Re: [PATCH] Input: apm-power - Rename to apm-power-input

[Dmitry Torokhov <dmitry.torokhov@gmail.com>]

Hi Nathan,

On Mon, Nov 09, 2026 at 01:25:26AM -0500, Nathan Chancellor wrote:
> When scripts/modules-check.sh is fixed to normalize the module's name
> like name-fix-token does in scripts/Makefile.lib before checking that it
> is unique [1], there is an error due to drivers/input/apm-power.ko and
> drivers/power/supply/apm_power.ko having the same module name.
...
> Rename drivers/input/apm-power.c to apm-power-input.c to make the module
> name unique and clear up the error.
...
> We would like to take [1] for 7.5 in kbuild-next, so we would appreciate
> an ack to take this directly or a stable branch that we can pull in.

drivers/input/apm-power.c only reacts to EV_PWR/KEY_SUSPEND input events
to queue APM_USER_SUSPEND via APM emulation. The only in-tree users that
generate those events are the legacy Sharp Zaurus ARM machines
(spitz/borzoi/akita on PXA alongside SHARPSL_PM, and collie on SA1100
via locomokbd). None of them were converted to Device Tree, both
mach-pxa board files and mach-sa1100 were marked deprecated in 7.3, and
Arnd (added to Cc) already has their removal queued up for 7.5 after
holding it back from 7.4 LTS:

https://lore.kernel.org/r/e4da2f9d-d3b9-4267-9b64-b0f2cb43402a@app.fastmail.com

None of the remaining ARM platforms or other APM_EMULATION targets emit
EV_PWR input events or enable CONFIG_INPUT_APMPOWER (and pxa_defconfig
already dropped CONFIG_PXA_SHARPSL in commit cc7852314abd, leaving
CONFIG_INPUT_APMPOWER inert there).

Since you are targeting 7.5 for the modules-check change anyway, let's
just drop drivers/input/apm-power.c altogether instead of renaming a
module whose last users are going away in the same cycle. I will send
out a removal patch shortly, and either you can take it directly through
kbuild-next or I can put it on an immutable branch for you to pull.

Thanks.

-- 
Dmitry