1329: Reduce bootloader memory usage r=rmja a=rmja
By not requiring that the provided buffer must be able to contain one "erase-size" bytes.
This PR is the successor of #1314 and replaces it entirely.
Co-authored-by: Rasmus Melchior Jacobsen <rmja@laesoe.org>
1321: executor: add Pender, rework Cargo features. r=Dirbaio a=Dirbaio
This introduces a `Pender` struct with enum cases for thread-mode, interrupt-mode and
custom callback executors. This avoids calls through function pointers when using only
the thread or interrupt executors. Faster, and friendlier to `cargo-call-stack`.
`embassy-executor` now has `arch-xxx` Cargo features to select the arch and to enable
the builtin executors (thread and interrupt).
Co-authored-by: Dario Nieuwenhuis <dirbaio@dirbaio.net>
1313: (embassy-stm32): rework bufferedUart to get rid of PeripheralMutex r=Dirbaio a=MathiasKoch
New implementation is very similar to the implementation of embassy-nrf & embassy-rp.
Also adds embedded-hal traits to bufferedUart.
**NB**: Still needs testing on actual hardware
Co-authored-by: Mathias <mk@blackbird.online>
1324: Add MCO support for L4 and F4 families r=Dirbaio a=m-dupont
Add MCO support for L4 and F4 as already done in F7.
When the 'HSI' source is selected as MCO source, 'HSI' is activated (`set_hsion(true)`) . This is done to operate the MCO in case 'MSI' is chosen as the clock source for the CPU. The same applies to PLL, etc.
1327: Avoid write before erase r=Dirbaio a=rmja
This introduces an additional marker to the state partition right after the magic which indicates whether the current progress is valid or not. Validation in tests that we never write without an erase is added.
There is currently a FIXME in the FirmwareUpdater. Let me know if we should take the erase value as a parameter. I opened a feature request in embedded-storage to get this value in the trait. Before this, the assumption about ERASE_VALUE=0xFF was the same.
I have made some thoughts about whether this is a breaking change between the app and firmware, i.e. whether adding the "Progress valid" field is breaking. My conclusion is that it is not a breaking change. For the case where an app uses this new FirmwareUpdater together with an old bootloader, what it now does, is that it:
1. Writes the progress valid field to all zeros. This field is not known in the old bootloader, so it actually writes a "current progress" index.
2. The entire state partition is erased - effectively removing any trace of 1.
3. Set magic
This should be compatible.
Co-authored-by: Mathieu Dupont <mdupont@cppm.in2p3.fr>
Co-authored-by: Rasmus Melchior Jacobsen <rmja@laesoe.org>
1328: Expose read/write/erase on partition r=Dirbaio a=rmja
This is needed now where FirmwareWriter is removed.
Co-authored-by: Rasmus Melchior Jacobsen <rmja@laesoe.org>
This introduces an additional marker to the state partition right after the magic which indicates whether the current progress is valid or not. Validation in tests that we never write without an erase is added.
There is currently a FIXME in the FirmwareUpdater. Let me know if we should take the erase value as a parameter. I opened a feature request in embedded-storage to get this value in the trait. Before this, the assumption about ERASE_VALUE=0xFF was the same.
1322: Remove FirmwareWriter r=Dirbaio a=rmja
FirmwareWriter currently has a "max-write-size" parameter, but this is a limitation that should be handled by chunking inside the NorFlash driver, and not "up here" in user code. In case that the driver (e.g. qspi driver) is unaware of any max-write limitations, one could simply add an intermediate NorFlash adapter providing the chunk'ing capability.
Co-authored-by: Rasmus Melchior Jacobsen <rmja@laesoe.org>
FirmwareWriter currently has a "max-write-size" parameter, but this is a limitation that should be handled by chunking inside the NorFlash driver, and not "up here" in user code. In case that the driver (e.g. qspi driver) is unaware of any max-write limitations, one could simply add an intermediate NorFlash adapter providing the chunk'ing capability.
1312: Let bootloader partition have read/write/erase operations r=Dirbaio a=rmja
This change should not have any breaking changes.
Co-authored-by: Rasmus Melchior Jacobsen <rmja@laesoe.org>
This introduces a `Pender` struct with enum cases for thread-mode, interrupt-mode and
custom callback executors. This avoids calls through function pointers when using only
the thread or interrupt executors. Faster, and friendlier to `cargo-call-stack`.
`embassy-executor` now has `arch-xxx` Cargo features to select the arch and to enable
the builtin executors (thread and interrupt).
1318: rp: Allow zero len reads for buffered uart r=Dirbaio a=timokroeger
Prevents the read methods from getting stuck forever.
cc `@MathiasKoch` can you test if this fixes the problem you described in the chat?
Co-authored-by: Timo Kröger <timokroeger93@gmail.com>
1311: Split bootloader implementation into multiple files r=lulf a=rmja
No other change has been made. I think that I have all the types re-exported in lib.rs as before.
Co-authored-by: Rasmus Melchior Jacobsen <rmja@laesoe.org>
This is a `core` patch to make wakers 1 word (the task pointer) instead of 2 (task pointer + vtable). It allows having the "waker optimization" we had a while back on `WakerRegistration/AtomicWaker`, but EVERYWHERE, without patching all crates.
Advantages:
- Less memory usage.
- Faster.
- `AtomicWaker` can actually use atomics to load/store the waker, No critical section needed.
- No `dyn` call, which means `cargo-call-stack` can now see through wakes.
Disadvantages:
- You have to patch `core`...
- Breaks all executors and other things that create wakers, unless they opt in to using the new `from_ptr` API.
How to use:
- Run this shell script to patch `core`. https://gist.github.com/Dirbaio/c67da7cf318515181539122c9d32b395
- Enable `build-std`
- Enable `build-std-features = core/turbowakers`
- Enable feature `turbowakers` in `embassy-executor`, `embassy-sync`.
- Make sure you have no other crate creating wakers other than `embassy-executor`. These will panic at runtime.
Note that the patched `core` is equivalent to the unpached one when the `turbowakers` feature is not enabled, so it should be fine to leave it there.
1294: Add support for `QSPI` in `stm32` r=Dirbaio a=Mirror0
Implemented with help of Tomasz Grześ <tomasz.grzes@gmail.com>.
Tested only on stm32f777zi.
Co-authored-by: Mateusz Butkiewicz <mateusz@github.butkiewicz.dev>
