linux-kernel - Re: [PATCH v4 4/4] drivers/rtc/rtc-s5m.c: add support for S2MPS15 RTC

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Thu, 29 Oct 2015 16:50:18 +0530
From:	Alim Akhtar <alim.akhtar@...sung.com>
To:	Krzysztof Kozlowski <k.kozlowski@...sung.com>
Cc:	lee.jones@...aro.org, broonie@...nel.org,
	linux-samsung-soc@...r.kernel.org, rtc-linux@...glegroups.com,
	linux-kernel@...r.kernel.org,
	Alexandre Belloni <alexandre.belloni@...e-electrons.com>
Subject: Re: [PATCH v4 4/4] drivers/rtc/rtc-s5m.c: add support for S2MPS15 RTC



On 10/29/2015 04:33 PM, Krzysztof Kozlowski wrote:
> 2015-10-29 16:47 GMT+09:00 Alim Akhtar <alim.akhtar@...sung.com>:
>> Hello Krzysztof,
>>
>>
>> On 10/29/2015 11:24 AM, Krzysztof Kozlowski wrote:
>>>
>>> On 29.10.2015 13:58, Alim Akhtar wrote:
>>>>
>>>> RTC found in s2mps15 is almost same as one found on s2mps14/13
>>>> with few differences in RTC_UPDATE register fields, like
>>>> bit fields are changed for WUDR and AUDR.
>>>> This patch add required changes to enable s2mps15 rtc timer.
>>>>
>>>> Cc: Alexandre Belloni <alexandre.belloni@...e-electrons.com>
>>>> Signed-off-by: Alim Akhtar <alim.akhtar@...sung.com>
>>>> ---
>>>>    drivers/rtc/rtc-s5m.c           |   20 ++++++++++++++++++--
>>>>    include/linux/mfd/samsung/rtc.h |    4 ++++
>>>>    2 files changed, 22 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/drivers/rtc/rtc-s5m.c b/drivers/rtc/rtc-s5m.c
>>>> index f2504b4eef34..0d106a99958f 100644
>>>> --- a/drivers/rtc/rtc-s5m.c
>>>> +++ b/drivers/rtc/rtc-s5m.c
>>>> @@ -188,6 +188,7 @@ static inline int
>>>> s5m_check_peding_alarm_interrupt(struct s5m_rtc_info *info,
>>>>                  ret = regmap_read(info->regmap, S5M_RTC_STATUS, &val);
>>>>                  val &= S5M_ALARM0_STATUS;
>>>>                  break;
>>>> +       case S2MPS15X:
>>>>          case S2MPS14X:
>>>>          case S2MPS13X:
>>>>                  ret = regmap_read(info->s5m87xx->regmap_pmic,
>>>> S2MPS14_REG_ST2,
>>>> @@ -223,6 +224,9 @@ static inline int s5m8767_rtc_set_time_reg(struct
>>>> s5m_rtc_info *info)
>>>>          if (info->device_type == S5M8763X || info->device_type ==
>>>> S5M8767X)
>>>>                  data |= S5M_RTC_TIME_EN_MASK;
>>>>
>>>> +       if (info->device_type == S2MPS15X)
>>>> +               data |= S2MPS15_RTC_WUDR_MASK;
>>>> +
>>>
>>>
>>>
>>> You are setting here bit 1 and bit 4. However vendor code sets only bit
>>> 4 (called WUDR there). That's confusing. Why the difference? Or maybe I
>>> am looking at wrong vendor tree (SM-G920F)?
>>>
>> Sorry, I don't have access to vendor code that you are looking into, so
>> won't be able to comment on that.
>
> Everyone have the access. It is on opensource.samsung.com, as required by GPL.
>
Ok, I didn't check there.
>>
>> I am testing this patch before sending them, what I have found is if you
>> don't update WUDR the time does not changes in rtc.
>> e.g.
>> if you don't do above changes then you will see below:
>> -----
>> # date --set="Oct 29 14:10:40 2015" <update the a new date>
>> Thu Oct 29 14:10:40 UTC 2015
>> #
>> # hwclock --systohc <copy system clock to hardware clock/rtc>
>> # hwclock <read back hwclock>
>> Thu Oct 29 12:52:32 2015  .922889 seconds
>> ----
>> which is not excepted.
>>
>> And with the above change I see:
>>
>> ----
>>   # date --set="Oct 29 14:30:10 2015" <update the a new date>
>> Thu Oct 29 14:30:10 UTC 2015
>> # date
>> Thu Oct 29 14:30:12 UTC 2015
>> # hwclock --systohc <copy system clock to hardware clock/rtc>
>> # hwclock <read back hwclock>
>> Thu Oct 29 14:30:21 2015  .333006 seconds
>> ----
>>
>> Which is as expected.
>
> Okay, but I said that vendor is setting only WUDR which is bit 4. You
bit:4 is AUDR (for alarm). not WUDR and bit: 1 is WDUR(for timer)
> are setting bit 1 and bit 4. Your reply - about the need of setting
> WUDR (bit 4 I guess) - does not explain my concerns. Do you have to
> set bit 1?
>
Yes, I think we need to set both for timer and alarm.
I have send you a snapshot of the rtc_update register in UM via internal 
email. PTAL.
>>
>>>>          ret = regmap_write(info->regmap, info->regs->rtc_udr_update,
>>>> data);
>>>>          if (ret < 0) {
>>>>                  dev_err(info->dev, "failed to write update reg(%d)\n",
>>>> ret);
>>>> @@ -252,6 +256,9 @@ static inline int s5m8767_rtc_set_alarm_reg(struct
>>>> s5m_rtc_info *info)
>>>>          case S5M8767X:
>>>>                  data &= ~S5M_RTC_TIME_EN_MASK;
>>>>                  break;
>>>> +       case S2MPS15X:
>>>> +               data |= S2MPS15_RTC_AUDR_MASK;
>>>> +               break;
>>>>          case S2MPS14X:
>>>>                  data |= S2MPS_RTC_RUDR_MASK;
>>>>                  break;
>>>
>>>
>>> Another difference: you are setting here exactly the same values as
>>> S2MPS13 - bit 1 and bit 4. However vendor code sets only bit 4 (called
>>> WUDR there)?
>>>
>> No they are not, AUDR on s2mps13 is bit:1 where as it is bit:4 on s2mps15.
>>>
>>> What exactly is necessary to update alarm and time registers on S2MPS15?
>>>
>> As explained above in example, it is required to update the 'write alarm
>> buffer(AUDR)' and 'write time buffer(WUDR)' bits in-order to get rtc
>> working.
>
> For updating time and alarm registers? Or only for updating alarm registers?
>
example was only for the timer register.

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