Dear Paul,

Sorry for the delay.

I uploaded another build from current master to the board_status repository.

I disabled the serial console and rebooted twice before uploading the data.

I think RAM initialization time is more or less the same.

Is there anything else I can do to debug this?

Best Regards,

- Eli

On 16/04/18 09:07, Paul Menzel wrote:
Dear Eli,


Am Montag, den 12.02.2018, 11:55 +0100 schrieb Elisenda Cuadros:

Sure, I will be happy to do it :-)
Thank you very much for uploading the data to the board status
repository [1].

I noticed, that the serial console is enabled in your run, which causes
the boot time to be quite slow as all messages have to be sent over the
serial cable.

Could you please do another upload with the serial console disabled?
You can just use latest commit from the master branch. That data would
be nice to have to keep track of possible boot time regressions.

Also, RAM initialization took 1.7 seconds according to the time stamps.

```
   2:before ram initialization                         178,042 (174,580)
   3:after ram initialization                          1,878,897 (1,700,855)
```

If it is really that long – I didn’t see any log messages, so it’s
shouldn’t be the serial console –, could you do at least two reboots to
see if it’s due to an empty cache of the RAM data?


Kind regards,

Paul


[1] asus/am1i-a/4.7-294-gac05cbdea4/2018-02-12T15_31_32Z