Firmware Release v1.20.1

[robert-hh]

@iwahdan Hellol Islam. Thank you for the update. I have a few question and comments:

1. Why was the update pushed to the branch called Release/v1.20.1 and not master. These many new branches confuse.
2. The bug in the LTE module I reported in Issue #338 still persists, but may not always pop up:
   In handling of the attach command inside modlte.c, the length of the command AT+SMDD (and others) is not set! Lines affected: 310, 654, 670. Compare to these place the use of the function lteppp_send_at_command in line 179, where the data length IS set. A test confirmed the bug: if I specify band=8, I get an error: band 8 not supported. If I omit band=8, attach works.
3. There is still the glitch in handling the bits=xx argument in adc.init. Before, the only accepted argument & value for adc.init() was bits=12, and even that was discarded. With that change, the function meets the documentation and works as expected. See PR #327

Edit: Setting dataLen in the calls to lteppp_send_at_command() makes attach wotk again with specifying the band.
Edit2: Thank you for keeping the espressif reset mode option

[iwahdan]

@robert-hh said in Firmware Release v1.20.1:

@iwahdan Hellol Islam. Thank you for the update. I have a few question and comments:

1. Why was the update pushed to the branch called Release/v1.20.1 and not master. These many new branches confuse.

we have restructured our GitHub public Repo to have a separate branches for 1.18 and 1.20 FWs (under Release) - Daily development will be done on a new default branch (Dev) (previously we used to do the main Dev privately and do weekly/monthly releases to public repo).
This will make it easier for people to Contribute to Our Firmware plus better traceability of opened Issues.
we recommend you open any pull request on that Branch (Dev).

2. The bug in the LTE module I reported in Issue #338 still persists, but may not always pop up:
   In handling of the attach command inside modlte.c, the length of the command AT+SMDD (and others) is not set! Lines affected: 310, 654, 670. Compare to these place the use of the function lteppp_send_at_command in line 179, where the data length IS set. A test confirmed the bug: if I specify band=8, I get an error: band 8 not supported. If I omit band=8, attach works.

Can you open a PR on Dev branch?

3. There is still the glitch in handling the bits=xx argument in adc.init. Before, the only accepted argument & value for adc.init() was bits=12, and even that was discarded. With that change, the function meets the documentation and works as expected. See PR #327

Just changed Base of your PR to Dev, can you please update?

Edit: Setting dataLen in the calls to lteppp_send_at_command() makes attach wotk again with specifying the band.

[rskoniec]

@iwahdan I'm not able to change pycom.wifi_on_boot

>>> os.uname()
(sysname='FiPy', nodename='FiPy', release='1.20.1', version='v1.11-12f4ce0 on 2019-10-06', machine='FiPy with ESP32', lorawan='1.0.2', sigfox='1.0.1', pybytes='1.1.2')
>>> import pycom
>>> pycom.wifi_on_boot()
False
>>> pycom.wifi_on_boot(True)
>>> pycom.wifi_on_boot()
False

[robert-hh]

@rskoniec said in Firmware Release v1.20.1:

pycom.wifi_on_boot(True)

Does it change if you use:

pycom.wifi_on_boot(True, True)

The code allows a second parameter, which is not documented. But as I understand it, it forces the immediate startup of WiFi. It should not affect the stored state.

[robert-hh]

@rskoniec Do you use the PyBytes variant? According to the code this variant always returns False (and it does not set this flag). See the code snippet from pycom_config.c:

bool config_set_wifi_on_boot (uint8_t wifi_on_boot) {
    if (pycom_config_block.wifi_config.wifi_on_boot != wifi_on_boot) {
        pycom_config_block.wifi_config.wifi_on_boot = wifi_on_boot;
#if (VARIANT != PYBYTES)
        return config_write();
#endif
    }
    return true;
}

bool config_get_wifi_on_boot (void) {
#if (VARIANT == PYBYTES)
    return false;
#endif
    return pycom_config_block.wifi_config.wifi_on_boot;
}

[River]

Hello @iwahdan, promissing update! Thanks to all people involved.

I will test new firmware flashing a module later today. I have two questions regarding this update:

1. It's now ULP co-processor supported? We have a product with custom hardware + lopy4 and we want to add capability to count pulses on two digital inputs while sleeping (measuring water consumption, etc..)
2. It's possible to change wifi mode to LR (Espressif's Long Range protocol) by code?

If still not supported (1 or 2) I don't know if may open an issue on github in order to pycom keep track of improvements, or maybe is better to leave written here.

Regards

[cmisztur]

Thank you.

Regarding 'legacy' vs 'pybytes' variants. Does the PyBytes variant auto connect to PyBytes platform and maintain connection? What is the difference between 'legacy' and 'pybytes'?

[robert-hh]

@cmisztur Sorry, I never looked at pybytes.

[cmisztur]

Breaking change for me from 1.20.0.rc13 to 1.20.1 is usage of machine.idle().
Below, line 7 is never reached.

import utime
import _thread

def t():
    print('thread start')
    utime.sleep_ms(4000)
    print('thread run')

_thread.stack_size(8192)
_thread.start_new_thread(t, ())

utime.sleep_ms(1000)

while 1:
    machine.idle()
    #utime.sleep_ms(50)

[cmisztur]

@robert-hh said in Firmware Release v1.20.1:

@cmisztur Sorry, I never looked at pybytes.

How are you flashing 1.20.1?

[robert-hh]

@cmisztur I build the image myself, and as part of that, i can call:

make BOARD=xxPy flash

There are also the pyupgrade tools, you can download from the pycom web site.

[robert-hh]

@cmisztur That finding with idle is strange. Comparing the Pycom variant with the Micropython.org shows no difference in the machine.idle() implementation. But the Micropython.org version works. That's how the function looks in both variants:

STATIC mp_obj_t machine_idle(void) {
    taskYIELD();
    return mp_const_none;
}

[robert-hh]

@cmisztur I compared the related functions in machine.idle() and utime.sleep_ms() for both v1.20.0.rc13 and v1.20.1. They are identical, and just call taskYIELD() (idle) resp. vTaskDelay() (sleep_ms). So the change must be cause by moving from esp_idf v3.1 to esp_if v3.2.

[Clemens]

I have some devices with firmware 1.20.0.rc13 before and now 1.20.1. Unfortunately the old software is not running with the new firmware and I got e.g.

File "main.py", line 11, in <module>
File "webserver.py", line 124, in <module>
File "/flash/lib/microDNSSrv.py", line 19, in Create
File "/flash/lib/microDNSSrv.py", line 161, in Start
OSError: Network card not available

or

File "/flash/lib/terkin/network/core.py", line 110, in start_services
File "/flash/lib/terkin/network/core.py", line 131, in start_httpserver
File "/flash/lib/terkin/api/http.py", line 70, in start
File "/flash/dist-packages/microWebSrv.py", line 219, in Start
OSError: Network card not available

and my former working OTA-Code is stopping also

  File "main.py", line 44, in <module>
  File "/flash/lib/OTA.py", line 176, in connect
NameError: name 'SSID' isn't defined

[robert-hh]

@Clemens I had a similar problem too on one Lopy4 device - Wifi was not available. Erasing the flash and reloading the firmware solved it.

[Clemens]

@robert-hh said in Firmware Release v1.20.1:

pycom.wifi_on_boot(True, True)

>>> pycom.wifi_on_boot()
False
>>> pycom.wifi_on_boot(True, True)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
OSError: the requested operation failed

[robert-hh]

@Clemens That wifi_on_boot() issue is a different topic and has been fixed with today's build for the Dev branch. AFIK it is not yet in the 'official' version.
Edit: It looks like there is a lag in the Pycom update. The updater does not show V1.20.1, and the download page for firmware packages is also not updated.

[rcolistete]

Please update the links for all Pycom devices in :
https://docs.pycom.io/advance/downgrade/
to include the version 1.20.1 for manual firmware download.

[Clemens]

One software is after the repeated firmware update with the checked option erase filesystem now running. The second one outputs still the same error.

[berni]

After upgrading to 1.20.1, I started having crashes when setting up the frequencies for the Nano-Gateway code:

Guru Meditation Error: Core  1 panic'ed (Cache disabled but cached memory region accessed)
Guru Meditation Error: Core  1 panic'ed (IllegalInstruction). Exception was unhandled.
Memory dump at 0x40201f50: bad00bad bad00bad bad00bad
Guru Meditation Error: Core  1 panic'ed (IllegalInstruction). Exception was unhandled.
Memory dump at 0x40201f50: bad00bad bad00bad bad00bad
Guru Meditation Error: Core  1 panic'ed (IllegalInstruction). Exception was unhandled.

This happens somehow randomly, after a few cycles of deep sleep/transmission, but even the Watchdog timer becomes useless.

Then I downgraded to 1.20.rc13. Now, the code doesn't work and stops in this line:

>>> from network import LoRa
>>> lora = LoRa(mode=LoRa.LORAWAN)
>>> lora.nvram_erase()
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
OSError: the requested operation failed
>>> lora.nvram_save()
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
OSError: the requested operation failed

Is there any way to go back to 1.20.rc13 and have these function working again? I suspect there must be some low level changes on the LoRa radio.