* pio: Move interrupt related (en|dis)abling/forcing methods to the statemachine
The SM knows its id using the type system so there's no need for checking it.
This commit also adds a `PioIRQ` enum to select the output IRQ.
* Move the interrupt control to Rx & Tx and make all accesses to inte atomic
* Adjust the ergonomics of IRQ handling.
* Elide lifetimes where they can be (clippy's advice)
* Prep for HAL 0.6.0 release
* Update changelog, readme and version number for HAL 0.5.0 release
* Bump HAL version in BSP deps
* Add the missing boards to the main README.md
* Implementation of the interpolator.
* corrected formatting
* fixed documentation code
* add clamp flag to LaneCtrl
* addition of an example for the interpolator
* put documentation behind ///
* rewording comment for clarity
* using more idiomatic fn new
There are a lot of non-binary crates depending on rp2040-hal. That way,
the default-features of rp2040-hal may be activated unintentionally
through an indirect dependency. Therefore, a binary crate which wants
to disable the `critical-section-impl` feature to provide its own one
could have a hard time to do so.
In contrast, the board support crates are usually only used by top-level
binary crates. So disabling the default features on those should usually
just work.
Binary crates depending on rp2040-hal directly, which don't use any
board support crate, might need to activate the feature manually. This
is reasonable because those binary crates need to replicate some
boilerplate from the board crates anyhow.
* Use rp2040-hal in all example (possibly through their bsp)
Some of the examples were using the cortex_m_rt::entry method which
misses the device specific spinlock re-initialisation.
This commits makes the usage more consistent by using rp2040_hal exported
macro as the only `entry` method used across examples.
This is a little bit hacky, as it relies on rp2040-hal actually
using (and therefore linking) some defmt code when the defmt feature is
enabled.
However this solution has the advantage that it only affects
dev-dependencies and therefore has no impact on the board crates
themselves.
Closes#420
* Implement rp2040-E5 workaround for usb enumeration.
* Expand documentation and add to pico_usb_serial & pico_usb_twitchy_mouse
* Fix errata-5 documentation around the bus-keep state.
* Update CHANGELOG.md
The workflow only shows "Runs action-rs/cargo@v1" which does not tell us
much about what's actually being done.
The list and clean up done after each build varies by only the debug/release
part of the path that is repeated several times (2).
This commit gives each step a more meaningful name and simplify the list &
cleanup step with a generic one liner that can be copy/pasted without
modification.
Without this change, building the vector_table example fails when
building from the rp2040-hal directory:
```
rp2040-hal$ cargo build --release --example vector_table
Compiling rp2040-hal v0.5.0 (rp2040-hal)
error[E0432]: unresolved import `pac::interrupt`
--> rp2040-hal/examples/vector_table.rs:30:5
|
30 | use pac::interrupt;
| ^^^^^---------
| | |
| | help: a similar name exists in the module (notice the capitalization): `Interrupt`
| no `interrupt` in the root
error: cannot determine resolution for the attribute macro `interrupt`
--> rp2040-hal/examples/vector_table.rs:148:3
|
148 | #[interrupt]
| ^^^^^^^^^
|
= note: import resolution is stuck, try simplifying macro imports
```
