Re: [PATCH v5 2/8] i2c: add I2C Address Translator (ATR) support
From: Tomi Valkeinen
Date: Thu Dec 08 2022 - 11:01:32 EST
Hi Andy,
On 08/12/2022 14:53, Andy Shevchenko wrote:
On Thu, Dec 08, 2022 at 12:40:00PM +0200, Tomi Valkeinen wrote:
From: Luca Ceresoli <luca@xxxxxxxxxxxxxxxx>
An ATR is a device that looks similar to an i2c-mux: it has an I2C
slave "upstream" port and N master "downstream" ports, and forwards
transactions from upstream to the appropriate downstream port. But is
is different in that the forwarded transaction has a different slave
address. The address used on the upstream bus is called the "alias"
and is (potentially) different from the physical slave address of the
downstream chip.
Add a helper file (just like i2c-mux.c for a mux or switch) to allow
implementing ATR features in a device driver. The helper takes care or
adapter creation/destruction and translates addresses at each transaction.
Besides comments given against diff between series versions, see below.
...
+static int i2c_atr_attach_client(struct i2c_adapter *adapter,
+ const struct i2c_board_info *info,
+ const struct i2c_client *client)
+{
+ struct i2c_atr_chan *chan = adapter->algo_data;
+ struct i2c_atr *atr = chan->atr;
+ struct i2c_atr_cli2alias_pair *c2a;
+ u16 alias_id;
+ int ret;
+
+ c2a = kzalloc(sizeof(*c2a), GFP_KERNEL);
+ if (!c2a)
+ return -ENOMEM;
+
+ ret = atr->ops->attach_client(atr, chan->chan_id, info, client,
+ &alias_id);
+ if (ret)
+ goto err_free;
+ if (alias_id == 0) {
+ ret = -EINVAL;
I'm wondering why attach_client can't return this error and provide a guarantee
that if no error, the alias_id is never be 0?
I think that's a valid point. I see no reason to check for alias_id == 0
here.
+ goto err_free;
+ }
+
+ c2a->client = client;
+ c2a->alias = alias_id;
+ list_add(&c2a->node, &chan->alias_list);
+
+ return 0;
+
+err_free:
+ kfree(c2a);
+ return ret;
+}
...
+ if (bus_handle) {
+ device_set_node(&chan->adap.dev, fwnode_handle_get(bus_handle));
I believe the correct way, while above still works, is
device_set_node(&chan->adap.dev, bus_handle);
fwnode_handle_get(dev_fwnode(&chan->adap.dev));
Hmm, why is that correct? Shouldn't you give device_set_node() an fwnode
that has been referenced?
But I agree that this looks a bit verbose. And...
+ } else {
+ struct fwnode_handle *atr_node;
+ struct fwnode_handle *child;
+ u32 reg;
+
+ atr_node = device_get_named_child_node(dev, "i2c-atr");
+
+ fwnode_for_each_child_node(atr_node, child) {
+ ret = fwnode_property_read_u32(child, "reg", ®);
+ if (ret)
+ continue;
+ if (chan_id == reg)
+ break;
+ }
+
+ device_set_node(&chan->adap.dev, child);
...OTOH, you set node with bumped reference here. So I leave all this to
the maintainers.
+ fwnode_handle_put(atr_node);
+ }
+ ret = i2c_add_adapter(&chan->adap);
+ if (ret) {
+ dev_err(dev, "failed to add atr-adapter %u (error=%d)\n",
+ chan_id, ret);
+ goto err_mutex_destroy;
+ }
+
+ snprintf(symlink_name, sizeof(symlink_name), "channel-%u",
+ chan->chan_id);
+
+ ret = sysfs_create_link(&chan->adap.dev.kobj, &dev->kobj, "atr_device");
+ if (ret)
+ dev_warn(dev, "can't create symlink to atr device\n");
+ ret = sysfs_create_link(&dev->kobj, &chan->adap.dev.kobj, symlink_name);
+ if (ret)
+ dev_warn(dev, "can't create symlink for channel %u\n", chan_id);
+
+ dev_dbg(dev, "Added ATR child bus %d\n", i2c_adapter_id(&chan->adap));
+
+ atr->adapter[chan_id] = &chan->adap;
+ return 0;
+
+err_mutex_destroy:
Now it's a bit misleading, wouldn't be better
err_put_fwnode:
?
Yes.
+ fwnode_handle_put(dev_fwnode(&chan->adap.dev));
+ mutex_destroy(&chan->orig_addrs_lock);
+ kfree(chan);
+ return ret;
+}
+EXPORT_SYMBOL_GPL(i2c_atr_add_adapter);
Wondering if we may put this into namespace from day 1.
Right, that's something I didn't look at all for v5. I have not heard
anyone else commenting about the NS, though. I could have a look at it
anyway, just to learn about NSes.
...
+/**
+ * i2c_atr_del_adapter - Remove a child ("downstream") I2C bus added by
+ * i2c_atr_del_adapter().
+ * @atr: The I2C ATR
+ * @chan_id: Index of the `adapter to be removed (0 .. max_adapters-1)
+ */
+void i2c_atr_del_adapter(struct i2c_atr *atr, u32 chan_id)
+{
+ char symlink_name[ATR_MAX_SYMLINK_LEN];
+
+ struct i2c_adapter *adap = atr->adapter[chan_id];
+ struct i2c_atr_chan *chan = adap->algo_data;
+ struct fwnode_handle *fwnode = dev_fwnode(&adap->dev);
+ struct device *dev = atr->dev;
+ if (!atr->adapter[chan_id]) {
Isn't it the same as
if (!adap)
?
Yes.
+ dev_err(dev, "Adapter %d does not exist\n", chan_id);
+ return;
+ }
+
+ dev_dbg(dev, "Removing ATR child bus %d\n", i2c_adapter_id(adap));
+
+ atr->adapter[chan_id] = NULL;
+
+ snprintf(symlink_name, sizeof(symlink_name), "channel-%u",
+ chan->chan_id);
+ sysfs_remove_link(&dev->kobj, symlink_name);
+ sysfs_remove_link(&chan->adap.dev.kobj, "atr_device");
+
+ i2c_del_adapter(adap);
+ fwnode_handle_put(fwnode);
+ mutex_destroy(&chan->orig_addrs_lock);
+ kfree(chan->orig_addrs);
+ kfree(chan);
+}
...
+struct i2c_atr {
+ /* private: internal use only */
What is private? The entire structure? Then why it's defined in
the include/linux/? Can't you make it opaque?
Good point, I see no reason to keep this in the public header.
i2c_atr_set/get_clientdata used it, but I can move their implementations
into the .c file.
+ struct i2c_adapter *parent;
+ struct device *dev;
+ const struct i2c_atr_ops *ops;
+
+ void *priv;
+
+ struct i2c_algorithm algo;
+ /* lock for the I2C bus segment (see struct i2c_lock_operations) */
+ struct mutex lock;
+ int max_adapters;
+
+ struct i2c_adapter *adapter[];
+};
...
+static inline void i2c_atr_set_clientdata(struct i2c_atr *atr, void *data)
+{
+ atr->priv = data;
+}
+
+static inline void *i2c_atr_get_clientdata(struct i2c_atr *atr)
+{
+ return atr->priv;
+}
The function names are misleading, because I would think this is about driver
data that has been set.
I would rather use name like
i2c_atr_get_priv()
i2c_atr_set_priv()
Indeed, set_clientdata is probably wrong. But i2c_atr_set_priv() sounds
like it's private to the i2c-atr itself. Maybe i2c_atr_set_driver_data?
Tomi