embedded-trainings-2020/advanced
Lotte Steenbrink cc0d844024 run cargo fmt
2020-07-03 17:00:01 +02:00
..
common usb-3, sb-3-solution.rs, get-descriptor-configuration.rs: handle SET_ADDRESS 2020-07-03 15:51:39 +02:00
firmware run cargo fmt 2020-07-03 17:00:01 +02:00
host/print-descs - spread out USB information 2020-06-30 19:02:21 +02:00
README.md update text in advanced/README.md 2020-07-03 16:53:54 +02:00
win-configured.txt - spread out USB information 2020-06-30 19:02:21 +02:00
win-enumeration.txt - spread out USB information 2020-06-30 19:02:21 +02:00

advanced

Advanced workshop

In this workshop we'll build a toy USB device application that gets enumerated and configured by the host.

The embedded application will run in a fully event driven fashion: only doing work when the host asks for it.

The nRF52840

Some details about the nRF52840 microcontroller that are relevant to this workshop.

  • single core ARM Cortex-M4 processor clocked at 64 MHz
  • 1 MB of Flash (at address 0x0000_0000)
  • 256 KB of RAM (at address 0x2000_0000)
  • IEEE 802.15.4 and BLE (Bluetooth Low Energy) compatible radio
  • USB controller (device function)

The nRF52840 Development Kit

The development board we'll use has two USB ports: J2 and J3 -- you can find a description of the board in the top-level README of this repository -- and an on-board J-Link programmer / debugger. USB port J2 is the J-Link's USB port. USB port J3 is the nRF52840's USB port.

Code organization

The advanced folder contains both "host" code, code that will run on the host, and "firmware" code, code that will run on the nRF52840 SoC. "host" and "firmware" source code has been placed in different Cargo workspaces so that each can be compiled with different compilation profiles. The host workspace will be natively compiled, whereas the firmware workspace will be cross-compiled for the ARM Cortex-M architecture.

$ cd advanced

$ tree -L 1 .
.
├── common
├── firmware
├── host
└── README.md

In addition to these two workspaces there's a third folder called "common". This folder contains no_std code that can be depended on by either "host" code or "firmware" code.

Listing USB devices

In the tools folder you'll find usb-list: a minimal cross-platform version of the lsusb tool. Run it (cargo run from tools/usb-list) to list all USB devices.

$ cargo run
Bus 002 Device 001: ID 1d6b:0003
Bus 001 Device 002: ID 0cf3:e300
Bus 001 Device 003: ID 0c45:6713
Bus 001 Device 001: ID 1d6b:0002

The goal is to get the nRF52840 SoC to show in this list. The embedded application will use the vendor ID (VID) and product ID (PID) defined in advanced/common/consts; the usb-list tool will highlight the USB device that matches that VID PID pair.

$ # expected output
$ cargo run
Bus 002 Device 001: ID 1d6b:0003
Bus 001 Device 002: ID 0cf3:e300
Bus 001 Device 003: ID 0c45:6713
Bus 001 Device 001: ID 1d6b:0002
Bus 001 Device 059: ID 2020:0717 <- nRF52840 on the nRF52840 Development Kit

Hello, world!

First, open the tools/dk-run folder and run cargo install --path . -f to install the dk-run tool.

Next open the advanced/firmware folder in VS Code. If have already opened the root of the repository (embedded-trainings-2020) then please also open the advanced/firmware folder: right click the left side panel, select "Add folder to workspace" and add the advanced/apps folder.

Now, on the left side panel, open the src/bin/hello.rs file from under the advanced/apps folder.

Give Rust Analyzer some time to analyze the file and its dependency graph. When it's done, a "Run" button will appear over the main function -- you may need to edit the file contents to make the "Run" button appear.

Click the "Run" button to run the application on the microcontroller.

If you are not using VS code run the cargo run --bin hello command from the advanced/firmware folder.

NOTE if you run into an error along the lines of "Debug power request failed" retry the operation and the error should disappear

The firmware workspace has been configured to cross-compile applications to the ARM Cortex-M architecture and then run them through the dk-run custom Cargo runner. The dk-run tool will load and run the embedded application on the microcontroller and collect logs from the microcontroller.

The dk-run process will terminate when the microcontroller enters the "halted" state. From the embedded application, one can enter the "halted" state using the asm::bkpt function. For convenience, an exit function is provided in the dk Hardware Abstraction Layer (HAL). This function is divergent like std::process::exit (fn() -> !) and can be used to halt the device and terminate the dk-run process.

Note that when the dk-run tool sees the device enter the halted state it will proceed to reset-halt the device. This is particularly important when writing USB applications because simply leaving the device in a halted state will make it appear as an unresponsive USB device to the host. Some OSes (e.g. Linux) will try to make an unresponsive device respond by power cycling the entire USB bus -- this will cause all other USB devices on the bus to be re-enumerated. Reset-halting the device will cause it to be electrically disconnected from the host USB bus and avoid the "power cycle the whole USB bus" scenario.

Checking the API documentation

We'll be using the dk Hardware Abstraction Layer. It's good to have handy its API documentation. You can generate the documentation for that crate from the command line:

$ cargo doc -p dk --open

Run this command from within the advanced/firmware folder. It will open the generated documentation in your default web browser.

RTIC hello

RTIC, Real Time on Integrated Circuits, is a framework for building evented, time sensitive applications.

Open the src/bin/rtic-hello.rs file.

RTIC applications are written in RTIC's Domain Specific Language (DSL). The DSL extends Rust syntax with custom attributes like #[init] and #[idle].

RTIC makes a clearer distinction between the application's initialization phase, the #[init] function, and the application's main loop or main logic, the #[idle] function. The initialization phase runs with interrupts disabled and interrupts are re-enabled before the idle function is executed.

rtic::app is a procedural macro that generates extra Rust code, in addition to the user's functions. The fully expanded version of the macro can be found in the file target/rtic-expansion.rs. This file will contain the expansion of the procedural macro for the last compiled RTIC application.

If you look at the rtic-expansion.rs file generated for the build of the rtic-hello example you can confirm that interrupts are disabled during the execution of the init function.

fn main() -> ! {
    rtic::export::interrupt::disable();
    let late = init(init::Context::new(/* .. */));
    rtic::export::interrupt::enable();
    idle(idle::Context::new(/* .. */))
}

Dealing with registers

Open the src/bin/events.rs file.

In this and the next section we'll look into RTIC's event handling features. To explore these features we'll use the action of connecting a USB cable to the DK's port J2 as the event we'd like to handle.

The example application enables the signaling of this "USB power" event in the init function. This is done using the low level register API generated by the svd2rust tool. The register API was generated from a SVD (System View Description) file, a file that describes all the peripherals and registers, and their memory layout, on a device. In our case the device was the nRF52840; a sample SVD file for this microcontroller can be found here.

In the svd2rust API, peripherals are represented as structs. The fields of each peripheral struct are the registers associated to that peripheral. Each register field exposes methods to read and write to the register in a single memory operation.

The read and write methods take closure arguments. These closures in turn grant access to a "constructor" value, usually named r or w, which provides methods to modify the bitfields of a register. At the same time the API of these "constructors" prevent you from modifying the reserved parts of the register: that is you cannot write arbitrary values into registers; you can only write valid values into registers.

Apart from the read and write methods there's a modify method that performs a read-modify-write operation on the register; this API is also closure-based. The svd2rust-generated API is documented in detail in the svd2rust crate starting at this section.

In Cortex-M devices interrupt handling needs to be enabled on two sides: on the peripheral side and on the core side. The register operations done in init take care of the peripheral side. The core side of the operation involves writing to the registers of the Nested Vector Interrupt Controller (NVIC) peripheral. This second part doesn't need to be done by the user in RTIC applications because the framework takes care of it.

Event handling

Below the idle function you'll see a #[task] handler, a function. This task is bound to the POWER_CLOCK interrupt signal and will be executed, function-call style, every time the interrupt signal is raised by the hardware.

"Run" the events application. Then connect a micro-USB cable to your PC/laptop then connect the other end to the DK (port J3). You'll see the "POWER event occurred" message after the cable is connected.

Note that all tasks will be prioritized over the idle function so the execution of idle will be interrupted (paused) by the on_power_event task. When the on_power_event task finishes (returns) the execution of the idle will be resumed. This will become more obvious in the next section.

Try this: add an infinite loop to the end of init so that it never returns. Now run the program and connect the USB cable. What behavior do you observe? How would you explain this behavior? (hint: look at the rtfm-expansion.rs file: under what conditions is the init function executed?)

Task state

Now let's say we want to change the previous program to count how many times the USB cable (port J3) has been connected and disconnected.

Tasks run from start to finish, like functions, in response to events. To preserve some state between the different executions of a task we can add a resource to the task. In RTIC, resources are the mechanism used to share data between different tasks in a memory safe manner but they can also be used to hold task state.

To get the desired behavior we'll want to store some counter in the state of the on_power_event task.

Open the src/bin/resource.rs file. The starter code shows the syntax to declare a resource, the Resources struct, and the syntax to associate a resource to a task, the resources list in the #[task] attribute.

In the starter code a resource is used to move (by value) the POWER peripheral from init to the on_power_event task. The POWER peripheral then becomes part of the state of the on_power_event task. The resources of a task are available via the Context argument of the task.

To elaborate more on this move action: in the svd2rust API, peripheral types like POWER are singletons (only a single instance of the type can ever exist). The consequence of this design is that holding a peripheral instance, like POWER, by value means that the function (or task) has exclusive access, or ownership, over the peripheral. This is the case of the init function: it owns the POWER peripheral but then transfer ownership over it to a task using the resource initialization mechanism.

We have moved the POWER peripheral into the task because we want to clear the USBDETECTED interrupt flag after it has been set by the hardware. If we miss this step the on_power_event task (function) will be called again once it returns and then again and again and again (ad infinitum).

Also note that in the starter code the idle function has been modified. Pay attention to the logs when you run the starter code.

Your task in this section will be to modify the program so that it prints the number of times the USB cable has been connected to the DK every time the cable is connected, as shown below.

(..)
INFO:resource -- on_power_event: cable connected 1 time
(..)
INFO:resource -- on_power_event: cable connected 2 times
(..)
INFO:resource -- on_power_event: cable connected 3 times

You can find a solution to this exercise in the resource-solution.rs file.

USB enumeration

The USB protocol is complex so we'll leave out many details and focus only on the concepts required to get enumeration and configuration working. There are also several USB specific terms so we recommend checking chapter 2, "Terms and Abbreviations", of the USB specification (linked at the bottom of this document) every now and then.

So what is enumeration? A USB device, like the nRF52840, can be one of these three states: the Default state, the Address state or the Configured state. After being powered the device will start in the Default state. The enumeration process will take the device from the Default state to the Address state. As a result of the enumeration process the device will be assigned an address, in the range 1..=127, by the host.

Each OS may perform the enumeration process slightly differently but the process will always involve these host actions:

  • USB reset. This will put the device in the Default state, regardless of what state it was in.
  • GET_DESCRIPTOR request to get the device descriptor.
  • SET_ADDRESS request to assign an address to the device.

These host actions will be perceived as events by the nRF52840. There are more USB concepts involved that we'll need to cover like descriptors, configurations, interfaces and endpoints but for now let's see how to handle USB events.

Dealing with USB events

The USBD peripheral on the nRF52840 contains a series of registers, called EVENTS registers, that indicate the reason for entering the USBD event handler. These events must be handled by the application to complete the enumeration process.

Open the src/bin/usb-1.rs file. In this starter code the USBD peripheral is initialized in init and a task, named main, is bound to the interrupt signal USBD. This task will be called every time a new USBD event needs to be handled. The main task uses a helper next_event function to check all the event registers; if any event is set (occurred) then the function returns the event, represented by the Event enum, wrapped in the Some variant. This Event is then passed to the on_event function for further processing.

Connect the USB cable to the port J3 then run the starter code.

In this section as a warm-up exercise you'll need to deal with the following USB events until you reach the EP0SETUP event.

  • USBRESET. This event indicates that the host issued a USB reset signal. According to the USB specification this will move the device from any state to the Default state. Where are not dealing with any other state so doing nothing in response to this event is OK for now.

  • EP0SETUP. The USBD peripheral has detected the SETUP stage of a control transfer. If you get to this point move to the next section.

  • EP0DATADONE. The USBD peripheral is signaling the end of the DATA stage of a control transfer. You won't encounter this event just yet.

When you are done you should see this output:

(..)
INFO:usb_1 -- USB: UsbEp0Setup
INFO:usb_1 -- goal reached; move to the next section

Do not overthink this exercise; it is not a trick question. There is very little to do and literally nothing to add.

You can find the solution in the usb-1-solution.rs file.

Before we continue we need to discuss how data transfers work under the USB protocol.

USB Endpoints

Under the USB protocol data transfers occur over endpoints.

Endpoints are similar to UDP or TCP ports on a PC in that they allow logical multiplexing of data over a single physical USB bus. USB endpoints, however, have directions: an endpoint can either be an IN endpoint or an OUT endpoint. The direction is always from the perspective of the host so in an IN endpoint data travels from the device to the host and in an OUT endpoint data travels from the host to the device.

Endpoints are identified by their address, a zero-based index, and direction. There are four types of endpoints: control endpoints, bulk endpoints, interrupt endpoints and isochronous endpoints. Each endpoint type has different properties: reliability, latency, etc. In this workshop we'll only need to deal with control endpoints.

All USB devices must use "endpoint 0" as the default control endpoint. "Endpoint 0" actually refers to two endpoints: endpoint 0 IN and endpoint 0 OUT. This endpoint pair is used to establish a control pipe, a bidirectional communication channel between the host and device where data is exchanged using a predefined format. The default control pipe over endpoint 0 is mandatory: it must always be present and must always be active.

For detailed information about endpoints check section 5.3.1, Device Endpoints, of the USB specification.

Control transfers

The control pipe handles control transfers, a special kind of data transfer used by the host to issue requests. A control transfer is a data transfer that occurs in three stages: a SETUP stage, an optional DATA stage and a STATUS stage.

During the SETUP stage the host sends 8 bytes of data that identify the control request. Depending on the issued request there may be a DATA stage or not; during the DATA stage data is transferred either from the device to the host or the other way around. During the STATUS stage the device acknowledges, or not, the whole control request.

For detailed information about control transfers check section 5.5, Control Transfers, of the USB specification.

SETUP stage

At the end of program usb-1 we received a EP0SETUP event. This event signals the end of the SETUP stage of a control transfer. The nRF52840 USBD peripheral will automatically receive the SETUP data and store it in the following registers: BMREQUESTTYPE, BREQUEST, WVALUE{L,H}, WINDEX{L,H} and WLENGTH{L,H}. These registers are documented in sections 6.35.13.31 to 6.35.13.38 of the nRF52840 Product Specification.

The format of this setup data is documented in section 9.3 of the USB specification. Your next task is to parse the setup data according to section 9.4, Standard Descriptor Requests, of the USB specification (tables 9-3, 9-4 and 9-5 are the most relevant part of the section).

Note that you won't need to be able to parse all the standard requests in this workshop. For now you should be able to parse the GET_DESCRIPTOR request, which is described in detail in section 9.4.3 of the USB specification.

When you need to write some no_std code that does not involve device-specific I/O you should consider writing it as a separate crate. Having this kind of code in a separate crate lets you test it on your development machine (e.g. x86_64) using the standard cargo test functionality.

So that's what we'll do here. In the advanced/common/usb folder you'll find starter code for writing a no_std SETUP data parser. The starter code contains some unit tests; you can run them with cargo test (from within the usb folder) or you can use Rust Analyzer's "Test" button if you have the file open in VS code.

To sum up the work to do here:

  1. write a SETUP data parser in advanced/common/usb. You only need to handle the GET_DESCRIPTOR request and make the get_descriptor_device test pass for now. Note that the parser already handles SET_ADDRESS requests.

  2. modify usb-1 to read (USBD registers) and parse the SETUP data when the EPSETUP event is received.

Note: If you're using a Mac, you need to catch SetAddress requests returned by the parser as these are sent before the first GetDescriptor request. You can handle them by doing nothing.

  1. when you have successfully received a GET_DESCRIPTOR request for a Device descriptor you are done and can move to the next section.

Alternatively, you can start from usb-2 instead of usb-1. In either case, the tasks are the same.

If you are logging like the usb-2 starter code does then you should see an output like this once you are done:

INFO:usb_2 -- USB: UsbReset @ 438.842772ms
INFO:usb_2 -- USB: UsbEp0Setup @ 514.984128ms
...
INFO:usb_2 -- SETUP: bmrequesttype: 128, brequest: 6, wlength: 64, windex: 0, wvalue: 256
INFO:usb_2 -- GET_DESCRIPTOR Device [length=64]
INFO:usb_2 -- Goal reached; move to the next section

wlength / length can vary depending on the OS, USB port (USB 2.0 vs USB 3.0) or the presence of a USB hub so you may see a different value.

You can find a solution to task (1) in advanced/common/usb/get-device-descriptor.rs.

You can find a solution to task (2) in advanced/firmware/src/bin/usb-2-solution.rs.

Device descriptor

After receiving a GET_DESCRIPTOR request during the SETUP stage the device needs to respond with a descriptor during the DATA stage.

A descriptor is a binary encoded data structure sent by the device to the host. The device descriptor, in particular, contains information about the device, like its product and vendor identifiers and how many configurations it has. The format of the device descriptor is specified in section 9.6.1, Device, of the USB specification.

As far as the enumeration process goes, the most relevant fields of the device descriptor are the number of configurations and bcdUSB, the version of the USB specification the devices adheres to. In bcdUSB you should report compatibility with USB 2.0.

What about (the number of) configurations?

A configuration is akin to an operation mode. USB devices usually have a single configuration that will be the only mode in which they'll operate, for example a USB mouse will always act as a USB mouse. Some devices, though, may provide a second configuration for the purpose of firmware upgrades. For example a printer may enter DFU (Device Firmware Upgrade) mode, a second configuration, so that a user can update its firmware; while in DFU mode the printer will not provide printing functionality.

The specification mandates that a device must have at least one available configuration so we can report a single configuration in the device descriptor.

DATA stage

The next step is to respond to the GET_DESCRIPTOR request with a device descriptor. To do this we'll use the dk::usb::Ep0In abstraction -- we'll look into what the abstraction does in a future section; for now we'll just use it.

An instance of this abstraction is available in the board value (#[init] function). The first step is to make this Ep0In instance available to the on_event function.

The Ep0In API has two methods: start and end (also see their API documentation). start is used to start a DATA stage; this method takes a slice of bytes ([u8]) as argument; this argument is the response data. The end method needs to be called after start, when the EP0DATADONE event is raised, to complete the control transfer. Ep0In will automatically issue the STATUS stage that must follow the DATA stage.

To goal of this section is to use Ep0In to respond to the GET_DESCRIPTOR Device request (and only to that request). The response must be a device descriptor with its fields set to these values:

  • bLength = 18, size of this descriptor (see table 9-8 of the USB spec)
  • bDescriptorType = 1, means device descriptor (see table 9-5 of the USB spec)
  • bcdUSB = 0x0200, means USB 2.0
  • bDeviceClass = bDeviceSubClass = bDeviceProtocol = 0, these are unimportant for enumeration
  • bMaxPacketSize0 = 64, this is the most performant option (minimizes exchanges between the device and the host) and it's assumed by the Ep0In abstraction
  • idVendor = common::VID, value expected by the usb-list tool (*)
  • idProduct = common::PID, value expected by the usb-list tool (*)
  • bcdDevice = 0x0100, this means version 1.0 but any value should do
  • iManufacturer = iProduct = iSerialNumber = 0, string descriptors not supported
  • bNumConfigurations = 1, must be at least 1 so this is the minimum value

(*) the common crate refers to the crate in the advanced/common folder. It is already part of the firmware crate dependencies.

Although you can create the device descriptor by hand as an array filled with magic values we strongly recommend you use the usb2::device::Descriptor abstraction. The crate is already in the dependency list of the project; you can open its API documentation with the following command: cargo doc -p usb2 --open.

Note that the device descriptor is 18 bytes long but the host may ask for fewer bytes (see wlength field in the SETUP data). In that case you must respond with the amount of bytes the host asked for. The opposite may also happen: wlength may be larger than the size of the device descriptor; in this case you must answer must be 18 bytes long (do not pad the response with zeroes).

If you need help getting the Ep0In value into the on_event function check out the src/bin/usb-3.rs file.

Once you have successfully responded to the GET_DESCRIPTOR Device request you should get logs like these (if you are logging like usb-3 does):

INFO:usb_3 -- USB: UsbReset @ 342.071532ms
INFO:usb_3 -- USB: UsbEp0Setup @ 414.855956ms
INFO:usb_3 -- SETUP: bmrequesttype: 128, brequest: 6, wlength: 64, windex: 0, wvalue: 256
INFO:usb_3 -- GET_DESCRIPTOR Device [length=64]
INFO:dk::usbd -- EP0IN: start 18B transfer
INFO:usb_3 -- USB: UsbEp0DataDone @ 415.222166ms
INFO:dk::usbd -- EP0IN: transfer done
INFO:usb_3 -- USB: UsbReset @ 465.637206ms
INFO:usb_3 -- USB: UsbEp0Setup @ 538.208007ms
INFO:usb_3 -- SETUP: bmrequesttype: 0, brequest: 5, wlength: 0, windex: 0, wvalue: 27
ERROR:usb_3 -- unknown request (goal achieved if GET_DESCRIPTOR Device was handled)
INFO:dk -- `dk::exit() called; exiting ...`

A solution to this exercise can be found in src/bin/usb-3-solution.rs

DMA

Let's zoom into the Ep0In abstraction next. You can use the "go to definition" to see the implementation of the Ep0In.start method. What this method does is start a DMA transfer to send bytes to the host. The data (bytes) is first copied into an internal buffer and then the DMA is configured to move the data from that internal buffer to the USBD peripheral.

The signature of the start method does not ensure that:

  • bytes won't be deallocated before the DMA transfer is over (e.g. bytes could be pointing into the stack), or that
  • bytes won't be modified right after the DMA transfer starts (this would be a data race in the general case).

For these two safety reasons the API is implemented using an internal buffer. The internal buffer has a 'static lifetime so it's guaranteed to never be deallocated -- this prevents issue (a). The busy flag prevents any further modification to the internal buffer -- from the public API -- while the DMA transfer is in progress.

Apart from thinking about lifetimes and explicit data races in the surface API one must internally use memory fences to prevent reordering of memory operations (e.g. by the compiler), which can also cause data races. DMA transfers run in parallel to the instructions performed by the processor and are "invisible" to the compiler.

In the implementation of the start method, data is copied from bytes to the internal buffer, memcpy operation, and then the DMA transfer is started with a write to the TASKS_STARTEPIN0 register. The compiler sees the start of the DMA transfer (register write) as an unrelated memory operation so it may move the memcpy to after the DMA transfer has started. This reordering results in a data race: the processor modifies the internal buffer while the DMA is reading data out from it.

To avoid this reordering a memory fence, dma_start, is used. The fence pairs with the store operation (register write) that starts the DMA transfer and prevents the previous memcpy, and any other memory operation, from being move to after the store operation.

Another memory fence, dma_end, is needed at the end of the DMA transfer. In the general case, this prevents instruction reordering that would result in the processor accessing the internal buffer before the DMA transfer has finished. This is particularly problematic with DMA transfers that modify a region of memory which the processor intends to read after the transfer.

Not relevant to the DMA operation but relevant to the USB specification, the start method sets a shortcut in the USBD peripheral to issue a STATUS stage right after the DATA stage is finished. Thanks to this it is not necessary to manually start a STATUS stage after calling the end method.

More standard requests

After responding to the GET_DESCRIPTOR Device request the host will start sending different requests. The parser in common/usb will need to be updated to handle these requests:

  1. SET_ADDRESS, see section 9.4.6 of the USB spec
  2. GET_DESCRIPTOR Configuration, see section 9.4.3 of the USB spec
  3. SET_CONFIGURATION, see section 9.4.7 of the USB spec -- this request is likely to only be observed on Linux during enumeration

We suggest you incrementally extend the parser to handle these requests in that order. So, for example, add SET_ADDRESS support first and then check how far the usb program goes.

The starter common/usb code contains unit tests for these other requests as well as extra Request variants for these requests. All of them have been commented out using a #[cfg(TODO)] attribute which you can remove once you need any new variant or new unit test.

If at any point you need them you can find solution to parsing the above three requests in the following files:

  • advanced/common/src/set-address.rs
  • advanced/common/src/get-descriptor-configuration.rs
  • advanced/common/src/set-configuration.rs

Each file contains just enough code to parse the request in its name and the GET_DESCRIPTOR Device request. So you can refer to set-configuration.rs without getting "spoiled" about how to parse the SET_ADDRESS request.

Error handling: stalling the endpoint

You may come across host requests other than the ones listed in the previous section. Here's what you should do if you encounter any of those.

The USB specification defines a device-side procedure for "stalling a endpoint", which amounts to the device telling the host that it doesn't support some request. This procedure should be used to deal with invalid requests, requests whose SETUP stage doesn't match any USB 2.0 standard request, and requests not supported by the device, for instance the SET_DESCRIPTOR request is not mandatory.

You can use the dk::usbd:ep0stall helper function or write 1 to the TASKS_EP0STALL register to stall endpoint 0. This is what you should use

The logic of the EP0SETUP event handling is going to get rather complex so we suggest this refactor to add error handling: move the event handling logic into a separate function that returns a Result. That function should return the Err variant when it encounters an invalid host request. So in code that may look like this:

fn on_event(/* parameters */) {
    match event {
        Event::EP0SETUP => {
            if ep0setup(/* arguments */).is_err() {
                log::error!("EP0: unexpected request; stalling the endpoint");
                // TODO stall the endpoint
            }
        }
    }
}

fn ep0setup(/* parameters */) -> Result<(), ()> {
    let req = Request::parse(/* arguments_*/)?;
    //                                       ^ early returns an `Err` if it occurs

    // TODO respond to the `req`; return `Err` if the request was invalid in this state

    Ok(())
}

Note that there's a difference between the error handling done here and the error handling commonly done in std programs. In std programs you usually bubble up errors to the top main function (using the ? operator), report the error, or chain of errors, and then exit the application with non-zero exit code. This approach is usually not appropriate for embedded programs as (1) main cannot return, (2) there may not be a console to print the error to and/or (3) stopping the program, and e.g. requiring the user to reset it to make it work again, may not be desirable behavior. For these reasons in embedded software errors tend to be handled as early as possible than propagated all the way up.

This does not preclude error reporting. The above snippet includes error reporting in the form of a log::error! statement. This log statement may not be included in the final release of the program as it may not be useful, or even visible, to an end user but error reporting is possible and certainly useful during development.

Device state

Eventually you'll receive a SET_ADDRESS request that will move the device from the Default state to the Address state and if you are working on Linux you'll receive a SET_CONFIGURATION request that will move the device from the Address state to the Configured state. Also, some requests are only valid in certain states, for example SET_CONFIGURATION is only valid if the device is in the Address state. For this reason the firmware will need to keep track of the device's current state.

The device state should be tracked using a resource so that it's preserved across many executions of the USBD event handler. The usb2 crate has a State enum with the 3 possible USB states: Default, Address and Configured. You can use that enum or roll your own.

Once you start tracking the device state don't forget to update the handling of the USBRESET event. This event changes the state of the USB device. See section 9.1, USB Device States, of the USB specification for more details.

A code hint

If you would like a hint about how device state and error handling should be integrated into your code, check out the src/bin/usb-4.rs file.

SET_ADDRESS

This request should come right after the GET_DESCRIPTOR Device request, though some OSes may issue a USB reset in between.

Section 9.4.6, Set Address, describes how to handle this request but below you can find a summary:

  • If the device is in the Default state, then

    • if the requested address was 0 (None in the usb API) then the device should stay in the Default state
    • otherwise the device should move to the Address state
  • If the device is in the Address state, then

    • if the requested address was 0 (None in the usb API) then the device should return to the Default state
    • otherwise the device should remain in the Address state but start using the new address
  • If the device is in the Configured state this request results in "unspecified" behavior according to the USB specication. You should stall the endpoint in this case.

According to the USB specification the device needs to respond to this request with a STATUS stage -- the DATA stage is omitted. The nRF52840 USBD peripheral will automatically issue the STATUS stage and switch to listening to the requested address (see the USBADDR register) so no interaction with the USBD peripheral is required for this request.

For more details, read the introduction of section 6.35.9 of the nRF52840 Product Specification 1.0 (pages 486 and 487).

GET_DESCRIPTOR Configuration

When the host issues a GET_DESCRIPTOR request to request a configuration descriptor the device needs to respond with the requested configuration descriptor plus all the interface and endpoint descriptors associated to that configuration descriptor during the DATA stage.

We have covered configurations and endpoints but what is an interface?

Interface

An interface is closest to a USB device's function. For example, a USB mouse may expose a single HID (Human Interface Device) interface to report user input to the host. USB devices can expose multiple interfaces within a configuration. For example, the nRF52840 Dongle could expose both a CDC ACM interface (AKA virtual serial port) and a HID interface; the first interface could be used for (log::info!-style) logs; and the second one could provide a RPC (Remote Procedure Call) interface to the host for controlling the nRF52840's radio.

An interface is made up of one or more endpoints. To give an example, a HID interface can use two (interrupt) endpoints, one IN and one OUT, for bidirectional communication with the host. A single endpoint cannot be used by more than one interface with the exception of the special "endpoint 0", which can be (and usually is) shared by all interfaces.

For detailed information about interfaces check section 9.6.5, Interface, of the USB specification.

Configuration descriptor

The configuration descriptor describes one of the device configurations to the host. The descriptor contains the following information about a particular configuration:

  • the total length of the configuration: this is the number of bytes required to transfer this configuration descriptor and the interface and endpoint descriptors associated to it
  • its number of interfaces
  • its configuration value -- this is not an index and can be any non-zero value
  • whether the configuration is self-powered
  • whether the configuration supports remote wakeup
  • its maximum power consumption

The format of the configuration descriptor is specified in section 9.6.3, Configuration, of the USB specification. What may not be obvious from that section is that number of interfaces must be greater than or equal to one and that the configuration value cannot be zero.

Interface descriptor

The interface descriptor describes one of the device interfaces to the host. The descriptor contains the following information about a particular interface:

  • its interface number -- this is a zero-based index
  • its alternate setting -- this allows configuring the interface
  • its number of endpoints
  • class, subclass and protocol -- these define the interface (HID, or TTY ACM, or DFU, etc.) according to the USB specification

The format of the interface descriptor is specified in section 9.6.5, Interface, of the USB specification. The most relevant parts of that section are: the number of endpoints can be zero and endpoint zero must not be accounted when counting endpoints.

Endpoint descriptor

We will not need to deal with endpoint descriptors in this workshop but they are specified in section 9.6.6, Endpoint, of the USB specification.

Response

So how should we respond to the host? As the goal is to be enumerated we'll respond with the minimum amount of information possible.

First, configuration descriptors are requested by index, not by their configuration value. The index specified in the host request should be checked. As we reported a single configuration in the device descriptor the index in the request must be zero. Any other value should be rejected by stalling the endpoint.

Next the response should continue a concatenation of the configuration descriptor, followed by interface descriptors and then by endpoint descriptors. The minimum amount of interfaces a device must have is one so we'll include a single interface descriptor in the response. The interface does not need to have any endpoint associated to it so we'll include zero endpoint descriptors in the response.

Thus the response will be one configuration descriptor and one interface descriptor. The two will be concatenated in a single packet so this response should be completed in a single DATA stage.

The configuration descriptor in the response should contain these fields:

  • bLength = 9, the size of this descriptor (see table 9-10 in the USB spec)
  • bDescriptorType = 2, configuration descriptor (see table 9-5 in the USB spec)
  • wTotalLength = 18 = one configuration descriptor (9 bytes) and one interface descriptor (9 bytes)
  • bNumInterfaces = 1, a single interface (the minimum value)
  • bConfigurationValue = 42, any non-zero value will do
  • iConfiguration = 0, string descriptors are not supported
  • bmAttributes { self_powered: true, remote_wakeup: false }, self-powered due to the debugger connection
  • bMaxPower = 250 (500 mA), this is the maximum allowed value but any (non-zero?) value should do

The interface descriptor in the response should contain these fields:

  • bLength = 9, the size of this descriptor (see table 9-11 in the USB spec)
  • bDescriptorType = 4, interface descriptor (see table 9-5 in the USB spec)
  • bInterfaceNumber = 0, this is the first, and only, interface
  • bAlternateSetting = 0, alternate settings are not supported
  • bNumEndpoints = 0, no endpoint associated to this interface (other than the control endpoint)
  • bInterfaceClass = bInterfaceSubClass = bInterfaceProtocol = 0, does not adhere to any specified USB interface
  • iInterface = 0, string descriptors are not supported

Again, we strongly recommend that you use the usb2::configuration::Descriptor and usb2::interface::Descriptor abstractions here. Each descriptor instance can be transformed into its byte representation using the bytes method -- the method returns an array. To concatenate both arrays you can use an stack-allocated heapless::Vec buffer. If you haven't the heapless crate before you can find example usage in the the src/bin/vec.rs file.

SET_CONFIGURATION (likely Linux only)

On Linux, the host will likely send a SET_CONFIGURATION request right after enumeration to put the device in the Configured state. For now you can reject (stall) the request. It is not necessary at this stage because the device has already been enumerated.

Idle state

Once you have handled all the previously covered requests the device should be enumerated and remain idle awaiting for a new host request. Your logs may look like this:

INFO:usb_4 -- USB: UsbReset @ 318.66455ms
INFO:usb_4 -- USB reset condition detected
INFO:usb_4 -- USB: UsbEp0Setup @ 391.418456ms
INFO:usb_4 -- EP0: GetDescriptor { descriptor: Device, length: 64 }
INFO:dk::usbd -- EP0IN: start 18B transfer
INFO:usb_4 -- USB: UsbEp0DataDone @ 391.723632ms
INFO:usb_4 -- EP0IN: transfer complete
INFO:dk::usbd -- EP0IN: transfer done
INFO:usb_4 -- USB: UsbReset @ 442.016601ms
INFO:usb_4 -- USB reset condition detected
INFO:usb_4 -- USB: UsbEp0Setup @ 514.709471ms
INFO:usb_4 -- EP0: SetAddress { address: Some(17) }
INFO:usb_4 -- USB: UsbEp0Setup @ 531.37207ms
INFO:usb_4 -- EP0: GetDescriptor { descriptor: Device, length: 18 }
INFO:dk::usbd -- EP0IN: start 18B transfer
INFO:usb_4 -- USB: UsbEp0DataDone @ 531.646727ms
INFO:usb_4 -- EP0IN: transfer complete
INFO:dk::usbd -- EP0IN: transfer done
INFO:usb_4 -- USB: UsbEp0Setup @ 531.829832ms
INFO:usb_4 -- EP0: GetDescriptor { descriptor: DeviceQualifier, length: 10 }
ERROR:usb_4 -- EP0IN: unexpected request; stalling the endpoint
INFO:usb_4 -- USB: UsbEp0Setup @ 532.226562ms
INFO:usb_4 -- EP0: GetDescriptor { descriptor: DeviceQualifier, length: 10 }
ERROR:usb_4 -- EP0IN: unexpected request; stalling the endpoint
INFO:usb_4 -- USB: UsbEp0Setup @ 532.592772ms
INFO:usb_4 -- EP0: GetDescriptor { descriptor: DeviceQualifier, length: 10 }
ERROR:usb_4 -- EP0IN: unexpected request; stalling the endpoint
INFO:usb_4 -- USB: UsbEp0Setup @ 533.020018ms
INFO:usb_4 -- EP0: GetDescriptor { descriptor: Configuration { index: 0 }, length: 9 }
INFO:dk::usbd -- EP0IN: start 9B transfer
INFO:usb_4 -- USB: UsbEp0DataDone @ 533.386228ms
INFO:usb_4 -- EP0IN: transfer complete
INFO:dk::usbd -- EP0IN: transfer done
INFO:usb_4 -- USB: UsbEp0Setup @ 533.569335ms
INFO:usb_4 -- EP0: GetDescriptor { descriptor: Configuration { index: 0 }, length: 18 }
INFO:dk::usbd -- EP0IN: start 18B transfer
INFO:usb_4 -- USB: UsbEp0DataDone @ 533.935546ms
INFO:usb_4 -- EP0IN: transfer complete
INFO:dk::usbd -- EP0IN: transfer done
INFO:usb_4 -- USB: UsbEp0Setup @ 534.118651ms
INFO:usb_4 -- EP0: SetConfiguration { value: Some(42) }
ERROR:usb_4 -- EP0IN: unexpected request; stalling the endpoint

Note that these logs are from a Linux host where a SET_CONFIGURATION request is sent after the SET_ADDRESS request. On other OSes you may not get that request before the bus goes idle. Also note that there are some GET_DESCRIPTOR DeviceQualifier requests in this case; you do not need to parse them in the usb crate as they'll be rejected (stalled) anyways.

You can find traces for other OSes in these files (they are next to this README):

  • win-enumeration.txt
  • macos-enumeration.txt (TODO)

At this point you can double check that enumeration worked by running the list-usb tool.

Bus 001 Device 013: ID 1366:1015 <- J-Link on the nRF52840 Development Kit
(..)
Bus 001 Device 016: ID 2020:0717 <- nRF52840 on the nRF52840 Development Kit

Both the J-Link and the nRF52840 should appear in the list.

You can find a working solution up to this point in src/bin/usb-4-solution.rs. Note that the solution uses the usb2 crate to parse SETUP packets and that crate supports parsing all standard requests.

Inspecting the descriptors

There's a tool in the advanced/host/ folder called print-descs. You can run this tool to print all the descriptors reported by your application.

$ print-descs
DeviceDescriptor {
    bLength: 18,
    bDescriptorType: 1,
    bcdUSB: 512,
    bDeviceClass: 0,
    bDeviceSubClass: 0,
    bDeviceProtocol: 0,
    bMaxPacketSize: 64,
    idVendor: 8224,
    idProduct: 1815,
    bcdDevice: 256,
    iManufacturer: 0,
    iProduct: 0,
    iSerialNumber: 0,
    bNumConfigurations: 1,
}
address: 22
config0: ConfigDescriptor {
    bLength: 9,
    bDescriptorType: 2,
    wTotalLength: 18,
    bNumInterfaces: 1,
    bConfigurationValue: 42,
    iConfiguration: 0,
    bmAttributes: 192,
    bMaxPower: 250,
    extra: None,
}
iface0: [
    InterfaceDescriptor {
        bLength: 9,
        bDescriptorType: 4,
        bInterfaceNumber: 0,
        bAlternateSetting: 0,
        bNumEndpoints: 0,
        bInterfaceClass: 0,
        bInterfaceSubClass: 0,
        bInterfaceProtocol: 0,
        iInterface: 0,
    },
]

The output above corresponds to the descriptor values we suggested. If you used different values, e.g. for bMaxPower, you'll a slightly different output.

Getting it configured

At this stage the device will be in the Address stage. It has been identified and enumerated by the host but cannot yet be used by host applications. The device must first move to the Configured state before the host can start, for example, HID communication or send non-standard requests over the control endpoint.

Windows and macOS will enumerate the device but not automatically configure it after enumeration. Here's what you should do to force the host to configure the device.

Linux

Nothing extra needs to be done on Linux. The host will automatically send a SET_CONFIGURATION request so proceed to the SET_CONFIGURATION section to see how to handle the request.

Windows

After getting the device enumerated and into the idle state, open the Zadig tool (covered in the setup instructions; see the top README) and use it to associate the nRF52840 USB device to the WinUSB driver. The nRF52840 will appear as a "unknown device" with a VID and PID that matches the ones defined in the common crate

Now modify the print-descs program to "open" the device -- this operation is commented out in the source code. With this modification print-descs will cause Windows to send a SET_CONFIGURATION request to configure the device. You'll need to run print-descs to test out the correct handling of the SET_CONFIGURATION request.

macOS

TODO uncomment the open line in print-descs and see if that forces the host to send a SET_CONFIGURATION request

SET_CONFIGURATION

Section 9.4.7, Set Configuration, of the USB spec describes how to handle this request but below you can find a summary:

  • If the device is in the Default state, you should stall the endpoint because the operation is not permitted in that state.

  • If the device is in the Address state, then

    • if the requested configuration value is 0 (None in the usb API) then stay in the Address state
    • if the requested configuration value is non-zero and valid (was previously reported in a configuration descriptor) then move to the Configured state
    • if the requested configuration value is not valid then stall the endpoint
  • If the device is in the Configured state, then

    • if the requested configuration value is 0 (None in the usb API) then return to the Address state
    • if the requested configuration value is non-zero and valid (was previously reported in a configuration descriptor) then move to the Configured state with the new configuration value
    • if the requested configuration value is not valid then stall the endpoint

In all the cases where you did not stall the endpoint (returned Err) you'll need to acknowledge the request by starting a STATUS stage. This can be done by writing 1 to the TASKS_EP0STATUS register.

NOTE: On Windows, you may get a GET_STATUS request before the SET_CONFIGURATION request and although you should respond to it, stalling the GET_STATUS request seems sufficient to get the device to the Configured state.

Expected output

Once you are correctly handling the SET_CONFIGURATION request you should get logs like these:

INFO:usb_5 -- USB: UsbReset @ 397.15576ms
INFO:usb_5 -- USB reset condition detected
INFO:usb_5 -- USB: UsbEp0Setup @ 470.00122ms
INFO:usb_5 -- EP0: GetDescriptor { descriptor: Device, length: 64 }
INFO:dk::usbd -- EP0IN: start 18B transfer
INFO:usb_5 -- USB: UsbEp0DataDone @ 470.306395ms
INFO:usb_5 -- EP0IN: transfer complete
INFO:dk::usbd -- EP0IN: transfer done
INFO:usb_5 -- USB: UsbReset @ 520.721433ms
INFO:usb_5 -- USB reset condition detected
INFO:usb_5 -- USB: UsbEp0Setup @ 593.292235ms
INFO:usb_5 -- EP0: SetAddress { address: Some(21) }
INFO:usb_5 -- USB: UsbEp0Setup @ 609.954832ms
INFO:usb_5 -- EP0: GetDescriptor { descriptor: Device, length: 18 }
INFO:dk::usbd -- EP0IN: start 18B transfer
INFO:usb_5 -- USB: UsbEp0DataDone @ 610.260008ms
INFO:usb_5 -- EP0IN: transfer complete
INFO:dk::usbd -- EP0IN: transfer done
INFO:usb_5 -- USB: UsbEp0Setup @ 610.443113ms
INFO:usb_5 -- EP0: GetDescriptor { descriptor: DeviceQualifier, length: 10 }
WARN:usb_5 -- EP0IN: stalled
INFO:usb_5 -- USB: UsbEp0Setup @ 610.809325ms
INFO:usb_5 -- EP0: GetDescriptor { descriptor: DeviceQualifier, length: 10 }
WARN:usb_5 -- EP0IN: stalled
INFO:usb_5 -- USB: UsbEp0Setup @ 611.175535ms
INFO:usb_5 -- EP0: GetDescriptor { descriptor: DeviceQualifier, length: 10 }
WARN:usb_5 -- EP0IN: stalled
INFO:usb_5 -- USB: UsbEp0Setup @ 611.511228ms
INFO:usb_5 -- EP0: GetDescriptor { descriptor: Configuration { index: 0 }, length: 9 }
INFO:dk::usbd -- EP0IN: start 9B transfer
INFO:usb_5 -- USB: UsbEp0DataDone @ 611.846922ms
INFO:usb_5 -- EP0IN: transfer complete
INFO:dk::usbd -- EP0IN: transfer done
INFO:usb_5 -- USB: UsbEp0Setup @ 612.030027ms
INFO:usb_5 -- EP0: GetDescriptor { descriptor: Configuration { index: 0 }, length: 18 }
INFO:dk::usbd -- EP0IN: start 18B transfer
INFO:usb_5 -- USB: UsbEp0DataDone @ 612.365721ms
INFO:usb_5 -- EP0IN: transfer complete
INFO:dk::usbd -- EP0IN: transfer done
INFO:usb_5 -- USB: UsbEp0Setup @ 612.640378ms
INFO:usb_5 -- EP0: SetConfiguration { value: Some(42) }
INFO:usb_5 -- entering the configured state

These logs are from a Linux host. You can find traces for other OSes in these files (they are next to this README):

  • win-configured.txt, this file only contains the logs produced by running print-descs
  • macos-configured.txt (TODO)

You can find a solution to this part of the exercise in src/bin/usb-5-solution.rs.

Next steps

We have covered only a few of the core features of the RTIC framework but the framework has many more features like software tasks, tasks that can be spawned by the software; message passing between tasks; and task scheduling, which allows the creation of periodic tasks. We encourage to check the RTIC book which describes the features we haven't covered here.

usb-device is a library for building USB devices. It has been built using traits (the pillar of Rust's generics) such that USB interfaces like HID and TTY ACM can be implemented in a device agnostic manner. The device details then are limited to a trait implementation. There's a work in progress implementation of the usb-device trait for the nRF52840 device in this PR and there are many usb-device "classes" like HID and TTY ACM that can be used with that trait implementation.

String descriptors

TODO more material if needed

Custom control transfers

TODO more material if needed

References