| Message ID | 20241217204526.1010989-3-s-ramamoorthy@ti.com |
|---|---|
| State | New |
| Headers | show |
| Series | TPS65219 Regulator Driver Cleanup Series | expand |
On Tue, Dec 17, 2024 at 02:45:24PM -0600, Shree Ramamoorthy wrote: > Follow the same naming convention in tps6594-regulator.c with > tpsxxx-regulator instead of tpsxxx-pmic. > static struct platform_driver tps65219_regulator_driver = { > .driver = { > - .name = "tps65219-pmic", > + .name = "tps65219-regulator", > .probe_type = PROBE_PREFER_ASYNCHRONOUS, This isn't just a naming convention thing AFACIT, the MFD registers the new name so the driver wouldn't previously have loaded. How did this ever work?
On Tue, Dec 17, 2024 at 03:16:44PM -0600, Andrew Davis wrote: > On 12/17/24 3:01 PM, Mark Brown wrote: > > This isn't just a naming convention thing AFACIT, the MFD registers the > > new name so the driver wouldn't previously have loaded. How did this > > ever work? > It matches based on the platform device .id_table[0] which does have > "tps65219-regulator" listed, the driver .name itself is only used > as a last fallback when there is no .id_table. Ah, there's not enough context in the diff to show the ID table.
diff --git a/drivers/regulator/tps65219-regulator.c b/drivers/regulator/tps65219-regulator.c index 14d843a99d0f..5decf69f327a 100644 --- a/drivers/regulator/tps65219-regulator.c +++ b/drivers/regulator/tps65219-regulator.c @@ -379,7 +379,7 @@ MODULE_DEVICE_TABLE(platform, tps65219_regulator_id_table); static struct platform_driver tps65219_regulator_driver = { .driver = { - .name = "tps65219-pmic", + .name = "tps65219-regulator", .probe_type = PROBE_PREFER_ASYNCHRONOUS, }, .probe = tps65219_regulator_probe,
Follow the same naming convention in tps6594-regulator.c with tpsxxx-regulator instead of tpsxxx-pmic. Signed-off-by: Shree Ramamoorthy <s-ramamoorthy@ti.com> --- drivers/regulator/tps65219-regulator.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)