RTC on CuBox-i

[Rudi]

Hi folks,


is it just my hardware or is the external RTC (pcf8523) no longer working with our current kernel ?


# hwclock -r
hwclock: RTC_RD_TIME: Invalid argument




--

Ruediger "Rudi" Ihle

[Thomas Genty]

Le 08/04/2014 20:15, Rudi a écrit :
> Hi folks,
Hello Rudi,

>
>
> is it just my hardware or is the external RTC (pcf8523) no longer
> working with our current kernel ?
>
>
> # hwclock -r
> hwclock: RTC_RD_TIME: Invalid argument
>
>
>

didn't get the same result
but it doesn't work here :
[Service]
Type=forking
RemainAfterExit=yes
TimeoutSec=0
ExecStart=/sbin/hwclock -s
ExecStop=/sbin/hwclock -w
StandardOutput=syslog

[Install]
WantedBy=multi-user.target
# hwclock -s
hwclock: settimeofday: Invalid argument
# hwclock -r
Thu Jan 1 00:59:59 1970 0.000000 seconds


Tom

[Rudi]

> # hwclock -s
> hwclock: settimeofday: Invalid argument
> # hwclock -r
> Thu Jan 1 00:59:59 1970 0.000000 seconds

Which CuBox-i model is this ? Looke a bit like you are accessing the other RTC...

Can you do a "ls -l /dev/rt*"



--

Ruediger "Rudi" Ihle

[Thomas Genty]

Le 08/04/2014 21:49, Rudi a écrit :
>
>> # hwclock -s
>> hwclock: settimeofday: Invalid argument
>> # hwclock -r
>> Thu Jan 1 00:59:59 1970 0.000000 seconds
>
> Which CuBox-i model is this ? Looke a bit like you are accessing the
> other RTC...
>

ciboxi4p

> Can you do a "ls -l /dev/rt*"
>
>
>

# ls -l /dev/rt*
lrwxrwxrwx 1 root root 4 /dev/rtc -> rtc0
crw------- 1 root root 254, 0 /dev/rtc0
crw------- 1 root root 254, 1 /dev/rtc1

# hwclock -f /dev/rtc0

Thu Jan 1 00:59:59 1970 0.000000 seconds

# hwclock -f /dev/rtc1
Thu Jan 1 00:29:47 1970 0.000000 seconds
# hwclock -f /dev/rtc1 -s
hwclock: settimeofday: Invalid argument
# hwclock -f /dev/rtc0 -s
hwclock: settimeofday: Invalid argument

[Rudi]

Hi Tom,

> # ls -l /dev/rt*
> lrwxrwxrwx 1 root root 4 /dev/rtc -> rtc0
> crw------- 1 root root 254, 0 /dev/rtc0
> crw------- 1 root root 254, 1 /dev/rtc1
>
> # hwclock -f /dev/rtc0
> Thu Jan 1 00:59:59 1970 0.000000 seconds
> # hwclock -f /dev/rtc1
> Thu Jan 1 00:29:47 1970 0.000000 seconds
> # hwclock -f /dev/rtc1 -s
> hwclock: settimeofday: Invalid argument
> # hwclock -f /dev/rtc0 -s
> hwclock: settimeofday: Invalid argument

Looks like our devices behave differently... What happens when you try set the RTCs
(i.e. hwclock -w") ?



--

Ruediger "Rudi" Ihle

[Thomas Genty]

Le 09/04/2014 08:38, Rudi a écrit :
> Hi Tom,

Hello Rudi,

it works here :

# hwclock -f /dev/rtc1 -s
hwclock: settimeofday: Invalid argument
# hwclock -f /dev/rtc0 -s
hwclock: settimeofday: Invalid argument

# hwclock -f /dev/rtc0 -w
# hwclock -f /dev/rtc1 -w
# hwclock -f /dev/rtc1
Wed Apr 9 07:39:33 2014 0.000000 seconds
# hwclock -f /dev/rtc0
Wed Apr 9 07:39:37 2014 0.000000 seconds

[Rudi]

Well, the "hwclock: settimeofday: Invalid argument" message shows up, when the RTC is
not set. The hwclock-sync service is supposed to write the system time to "rtc0" on
shutdown and reload it on boot. So normally you should not see this message for rtc0
unless you have a virgin CuBox or something went wrong with the battery backup.



--

Ruediger "Rudi" Ihle

[Thomas Genty]

I see no error now after a reboot
nor after a cold boot
maybe , as you suggested, it was a vigin Cubox ;)
>

[Rudi]

Hi Tom,

> I see no error now after a reboot
> nor after a cold boot
> maybe , as you suggested, it was a vigin Cubox ;)

Look what happens after a day or so. If there is a problem with
the RTC battery or this somewhat strange oscillator stuff touched
in the Feb 26 commit, the message might show up again.

https://github.com/SolidRun/linux-linaro-stable-mx6/commits/linux-linaro-lsk-mx6


The problem I have is a different one. My CuBox-i definitively has a hardware
issue...

- The PHY is not detected anymore (not even in u-boot). However, the
network activity LED blinks when a cable is connected.

- The SATA port never worked. Always got the message that no device is
attached.

- The microUSB can be made non-functional by applying mechanical pressure to
the PCB. This sometime happens when the screws in the bottom of the housing
are fittet to tight.





--

Ruediger "Rudi" Ihle

[Thomas Genty]

Le 09/04/2014 10:09, Rudi a écrit :
> Hi Tom,
Hell oRudi,
>

>> I see no error now after a reboot
>> nor after a cold boot
>> maybe , as you suggested, it was a vigin Cubox ;)
>
> Look what happens after a day or so. If there is a problem with
> the RTC battery or this somewhat strange oscillator stuff touched
> in the Feb 26 commit, the message might show up again.
>
> https://github.com/SolidRun/linux-linaro-stable-mx6/commits/linux-linaro-lsk-mx6

Ok I can wait 2 days or more and see what happens.

>
> The problem I have is a different one. My CuBox-i definitively has a hardware
> issue...
>
> - The PHY is not detected anymore (not even in u-boot). However, the
> network activity LED blinks when a cable is connected.

nothing here

>
> - The SATA port never worked. Always got the message that no device is
> attached.

didn't get the time to look at this ...

>
> - The microUSB can be made non-functional by applying mechanical pressure to
> the PCB. This sometime happens when the screws in the bottom of the housing
> are fittet to tight.
>

with the problem I had with HB1, I do carefully things ;-)

Rabeeh, can you comment on all this problems ?
>
>
>
Tom