user/sven/linux.git/drivers/acpi/processor_idle.c, branch v5.18.16

Revert "ACPI: processor: idle: fix lockup regression on 32-bit ThinkPad T40"

2022-04-21T17:58:14Z

This reverts commit bfe55a1f7fd6bfede16078bf04c6250fbca11588. This was presumably misdiagnosed as an inability to use C3 at all when I suspect the real problem is just misconfiguration of C3 vs. ARB_DIS. Signed-off-by: Ville Syrjälä Cc: 5.16+ # 5.16+ Tested-by: Woody Suwalski Signed-off-by: Rafael J. Wysocki

ACPI: processor: idle: Avoid falling back to C3 type C-states

2022-04-21T17:55:41Z

The "safe state" index is used by acpi_idle_enter_bm() to avoid entering a C-state that may require bus mastering to be disabled on entry in the cases when this is not going to happen. For this reason, it should not be set to point to C3 type of C-states, because they may require bus mastering to be disabled on entry in principle. This was broken by commit d6b88ce2eb9d ("ACPI: processor idle: Allow playing dead in C3 state") which inadvertently allowed the "safe state" index to point to C3 type of C-states. This results in a machine that won't boot past the point when it first enters C3. Restore the correct behaviour (either demote to C1/C2, or use C3 but also set ARB_DIS=1). I hit this on a Fujitsu Siemens Lifebook S6010 (P3) machine. Fixes: d6b88ce2eb9d ("ACPI: processor idle: Allow playing dead in C3 state") Cc: 5.16+ # 5.16+ Signed-off-by: Ville Syrjälä Tested-by: Woody Suwalski [ rjw: Subject and changelog adjustments ] Signed-off-by: Rafael J. Wysocki

Revert "ACPI: processor: idle: Only flush cache on entering C3"

2022-04-04T18:51:15Z

Revert commit 87ebbb8c612b ("ACPI: processor: idle: Only flush cache on entering C3") that broke the assumptions of the acpi_idle_play_dead() callers. Namely, the CPU cache must always be flushed in acpi_idle_play_dead(), regardless of the target C-state that is going to be requested, because this is likely to be part of a CPU offline procedure or preparation for entering a system-wide sleep state and the lack of synchronization between the CPU cache and RAM may lead to problems going forward, for example when the CPU is brought back online. In particular, it breaks resume from suspend-to-RAM on Lenovo ThinkPad C13 which fails occasionally until the problematic commit is reverted. Signed-off-by: Akihiko Odaki [ rjw: Changelog ] Signed-off-by: Rafael J. Wysocki

ACPI: processor idle: Check for architectural support for LPI

2022-03-03T19:20:06Z

When `osc_pc_lpi_support_confirmed` is set through `_OSC` and `_LPI` is populated then the cpuidle driver assumes that LPI is fully functional. However currently the kernel only provides architectural support for LPI on ARM. This leads to high power consumption on X86 platforms that otherwise try to enable LPI. So probe whether or not LPI support is implemented before enabling LPI in the kernel. This is done by overloading `acpi_processor_ffh_lpi_probe` to check whether it returns `-EOPNOTSUPP`. It also means that all future implementations of `acpi_processor_ffh_lpi_probe` will need to follow these semantics as well. Reviewed-by: Sudeep Holla Signed-off-by: Mario Limonciello Signed-off-by: Rafael J. Wysocki

ACPI: processor: idle: fix lockup regression on 32-bit ThinkPad T40

2022-02-16T17:17:16Z

Add and ACPI idle power level limit for 32-bit ThinkPad T40. There is a regression on T40 introduced by commit d6b88ce2, starting with kernel 5.16: commit d6b88ce2eb9d2698eb24451eb92c0a1649b17bb1 Author: Richard Gong Date: Wed Sep 22 08:31:16 2021 -0500 ACPI: processor idle: Allow playing dead in C3 state The above patch is trying to enter C3 state during init, what is causing a T40 system freeze. I have not found a similar issue on any other of my 32-bit machines. The fix is to add another exception to the processor_power_dmi_table[] list. As a result the dmesg shows as expected: [2.155398] ACPI: IBM ThinkPad T40 detected - limiting to C2 max_cstate. Override with "processor.max_cstate=9" [2.155404] ACPI: processor limited to max C-state 2 The fix is trivial and affects only vintage T40 systems. Fixes: d6b88ce2eb9d ("CPI: processor idle: Allow playing dead in C3 state") Signed-off-by: Woody Suwalski Reviewed-by: Hans de Goede Cc: 5.16+ # 5.16+ [ rjw: New subject ] Signed-off-by: Rafael J. Wysocki

Merge branches 'acpi-ec' and 'acpi-processor'

2022-01-10T15:57:59Z

Merge ACPI EC driver updates and ACPI processor driver updates for 5.17-rc1: - Rework flushing of EC work while suspended to idle and clean up the handling of events in the ACPI EC driver (Rafael Wysocki). - Prohibit ec_sys module parameter write_support from being used when the system is locked down (Hans de Goede). - Make the ACPI processor thermal driver use cpufreq_cpu_get() to check for presence of cpufreq policy (Manfred Spraul). - Avoid unnecessary CPU cache flushing in the ACPI processor idle driver (Kirill A. Shutemov). - Replace kernel.h with the necessary inclusions in the ACPI processor driver (Andy Shevchenko). - Use swap() instead of open coding it in the ACPI processor idle driver (Guo Zhengkui). * acpi-ec: ACPI: EC: Mark the ec_sys write_support param as module_param_hw() ACPI: EC: Relocate acpi_ec_create_query() and drop acpi_ec_delete_query() ACPI: EC: Make the event work state machine visible ACPI: EC: Avoid queuing unnecessary work in acpi_ec_submit_event() ACPI: EC: Rename three functions ACPI: EC: Simplify locking in acpi_ec_event_handler() ACPI: EC: Rearrange the loop in acpi_ec_event_handler() ACPI: EC: Fold acpi_ec_check_event() into acpi_ec_event_handler() ACPI: EC: Pass one argument to acpi_ec_query() ACPI: EC: Call advance_transaction() from acpi_ec_dispatch_gpe() ACPI: EC: Rework flushing of EC work while suspended to idle * acpi-processor: ACPI: processor: thermal: avoid cpufreq_get_policy() ACPI: processor: idle: Only flush cache on entering C3 ACPI: processor idle: Use swap() instead of open coding it ACPI: processor: Replace kernel.h with the necessary inclusions

ACPI: processor: idle: Only flush cache on entering C3

2021-12-17T17:54:56Z

According to ACPI 6.4, Section 8.2, CPU cache flushing required on entering the C3 power state. Avoid flushing the cache on entering other C-states. Signed-off-by: Kirill A. Shutemov [ rjw: Changelog edits ] Signed-off-by: Rafael J. Wysocki

ACPI: Use acpi_fetch_acpi_dev() instead of acpi_bus_get_device()

2021-12-17T17:45:51Z

Modify the ACPI code to use acpi_fetch_acpi_dev() instead of acpi_bus_get_device() where applicable. Signed-off-by: Rafael J. Wysocki Reviewed-by: Mika Westerberg Reviewed-by: Hans de Goede

ACPI: processor idle: Use swap() instead of open coding it

2021-11-24T16:46:12Z

Address the following coccicheck warning: ./drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c:914:40-41: WARNING opportunity for swap(). by using swap() for the swapping of variable values and drop the tmp variable that is not needed any more. Signed-off-by: Guo Zhengkui [ rjw: Subject and changelog rewrite ] Signed-off-by: Rafael J. Wysocki

ACPI: processor idle: Allow playing dead in C3 state

2021-09-24T16:26:27Z

When some cores are disabled on AMD platforms, the system will no longer be able to enter suspend-to-idle s0ix. Update to allow playing dead in C3 state so that the CPUs can enter the deepest state on AMD platforms. BugLink: https://gitlab.freedesktop.org/drm/amd/-/issues/1708 Suggested-by: Mario Limonciello Signed-off-by: Richard Gong [ rjw: Fixed coding style ] Signed-off-by: Rafael J. Wysocki