Rust ESP32 Community Meeting 2024-02-15

[SergioGasquez]

Please add topics and links to this discussion.

The meeting is scheduled for 15th February 2024 - 5 pm CET/GMT+1 (a.k.a. Brno time) via Google Meet. The meeting invite is automatically sent to attendees of previous meetings. If you'd like to join, please notify us at https://matrix.to/#/#esp-rs-community-meetins:matrix.org

You can find information from the previous meeting at https://github.com/esp-rs/rust/discussions

If you find a topic exciting, click Arrow Icon or Emoji, and your vote will give the preference to the topic.

[georgik]

Rust Releases

• 1.76.0.1 - latest release with LLVM 17 and GCC 13 - https://github.com/esp-rs/rust-build/releases/tag/v1.76.0.1
• 1.77.0.0 - scheduled official release with LLVM 17 and GCC 13 - 24th April 2024

ESP-IDF Releases

• Available releases of ESP-IDF: 5.2-RC1, 5.1.2, 5.0.5, 4.4.6

Wokwi

• Wokwi + IDE - @Jozott00 created PR with several improvements of Wokwi IntelliJ/CLion/RustRover plugin, added support for esp-rs style of projects

[jessebraham]

ESPFLASH

• Remove support for the ESP8266 espflash#576
• Remove direct boot espflash#577
• Update dependencies and bump MSRV espflash#578

ESP-PACS

• Remove the esp8266 package esp-pacs#189
• ESP32-P4: Patch the USB_DEVICE peripheral esp-pacs#188
• Fix typo in field name for TRACE peripheral on C6/H2 esp-pacs#190
• Update xtask dependencies, regenerate PACs using latest svd2rust esp-pacs#191
• Patch the UART0 peripheral for ESP32-P4 esp-pacs#193
• Patch the HP_SYS_CLKRST peripheral for ESP32-P4 esp-pacs#194
• Update to latest SVDs and remove old patches esp-pacs#195

ESP-HAL

• "Fix" clippy checks in CI, resolve a bunch of clippy warnings esp-hal#1134
• Refactor the Trace driver to be generic around its peripheral esp-hal#1140
• Document the features of the remaining packages esp-hal#1143
• Implement UART support for ESP32-P4 esp-hal#1150
• Update PACs to their latest (unpubished) versions an make required changes esp-hal#1152
• Add the xtask package, implement automation for building examples esp-hal#1157
• Add an xtask subcommand for building documentation esp-hal#1160
• Remove clippy lint exceptions from CI workflow esp-hal#1167

MISCELLANEOUS

• Remove support for the ESP8266 esp-println#82
• Add the PAC to the repo, add deprecation notice to README.md esp8266-hal#25
• Published new SVDs
  • https://github.com/espressif/svd/releases/tag/2024-02-08

[jessebraham]

Note to self: give an update on the HAL unification

[bjoernQ]

esp-hal

• support stack-protector (also acts as "kind of" stack overflow protection)
  • Support Rust's stack-protector feature esp-hal#1135
• default to 32k ICACHE for ESP32-S3 - nice performance boost for esp-wifi
  • ESP32-S3: Use 32K ICACHE esp-hal#1169
• fix a subtle bug with circular DMA
  • Fix circular DMA esp-hal#1144
• remove unused dependency
  • Remove (unused) heapless dependency esp-hal#1148

esp-println

• have our logo in docs
  • Have our logo on docs esp-println#83
• release 0.9.0
  • https://crates.io/crates/esp-println/0.9.0

esp-backtrace

• make the output look nicer when on nightly compiler
  • Nicer output on nightly esp-backtrace#50
• fix docs.rs build
  • Fix docs.rs build esp-backtrace#51
• fixed some problems caused by the added complexity of defmt support
  • Fix: println without colors not printing esp-backtrace#53
• release 0.11.0
  • https://crates.io/crates/esp-backtrace/0.11.0

esp-wifi

• embedded-svc is now optional
  • Make embedded svc really optional esp-wifi-sys#429
• very nice performance improvement
  • Performance improvement esp-wifi-sys#430
• spend a lot of time looking into possible performance bottlenecks especially on Xtensa
  - interrupt context save and restore? No
  - dispatching to the interrupt handlers? Should get improved but not a real problem in esp-wifi
  - general improvements to the scheduler - e.g. skip a task if we know it cannot make progress? Would be good for the main task but doesn't get us better wifi performance
  - chip specific "hacks" - found something to boost C6 a lot but doesn't make any difference for other chips

espflash

• better output of resolved symbols
  • Improve resolving symbols espflash#581

[SergioGasquez]

awesome-esp-rust

• Add dark sky meter project awesome-esp-rust#34

book

• Add matrix room links book#141

esp-hal

• Autodetect xtal-freq esp-hal#1165

esp-template

• Update dependencies esp-template#143

espflash/esp-flasher-stub

• Working on fixing 26MHz targets: See Incorrect monitoring in 26MHz targets espflash#582

espup

• Released 0.11.0
• Prepare 0.11.0 release espup#416
• Begin next release cycle espup#417

[MabezDev]

compiler

• 1.76.0.0 and 1.76.0.1 rebased, 1.76.0.1 released, thanks @georgik!
  • There is a small regression in Xtensa llvm master for debug builds, hence I rolled back the LLVM version for the 1.76.0.1 release.
• 1.76 includes Pass +forced-atomics feature for riscv32{i,im,imc}-unknown-none-elf rust-lang/rust#114499 for the RISC-V targets.

probe-rs

• Halt the core after attaching so sequences can run freely probe-rs/probe-rs#2151 - ensure that sequences are run when halted.
• Ensure we have valid descriptor data when initially accessing espusbjtag probe-rs/probe-rs#2173 - ensure that the USB device is ready before communicating.
• ESP SoC Reset sequences probe-rs/probe-rs#2164 - WIP to allow flashing when memprot has been enabled. It works, but if the chip resets underneath the probe, the debug module gets confused and becomes unresponsive. The work around is to pass --catch-reset to that if the chip does reset the debugger will catch it and keep the state consistent. Unsure if we should merge as is, or try and improve the handling when --catch-reset is not passed.

xtensa-lx

• Fold in xtensa-lx-rt and core-isa parser xtensa-lx#30 - merge xtensa-lx-rt and xtensa-lx into one repo.
• Make saving and restoring float registers optional xtensa-lx#33 - make saving/restoring float registers optional.
• Remove esp8266 support xtensa-lx#31 - remove esp8266 support.
• Update spin to 0.9.8 xtensa-lx#32 - update spin.

[JurajSadel]

esp-hal

• lp-i2c
  • small progress, at least clocks are working correctly now
  • the initial sequence is correct, however it's not sending data
• lp-timers - started to work on this as a distraction from ^^

Other

• updated our DevCon demos to the latest hal/wifi versions
• https://docs.rs/crate/esp32c3-hal/latest and https://docs.rs/crate/esp32c2-hal/latest are broken