search

meshtastic / firmware

Meshtastic device firmware
https://meshtastic.org
GNU General Public License v3.0
3.63k stars 909 forks source link

Heltec Lora32 is not really the same as ttgo lora32 #17

Closed geeksville closed 4 years ago

geeksville commented 4 years ago

theory, it is the flash partitioning.

I think I have a pretty good theory on what is wrong with programming your device (I just reproed it on a similar device of mine). Some kind soul reported that the HELTEC image works fine on the TTGO LoRa32, but I think that somewhat depends on luck because I think TTGO and Heltec partition their flash somewhat differently. I'll need to make a small script to let you program your device reliably. I'll do that tonight.

geeksville commented 4 years ago

cc @bcanderson

BCAnderson commented 4 years ago

I have to apologize but I think building this is beyond my capabilities. looks like I'm no help there, but I'd be happy to try whatever you like on my board.

geeksville commented 4 years ago

no worries. I'll work on this tonight/tomorrow.

geeksville commented 4 years ago

Hi @bcanderson,

Can you try running the following three commands on your board (using the binary file I'm attaching)? This will allow me to test my theory. After running these commands, power cycle the board and see if it boots?

Note: the two files you flash have different starting addresses

kevinh@kevin-server:~/development/meshtastic/meshtastic-esp32$ esptool.py --baud 921600 erase_flash
esptool.py v2.6
Found 2 serial ports
Serial port /dev/ttyUSB0
Connecting.....
Detecting chip type... ESP32
Chip is ESP32D0WDQ6 (revision 1)
Features: WiFi, BT, Dual Core, 240MHz, VRef calibration in efuse, Coding Scheme None
MAC: 24:6f:28:b5:36:7c
Uploading stub...
Running stub...
Stub running...
Changing baud rate to 921600
Changed.
Erasing flash (this may take a while)...
Chip erase completed successfully in 16.2s
Hard resetting via RTS pin...
kevinh@kevin-server:~/development/meshtastic/meshtastic-esp32$ esptool.py --baud 921600 write_flash 0x1000 images/system-info.bin 
esptool.py v2.6
Found 2 serial ports
Serial port /dev/ttyUSB0
Connecting......
Detecting chip type... ESP32
Chip is ESP32D0WDQ6 (revision 1)
Features: WiFi, BT, Dual Core, 240MHz, VRef calibration in efuse, Coding Scheme None
MAC: 24:6f:28:b5:36:7c
Uploading stub...
Running stub...
Stub running...
Changing baud rate to 921600
Changed.
Configuring flash size...
Auto-detected Flash size: 8MB
Compressed 61440 bytes to 11950...
Wrote 61440 bytes (11950 compressed) at 0x00001000 in 0.1 seconds (effective 3569.9 kbit/s)...
Hash of data verified.

Leaving...
Hard resetting via RTS pin...

You should use the correct file for your region though...

kevinh@kevin-server:~/development/meshtastic/meshtastic-esp32$ esptool.py --baud 921600 write_flash 0x10000 release/latest/firmware-HELTEC-US-0.1.3.bin 
esptool.py v2.6
Found 2 serial ports
Serial port /dev/ttyUSB0
Connecting....
Detecting chip type... ESP32
Chip is ESP32D0WDQ6 (revision 1)
Features: WiFi, BT, Dual Core, 240MHz, VRef calibration in efuse, Coding Scheme None
MAC: 24:6f:28:b5:36:7c
Uploading stub...
Running stub...
Stub running...
Changing baud rate to 921600
Changed.
Configuring flash size...
Auto-detected Flash size: 8MB
Compressed 1202448 bytes to 663131...
Wrote 1202448 bytes (663131 compressed) at 0x00010000 in 10.7 seconds (effective 900.7 kbit/s)...
Hash of data verified.

Leaving...
Hard resetting via RTS pin...

system-info.zip

claesg commented 4 years ago

Hi, Just let me know if you want me to run the same thing on my TTGO Lora32 which is the one that can run the Heltec build.

illperipherals commented 4 years ago

Nice, that worked on my TTGO LoRa boards that came in today. Was otherwise going into a crash loop with: 


configsip: 0, SPIWP:0xee
clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
mode:DIO, clock div:1
load:0x3fff0010,len:4
load:0x3fff0014,len:588
load:0x40078000,len:0
load:0x40078000,len:9768
entry 0x400789b8
Guru Meditation Error: Core  0 panic'ed (LoadProhibited). Exception was unhandled.
Core 0 register dump:
PC      : 0x40084137  PS      : 0x00060e30  A0      : 0x80085ee5  A1      : 0x3ffbbd30
A2      : 0x3ffbbd58  A3      : 0xbde04382  A4      : 0xfdd04382  A5      : 0x00000000
A6      : 0x00000000  A7      : 0x00000000  A8      : 0x00013ffc  A9      : 0x3ffbbd10
A10     : 0x00000000  A11     : 0x3ff42000  A12     : 0x00000040  A13     : 0x3ff420c0
A14     : 0x3ffba7a8  A15     : 0x00000004  SAR     : 0x0000000f  EXCCAUSE: 0x0000001c
EXCVADDR: 0xfdd04382  LBEG    : 0x4000c46c  LEND    : 0x4000c477  LCOUNT  : 0x00000000

Backtrace: 0x40084137:0x3ffbbd30 0x40085ee2:0x3ffbbd50 0x400e7699:0x3ffbbd80 0x400e4078:0x3ffbbda0 0x400e3fad:0x3ffbbdc0 0x400e4b67:0x3ffbbde0 0x400e729a:0x3ffbbe10 0x400913c9:0x3ffbbe30

Rebooting...
ets Jun  8 2016 00:22:57```
BCAnderson commented 4 years ago

"> Nice, that worked on my TTGO LoRa boards that came in today. Was otherwise going into a crash loop with:"

My TTGO LoRa board is stuck in the same (or similar) crash loop. What did you do??

illperipherals commented 4 years ago

I just ran the three commands above with the bin that is attached.


esptool.py --port COM10 --baud 921600 erase_flash
esptool.py --port COM10 --baud 921600 write_flash 0x1000 system-info.bin
esptool.py --port COM10 --baud 921600 write_flash 0x10000 firmware-HELTEC-US-0.1.6.bin
BCAnderson commented 4 years ago

I did this: esptool.py --port /dev/ttyUSB0 --baud 921600 erase_flash esptool.py --port /dev/ttyUSB0 --baud 921600 write_flash 0x1000 system-info.bin esptool.py --port /dev/ttyUSB0 --baud 921600 write_flash 0x10000 firmware-HELTEC-US-0.1.6.bin and got this:

"ets Jun 8 2016 00:22:57

rst:0xc (SW_CPU_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT) configsip: 188777542, SPIWP:0xee clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00 mode:DIO, clock div:2 load:0x3fff0018,len:4 load:0x3fff001c,len:1044 load:0x40078000,len:8896 load:0x40080400,len:5828 entry 0x400806ac Guru Meditation Error: Core 0 panic'ed (IllegalInstruction). Exception was unhandled. Memory dump at 0x40153630: 74010404 51010444 50511501 Core 0 register dump: PC : 0x40153634 PS : 0x00050030 A0 : 0x800e5bbe A1 : 0x3ffbbdc0
A2 : 0x8001f880 A3 : 0x8001f880 A4 : 0x400fe632 A5 : 0xbffa2b3e
A6 : 0x00060c20 A7 : 0x00000000 A8 : 0x800e500c A9 : 0x3ffbbda0
A10 : 0x15401540 A11 : 0x00000001 A12 : 0x3ffd1589 A13 : 0x00000001
A14 : 0xffffffff A15 : 0x00000000 SAR : 0x0000000f EXCCAUSE: 0x00000000
EXCVADDR: 0x00000000 LBEG : 0x4000c2e0 LEND : 0x4000c2f6 LCOUNT : 0xffffffff

Backtrace: 0x40153634:0x3ffbbdc0 0x400e5bbb:0x3ffbbde0 0x400e82ee:0x3ffbbe10 0x400913c9:0x3ffbbe30

Rebooting..."

So no change for me. :(

geeksville commented 4 years ago

@BCAnderson later today I'll push out a 0.1.7 build that takes advantage of a new stacktrace feature girts added. If you try running that load and give back the stacktrace in this bug we might be able to remote debug your board from here.

BCAnderson commented 4 years ago

Howdy, I tried 1.7 esptool.py --port /dev/ttyUSB0 --baud 921600 erase_flash esptool.py --port /dev/ttyUSB0 --baud 921600 write_flash 0x1000 system-info.bin esptool.py --port /dev/ttyUSB0 --baud 921600 write_flash 0x10000 firmware-HELTEC-US-0.1.7.bin

And I got this:

ets Jun 8 2016 00:22:57

rst:0xc (SW_CPU_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT) configsip: 188777542, SPIWP:0xee clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00 mode:DIO, clock div:2 load:0x3fff0018,len:4 load:0x3fff001c,len:1044 load:0x40078000,len:8896 load:0x40080400,len:5828 entry 0x400806ac Guru Meditation Error: Core 0 panic'ed (IllegalInstruction). Exception was unhandled. Core 0 register dump: PC : 0x55445400 PS : 0x00060030 A0 : 0x800e5ba8 A1 : 0x3ffbbf20
A2 : 0x40111450 A3 : 0x00000001 A4 : 0x3ffd20a4 A5 : 0x00000001
A6 : 0xffffffff A7 : 0x00000000 A8 : 0x800ff1f5 A9 : 0x3ffbbf00
A10 : 0x3ffba3d8 A11 : 0x00000000 A12 : 0xffffffff A13 : 0x00000000
A14 : 0x0000001f A15 : 0x00000000 SAR : 0x0000000f EXCCAUSE: 0x00000000
EXCVADDR: 0x00000000 LBEG : 0x4000c2e0 LEND : 0x4000c2f6 LCOUNT : 0xffffffff

Backtrace: 0x55445400:0x3ffbbf20 0x400e5ba5:0x3ffbbf40 0x400e6757:0x3ffbbf60 0x400e8e8a:0x3ffbbf90 0x400913c9:0x3ffbbfb0

Rebooting...

It's a different boot error but it still continually reboots. How do I use the tool to capture the system portion of -flash?

geeksville commented 4 years ago

@BCAnderson you shouldn't need to run that tool, because I grabbed the image here for you to use.

Alas, I think I'll need to come back to this tomorrow because (as unlikely as it sounds) I think I might actually be slightly coronamotized - and I'm going to go lay down.

BCAnderson commented 4 years ago

Howdy Kevin no change with 0.1.8

ets Jun 8 2016 00:22:57

rst:0xc (SW_CPU_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT) configsip: 188777542, SPIWP:0xee clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00 mode:DIO, clock div:2 load:0x3fff0018,len:4 load:0x3fff001c,len:1044 load:0x40078000,len:8896 load:0x40080400,len:5828 entry 0x400806ac Guru Meditation Error: Core 0 panic'ed (LoadProhibited). Exception was unhandled. Core 0 register dump: PC : 0x400ff646 PS : 0x00060030 A0 : 0x800e6020 A1 : 0x3ffbc080
A2 : 0x15404101 A3 : 0x00000001 A4 : 0x05015500 A5 : 0x00000001
A6 : 0xffffffff A7 : 0x00000000 A8 : 0x800e6104 A9 : 0x3ffbc060
A10 : 0x00000011 A11 : 0x00000000 A12 : 0x00000000 A13 : 0x3ffbd908
A14 : 0x0000001f A15 : 0x00000000 SAR : 0x0000000f EXCCAUSE: 0x0000001c
EXCVADDR: 0x05015500 LBEG : 0x4000c2e0 LEND : 0x4000c2f6 LCOUNT : 0xffffffff

Backtrace: 0x400ff646:0x3ffbc080 0x400e601d:0x3ffbc0a0 0x400e6bcf:0x3ffbc0c0 0x400e9302:0x3ffbc0f0 0x400913c9:0x3ffbc110

Rebooting...

geeksville commented 4 years ago

@BCAnderson okay - I'm back from a six day flu/covid break. I'll check this stack trace and comment later today.

geeksville commented 4 years ago

@BCAnderson sorry to ask this, but can I ask you to try this one more time with 0.1.9 and send in the stack trace? I would do it myself, but I don't have a board from your mfg.

for some reason, just using the git tags is not allowing me to generate a build that makes sense for that stack trace. For instance, for that trace I get:

kevinh@kevin-server:~/development/meshtastic/meshtastic-esp32$ ./bin/exception_decoder.py /tmp/stack 
stack:
0x400d8df1: OLEDDisplay::getStringWidth(char const*, unsigned short) at /home/kevinh/development/meshtastic/meshtastic-esp32/.pio/libdeps/esp32/ESP8266_SSD1306/src/OLEDDisplay.cpp:813
0x400d973f: OLEDDisplayUi::tick() at /home/kevinh/development/meshtastic/meshtastic-esp32/.pio/libdeps/esp32/ESP8266_SSD1306/src/OLEDDisplayUi.cpp:219 (discriminator 4)
0x400d97e1: OLEDDisplayUi::update() at /home/kevinh/development/meshtastic/meshtastic-esp32/.pio/libdeps/esp32/ESP8266_SSD1306/src/OLEDDisplayUi.cpp:219 (discriminator 4)
0x400d64f5: meshtastic::drawNodeInfo(OLEDDisplay*, OLEDDisplayUiState*, short, short) at /home/kevinh/development/meshtastic/meshtastic-esp32/src/screen.cpp:337
0x400d5442: initBluetooth() at /home/kevinh/development/meshtastic/meshtastic-esp32/src/main.cpp:277 (discriminator 8)
0x400e735b: _Unwind_ForcedUnwind at /builds/idf/crosstool-NG/.build/src/gcc-5.2.0/libgcc/unwind.inc:209
0x400913c9: vPortTaskWrapper at /home/runner/work/esp32-arduino-lib-builder/esp32-arduino-lib-builder/esp-idf/components/freertos/port.c:355 (discriminator 1)

So I'm now doing what I should have done all along, I'm including the elf file (which contains symbols with all builds, so in the future we can always generate the correct stack trace:

https://github.com/geeksville/Meshtastic-esp32/commit/3a756b0e088bb8a31769a01b32052f2482385b1c

geeksville commented 4 years ago

installer script item moved to https://github.com/meshtastic/Meshtastic-esp32/issues/27

BCAnderson commented 4 years ago

Welcome back Kevin

No change with 0.1.9

ets Jun 8 2016 00:22:57

rst:0xc (SW_CPU_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT) configsip: 188777542, SPIWP:0xee clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00 mode:DIO, clock div:2 load:0x3fff0018,len:4 load:0x3fff001c,len:1044 load:0x40078000,len:8896 load:0x40080400,len:5828 entry 0x400806ac Guru Meditation Error: Core 0 panic'ed (LoadProhibited). Exception was unhandled. Core 0 register dump: PC : 0x400ff756 PS : 0x00060030 A0 : 0x800e6130 A1 : 0x3ffbc080
A2 : 0x10000000 A3 : 0x00000001 A4 : 0x00000000 A5 : 0x00000001
A6 : 0xffffffff A7 : 0x00000000 A8 : 0x800e6214 A9 : 0x3ffbc060
A10 : 0x00000011 A11 : 0x00000000 A12 : 0x00000000 A13 : 0x3ffbd908
A14 : 0x0000001f A15 : 0x00000000 SAR : 0x0000000f EXCCAUSE: 0x0000001c
EXCVADDR: 0x00000000 LBEG : 0x4000c2e0 LEND : 0x4000c2f6 LCOUNT : 0xffffffff

Backtrace: 0x400ff756:0x3ffbc080 0x400e612d:0x3ffbc0a0 0x400e6cdf:0x3ffbc0c0 0x400e9412:0x3ffbc0f0 0x400913c9:0x3ffbc110

Rebooting...

BCAnderson commented 4 years ago

I apologize for my ignorance, but how do I get the stack trace?

geeksville commented 4 years ago

re: stacktrace No worries, that was the "backtrace" line you just provided. And wow. I thought I just had a mismatch of my elf file compared to what was failing on your clone board. And nope, that really was the failure. I'm currently thinking about what this means.

kevinh@kevin-server:~/development/meshtastic/meshtastic-esp32$ ./bin/exception_decoder.py -e release/latest/firmware-HELTEC-US-0.1.9.elf /tmp/stack 
stack:
0x400d8df1: OLEDDisplay::drawInternal(short, short, short, short, unsigned char const*, unsigned short, unsigned short) at /home/kevinh/development/meshtastic/meshtastic-esp32/.pio/libdeps/esp32/ESP8266_SSD1306/src/OLEDDisplay.cpp:813
  \-> inlined by: OLEDDisplay::drawStringInternal(short, short, char*, unsigned short, unsigned short) at /home/kevinh/development/meshtastic/meshtastic-esp32/.pio/libdeps/esp32/ESP8266_SSD1306/src/OLEDDisplay.cpp:514
0x400d973f: OLEDDisplayUi::drawIndicator() at /home/kevinh/development/meshtastic/meshtastic-esp32/.pio/libdeps/esp32/ESP8266_SSD1306/src/OLEDDisplayUi.cpp:219 (discriminator 4)
0x400d97e1: OLEDDisplayUi::drawIndicator() at /home/kevinh/development/meshtastic/meshtastic-esp32/.pio/libdeps/esp32/ESP8266_SSD1306/src/OLEDDisplayUi.cpp:219 (discriminator 4)
0x400d64f5: meshtastic::drawNodeInfo(OLEDDisplay*, OLEDDisplayUiState*, short, short) at /home/kevinh/development/meshtastic/meshtastic-esp32/src/screen.cpp:216
  \-> inlined by: drawNodeInfo at /home/kevinh/development/meshtastic/meshtastic-esp32/src/screen.cpp:359
0x400d5442: setup() at /home/kevinh/development/meshtastic/meshtastic-esp32/src/main.cpp:213
0x400e735b: _Unwind_FindEnclosingFunction at /builds/idf/crosstool-NG/.build/src/gcc-5.2.0/libgcc/config/xtensa/unwind-dw2-xtensa.c:193
0x400913c9: vPortTaskWrapper at /home/runner/work/esp32-arduino-lib-builder/esp32-arduino-lib-builder/esp-idf/components/freertos/port.c:355 (discriminator 1)
girtsf commented 4 years ago

Maybe the nodeinfo changes from underneath it as it is drawing?

geeksville commented 4 years ago

@girtsf yeah, but every time and this occurs early in the boot, before printing anything else.

geeksville commented 4 years ago

actually never mind. There must be something wrong with how I'm running exception_decoder. I just used a completely different backtrace and got the same output. Looking...

geeksville commented 4 years ago

yeah - I was dumb. fixing.

geeksville commented 4 years ago

better - this stack trace makes sense. Though it seems like a totally normal startup except we blew up the first time calling the esp library. Puzzling... I'll think about this later today.

kevinh@kevin-server:~/development/meshtastic/meshtastic-esp32$ ./bin/exception_decoder.py -e release/latest/firmware-HELTEC-US-0.1.9.elf ~/tmp/stack 
stack:
0x400ff756: esp_log_level_set at /home/runner/work/esp32-arduino-lib-builder/esp32-arduino-lib-builder/esp-idf/components/log/log.c:125
0x400e612d: initArduino at /home/kevinh/.platformio/packages/framework-arduinoespressif32/cores/esp32/esp32-hal-misc.c:202
0x400e6cdf: app_main at /home/kevinh/.platformio/packages/framework-arduinoespressif32/cores/esp32/main.cpp:26
0x400e9412: main_task at /home/runner/work/esp32-arduino-lib-builder/esp32-arduino-lib-builder/esp-idf/components/esp32/cpu_start.c:506
0x400913c9: vPortTaskWrapper at /home/runner/work/esp32-arduino-lib-builder/esp32-arduino-lib-builder/esp-idf/components/freertos/port.c:355 (discriminator 1)
kevinh@kevin-server:~/development/meshtastic/meshtastic-esp32$
Gotfrogs commented 4 years ago

I am not a programmer but did manage to complete a project using two Heltec WiFi LoRa 32 (V2) boards and sample code from online. I would like to make more of them but also save some cash. Can I use my same code on the TTGO LORA32 version of the board?

geeksville commented 4 years ago

@Gotfrogs sorry - I didn't notice this comment until just now (for questions like this, probably faster response to post in the development forum on meshtastic.discourse.group)

Are you asking about meshtastic? If so, it already runs on either the Heltec Lora32 or the TTGO lora32 (there are multiple build targets in our platformio.ini and we generated prebuilt bins for both of those boards). If you are asking about your own project, the two boards are almost the same - but a few gpios are different. See our configuration.h