On Sat, 2024-09-28 at 17:05 +0100, Jonathan Cameron wrote: > On Mon, 16 Sep 2024 16:56:39 +0200 > Emil Gedenryd <emil.gedenryd@xxxxxxxx> wrote: > > > TI's opt3002 light sensor shares most properties with the opt3001 > > model, with the exception of supporting a wider spectrum range. > > > > Add support for TI's opt3002 by extending the TI opt3001 driver. > > > > Datasheet: https://www.ti.com/product/OPT3002 > > Signed-off-by: Emil Gedenryd <emil.gedenryd@xxxxxxxx> > Hi Emil, > > A few things inline, > > Thanks, > > Jonathan Hi Jonathan, Thank you for the comments, I'll start working on fixing them during the week. Best Regards, Emil > > > > > @@ -610,7 +704,7 @@ static int opt3001_read_id(struct opt3001 *opt) > > ret = i2c_smbus_read_word_swapped(opt->client, OPT3001_DEVICE_ID); > > if (ret < 0) { > > dev_err(opt->dev, "failed to read register %02x\n", > > - OPT3001_DEVICE_ID); > > + OPT3001_DEVICE_ID); > > Unrelated change so in theory should be in a separate patch but > meh, it's trivial so leave it here if you like. Good to know! I'll keep it since it's minor but I'll remember this for the future. > > > return ret; > > } > > > @@ -755,22 +850,25 @@ static int opt3001_probe(struct i2c_client *client) > > opt = iio_priv(iio); > > opt->client = client; > > opt->dev = dev; > > + opt->chip_info = device_get_match_data(&client->dev); > > Use the i2c specific way to to do this. > https://elixir.bootlin.com/linux/v6.11/source/drivers/i2c/i2c-core-base.c#L120 > i2c_get_match_data() because it will also handle falling back to matching > via the i2c_match_id() path against the old style match tables in a few > cases. Okay! > > > > mutex_init(&opt->lock); > > init_waitqueue_head(&opt->result_ready_queue); > > i2c_set_clientdata(client, iio); > > > > - ret = opt3001_read_id(opt); > > - if (ret) > > - return ret; > > + if (opt->chip_info->has_id) { > > + ret = opt3001_read_id(opt); > > + if (ret) > > + return ret; > > + } > > > > ret = opt3001_configure(opt); > > if (ret) > > return ret; > > > > iio->name = client->name; > > - iio->channels = opt3001_channels; > > - iio->num_channels = ARRAY_SIZE(opt3001_channels); > > + iio->channels = *opt->chip_info->channels; > > + iio->num_channels = ARRAY_SIZE(*opt->chip_info->channels); > This won't work as it has no way to perform a sizeof > through a pointer. > > Add a num_channels filed to your opt3001_chip_info structure > as then it can be ARRAY_SIZE(&opt3001_channels) which can work. Good catch and thanks for the suggestion, I'll do it that way! > > > iio->modes = INDIO_DIRECT_MODE; > > iio->info = &opt3001_info; > > > > @@ -825,14 +923,36 @@ static void opt3001_remove(struct i2c_client *client) > > } > > } > > > > +static const struct opt3001_chip_info opt3001_chip_information = { > > + .channels = &opt3001_channels, > > + .chan_type = IIO_LIGHT, > > + .scales = &opt3001_scales, > > + .factor_whole = 10, > > + .factor_integer = 1000, > > + .factor_decimal = 1000, > > + .has_id = true, > > +}; > > + > > +static const struct opt3001_chip_info opt3002_chip_information = { > > + .channels = &opt3002_channels, > > + .chan_type = IIO_INTENSITY, > > + .scales = &opt3002_scales, > > + .factor_whole = 12, > > + .factor_integer = 10, > > + .factor_decimal = 100000, > > + .has_id = false, > > +}; > > + > > static const struct i2c_device_id opt3001_id[] = { > > - { "opt3001" }, > > + { "opt3001", (kernel_ulong_t)&opt3001_chip_information }, > > + { "opt3002", (kernel_ulong_t)&opt3002_chip_information }, > > { } /* Terminating Entry */ > > }; > > MODULE_DEVICE_TABLE(i2c, opt3001_id); > > > > static const struct of_device_id opt3001_of_match[] = { > > - { .compatible = "ti,opt3001" }, > > + { .compatible = "ti,opt3001", .data = &opt3001_chip_information }, > > + { .compatible = "ti,opt3002", .data = &opt3002_chip_information }, > > { } > > }; > > MODULE_DEVICE_TABLE(of, opt3001_of_match); > > >
