Re: [PATCH v11] i2c: virtio: add a virtio i2c frontend driver

[Viresh Kumar]

On 01-07-21, 11:24, Jie Deng wrote:
> Changes v10 -> v11
> 	- Remove vi->adap.class = I2C_CLASS_DEPRECATED.
> 	- Use #ifdef CONFIG_PM_SLEEP to replace the "__maybe_unused".
> 	- Remove "struct mutex lock" in "struct virtio_i2c".
> 	- Support zero-length request.
> 	- Remove unnecessary logs.
> 	- Remove vi->adap.timeout = HZ / 10, just use the default value.
> 	- Use BIT(0) to define VIRTIO_I2C_FLAGS_FAIL_NEXT.
> 	- Add the virtio_device index to adapter's naming mechanism.

Thanks Jie.

I hope you are going to send a fix for specification as well (for the
zero-length request) ?

> diff --git a/drivers/i2c/busses/i2c-virtio.c b/drivers/i2c/busses/i2c-virtio.c
> +static int virtio_i2c_send_reqs(struct virtqueue *vq,
> +				struct virtio_i2c_req *reqs,
> +				struct i2c_msg *msgs, int nr)
> +{
> +	struct scatterlist *sgs[3], out_hdr, msg_buf, in_hdr;
> +	int i, outcnt, incnt, err = 0;
> +
> +	for (i = 0; i < nr; i++) {
> +		/*
> +		 * Only 7-bit mode supported for this moment. For the address format,
> +		 * Please check the Virtio I2C Specification.
> +		 */
> +		reqs[i].out_hdr.addr = cpu_to_le16(msgs[i].addr << 1);
> +
> +		if (i != nr - 1)
> +			reqs[i].out_hdr.flags = cpu_to_le32(VIRTIO_I2C_FLAGS_FAIL_NEXT);
> +
> +		outcnt = incnt = 0;
> +		sg_init_one(&out_hdr, &reqs[i].out_hdr, sizeof(reqs[i].out_hdr));
> +		sgs[outcnt++] = &out_hdr;
> +
> +		reqs[i].buf = i2c_get_dma_safe_msg_buf(&msgs[i], 1);
> +		if (!reqs[i].buf)
> +			break;
> +
> +		sg_init_one(&msg_buf, reqs[i].buf, msgs[i].len);

The len can be zero here for zero-length transfers.

> +
> +		if (msgs[i].flags & I2C_M_RD)
> +			sgs[outcnt + incnt++] = &msg_buf;
> +		else
> +			sgs[outcnt++] = &msg_buf;
> +
> +		sg_init_one(&in_hdr, &reqs[i].in_hdr, sizeof(reqs[i].in_hdr));
> +		sgs[outcnt + incnt++] = &in_hdr;

Why are we still sending the msg_buf if the length is 0? Sending the
buffer makes sense if you have some data to send, but otherwise it is
just an extra sg element, which isn't required to be sent.

> +
> +		err = virtqueue_add_sgs(vq, sgs, outcnt, incnt, &reqs[i], GFP_KERNEL);
> +		if (err < 0) {
> +			i2c_put_dma_safe_msg_buf(reqs[i].buf, &msgs[i], false);
> +			break;
> +		}
> +	}
> +
> +	return i;

I just noticed this now, but this function even tries to send data
partially, which isn't right. If the caller (i2c device's driver)
calls this for 5 struct i2c_msg instances, then all 5 need to get
through or none.. where as we try to send as many as possible here.

This looks broken to me. Rather return an error value here on success,
or make it complete failure.

Though to be fair I see i2c-core also returns number of messages
processed from i2c_transfer().

Wolfram, what's expected here ? Shouldn't all message transfer or
none?

> +#ifdef CONFIG_PM_SLEEP
> +static int virtio_i2c_freeze(struct virtio_device *vdev)
> +{
> +	virtio_i2c_del_vqs(vdev);
> +	return 0;
> +}
> +
> +static int virtio_i2c_restore(struct virtio_device *vdev)
> +{
> +	return virtio_i2c_setup_vqs(vdev->priv);
> +}
> +#endif
> +
> +static struct virtio_driver virtio_i2c_driver = {
> +	.id_table	= id_table,
> +	.probe		= virtio_i2c_probe,
> +	.remove		= virtio_i2c_remove,
> +	.driver	= {
> +		.name	= "i2c_virtio",
> +	},
> +#ifdef CONFIG_PM_SLEEP

You could avoid this pair of ifdef by creating dummy versions of below
routines for !CONFIG_PM_SLEEP case. Up to you.

> +	.freeze = virtio_i2c_freeze,
> +	.restore = virtio_i2c_restore,
> +#endif
> +};

-- 
viresh