platform/x86: asus-wmi: temporarily revert to setting a charge limit
Denis Benato
maintainerPATCH 1/1 platform/x86: asus-wmi: temporarily revert to setting a charge limit
A userspace regression has been observed leaving the battery charging threshold unconfigured, so while the situation is being sorted revert the change keeping the infrastructure in place to return to the preferred behaviour whenever possible.
--- drivers/platform/x86/asus-wmi.c | 26 +++++++++++++++++++++++--- 1 file changed, 23 insertions(+), 3 deletions(-)
drivers/platform/x86/asus-wmi.c +23 -3
@@ -1618,6 +1618,8 @@ static DEVICE_ATTR_RW(charge_control_end_threshold);static int asus_wmi_battery_add (struct power_supply *battery,struct acpi_battery_hook *hook) { +int ret, rv; +/* The WMI method does not provide a way to specific a battery, so we * just assume it is the first battery. * Note: On some newer ASUS laptops (Zenbook UM431DA), the primary/first
@@ -1635,12 +1637,30 @@ static int asus_wmi_battery_add(struct power_supply *battery, struct acpi_batter/* The charge threshold is only reset when the system is power cycled, * and we can't read the current threshold, however the majority of -* platforms retains it, therefore signal the threshold as unknown -* until user explicitly sets it to a new value. +* platforms retains it. +* +* Setting a negative value would signal the threshold as unknown +* until user explicitly sets it to a new value, however to avoid +* regressing userspace, we initialize it to a value of 100. */ - charge_end_threshold =-1 ; + charge_end_threshold =100 ; + ret =asus_wmi_set_devstate (ASUS_WMI_DEVID_RSOC , charge_end_threshold, &rv); +if (ret) { +pr_err ("Failed to reset battery charge threshold\n" ); + goto asus_wmi_battery_add_err; + } + +if (rv !=1 ) { +pr_err ("Error in battery charge threshold reset\n" ); + ret = -EIO ; + goto asus_wmi_battery_add_err; + }return 0 ; +asus_wmi_battery_add_err: +device_remove_file (&battery->dev, + &dev_attr_charge_control_end_threshold); +return ret; }static int asus_wmi_battery_remove (struct power_supply *battery,struct acpi_battery_hook *hook)
Signature · 1 line
2.47.3
Ilpo Järvinen
maintainerQuoted 8 lines · Denis Benato
A userspace regression has been observed leaving the battery charging threshold unconfigured, so while the situation is being sorted revert the change keeping the infrastructure in place to return to the preferred behaviour whenever possible.
Reported-by ?
No blank lines in tags block please.
How stable people will know which kernel this should go? The easiest way is put Fixes tag for that purpose. If don't want to say this "fixes" things, also Cc: stable... # [instruction here] is possible (more information about formatting in stable-kernel-rules.rst).
Signature · 1 line
i.
Quoted 53 lines
Signed-off-by Denis Benato <[email protected]>--- drivers/platform/x86/asus-wmi.c | 26 +++++++++++++++++++++++--- 1 file changed, 23 insertions(+), 3 deletions(-)drivers/platform/x86/asus-wmi.c +23
-3@@ -1618,6 +1618,8 @@ static DEVICE_ATTR_RW(charge_control_end_threshold);static int asus_wmi_battery_add (struct power_supply *battery,struct acpi_battery_hook *hook) { +int ret, rv; +/* The WMI method does not provide a way to specific a battery, so we * just assume it is the first battery. * Note: On some newer ASUS laptops (Zenbook UM431DA), the primary/first @@ -1635,12 +1637,30 @@ static int asus_wmi_battery_add(struct power_supply *battery, struct acpi_batter/* The charge threshold is only reset when the system is power cycled, * and we can't read the current threshold, however the majority of -* platforms retains it, therefore signal the threshold as unknown -* until user explicitly sets it to a new value. +* platforms retains it. +* +* Setting a negative value would signal the threshold as unknown +* until user explicitly sets it to a new value, however to avoid +* regressing userspace, we initialize it to a value of 100. */ - charge_end_threshold =-1 ; + charge_end_threshold =100 ; + ret =asus_wmi_set_devstate (ASUS_WMI_DEVID_RSOC , charge_end_threshold, &rv); +if (ret) { +pr_err ("Failed to reset battery charge threshold\n" ); + goto asus_wmi_battery_add_err; + } + +if (rv !=1 ) { +pr_err ("Error in battery charge threshold reset\n" ); + ret = -EIO ; + goto asus_wmi_battery_add_err; + }return 0 ; +asus_wmi_battery_add_err: +device_remove_file (&battery->dev, + &dev_attr_charge_control_end_threshold); +return ret; }static int asus_wmi_battery_remove (struct power_supply *battery,struct acpi_battery_hook *hook)
Hi Denis,
Upower 1.91.3 has been released, including a fix for issue #347 [1]. This release addresses the -ENODATA encountered when accessing charge limits on Asus laptops, which previously caused UPower and GNOME to incorrectly report charge threshold support as unavailable. The fix updates the udev rule to verify file existence rather than reading the attribute directly. The corresponding commit can be found at:
https://gitlab.freedesktop.org/upower/upower/-/commit/83f86c75d28a57854a4b5986cbfb701249a5b6ad
[1] https://gitlab.freedesktop.org/upower/upower/-/work_items/347
Thank you :)
Quoted 60 lines · Denis Benato
A userspace regression has been observed leaving the battery charging threshold unconfigured, so while the situation is being sorted revert the change keeping the infrastructure in place to return to the preferred behaviour whenever possible.
Signed-off-by Denis Benato <[email protected]>--- drivers/platform/x86/asus-wmi.c | 26 +++++++++++++++++++++++--- 1 file changed, 23 insertions(+), 3 deletions(-)drivers/platform/x86/asus-wmi.c +23
-3@@ -1618,6 +1618,8 @@ static DEVICE_ATTR_RW(charge_control_end_threshold);static int asus_wmi_battery_add (struct power_supply *battery,struct acpi_battery_hook *hook) { +int ret, rv; +/* The WMI method does not provide a way to specific a battery, so we * just assume it is the first battery. * Note: On some newer ASUS laptops (Zenbook UM431DA), the primary/first @@ -1635,12 +1637,30 @@ static int asus_wmi_battery_add(struct power_supply *battery, struct acpi_batter/* The charge threshold is only reset when the system is power cycled, * and we can't read the current threshold, however the majority of -* platforms retains it, therefore signal the threshold as unknown -* until user explicitly sets it to a new value. +* platforms retains it. +* +* Setting a negative value would signal the threshold as unknown +* until user explicitly sets it to a new value, however to avoid +* regressing userspace, we initialize it to a value of 100. */ - charge_end_threshold =-1 ; + charge_end_threshold =100 ; + ret =asus_wmi_set_devstate (ASUS_WMI_DEVID_RSOC , charge_end_threshold, &rv); +if (ret) { +pr_err ("Failed to reset battery charge threshold\n" ); + goto asus_wmi_battery_add_err; + } + +if (rv !=1 ) { +pr_err ("Error in battery charge threshold reset\n" ); + ret = -EIO ; + goto asus_wmi_battery_add_err; + }return 0 ; +asus_wmi_battery_add_err: +device_remove_file (&battery->dev, + &dev_attr_charge_control_end_threshold); +return ret; }static int asus_wmi_battery_remove (struct power_supply *battery,struct acpi_battery_hook *hook)
Denis Benato
maintainerQuoted 10 lines · Kate Hsuan
Hi Denis,
Upower 1.91.3 has been released, including a fix for issue #347 [1]. This release addresses the -ENODATA encountered when accessing charge limits on Asus laptops, which previously caused UPower and GNOME to incorrectly report charge threshold support as unavailable. The fix updates the udev rule to verify file existence rather than reading the attribute directly. The corresponding commit can be found at:
https://gitlab.freedesktop.org/upower/upower/-/commit/83f86c75d28a57854a4b5986cbfb701249a5b6ad
[1] https://gitlab.freedesktop.org/upower/upower/-/work_items/347
Thank you :)
Thank you very much Hsuan!
Ilpo: what do we do now? I am preparing the fixed patch to revert this for stable and will soon send it.
How much do we wait to revert the revert? Or do I simply Cc stable and you don't pick the revert in the newer kernel? In any case just say what I need to do and I'll do it.
Thanks,
Denis
Quoted 58 lines · Denis Benato
Quoted 56 lines
A userspace regression has been observed leaving the battery charging threshold unconfigured, so while the situation is being sorted revert the change keeping the infrastructure in place to return to the preferred behaviour whenever possible.
Signed-off-by Denis Benato <[email protected]>---
drivers/platform/x86/asus-wmi.c| 26 +++++++++++++++++++++++---1 file changed, 23 insertions(+), 3 deletions(-)drivers/platform/x86/asus-wmi.c +23
-3@@ -1618,6 +1618,8 @@ static DEVICE_ATTR_RW(charge_control_end_threshold); static int asus_wmi_battery_add(struct power_supply *battery, struct acpi_battery_hook *hook) { + int ret, rv; + /* The WMI method does not provide a way to specific a battery, so we* just assume it is the first battery. * Note: On some newer ASUS laptops (Zenbook UM431DA), the primary/first @@ -1635,12 +1637,30 @@ static int asus_wmi_battery_add(struct power_supply *battery, struct acpi_batter /* The charge threshold is only reset when the system is power cycled, * and we can't read the current threshold, however the majority of - * platforms retains it, therefore signal the threshold as unknown - * until user explicitly sets it to a new value. + * platforms retains it. + * + * Setting a negative value would signal the threshold as unknown + * until user explicitly sets it to a new value, however to avoid + * regressing userspace, we initialize it to a value of 100. */ - charge_end_threshold = -1; + charge_end_threshold = 100; + ret = asus_wmi_set_devstate(ASUS_WMI_DEVID_RSOC, charge_end_threshold, &rv); + if (ret) { + pr_err("Failed to reset battery charge threshold\n"); + goto asus_wmi_battery_add_err; + } + + if (rv != 1) { + pr_err("Error in battery charge threshold reset\n"); + ret = -EIO; + goto asus_wmi_battery_add_err; + } return 0; +asus_wmi_battery_add_err: + device_remove_file(&battery->dev, + &dev_attr_charge_control_end_threshold); + return ret; } static int asus_wmi_battery_remove(struct power_supply *battery, struct acpi_battery_hook *hook)