Add basic Embassy support #53

Serhii-the-Dev · 2024-09-08T21:23:03Z

Resolves #49

Embassy time driver implementation is mostly a copy-pasta of the STM32 driver except it's not implementing RTC functionality required for low-power/deep-sleep scenarios in order to keep it simple (AFAIK current HAL implementation also lacks of RTC support).

This PR brings a set of new features:

embassy - enables Embassy support in general
time-driver-tim<1, 2, 14> - enables Embassy time driver support for a specified general-purpose timer instance

Both features are required to use the Embassy executor, so any of the time-driver-tim<1, 2, 14> features enables the embassy feature in general.

Despite that the driver uses unsafe access to the selected timer's registry, in order to guard an access to a TIMERx peripheral, the driver initialization requires to move a timer<1, 2, 14> so it couldn't be used anywhere else causing undefined behavior, ex:

embassy::init(p.timer1, &clocks, &mut rcu.apb1);

After that the p.timer1 could not be moved, to create a Timer::timer1, for example.

Please take into account that Embassy support comes with a price, as a binary relying on the embassy feature and using the async executor would have additional ~2KiB added to it's size (also an extra memory would be required for the tasks arena):

> cargo bloat --release --features gd32f130x8,time-driver-tim1 --example embassy
...
File  .text   Size            Crate Name
0.4%  41.9% 1.4KiB embassy_executor embassy_executor::arch::thread::Executor::run
0.2%  20.2%   706B embassy_executor embassy_executor::raw::TaskStorage<F>::poll
0.1%  11.6%   406B     gd32f1x0_hal TIMER1
0.1%  10.5%   366B embassy_executor embassy_executor::raw::TaskStorage<F>::poll
0.0%   4.7%   164B        [Unknown] __aeabi_memclr8
0.0%   2.3%    80B embassy_executor embassy_executor::arch::thread::Executor::new
0.0%   1.3%    46B embassy_executor embassy_executor::raw::util::UninitCell<T>::write_in_place
0.0%   1.1%    40B      cortex_m_rt Reset
0.0%   0.7%    24B        [Unknown] HardFaultTrampoline
0.0%   0.6%    20B          embassy embassy::__cortex_m_rt_main
0.0%   0.5%    18B              std core::option::unwrap_failed
0.0%   0.5%    16B     gd32f1x0_hal gd32f1x0_hal::gpio::gpioc::<impl gd32f1x0_hal::gpio::GpioRegExt for gd32f1::gd32f130::gpioc::RegisterBlock>::is_set_low
0.0%   0.5%    16B     gd32f1x0_hal gd32f1x0_hal::gpio::gpioc::<impl gd32f1x0_hal::gpio::GpioRegExt for gd32f1::gd32f130::gpioc::RegisterBlock>::is_low
0.0%   0.4%    14B     gd32f1x0_hal gd32f1x0_hal::gpio::gpioc::<impl gd32f1x0_hal::gpio::GpioRegExt for gd32f1::gd32f130::gpioc::RegisterBlock>::set_high
0.0%   0.4%    14B     gd32f1x0_hal gd32f1x0_hal::gpio::gpioc::<impl gd32f1x0_hal::gpio::GpioRegExt for gd32f1::gd32f130::gpioc::RegisterBlock>::set_low
0.0%   0.3%    10B         cortex_m _critical_section_1_0_release
0.0%   0.2%     8B embassy_executor embassy_executor::raw::SyncExecutor::alarm_callback
0.0%   0.2%     8B              std core::panicking::panic_const::panic_const_async_fn_resumed
0.0%   0.2%     8B              std core::panicking::panic
0.0%   0.2%     8B              std core::panicking::panic_fmt
0.0%   1.1%    38B                  And 9 smaller methods. Use -n N to show more.
0.8% 100.0% 3.4KiB                  .text section size, the file size is 405.7KiB

Serhii-the-Dev · 2024-09-19T18:18:59Z

@qwandor I must ask to wait with a merge as I've just discovered some strange issues with this PR: a couple of fresh new GD32F130C8 are starting to reboot after a few alarm cycles on any timer/channel instance. I tried to fully erase my GD32F130G8's flash, and uploaded an example, now it's reproducing on this chip as well. Moreover, openocd is not able to program the GD32F130G8 anymore, however probe-rs is working just fine. Alas, no meaningful ideas what's going on so far, I'll keep investigation.

qwandor

Weird! Good luck. I haven't finished reviewing this yet, but some comments in the meantime...

qwandor · 2024-09-09T11:00:31Z

Cargo.toml

+  "dep:embassy-time",
+  "dep:embassy-time-driver",
+]
+time-driver-tim1=["embassy"]


Nit: please format these lines with a space before and after the = to be consistent with the rest of the file.

qwandor · 2024-09-09T11:10:20Z

Cargo.toml

+  "executor-thread",
+  "integrated-timers",
+], optional = true }
+embassy-time = { version = "0.3.2", optional = true }


Do you need this dependency? The documentation for embassy-time-driver says "If you are writing a driver, you should depend only on this crate, not on the main embassy-time crate.".

qwandor · 2024-09-09T11:11:20Z

Cargo.toml

@@ -26,6 +27,14 @@ embedded-io = "0.6.1"
 gd32f1 = { version = "0.9.1", features = ["critical-section"] }
 nb = "1.1.0"
 void = { version = "1.0.2", default-features = false, optional = true }
+embassy-executor = { version = "0.6", features = [


Is this needed for the implementation or just for the example? If it's only for the example then you can move it to the dev-dependencies section below.

Serhii-the-Dev · 2024-09-21T17:37:01Z

@qwandor Thank you for pointing those out, I've fixed the issues you mentioned. Also I realized that my MCUs are restarting due to the free watchdog timer timeout. According to the GD32F1x0 user manual, the watchdog timer is always running which makes me confused why the examples were running before without chip reset.
I've added a watchdog timer loop into the embassy example however it's still looking odd.
P.S. I've just found your old thread describing the similar issue 😄 So after the SWWDG option was set to zero, the watchdog is disabled and everything is working as expected.

qwandor · 2024-09-27T10:58:56Z

src/embassy/time_driver.rs

+#[cfg(feature = "time-driver-tim14")]
+const ALARM_COUNT: usize = 1;
+
+fn timer() -> &'static RegisterBlock {


This function shouldn't be safe, because it can allow creating aliases to timer register blocks which may already be owned by other parts of the program.

Or better, rather than constructing the reference here, use the Timer passed in to EmbassyTimeDriver::init, as that way you know you have ownership of it.

qwandor · 2024-09-27T11:00:10Z

src/embassy/time_driver.rs

+    callback: Cell<Option<(fn(*mut ()), *mut ())>>,
+}
+
+unsafe impl Send for AlarmState {}


If this is sound then please add a safety comment explaining how AlarmState is guaranteed to meet the safety requirements of Send.

qwandor · 2024-09-27T11:13:31Z

src/embassy/time_driver.rs

+    ((period as u64) << 15) + ((counter ^ ((period & 1) << 15)) as u64)
+}
+
+const ALARM_STATE_NEW: AlarmState = AlarmState::new();


Why this constant, rather than just calling AlarmState::new() where you need it? Having both seems like unneccesary duplication.

qwandor · 2024-09-27T11:16:02Z

src/embassy/time_driver.rs

+    }
+
+    fn get_alarm<'a>(&'a self, cs: CriticalSection<'a>, alarm: AlarmHandle) -> &'a AlarmState {
+        // safety: we're allowed to assume the AlarmState is created by us, and


Do you mean the AlarmHandle? This is a safe method, so you can't assume anything about what arguments it is called with. Either make this method unsafe and document the safety requirement, or use a safe method below to panic if the handle is out of bounds rather than causing undefined behaviour.

qwandor · 2024-09-27T11:25:31Z

src/embassy/time_driver.rs

+        let period = self.period.load(Ordering::Relaxed);
+        compiler_fence(Ordering::Acquire);
+        // Timer1 has a 32-bit counter, while other timers resolution is limited to 16 bits
+        let counter = r.cnt().read().cnt().bits() as u32;


Can you use .into() here rather than as u32? That way it's clear that no bits are being discarded.

qwandor · 2024-09-27T11:29:19Z

src/embassy/time_driver.rs

+    fn now(&self) -> u64 {
+        let r = timer();
+        let period = self.period.load(Ordering::Relaxed);
+        compiler_fence(Ordering::Acquire);


Would using Ordering::Acquire for the period load operation on the previous line have an equivalent effect?

qwandor · 2024-09-27T11:30:08Z

src/embassy/time_driver.rs

+    }
+
+    unsafe fn allocate_alarm(&self) -> Option<AlarmHandle> {
+        critical_section::with(|_| {


Why does this need a critical section, when it's only accessing atomics?

Serhii-the-Dev and others added 2 commits September 9, 2024 00:21

Add basic Embassy support

261b79f

Avoid casting function pointer.

b692fd2

Serhii-the-Dev marked this pull request as draft September 19, 2024 18:18

qwandor reviewed Sep 19, 2024

View reviewed changes

Serhii-the-Dev added 2 commits September 21, 2024 20:08

Move example-specific crates to dev dependencies.

bdd0402

Add watchdog timer to the Embassy example.

e083ac9

Serhii-the-Dev marked this pull request as ready for review September 21, 2024 17:37

Serhii-the-Dev requested a review from qwandor September 21, 2024 19:34

qwandor reviewed Sep 27, 2024

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add basic Embassy support #53

Add basic Embassy support #53

Serhii-the-Dev commented Sep 8, 2024 •

edited

Loading

Serhii-the-Dev commented Sep 19, 2024 •

edited

Loading

qwandor left a comment

qwandor Sep 9, 2024

qwandor Sep 9, 2024

qwandor Sep 9, 2024

Serhii-the-Dev commented Sep 21, 2024 •

edited

Loading

qwandor Sep 27, 2024

qwandor Sep 27, 2024

qwandor Sep 27, 2024

qwandor Sep 27, 2024

qwandor Sep 27, 2024

qwandor Sep 27, 2024

qwandor Sep 27, 2024

Add basic Embassy support #53

Are you sure you want to change the base?

Add basic Embassy support #53

Conversation

Serhii-the-Dev commented Sep 8, 2024 • edited Loading

Serhii-the-Dev commented Sep 19, 2024 • edited Loading

qwandor left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Serhii-the-Dev commented Sep 21, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Serhii-the-Dev commented Sep 8, 2024 •

edited

Loading

Serhii-the-Dev commented Sep 19, 2024 •

edited

Loading

Serhii-the-Dev commented Sep 21, 2024 •

edited

Loading