Re: [PATCH] ACPI/Battery: Retry to get Battery information if failed during probing

[Lan Tianyu]

On 2014å06æ12æ 15:26, David Rientjes wrote:
> On Thu, 12 Jun 2014, Lan Tianyu wrote:
> 
>>>> Some machines'(E,G Lenovo Z480) ECs are not stable during boot up
>>>> and causes battery driver fails to be probed due to failure of getting
>>>> battery information from EC sometimes. After several retries, the
>>>> operation will work. This patch is to retry to get battery information 5
>>>> times if the first try fails.
>>>>
>>>> Reported-and-tested-by: naszar <naszar@xxxxx>
>>>> Reference: https://bugzilla.kernel.org/show_bug.cgi?id=75581
>>>> Cc: stable@xxxxxxxxxxxxxxx
>>>> Signed-off-by: Lan Tianyu <tianyu.lan@xxxxxxxxx>
>>>> ---
>>>>  drivers/acpi/battery.c | 12 +++++++++++-
>>>>  1 file changed, 11 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c
>>>> index e48fc98..485009d 100644
>>>> --- a/drivers/acpi/battery.c
>>>> +++ b/drivers/acpi/battery.c
>>>> @@ -34,6 +34,7 @@
>>>>  #include <linux/dmi.h>
>>>>  #include <linux/slab.h>
>>>>  #include <linux/suspend.h>
>>>> +#include <linux/delay.h>
>>>>  #include <asm/unaligned.h>
>>>>  
>>>>  #ifdef CONFIG_ACPI_PROCFS_POWER
>>>> @@ -1119,7 +1120,7 @@ static struct dmi_system_id bat_dmi_table[] = {
>>>>  
>>>>  static int acpi_battery_add(struct acpi_device *device)
>>>>  {
>>>> -	int result = 0;
>>>> +	int result = 0, retry = 5;
>>>>  	struct acpi_battery *battery = NULL;
>>>>  
>>>>  	if (!device)
>>>> @@ -1135,7 +1136,16 @@ static int acpi_battery_add(struct acpi_device *device)
>>>>  	mutex_init(&battery->sysfs_lock);
>>>>  	if (acpi_has_method(battery->device->handle, "_BIX"))
>>>>  		set_bit(ACPI_BATTERY_XINFO_PRESENT, &battery->flags);
>>>> +
>>>> +retry_get_info:
>>>>  	result = acpi_battery_update(battery, false);
>>>> +
>>>> +	if (result && retry) {
>>>> +		msleep(20);
>>>
>>
>> Hi David:
>> 	Thanks for review.
>>
>>> We're really going to wait up to 20 * 5 = 100ms for acpi_battery_update() 
>>> to succeed?
>>
>> No, this depends which retry acpi_battery_update() will succeed. For
>> most machines, there will be no delay.
>>
> 
> Right, but you're willing to wait up to 100ms for it to succeed?  You're 
> implementing x retries with y ms sleep in between, I'm asking how it is 
> determined that the optimal values are x = 5 and y = 20. More directly:
> is it possible to succeed at 101ms?

The retry time is set by randomly and not accurate because don't know
when EC will work normally. Set the retry time to 5 just in order to
make sure battery driver probing sucessfully every time,

>  Is it really likely to succeed after 
> the first 20ms?
> 

Yes, it's possible.
>From naszar's test log, acpi_battery_update() failed only once. But not
sure that this happens every time, treat it conservatively and set the
retry time to 5.
https://bugzilla.kernel.org/attachment.cgi?id=139081&action=edit

-- 
Best regards
Tianyu Lan
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/