battery: Add the ThinkPad "Not Charging" quirk

The EC/ACPI firmware on Lenovo ThinkPads used to report a status of "Unknown" when the battery is between the charge start and charge stop thresholds. On Windows, it reports "Not Charging" so the quirk has been added to also report correctly.

Now the "status" attribute returns "Not Charging" when the battery on ThinkPads is not physicaly charging.

Tested-by Kevin Locke <kevin(a)kevinlocke.name>

Tested-by Christoph Böhmwalder <christoph(a)boehmwalder.at>

Signed-off-by Ognjen Galic <smclt30p(a)gmail.com>

drivers/acpi/battery.c +7 -0

 static unsigned int cache_time = 1000;
 module_param(cache_time, uint, 0644);
 MODULE_PARM_DESC(cache_time, "cache time in milliseconds");

@@ -222,6 +223,13 @@ static int acpi_battery_get_property(struct power_supply *psy,
                        val->intval = POWER_SUPPLY_STATUS_CHARGING;
                else if (acpi_battery_is_charged(battery))
                        val->intval = POWER_SUPPLY_STATUS_FULL;
+               /*
+                * On the Lenovo ThinkPad ACPI implementation, when
+                * neither bits 0 or 1 are set, that state is
+                * considered as "Not Charging".
+                */
+               else if (battery_quirk_thinkpad_notcharging)
+                       val->intval = POWER_SUPPLY_STATUS_NOT_CHARGING;
                else
                        val->intval = POWER_SUPPLY_STATUS_UNKNOWN;
                break;

Quoted 11 lines · Ognjen Galic

The EC/ACPI firmware on Lenovo ThinkPads used to report a status of "Unknown" when the battery is between the charge start and charge stop thresholds. On Windows, it reports "Not Charging" so the quirk has been added to also report correctly.

Now the "status" attribute returns "Not Charging" when the battery on ThinkPads is not physicaly charging.

Tested-by Kevin Locke <kevin(a)kevinlocke.name>

Tested-by Christoph Böhmwalder <christoph(a)boehmwalder.at>

Signed-off-by Ognjen Galic <smclt30p(a)gmail.com>

It doesn't look like this is related to the [1/3] and [3/3], so why do you make it part of the series?

drivers/acpi/battery.c static async_cookie_t async_cookie; · line 74

Quoted 13 lines

---
 drivers/acpi/battery.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

drivers/acpi/battery.c +1 -0

index 2951d07..81e9b4e 100644
@@ -71,6 +71,7 @@ static async_cookie_t async_cookie;
 static bool battery_driver_registered;
 static int battery_bix_broken_package;
 static int battery_notification_delay_ms;
+static int battery_quirk_thinkpad_notcharging;

Drop "thinkpad" from this name as somebody may need to use the quirk for a machine from a different vendor in the future.

Quoted 18 lines

drivers/acpi/battery.c +7 -0

 static unsigned int cache_time = 1000;
 module_param(cache_time, uint, 0644);
 MODULE_PARM_DESC(cache_time, "cache time in milliseconds");

@@ -222,6 +223,13 @@ static int acpi_battery_get_property(struct power_supply *psy,
                        val->intval = POWER_SUPPLY_STATUS_CHARGING;
                else if (acpi_battery_is_charged(battery))
                        val->intval = POWER_SUPPLY_STATUS_FULL;
+               /*
+                * On the Lenovo ThinkPad ACPI implementation, when
+                * neither bits 0 or 1 are set, that state is
+                * considered as "Not Charging".
+                */
+               else if (battery_quirk_thinkpad_notcharging)
+                       val->intval = POWER_SUPPLY_STATUS_NOT_CHARGING;
                else
                        val->intval = POWER_SUPPLY_STATUS_UNKNOWN;
                break;

drivers/acpi/battery.c battery_notification_delay_quirk(const struct dmi_system_id *d) · line 1313

Quoted 5 lines

drivers/acpi/battery.c +2 -0

@@ -1301,6 +1309,13 @@ battery_notification_delay_quirk(const struct dmi_system_id *d)
        return 0;
 }
 
+static int __init
+battery_quirk_not_charging(const struct dmi_system_id *d)

Don't break the above line (it will be over 80 chars long, but that's fine).

drivers/acpi/battery.c static const struct dmi_system_id bat_dmi_table[] __initconst = { · line 1336

Quoted 13 lines

drivers/acpi/battery.c +6 -0

+{
+       battery_quirk_thinkpad_notcharging = 1;
+       return 0;
+}
+
 static const struct dmi_system_id bat_dmi_table[] __initconst = {
        {
                .callback = battery_bix_broken_package_quirk,

@@ -1318,6 +1333,14 @@ static const struct dmi_system_id bat_dmi_table[] __initconst = {
                        DMI_MATCH(DMI_PRODUCT_NAME, "Aspire V5-573G"),
                },
        },
+       {

Add a comment to describe this quirk (which it is needed in the first place).

Quoted 12 lines

drivers/acpi/battery.c +7 -1

+               .callback = battery_quirk_not_charging,
+               .ident = "Lenovo ThinkPad",
+               .matches = {
+                       DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+                       DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad"),
+               },
+       },
        {},
 };
 
--

Quoted 8 lines · Ognjen Galic

The EC/ACPI firmware on Lenovo ThinkPads used to report a status of "Unknown" when the battery is between the charge start and charge stop thresholds. On Windows, it reports "Not Charging" so the quirk has been added to also report correctly.

Now the "status" attribute returns "Not Charging" when the battery on ThinkPads is not physicaly charging.

Quoted 3 lines

Tested-by Kevin Locke <[email protected]>

Tested-by Christoph Böhmwalder <[email protected]>

Signed-off-by Ognjen Galic <[email protected]>

It doesn't look like this is related to the [1/3] and [3/3], so why do you make it part of the series?

drivers/acpi/battery.c static async_cookie_t async_cookie; · line 74

Quoted 13 lines

---
 drivers/acpi/battery.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

drivers/acpi/battery.c +1 -0

index 2951d07..81e9b4e 100644
@@ -71,6 +71,7 @@ static async_cookie_t async_cookie;
 static bool battery_driver_registered;
 static int battery_bix_broken_package;
 static int battery_notification_delay_ms;
+static int battery_quirk_thinkpad_notcharging;

Drop "thinkpad" from this name as somebody may need to use the quirk for a machine from a different vendor in the future.

Quoted 18 lines

drivers/acpi/battery.c +7 -0

 static unsigned int cache_time = 1000;
 module_param(cache_time, uint, 0644);
 MODULE_PARM_DESC(cache_time, "cache time in milliseconds");

@@ -222,6 +223,13 @@ static int acpi_battery_get_property(struct power_supply *psy,
                        val->intval = POWER_SUPPLY_STATUS_CHARGING;
                else if (acpi_battery_is_charged(battery))
                        val->intval = POWER_SUPPLY_STATUS_FULL;
+               /*
+                * On the Lenovo ThinkPad ACPI implementation, when
+                * neither bits 0 or 1 are set, that state is
+                * considered as "Not Charging".
+                */
+               else if (battery_quirk_thinkpad_notcharging)
+                       val->intval = POWER_SUPPLY_STATUS_NOT_CHARGING;
                else
                        val->intval = POWER_SUPPLY_STATUS_UNKNOWN;
                break;

drivers/acpi/battery.c battery_notification_delay_quirk(const struct dmi_system_id *d) · line 1313

Quoted 5 lines

drivers/acpi/battery.c +2 -0

@@ -1301,6 +1309,13 @@ battery_notification_delay_quirk(const struct dmi_system_id *d)
        return 0;
 }
 
+static int __init
+battery_quirk_not_charging(const struct dmi_system_id *d)

Don't break the above line (it will be over 80 chars long, but that's fine).

drivers/acpi/battery.c static const struct dmi_system_id bat_dmi_table[] __initconst = { · line 1336

Quoted 13 lines

drivers/acpi/battery.c +6 -0

+{
+       battery_quirk_thinkpad_notcharging = 1;
+       return 0;
+}
+
 static const struct dmi_system_id bat_dmi_table[] __initconst = {
        {
                .callback = battery_bix_broken_package_quirk,

@@ -1318,6 +1333,14 @@ static const struct dmi_system_id bat_dmi_table[] __initconst = {
                        DMI_MATCH(DMI_PRODUCT_NAME, "Aspire V5-573G"),
                },
        },
+       {

Add a comment to describe this quirk (which it is needed in the first place).

drivers/acpi/battery.c static const struct dmi_system_id bat_dmi_table[] __initconst = {

Quoted 11 lines

drivers/acpi/battery.c +7 -1

+               .callback = battery_quirk_not_charging,
+               .ident = "Lenovo ThinkPad",
+               .matches = {
+                       DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+                       DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad"),
+               },
+       },
        {},
 };
 
--

Thanks,
Rafael

Quoted 13 lines · Rafael J. Wysocki

Quoted 11 lines · Ognjen Galic

The EC/ACPI firmware on Lenovo ThinkPads used to report a status of "Unknown" when the battery is between the charge start and charge stop thresholds. On Windows, it reports "Not Charging" so the quirk has been added to also report correctly.

Now the "status" attribute returns "Not Charging" when the battery on ThinkPads is not physicaly charging.

Tested-by Kevin Locke <kevin(a)kevinlocke.name>

Tested-by Christoph Böhmwalder <christoph(a)boehmwalder.at>

Signed-off-by Ognjen Galic <smclt30p(a)gmail.com>

Quoted 3 lines

It doesn't look like this is related to the [1/3] and [3/3], so why do you make it part of the series?

I made it the same series because it is practically the same feature set. Without this patch and with 1/3 and 3/3 applied, there is a bug where the status attribute would show "Unknown" for a battery that is between the start and stop thresholds while attached to AC.

Quoted 15 lines · Ognjen Galic

Quoted 14 lines · Rafael J. Wysocki

Quoted 11 lines · Ognjen Galic

The EC/ACPI firmware on Lenovo ThinkPads used to report a status of "Unknown" when the battery is between the charge start and charge stop thresholds. On Windows, it reports "Not Charging" so the quirk has been added to also report correctly.

Now the "status" attribute returns "Not Charging" when the battery on ThinkPads is not physicaly charging.

Tested-by Kevin Locke <[email protected]>

Tested-by Christoph Böhmwalder <[email protected]>

Signed-off-by Ognjen Galic <[email protected]>

Quoted 8 lines

Quoted 3 lines

It doesn't look like this is related to the [1/3] and [3/3], so why do you make it part of the series?

I made it the same series because it is practically the same feature set. Without this patch and with 1/3 and 3/3 applied, there is a bug where the status attribute would show "Unknown" for a battery that is between the start and stop thresholds while attached to AC.

OK, fair enough.

I would reorder it as the first patch in the series, then, because the other two are more related to each other.

Thanks,
Rafael

Quoted 2 lines · Rafael J. Wysocki

I would reorder it as the first patch in the series, then, because the other two are more related to each other.

Do I really need now to re-order and re-send the whole patchset for it to get merged?

Quoted 4 lines · Ognjen Galić

Quoted 2 lines · Rafael J. Wysocki

I would reorder it as the first patch in the series, then, because the other two are more related to each other.

Quoted 2 lines

Do I really need now to re-order and re-send the whole patchset for it to get merged?

Through which tree it's supposed to go?

--
With Best Regards,
Andy Shevchenko

Quoted 7 lines · Ognjen Galić

Quoted 2 lines · Rafael J. Wysocki

I would reorder it as the first patch in the series, then, because the other two are more related to each other.

Do I really need now to re-order and re-send the whole patchset for it to get merged?

Through which tree it's supposed to go?

Quoted 6 lines · Andy Shevchenko

Quoted 4 lines · Ognjen Galić

Quoted 2 lines · Rafael J. Wysocki

I would reorder it as the first patch in the series, then, because the other two are more related to each other.

Quoted 4 lines

Quoted 2 lines

Do I really need now to re-order and re-send the whole patchset for it to get merged?

Through which tree it's supposed to go?

Wait, wait. I haven't commented patches [1,3/3] yet, so no ACKs assumed, please.

Quoted 11 lines · Rafael J. Wysocki

Quoted 9 lines · Andy Shevchenko

Quoted 7 lines · Ognjen Galić

Quoted 2 lines · Rafael J. Wysocki

I would reorder it as the first patch in the series, then, because the other two are more related to each other.

Do I really need now to re-order and re-send the whole patchset for it to get merged?

Quoted 4 lines

Quoted 1 line

Through which tree it's supposed to go?

Wait, wait. I haven't commented patches [1,3/3] yet, so no ACKs assumed, please.

I would prefer if you can take it through linux-pm.

Whenever you are okay with the series let me know I would review PDx86 bits.

Quoted 30 lines · Rafael J. Wysocki

Quoted 15 lines · Ognjen Galic

Quoted 14 lines · Rafael J. Wysocki

Quoted 11 lines · Ognjen Galic

The EC/ACPI firmware on Lenovo ThinkPads used to report a status of "Unknown" when the battery is between the charge start and charge stop thresholds. On Windows, it reports "Not Charging" so the quirk has been added to also report correctly.

Now the "status" attribute returns "Not Charging" when the battery on ThinkPads is not physicaly charging.

Tested-by Kevin Locke <kevin(a)kevinlocke.name>

Tested-by Christoph Böhmwalder <christoph(a)boehmwalder.at>

Signed-off-by Ognjen Galic <smclt30p(a)gmail.com>

Quoted 8 lines

Quoted 3 lines

It doesn't look like this is related to the [1/3] and [3/3], so why do you make it part of the series?

I made it the same series because it is practically the same feature set. Without this patch and with 1/3 and 3/3 applied, there is a bug where the status attribute would show "Unknown" for a battery that is between the start and stop thresholds while attached to AC.

OK, fair enough.

I would reorder it as the first patch in the series, then, because the other two are more related to each other.

Quoted 3 lines

Thanks,
Rafael

Do I really need now to re-order and re-send the whole patchset for it to get merged?