Re: [PATCH 1/1] ARM: dts: Use more descriptive names for Exynos5420 PDs

From: Sergei Shtylyov
Date: Tue Feb 10 2015 - 06:55:24 EST


On 2/10/2015 2:46 PM, Krzysztof Kozlowski wrote:

All the device nodes for the Exynos5420 power-domains have a quite
generic "power-domain" name.

And this is in conformance to the ePAPR standard.

True, I forgot that the ePAPR recommends that the node names should be
somewhat generic but OTOH this is the only Exynos DTSI file that follows
the standard for the power domain device nodes. All other Exynos DTSI
use a prefix to differentiate between each power domain.

So in case of an error, the Exynos PD
driver shows the following (not very useful) message:

"Power domain power-domain disable failed"

Why not fix the message instead to use the full device name?

Well, the full node name is also not very useful IMHO since you have
to check the DTSI or SoC manual to map the device node unit-address to
the corresponding power domain.

I used $subject when debugging an HDMI issue and instead of dropping
it, I just posted it in case someone considered useful. I don't really
mind if the patch is nacked / not picked.

Additionally (on Arndale Octa):

$ cat /sys/kernel/debug/pm_genpd/pm_genpd_summary
domain status slaves
/device runtime status
power-domain on
/devices/platform/amba/3880000.adma suspended
power-domain off
power-domain off
power-domain off
power-domain off
power-domain off

This really is not helpful. From the power domain debugfs code it is
complicated to extract of_node of power domain.

You shouldn't need it.

It is easier to print
the name of power domain. But wait... all names are the same! :) So
why do we have the name in the first place?

I'm not sure why the full platform device names aren't printed -- they should all be different.

WBR, Sergei

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at