On 07/05/2017 08:18 PM, Pavel Machek wrote:

Hi!

quoted

Although this callback is optional and documented as such, it requires
a platform_suspend_ops::begin callback to be implemented in order to
provide an accurate suspend/resume state within the driver that
implements this platform_suspend_ops.

An enumeration: platform_target_state is defined which currently defines
the standard ACPI_S[1-4] states and can be extended with platform
specific suspend states.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
 include/linux/suspend.h | 25 +++++++++++++++++++++++++
 kernel/power/suspend.c  | 15 +++++++++++++++
 2 files changed, 40 insertions(+)

diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 0b1cf32edfd7..6e6cc0778816 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h

@@ -50,6 +50,16 @@ enum suspend_stat_step {
 	SUSPEND_RESUME
 };
 
+enum platform_target_state {
+	PLATFORM_STATE_UNKNOWN = -1,
+	PLATFORM_STATE_WORKING = 0,
+	PLATFORM_STATE_ACPI_S1,
+	PLATFORM_STATE_ACPI_S2,
+	PLATFORM_STATE_ACPI_S3,
+	PLATFORM_STATE_ACPI_S4,
+	/* Add platform specific states here */
+};
+

As I tried to explain in the email thread, having list with all the possible platform
states is no-go. We have about 1000 platforms supported...

FYI, the recent (relatively recent) CPU hotplug conversion from
notifiers to a state machine has a similar pattern whereby pieces of
code needing to hook into the CPU hotplug state machine add their own
enum values as they need. So far it's been working for them, and there
were tons of CPU hotplug notifiers in the kernel.

Anyhow, let me implement Rafael's suggestions and we can see how we move
from there.
-- 
Florian