Messages in this thread | | | Subject | Re: [PATCH] thermal/core: Make 'forced_passive' as obsolete candidate | From | Daniel Lezcano <> | Date | Sat, 12 Dec 2020 10:11:31 +0100 |
| |
On 12/12/2020 04:50, Matthew Garrett wrote: > On Fri, Dec 11, 2020 at 02:17:55PM +0100, Daniel Lezcano wrote: >> On 08/12/2020 16:30, Daniel Lezcano wrote: >>> The passive file in sysfs forces the usage of a passive trip point set >>> by the userspace when a broken BIOS does not provide the mitigation >>> temperature for such thermal zone. The hardware evolved a lot since >>> 2008 as a good thermal management is no longer an option. >>> >>> Linux on the other side also provides now a way to load fixed ACPI >>> table via the option ACPI_TABLE_UPGRADE, so additionnal trip point >>> could be added there. >>> >>> Set the option obsolete and plan to remove it, so the corresponding >>> code can be removed from the core code and allow more cleanups the >>> thermal framework deserves. >>> >>> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> >>> --- >> >> Is there any concern about this change ? > > Yes - what's the reason to do so?
I'm cleaning up the thermal core code, so questioning every old ABI.
> The code isn't specific to ACPI, > so being able to override ACPI tables doesn't seem to justify it.
I agree, the code is no specific to ACPI.
What non-ACPI architecture, without device tree or platform data would need the 'passive' option today ?
-- <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs
Follow Linaro: <http://www.facebook.com/pages/Linaro> Facebook | <http://twitter.com/#!/linaroorg> Twitter | <http://www.linaro.org/linaro-blog/> Blog
| |