.. and switch makerfabs tft7 over to it as a test.
We have our existing way of "reserving" PSRAM for esp-idf (we actually
control it all but add back the "reserved" part). However, this does
not work with off the shelf esp_lcd, which only will allocate a
framebuffer in PSRAM if CONFIG_SPIRAM_USE_CAPS_ALLOC (or CONFIG_SPIRAM_USE_ALLOC)
is defined, not if CONFIG_SPIRAM_USE_MEMMAP is.
This new way is possibly compatible with more esp-idf code, but it complicates
CircuitPython's initial startup since nothing until port_heap_init is
permitted to use the CP heap or supervisor allocator. In practice this
seems to be OK today.
Right now this doesn't change the setting across all boards with PSRAM and so
it does not revert esp-idf to its prior state. Instead, what I'm thinking is
that we can do it during or just after the IDF5 update when sdkconfig files
will be getting an overhaul anyway.
With DEBUG build and a consle UART, this would occur early during startup on makerfabs tft7:
```
I (0) cpu_start: Starting scheduler on APP CPU.
I (10) uart: queue free spaces: 20
***ERROR*** A stack overflow in task uart_event_task has been detected.
Backtrace: 0x403786f2:0x3fce9f40 0x403820a9:0x3fce9f60 0x403850da:0x3fce9f80 0x40383a7d:0x3fcea000 0x40382158:0x3fcea030 0x4038214e:0xa5a5a5a5 |<-CORRUPTED
```
Decoded backtrace was not enlightening:
```
0x403786f2: panic_abort at /home/jepler/src/circuitpython/ports/espressif/build-makerfabs_tft7/esp-idf/../../esp-idf/components/esp_system/panic.c:408
0x403820a9: esp_system_abort at /home/jepler/src/circuitpython/ports/espressif/build-makerfabs_tft7/esp-idf/../../esp-idf/components/esp_system/esp_system.c:137
0x403850da: vApplicationStackOverflowHook at /home/jepler/src/circuitpython/ports/espressif/build-makerfabs_tft7/esp-idf/../../esp-idf/components/freertos/port/xtensa/port.c:407
0x40383a7d: vTaskSwitchContext at /home/jepler/src/circuitpython/ports/espressif/build-makerfabs_tft7/esp-idf/../../esp-idf/components/freertos/tasks.c:3505
0x40382158: _frxt_dispatch at /home/jepler/src/circuitpython/ports/espressif/build-makerfabs_tft7/esp-idf/../../esp-idf/components/freertos/port/xtensa/portasm.S:436
0x4038214e: _frxt_int_exit at /home/jepler/src/circuitpython/ports/espressif/build-makerfabs_tft7/esp-idf/../../esp-idf/components/freertos/port/xtensa/portasm.S:231
0x00000000: ?? ??:0
```
Adding an additional 512 bytes of stack allowed CircuitPython to start
successfully.
and re-organize so that esp32 s2/s3 don't do as much at reset
.. it's not necessary (because most data is in esp-idf managed memory)
and doing this saves me from having to debug why reconstruct isn't working
properly on that platform.
This needs to be tested on other platforms again before being merged!
ESP32-S3 defines two additional general use pins in
ports/espressif/peripherals/esp32s3/pins.h, for which
support is missing in the microcontroller module HAL.
esp_ping_new_session can fail, particularly if ping is called quickly
many times in succession.
This is because `esp_ping_new_session` has to do a bunch of stuff
including creating a task and a socket. Calling `esp_ping_delete_session`
doesn't clean up these resources immediately. Instead, it signals the
task to clean up resources and exit 'soon', but 'soon' is defined as 1
second.
When the calls are frequent, the in-use sockets and tasks fill up
available slots—I didn't actually check which resource gets used
up first.
With this change, the ping call will raise an exception instead of
continuing with a call to esp_ping_start that crashes.
Closes#5980 based on my testing on an ESP32S3-N8R2.
Fixes polling thread looping forever hangs preventing new connections.
Don't lose listening sockets on mp resets and re-init.
Keep better separation of "system" and "user" sockets.
Track socket states to prevent re-use of sockets before closed.
Close REST socket when transaction completes. No post-init.
Remove unnecessary state flags.
Scott Shawcroft
MicroDev
Dan Halbert
Bill Sideris
Jeff Epler
Bob Abeles
Bojan Potočnik
anecdata
Milind
Ted Hess