USB support by Disasm · Pull Request #295 · nrf-rs/nrf-hal

Disasm · 2021-02-17T04:21:16Z

This PR is based on #144 and previous attempts.

the addresses look like MMIO registers

or at least they seem to work

to ensure the external oscillator has been enabled

@unrelentingtech

Fixes compilation by pointing to current versions and takes in @unrelentingtech's fork of usb-device and suggestion in nrf-rs#144 (review) I folded in the hello world portion from japaric's PR summary text. refs: rust-embedded-community/usb-device#51 closes: nrf-rs#144

refs: nrf-rs#144

Disasm · 2021-02-17T06:02:41Z

Current problems:

IN packets sometimes get lost. This happens because there is no status tracking implemented for IN endpoints and write never returns WouldBlock.
full IN SETUP transactions with size == 0 (mod ep0_size) are not handled correctly. The size of the transaction should be tracked and status stage should be signaled when the transaction finishes (instead of sending a ZLP).
backing buffers are no longer used and should be removed.

wez

Looks like you're making great progress! I got about 60% of the way that you did; I'm pleased to see some similarities with the stuff in my local hack branch, and happy that spending the time on that means that I can more easily reason about your PR!

wez · 2021-02-17T06:43:51Z

+                    return true;
+                }
+
+                if ep0_state.direction == UsbDirection::Out {


makes sense!

Co-authored-by: Wez Furlong <wez@wezfurlong.org>

Disasm · 2021-02-26T16:30:24Z

Current status: serial stress test works better but still fails. One of the IN packets is not followed by EPDATASTATUS change, so "busy" flag is not erased and driver ends up in a locked state.

This helps in a situation when `poll()` is not called periodically, for example, when main loop is busy writing data to an endpoint. This situation happens in the usb_serial example.

Disasm · 2021-06-04T21:26:28Z

There was a problem in the serial example: it doesn't call pool() while trying to send the whole buffer. Since pool() was the only place where IN endpoints are unblocked, this resulted in an endpoint that was blocked forever. The problem was resolved by checking for the "IN complete" event inside write().

Disasm · 2021-06-05T13:07:39Z

I tested this version of the driver:

usb-device tests passed
serial test passed
almost all USB compliance tests passed: two problems relate to usb-device (alt.settings and LPM are not supported)

jonas-schievink

Did a first pass, this looks pretty good so far!

jonas-schievink · 2021-06-17T11:49:43Z

+            regs.events_endepout[i].reset();
+            dma_end();
+
+            // TODO: ISO


What's the current behavior when trying to use an ISO endpoint?

Dirbaio

Looks great! Excited to finally have USB support in :)

Commented a few minor nits

Dirbaio · 2021-07-07T11:52:11Z

+
+pub use nrf_usbd::Usbd;
+
+#[allow(dead_code)] // fields are unused and only hold ownership


minor nit: I think if you name the field _clocks the warning goes away

Dirbaio · 2021-07-07T11:53:46Z

+#[allow(dead_code)] // fields are unused and only hold ownership
+pub struct UsbPeripheral<'a> {
+    usbd: USBD,
+    clocks: &'a (),


&() is not ZST even if () is. Maybe use PhantomData<&'a ()>?

Dirbaio

🚀

bors r+

bors · 2021-07-07T12:56:07Z

Build succeeded:

ci

jonas-schievink and others added 17 commits February 17, 2021 06:53

wip

0fbc358

errata: use volatile operations

86e84a8

the addresses look like MMIO registers

4 byte align bulk/interrupt endpoint buffers

261eeae

don't return PollResult::Data when there's no data

faf0cb3

EP0: shoehorn status stage

2c3693b

make bulk reads works

b9e0eb4

or at least they seem to work

demand a Clocks token

714b4f8

to ensure the external oscillator has been enabled

remove stray print statement

788ff0f

make the example compile again

b84300b

remove patch

d1d70b4

rebase fixes

7fb2998

fixup CI for newly add usb test

1c87602

refs: nrf-rs#144

Reset shorts on SETUP to prevent spurious status stages

f36a7e0

Introduce EP0State

fa4011e

Rewrite read/write logic

c6d6aeb

Add hack to detect interrupted IN transfers

b938651

Disasm force-pushed the usb branch from a50be14 to b938651 Compare February 17, 2021 04:59

Inhibit SetAddress response

1b3e38b

wez reviewed Feb 17, 2021

View reviewed changes

Disasm and others added 7 commits February 17, 2021 20:04

Apply suggestions from code review

e78f3d0

Co-authored-by: Wez Furlong <wez@wezfurlong.org>

Replace unsafe borrow with get() and set()

a7a7e84

Issue a status stage when the transfer data is fully written

84ccc05

Don't prepare DMA transfer in read_control_setup

5d0c0eb

Don't prepare DMA transfers in reset

048e51d

Remove backing buffers

7ad844a

Track IN endpoint state

48b50e6

Remove redundant clear operation

6ffd9e1

Check EPDATASTATUS inside write()

922ced5

This helps in a situation when `poll()` is not called periodically, for example, when main loop is busy writing data to an endpoint. This situation happens in the usb_serial example.

Disasm marked this pull request as ready for review June 4, 2021 21:27

Disasm added 5 commits June 5, 2021 00:32

Merge remote-tracking branch 'upstream/master' into usb

07b01ec

Cleanup

4f33ab8

Implement is_stalled()

2e469df

Add serial and test_class examples

f55b541

Fix warning

d93fc5e

Dirbaio reviewed Jun 10, 2021

View reviewed changes

Comment thread examples/usb/examples/serial.rs Outdated

Cleanup examples

74b797e

jonas-schievink suggested changes Jun 17, 2021

View reviewed changes

Jonas Schievink and others added 5 commits July 6, 2021 16:59

Fix USB demo

9ce50b5

Clean up Cargo features, support nRF52833

6ede424

Add a 1ms delay to force_reset

dbe0b5a

Use nrf-usbd

844b782

Use crates.io version of nrf-usbd

3ea49b7

Dirbaio requested changes Jul 7, 2021

View reviewed changes

Address review comments

9185196

Dirbaio approved these changes Jul 7, 2021

View reviewed changes

bors Bot merged commit 5099e8f into nrf-rs:master Jul 7, 2021

jonas-schievink mentioned this pull request Jul 7, 2021

Add USB tests #292

Closed


		pub use nrf_usbd::Usbd;

		#[allow(dead_code)] // fields are unused and only hold ownership

Conversation

Disasm commented Feb 17, 2021

Uh oh!

Disasm commented Feb 17, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

wez left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

wez Feb 17, 2021

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Disasm commented Feb 26, 2021

Uh oh!

Disasm commented Jun 4, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Disasm commented Jun 5, 2021

Uh oh!

Uh oh!

jonas-schievink left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

jonas-schievink Jun 17, 2021

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Dirbaio left a comment

Choose a reason for hiding this comment

Uh oh!

Dirbaio Jul 7, 2021

Choose a reason for hiding this comment

Uh oh!

Dirbaio Jul 7, 2021

Choose a reason for hiding this comment

Uh oh!

Dirbaio left a comment

Choose a reason for hiding this comment

Uh oh!

bors Bot commented Jul 7, 2021

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Disasm commented Feb 17, 2021 •

edited

Loading

Disasm commented Jun 4, 2021 •

edited

Loading