| Message ID | d4218947e354aee45589acb1060ce41727a26750.1681646904.git.jahau@rocketmail.com |
|---|---|
| State | New |
| Headers | show |
| Series | [v2,1/9] mfd: rt5033: Drop rt5033-battery sub-device | expand |
On 16/04/2023 14:44, Jakob Hauser wrote: > Lowercase is preferred for node names. > This will break all existing users. In-tree and out-of-tree. Where is the binding update? Best regards, Krzysztof
Hi Krzysztof, On 16.04.23 20:32, Krzysztof Kozlowski wrote: > On 16/04/2023 14:44, Jakob Hauser wrote: >> Lowercase is preferred for node names. > > This will break all existing users. In-tree and out-of-tree. Where is > the binding update? In my reply to Rob's comments in v1 I was pointing out that this will affect an existing driver. There was no reaction. As far as I can see, there is no in-tree usage yet. Though I can't tell about out-of-tree usage. Although if there is, adding the rt5033-charger driver might already causes the need for changes. Well, to stay on the safe side, I'll drop this patch in v3 and will change the bindings (patch 9) back to uppercase. Kind regards, Jakob
On 18/04/2023 23:24, Jakob Hauser wrote: > Hi Krzysztof, > > On 16.04.23 20:32, Krzysztof Kozlowski wrote: >> On 16/04/2023 14:44, Jakob Hauser wrote: >>> Lowercase is preferred for node names. >> >> This will break all existing users. In-tree and out-of-tree. Where is >> the binding update? > > In my reply to Rob's comments in v1 I was pointing out that this will > affect an existing driver. There was no reaction. > > As far as I can see, there is no in-tree usage yet. Though I can't tell > about out-of-tree usage. Although if there is, adding the rt5033-charger > driver might already causes the need for changes. > > Well, to stay on the safe side, I'll drop this patch in v3 and will > change the bindings (patch 9) back to uppercase. > Your v1 binding patch did not explain that you document existing ABI, so you got comments like for a new binding. This is not really new binding, is it? Best regards, Krzysztof
Hi Krzysztof, On 19.04.23 10:40, Krzysztof Kozlowski wrote: > On 18/04/2023 23:24, Jakob Hauser wrote: >> On 16.04.23 20:32, Krzysztof Kozlowski wrote: >>> On 16/04/2023 14:44, Jakob Hauser wrote: >>>> Lowercase is preferred for node names. >>> >>> This will break all existing users. In-tree and out-of-tree. Where is >>> the binding update? >> >> In my reply to Rob's comments in v1 I was pointing out that this will >> affect an existing driver. There was no reaction. >> >> As far as I can see, there is no in-tree usage yet. Though I can't tell >> about out-of-tree usage. Although if there is, adding the rt5033-charger >> driver might already causes the need for changes. >> >> Well, to stay on the safe side, I'll drop this patch in v3 and will >> change the bindings (patch 9) back to uppercase. > > Your v1 binding patch did not explain that you document existing ABI, so > you got comments like for a new binding. This is not really new binding, > is it? The bindings for the mfd and regulator are new, even though the drivers are already existing. Sorry for not being clear on this in v1. This is due to historic reasons of the old patchset, more information on that further down. The current situation in the mainline kernel is as follows. drivers ------- rt5033: drivers/mfd/rt5033.c rt5033-regulator: drivers/regulator/rt5033-regulator.c rt5033-charger: not existent rt5033-battery: drivers/power/supply/rt5033_battery.c rt5033-leds: not existent bindings -------- rt5033: not existent rt5033-regulator: not existent rt5033-charger: not existent rt5033-battery: .../bindings/power/supply/richtek,rt5033-battery.yaml rt5033-leds: not existent The reason for that discrepancy: RT5033 mfd, regulator and fuelgauge drivers were applied but charger & documentation didn't make [1]. They were submitted again but it phased out at that point, last known state is [2]. The LEDs are also a phased out patchset [3]. The fuelgauge binding was added not so long ago by Stephan [4]. [1] https://lore.kernel.org/all/1421899601-19327-1-git-send-email-beomho.seo@samsung.com/T/#t [2] https://lore.kernel.org/lkml/1425864191-4121-1-git-send-email-beomho.seo@samsung.com/T/#t [3] https://lore.kernel.org/linux-leds/1448446948-13729-1-git-send-email-ingi2.kim@samsung.com/T/#u [4] https://lore.kernel.org/linux-devicetree/20210517105113.240379-1-stephan@gerhold.net/T/#m197b5719a5d37b17ba4ff9f3b3ff4bd4efcda71e Kind regards, Jakob
diff --git a/drivers/regulator/rt5033-regulator.c b/drivers/regulator/rt5033-regulator.c index 2ba74f205543..dee272dc81df 100644 --- a/drivers/regulator/rt5033-regulator.c +++ b/drivers/regulator/rt5033-regulator.c @@ -41,8 +41,8 @@ static const struct regulator_ops rt5033_buck_ops = { static const struct regulator_desc rt5033_supported_regulators[] = { [RT5033_BUCK] = { - .name = "BUCK", - .of_match = of_match_ptr("BUCK"), + .name = "buck", + .of_match = of_match_ptr("buck"), .regulators_node = of_match_ptr("regulators"), .id = RT5033_BUCK, .ops = &rt5033_buck_ops, @@ -57,8 +57,8 @@ static const struct regulator_desc rt5033_supported_regulators[] = { .vsel_mask = RT5033_BUCK_CTRL_MASK, }, [RT5033_LDO] = { - .name = "LDO", - .of_match = of_match_ptr("LDO"), + .name = "ldo", + .of_match = of_match_ptr("ldo"), .regulators_node = of_match_ptr("regulators"), .id = RT5033_LDO, .ops = &rt5033_buck_ops, @@ -73,8 +73,8 @@ static const struct regulator_desc rt5033_supported_regulators[] = { .vsel_mask = RT5033_LDO_CTRL_MASK, }, [RT5033_SAFE_LDO] = { - .name = "SAFE_LDO", - .of_match = of_match_ptr("SAFE_LDO"), + .name = "safe_ldo", + .of_match = of_match_ptr("safe_ldo"), .regulators_node = of_match_ptr("regulators"), .id = RT5033_SAFE_LDO, .ops = &rt5033_safe_ldo_ops,
Lowercase is preferred for node names. Cc: Beomho Seo <beomho.seo@samsung.com> Cc: Chanwoo Choi <cw00.choi@samsung.com> Cc: Axel Lin <axel.lin@ingics.com> Cc: ChiYuan Huang <cy_huang@richtek.com> Signed-off-by: Jakob Hauser <jahau@rocketmail.com> --- drivers/regulator/rt5033-regulator.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)