initialization problem during boot

[paoloach]

Hi, I tried to run the example examples/sysparam_editor but as soon as it start I got the error:

[NOTE: No current sysparam region (initialization problem during boot?)]

and then a stack trace

==> @R@0L,:+2QQ3*|ppPP@2A!|Pl Setting '@R@0L,' to binary data... rtos assert /home/paolo/esp2866/esp-open-rtos/FreeRTOS/Source/queue.c 1406 abort() invoked at 0x40204743.

Stack: SP=0x3fff4270 0x3fff4270: 0000000d 3fff43b8 00000000 40204746 0x3fff4280: 3fff3a60 00000000 3fff02d8 40204f68 0x3fff4290: 00000000 00000000 3fff0378 4020489e

this is the command I used:

make -j4 -C examples/sysparam_editor flash ESPPORT=/dev/ttyUSB0 FLASH_SIZE=4

and the esptool flash_id results is

Manufacturer: e0 Device: 4016 Detected flash size: 4MB

Note: I have the same problem of system param region not initialized with the example/wificfg

In the user_init of my application I replicated the code into app_main.c:sdk_user_start function that I see it is used to initialize the sysparam region and it seems to work fine but I don't understand why it doesn't work in the sdk_user_start function.

Regards, Paolo

[ourairquality]

It should not be necessary to initialize the sysparam in user_start. The problem is that the region did not initialize correctly, so it did not create the semaphore (we can fix that). I suggest tracing through sysparam_init() to see where it is failing and that might give some clues.

[paoloach]

Hi, thank for the response, but now magically all works fine. I had begun to use some static variable in order to understand the real execution flow and I noticed the problem was gone. It reverted all my modification and again the problem was gone. The only modification from yesterday is I disconected the esp2866 board, and so I powered off it, instead of simply reset it.

[kanflo]

Feel free to reopen the issue if the problem comes back.

[MatzeS]

If somebody experiences the same problems: For me, running esptool erase_flash and reflashing the firmware resolved the issue.