_num == 1)
> > +			i = 0;
> > +		else
> > +			i = 1;
> > +
> > +		if (!priv->xfer.msg[i].buf)
> > +			return -EINVAL;
> > +
> > +		/*
> > +		 * Actual read data len will be always the same as
> > +		 * requested len. 0xff (line pull-up) will be returned
> > +		 * if slave has no data to return. Thus don't read
> > +		 * MLXCPLD_LPCI2C_NUM_DAT_REG reg from CPLD.
> > +		 */
> > +		datalen = priv->xfer.data_len;
> > +
> > +		mlxcpld_i2c_read_comm(priv, MLXCPLD_LPCI2C_DATA_REG,
> > +				      priv->xfer.msg[i].buf, datalen);
> > +
> > +		return datalen;
> > +
> > +	case MLXCPLD_LPCI2C_NACK_IND:
> > +		return -EAGAIN;
> > +
> > +	default:
> > +		return -EINVAL;
> > +	}
> > +}
> > +
> 
> [snip]
> 
> > +
> > +/* Generic lpc-i2c transfer.
> 
> Just for my information, how do you unwrap 'lpc' acronym found in the code?
> 

On Mellanox systems I2C controller logic is implemented in Lattice CPLD device.
This CPLD device is connected to CPU through LPC bus, which Intel's Low Pin Count bus:
http://www.intel.com/design/chipsets/industry/25128901.pdf
So, CPLD provides some sort of LPC to I2C bridging.

> > + * Returns the number of processed messages or error (<0).
> > + */
> > +static int mlxcpld_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs,
> > +			    int num)
> > +{
> > +	struct mlxcpld_i2c_priv *priv = i2c_get_adapdata(adap);
> > +	u8 comm_len = 0;
> > +	int err;
> > +
> > +	err = mlxcpld_i2c_check_msg_params(priv, msgs, num, &comm_len);
> > +	if (err) {
> > +		dev_err(priv->dev, "Incorrect message\n");
> > +		return err;
> > +	}
> > +
> > +	/* Check bus state */
> > +	if (mlxcpld_i2c_wait_for_free(priv)) {
> > +		dev_err(priv->dev, "LPCI2C bridge is busy\n");
> > +
> > +		/*
> > +		 * Usually it means something serious has happened.
> > +		 * We can not have unfinished previous transfer
> > +		 * so it doesn't make any sense to try to stop it.
> > +		 * Probably we were not able to recover from the
> > +		 * previous error.
> > +		 * The only reasonable thing - is soft reset.
> > +		 */
> > +		mlxcpld_i2c_reset(priv);
> > +		if (mlxcpld_i2c_check_busy(priv)) {
> > +			dev_err(priv->dev, "LPCI2C bridge is busy after
> reset\n");
> > +			return -EIO;
> > +		}
> > +	}
> > +
> > +	mlxcpld_i2c_set_transf_data(priv, msgs, num, comm_len);
> > +
> > +	mutex_lock(&priv->lock);
> > +
> > +	/* Do real transfer. Can't fail */
> > +	mlxcpld_i2c_xfer_msg(priv);
> 
> Please add an empty line here to improve readability.
> 
> > +	/* Wait for transaction complete */
> > +	err = mlxcpld_i2c_wait_for_tc(priv);
> > +
> > +	mutex_unlock(&priv->lock);
> > +
> > +	return err < 0 ? err : num;
> > +}
> > +
> > +static u32 mlxcpld_i2c_func(struct i2c_adapter *adap) {
> > +	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL |
> > +I2C_FUNC_SMBUS_BLOCK_DATA; }
> > +
> > +static const struct i2c_algorithm mlxcpld_i2c_algo = {
> > +	.master_xfer	= mlxcpld_i2c_xfer,
> > +	.functionality	= mlxcpld_i2c_func
> > +};
> > +
> > +static struct i2c_adapter mlxcpld_i2c_adapter = {
> > +	.owner          = THIS_MODULE,
> > +	.name           = "i2c-mlxcpld",
> > +	.class          = I2C_CLASS_HWMON | I2C_CLASS_SPD,
> > +	.algo           = &mlxcpld_i2c_algo,
> > +};
> > +
> > +static int mlxcpld_i2c_probe(struct platform_device *pdev) {
> > +	struct mlxcpld_i2c_priv *priv;
> > +	int err;
> > +
> > +	priv = devm_kzalloc(&pdev->dev, sizeof(struct mlxcpld_i2c_priv),
> > +			    GFP_KERNEL);
> > +	if (!priv)
> > +		return -ENOMEM;
> > +
> > +	mutex_init(&priv->lock);
> > +	platform_set_drvdata(pdev, priv);
> > +
> > +	priv->dev = &pdev->dev;
> > +
> > +	/* Register with i2c layer */
> > +	priv->adap = mlxcpld_i2c_adapter;
> > +	priv->adap.dev.parent = &pdev->dev;
> > +	priv->adap.retries = MLXCPLD_I2C_RETR_NUM;
> > +	priv->adap.nr = MLXCPLD_I2C_BUS_NUM;
> 
> So since you want to allow the registration of only one such controller on a
> board (by the way in the opposite case it is also expected that "struct
> i2c_adapter" is allocated dynamically to avoid rewriting of data), you probably
> know the good reason behind it.
> 
> But in that case you may consider to move .retries and .nr instantiation to the
> "static struct i2c_adapter" declaration above.
> 
> > +	priv->adap.timeout = usecs_to_jiffies(MLXCPLD_I2C_XFER_TO);
> > +	priv->base_addr = MLXPLAT_CPLD_LPC_I2C_BASE_ADDR;
> > +	i2c_set_adapdata(&priv->adap, priv);
> > +
> > +	err = i2c_add_numbered_adapter(&priv->adap);
> > +	if (err) {
> > +		dev_err(&pdev->dev, "Failed to add %s adapter (%d)\n",
> > +			MLXCPLD_I2C_DEVICE_NAME, err);
> > +		goto fail_adapter;
> > +	}
> > +
> > +	return 0;
> > +
> > +fail_adapter:
> > +	mutex_destroy(&priv->lock);
> > +	return err;
> > +}
> > +
> > +static int mlxcpld_i2c_remove(struct platform_device *pdev) {
> > +	struct mlxcpld_i2c_priv *priv = platform_get_drvdata(pdev);
> > +
> > +	i2c_del_adapter(&priv->adap);
> > +	mutex_destroy(&priv->lock);
> > +
> > +	return 0;
> > +}
> > +
> > +static struct platform_driver mlxcpld_i2c_driver = {
> > +	.probe		= mlxcpld_i2c_probe,
> > +	.remove		= mlxcpld_i2c_remove,
> > +	.driver = {
> > +		.name = MLXCPLD_I2C_DEVICE_NAME,
> > +	},
> > +};
> > +
> > +module_platform_driver(mlxcpld_i2c_driver);
> > +
> > +MODULE_AUTHOR("Michael Shych <michaels@xxxxxxxxxxxx>");
> > +MODULE_DESCRIPTION("Mellanox I2C-CPLD controller driver");
> > +MODULE_LICENSE("Dual BSD/GPL"); MODULE_ALIAS("platform:i2c-
> mlxcpld");
> >
> 
> Generally the driver looks good and simple.
> 
> --
> With best wishes,
> Vladimir

Thank you very much,
Vadim.