Re: git: 427747fc16c9 - main - acpi_cpu: No longer disable idle during suspend
- In reply to: Adrian Chadd : "Re: git: 427747fc16c9 - main - acpi_cpu: No longer disable idle during suspend"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Thu, 12 Nov 2026 15:10:46 UTC
note - suspend/resume is now flakey on a couple of my laptops. I don't know
what combination of kern or drm-kmod changes have caused it.
please list out the set of hardware you're testing these changes on.
-adrian
On Wed, 11 Nov 2026 at 10:20, Adrian Chadd <adrian@freebsd.org> wrote:
> Hey err!
>
> I haven't tested this yet on my laptops (which all run ACPI.)
>
> Are y'all absolutely /sure/ that this can't possibly ever try to idle
> during suspend on ACPI platforms?
> And thus this was truly dead code?
>
> Would you mind listing out what hardware / config you're testing on when
> you do these changes?
>
>
>
> -adrian
>
>
> On Wed, 11 Nov 2026 at 02:17, Aymeric Wibo <obiwac@freebsd.org> wrote:
>
>> The branch main has been updated by obiwac:
>>
>> URL:
>> https://cgit.FreeBSD.org/src/commit/?id=427747fc16c9be5e4d4eb31a797d7854c4fd839f
>>
>> commit 427747fc16c9be5e4d4eb31a797d7854c4fd839f
>> Author: Aymeric Wibo <obiwac@FreeBSD.org>
>> AuthorDate: 2026-02-25 09:39:05 +0000
>> Commit: Aymeric Wibo <obiwac@FreeBSD.org>
>> CommitDate: 2026-11-11 10:15:46 +0000
>>
>> acpi_cpu: No longer disable idle during suspend
>>
>> This had been introduced in 3331373ce733 ("Disable CPU idle states
>> during suspend and reenable them during resume.") to work around
>> C-state
>> selection logic which would skip C-states which returned immediately,
>> considering them "broken".
>>
>> This workaround is incompatible with suspend-to-idle however, which
>> requires CPUs to be in their idle states, and is no longer relevant
>> either way since b0baaaaecf1a ("Avoid comparing negative signed to
>> positive unsignad values.") removed the C-state skipping logic the
>> previous change was working around.
>>
>> Reviewed by: olce
>> Sponsored by: The FreeBSD Foundation
>> Pull Request: https://ron-dev.freebsd.org/FreeBSD/src/pulls/71
>> Differential Revision: https://reviews.freebsd.org/D60649
>> ---
>> sys/dev/acpica/acpi_cpu.c | 30 ++----------------------------
>> 1 file changed, 2 insertions(+), 28 deletions(-)
>>
>> diff --git a/sys/dev/acpica/acpi_cpu.c b/sys/dev/acpica/acpi_cpu.c
>> index d3ac92f920ab..9a523611a7f9 100644
>> --- a/sys/dev/acpica/acpi_cpu.c
>> +++ b/sys/dev/acpica/acpi_cpu.c
>> @@ -159,8 +159,6 @@ ACPI_SERIAL_DECL(cpu, "ACPI CPU");
>>
>> static int acpi_cpu_probe(device_t dev);
>> static int acpi_cpu_attach(device_t dev);
>> -static int acpi_cpu_suspend(device_t dev);
>> -static int acpi_cpu_resume(device_t dev);
>> static int acpi_pcpu_get_id(device_t dev, uint32_t acpi_id,
>> u_int *cpu_id);
>> static void acpi_cpu_madt_handler(ACPI_SUBTABLE_HEADER *entry, void
>> *arg);
>> @@ -199,8 +197,8 @@ static device_method_t acpi_cpu_methods[] = {
>> DEVMETHOD(device_attach, acpi_cpu_attach),
>> DEVMETHOD(device_detach, bus_generic_detach),
>> DEVMETHOD(device_shutdown, acpi_cpu_shutdown),
>> - DEVMETHOD(device_suspend, acpi_cpu_suspend),
>> - DEVMETHOD(device_resume, acpi_cpu_resume),
>> + DEVMETHOD(device_suspend, bus_generic_suspend),
>> + DEVMETHOD(device_resume, bus_generic_resume),
>>
>> /* Bus interface */
>> DEVMETHOD(bus_add_child, acpi_cpu_add_child),
>> @@ -531,30 +529,6 @@ is_idle_disabled(struct acpi_cpu_softc *sc)
>> }
>> #endif
>>
>> -/*
>> - * Disable any entry to the idle function during suspend and re-enable it
>> - * during resume.
>> - */
>> -static int
>> -acpi_cpu_suspend(device_t dev)
>> -{
>> - int error;
>> -
>> - error = bus_generic_suspend(dev);
>> - if (error)
>> - return (error);
>> - disable_idle(device_get_softc(dev));
>> - return (0);
>> -}
>> -
>> -static int
>> -acpi_cpu_resume(device_t dev)
>> -{
>> -
>> - enable_idle(device_get_softc(dev));
>> - return (bus_generic_resume(dev));
>> -}
>> -
>> /*
>> * Find the processor associated with a given ACPI ID.
>> */
>>
>>