669: Add SDMMC v1 and SDIO support r=Dirbaio a=chemicstry
SDMMC v2 peripheral is an extension of SDMMC v1 (or SDIO) so I managed to reuse most of the code, with some cfg's.
Apart from small differeces in registers, the biggest change is that v2 uses internal DMA, while v1 has to use shared DMA peripheral. This makes code a bit uglier, because DMA channel for v1 has to be passed around. Not sure if it's possible to make it any cleaner.
This also adds `TransferOptions` structure to DMA, because SDMMC v1 requires setting peripheral flow control and burst transfers. Let me know if some alternative way would be prefered.
I tested this on STM32F429ZIT6 (with sd card) and STM32H745ZIT6 (with oscilloscope).
Depends on: https://github.com/embassy-rs/stm32-data/pull/130
Co-authored-by: chemicstry <chemicstry@gmail.com>
673: Inline GPIO functions r=Dirbaio a=nviennot
All GPIO functions are monomorphized (per pin). Inlining these make the ROM smaller when using opt-level="z"
Co-authored-by: Nicolas Viennot <nicolas@viennot.biz>
* Keeps existing API for usart, but wraps it in Tx and Rx sub-types
* Adds split() method similar to nRF for getting indepdendent TX and RX
parts
* Implements e-h traits for TX and RX types
* Add stm32h7 example
Previously, every dma interrupt handler called the same `on_irq`
function which had to check the state of every dma channel.
Now, each dma interrupt handler only calls an `on_irq` method for its
corresponding channel or channels.
Approach is similar to USB OTG.
STM32L49x and STML4Ax have CAN1 and CAN2.
All others STM32L4xx do only have CAN1.
STM32F72x and STM32F73x are the only F7 devices with only CAN1.
652: Use new stm32-data registers and fix AHB clock calculation r=Dirbaio a=msamsonoff
This is the follow-on to my PR against stm32-data that added new register enums for the G0. I have updated the G0 RCC module to use those new enums.
I have also fixed an issue with the calculation of the AHB clock rate. 32 is not available as an AHB prescaler. The sequence jumps from 16 to 64. The original bit shifting math did not account for this gap. I have replaced it with a `match` instead.
653: Fixes for rustdoc building. r=Dirbaio a=Dirbaio
Co-authored-by: Matthew W. Samsonoff <matt.samsonoff@gmail.com>
Co-authored-by: Dario Nieuwenhuis <dirbaio@dirbaio.net>
The original code for calculating the AHB clock did not account for the gap in
prescaler values (32 is not an available value.) The bit shifting and math has
been replaced by a `match`.
These files end up containing chip-specific info, so they deduplicate
really badly when building docs for all 1200 chips.
The doc generator removes files starting with `_`.
* On v1 interrupts cannot be cleared individually.
Instead they are cleared implicitly by reading or writing DR (which we do now).
* Multiple error flags can be set at the same time:
Handle them all in one go intstead of re-entering the ISR for each one so that
we do not lose any error flags on v1 hardware.
* Wake when the RX buffer becomes full: This allows fast running chips to pull data
from the buffer before receiving the next byte.
608: stm32f4: add adc + example r=Dirbaio a=ain101
Example tested on stm32f407vg Discovery Board.
minimal adc: no vref, dma, complex sequence
Co-authored-by: Frederik <frederik@frederik.at>
601: [part 1/n] Change macrotables to build.rs codegen r=lulf a=Dirbaio
This PR replaces the "macrotables" (the macros like `stm32_data::peripherals!`) with a `const METADATA`.
Macrotables had some problems:
- Hard to debug
- Somewhat footgunny (typo the "pattern" and then nothing matches and the macro now expands to nothing, silently!)
- Limited power
- Can't count, so we had to add a [special macrotable for that](f50f3f0a73/embassy-stm32/src/dma/bdma.rs (L26)).
- Can't remove duplicates, so we had to fallback to [Rust code in build.rs](f50f3f0a73/embassy-stm32/build.rs (L105-L145))
- Can't include the results as a listto another macro, so again [build.rs](https://github.com/embassy-rs/embassy/blob/master/embassy-stm32/build.rs#L100-L101).
They work fine for the 95% of cases, but for the remaining 5% we need Rust code in build.rs. So we might as well do everything with Rust code, so everything is consistent.
The new approach generates a `const METADATA: Metadata = Metadata { ... }` with [these structs](https://github.com/embassy-rs/embassy/blob/unmacrotablize/stm32-metapac-gen/src/assets/metadata.rs) in `stm32-metapac`. `build.rs` can then read that and generate whatever code.
Co-authored-by: Dario Nieuwenhuis <dirbaio@dirbaio.net>
602: Add stm32 USB OTG peripherals r=Dirbaio a=chemicstry
Fixes#557. This is similar to #580, but for synopsys IP.
I could add examples to other chips, but I have no way of testing them. The F4 example is tested and working.
Co-authored-by: chemicstry <chemicstry@gmail.com>
The rcc code was taken from stm32-rs which uses 'x' features, but
embassy uses features with full chip names.
Add these 'x' wildcards as cfgs and use them in rcc.
They will be useful for USB too.
545: Add adapter for implementing async traits for blocking types r=lulf a=lulf
This allows writing drivers relying on async traits, while still
functioning with implementations that already implement the embedded-hal
traits.
Co-authored-by: Ulf Lilleengen <lulf@redhat.com>
563: Initial ADC support for on STM32F1xx r=Dirbaio a=sjoerdsimons
Add an ADC implementation for F1 based chips. Primarily tested using ADC1, proper functionality for ADC2 probably needs some extra work as it's mainly a slave and can't e.g. measure vrefint by itself.
Needs https://github.com/embassy-rs/stm32-data/pull/115
Co-authored-by: Sjoerd Simons <sjoerd@collabora.com>
Co-authored-by: Dario Nieuwenhuis <dirbaio@dirbaio.net>
558: Port buffered uart to v1 stm32 hardware r=Dirbaio a=DCNick3
#526 seems to suggest that it will be rewritten for DMA support, but I am not sure how to implement it and the port was quite straightforward, so here it is. It might be immediately useful before DMA version will be implemented
Note that I have not tested this on v2 hardware
Co-authored-by: Nikita Strygin <nikita6@bk.ru>
Previously disable_channel enabled rather than disabled the requested
channel due to an apparent copy-paste error. Refactor to eliminate this
sort of issue by construction.
This allows writing drivers relying on async traits, while still
functioning with implementations that already implement the embedded-hal
traits.
Add examples to stm32l4 for using this feature.
540: Initial support for STM32F3 r=Dirbaio a=VasanthakumarV
The [companion PR](https://github.com/embassy-rs/stm32-data/pull/109) in `stm32-data` should be merged before this PR.
The examples were tested on an STM32F303VC MCU.
Co-authored-by: VasanthakumarV <vasanth260m12@gmail.com>
Co-authored-by: Dario Nieuwenhuis <dirbaio@dirbaio.net>
`async fn`s do nothing until polled, but we want the DMA transfer to
immediately start in this case. Drivers rely on it. Some require special
orders, such as "start DMA, start SPI, then wait for DMA" which is awkward
to do without eager start.
Also use a manually-impl'd future, this allows getting rid of the "double"
Unborrow channel clone.
