On Wed, Sep 15, 2021 at 2:05 PM Rafael J. Wysocki [off-list ref] wrote:

On Wed, Sep 15, 2021 at 4:23 AM Florian Fainelli [off-list ref] wrote:

quoted

An implementation of suspend_ops is allowed to reject the PM_SUSPEND_MEM
suspend type from its ->valid() callback, we should not assume that it
is always present as this is not a correct reflection of what a firmware
interface may support.

Fixes: 406e79385f32 ("PM / sleep: System sleep state selection interface rework")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
 kernel/power/suspend.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c
index eb75f394a059..02e306ad8db8 100644
--- a/kernel/power/suspend.c
+++ b/kernel/power/suspend.c

@@ -171,8 +171,7 @@ static bool valid_state(suspend_state_t state)

 void __init pm_states_init(void)
 {
-       /* "mem" and "freeze" are always present in /sys/power/state. */
-       pm_states[PM_SUSPEND_MEM] = pm_labels[PM_SUSPEND_MEM];
+       /* "freeze" is always present in /sys/power/state. */
        pm_states[PM_SUSPEND_TO_IDLE] = pm_labels[PM_SUSPEND_TO_IDLE];
        /*
         * Suspend-to-idle should be supported even without any suspend_ops,

@@ -214,6 +213,7 @@ void suspend_set_ops(const struct platform_suspend_ops *ops)
        }
        if (valid_state(PM_SUSPEND_MEM)) {
                mem_sleep_states[PM_SUSPEND_MEM] = mem_sleep_labels[PM_SUSPEND_MEM];
+               pm_states[PM_SUSPEND_MEM] = pm_labels[PM_SUSPEND_MEM];
                if (mem_sleep_default >= PM_SUSPEND_MEM)
                        mem_sleep_current = PM_SUSPEND_MEM;
        }
--

Applied as 5.16 material, thanks!

And reverted as per
https://lore.kernel.org/linux-pm/d36f79b4-472c-4852-7370-a011f9f556ce@intel.com/ (local)