Re: [PATCH 1/1] power: supply: sbs-battery: Add ability to force load a battery via the devicetree

[Richard Tresidder <rtresidd@electromag.com.au>]

Hi Guenter
   See below
Richard Tresidder

Cheers

   Richard Tresidder

**
On 30/07/2019 12:09 pm, Guenter Roeck wrote:
> On Mon, Jul 29, 2019 at 8:02 PM Richard Tresidder
> <rtresidd@electromag.com.au> wrote:
>> Hi Nick and Guenter
>> Just adding you to this one also seeing as you're looking at that other
>> sbs_battery patch for me.
>> Not sure why the get maintainers didn't list you for this one.
>>
>> Cheers
>>      Richard Tresidder
>>> Add the ability to force load a hot pluggable battery during boot where
>>> there is no gpio detect method available and the module is statically
>>> built. Normal polling will then occur on that battery when it is inserted.
>>>
>>> Signed-off-by: Richard Tresidder <rtresidd@electromag.com.au>
>>> ---
>>>
>>> Notes:
>>>       Add the ability to force load a hot pluggable battery during boot where
>>>       there is no gpio detect method available and the module is statically
>>>       built. Normal polling will then occur on that battery when it is inserted.
>>>
>>>    drivers/power/supply/sbs-battery.c | 6 +++++-
>>>    1 file changed, 5 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/power/supply/sbs-battery.c b/drivers/power/supply/sbs-battery.c
>>> index 048d205..ea8ba3e 100644
>>> --- a/drivers/power/supply/sbs-battery.c
>>> +++ b/drivers/power/supply/sbs-battery.c
>>> @@ -161,6 +161,7 @@ struct sbs_info {
>>>        int                             poll_time;
>>>        u32                             i2c_retry_count;
>>>        u32                             poll_retry_count;
>>> +     bool                            force_load;
>>>        struct delayed_work             work;
>>>        struct mutex                    mode_lock;
>>>        u32                             flags;
>>> @@ -852,6 +853,9 @@ static int sbs_probe(struct i2c_client *client,
>>>        if (rc)
>>>                chip->poll_retry_count = 0;
>>>
>>> +     chip->force_load = of_property_read_bool(client->dev.of_node,
>>> +                                             "sbs,force-load");
>>> +
> Maybe it is documented in another patch, which I have not seen. If it
> isn't, it will have to be documented and reviewed by a devicetree
> maintainer. Either case, I don't immediately see why the variable
> needs to reside in struct sbs_info; it seems to be used only in the
> probe function.
Good point, we don't actually need to store the value, it can just be a 
local.
Yep I had done a device tree documentation patch, but confused myself 
when submitting due to it having to be in a separate patch
I think I need to create 2 different patches but submit in the same 
email via send-patch as a multipart patch..
I'll try to fix that when I send it in again.
>
>>>        if (pdata) {
>>>                chip->poll_retry_count = pdata->poll_retry_count;
>>>                chip->i2c_retry_count  = pdata->i2c_retry_count;
>>> @@ -890,7 +894,7 @@ static int sbs_probe(struct i2c_client *client,
>>>         * Before we register, we might need to make sure we can actually talk
>>>         * to the battery.
>>>         */
>>> -     if (!(force_load || chip->gpio_detect)) {
>>> +     if (!(force_load || chip->gpio_detect || chip->force_load)) {
>>>                rc = sbs_read_word_data(client, sbs_data[REG_STATUS].addr);
>>>
>>>                if (rc < 0) {
>