00:24 <naoki> I'm starting to worry about whether I should really add rockchip,rk3582. But it's not necessary to do the same thing.

00:24 <naoki> ^... do as same as i.MX

00:37 <naoki> Btw, why v(q)mmc-supply is not written in sdhci?

03:32 <naoki> How can I specify a regulator without a specific name?

03:34 <naoki> regulator-XvY-0, regulator-XvY-1, ...?

04:07 <naoki> I decided to use the -[0-9]+ suffix

04:09 <naoki> I will not sort the regulators in pmic by node name

04:15 <naoki> I will not change phandle and regulator-name

04:29 <naoki> ... I guess the regulator name of pmic(rk8xx) cannot be changed...

08:03 <naoki> https://github.com/RadxaNaoki/linux/blob/d61808b5c7aca74539ec09d16a76af01e17e694d/arch/arm64/boot/dts/rockchip/rk3582-radxa-e52c.dts Is this correct naming...?

08:44 <Kwiboo> naoki: for regulator node names I tried to use regulator-XvY-<name excluding voltage> for e.g. https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/arm64/boot/dts/rockchip/rk3568-rock-3b.dts thought it best matched both scematic and regulator node name pattern

08:47 <Kwiboo> also wonder if rk3582 boards should possible have compatible: <board>, rockchip,rk3582, rockchip,rk3588s ? they are basically just that, rk3588s with few ip-cores disabled/removed/bad/faulty

09:13 <naoki> Kwiboo: "But usually more than one suffix, vcc+3v3+pcie_2x1l2, means you created a very specific name."

09:16 <naoki> change for irq-gic-v3-its.c is rejected

09:17 <naoki> I am not the person who advocated for the introduction of rk3582, so I have not thought about it any further.

09:18 <naoki> Is rockchip,rk3588s not enough?

09:40 <naoki> I could not understand "Please use the relevant property for that purpose ("dma-noncoherent") at the distributor and ITS levels. We're not adding extra compatibles for this anymore, and you might as well fix the core dtsi to expose such property."

10:00 <Kwiboo> naoki: could be possible that the dma-noncoherent prop was added to dt-binding after the workaround that checked the rk3588 compatible was added to driver

10:03 <Kwiboo> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/Documentation/devicetree/bindings/interrupt-controller/arm,gic-v3.yaml?id=5e5c636c69bdba04033161bbb111fbb6f1f6661e

10:03 <Kwiboo> https://lore.kernel.org/all/20231006125929.48591-2-lpieralisi@kernel.org/

10:10 <Kwiboo> so probably the rk3588 base dtsi can probably be updated to use the new prop and the code should not need to be changed