dwc2: submit SETUP on SETUP_DONE, handle v3.10a spurious RX_COMPLETE (STM32L476)#3669
Merged
Conversation
…LETE on core v3.10a (STM32L476)
DWC2 core rev 3.10a pushes an extra EP0 RX_COMPLETE (RXFLVL PKTSTS 0x3)
that is not a real OUT data completion, in two cases flagged on DOEPINT:
- STPKTRX (Setup Packet Received): between SETUP_RX and SETUP_DONE
- STSPHSRX (Status Phase Received, control write): after the OUT data
stage when the host starts the IN status phase
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
hathach
referenced
this pull request
Jun 3, 2026
- GRXSTSP register has internal FIFO, receiving events won't mix up (STATUS OUT & next SETUP) - Improve efficiency, remove 2nd IRQ overhead Signed-off-by: HiFiPhile <admin@hifiphile.com>
Contributor
There was a problem hiding this comment.
Pull request overview
This PR adjusts the DWC2 device controller’s EP0 control transfer handling by reporting SETUP at SETUP_DONE (instead of SETUP_RX) and adds a DWC2 core rev 3.10a (STM32L476-class) workaround to ignore spurious EP0 RX_COMPLETE conditions that can otherwise double-count control-OUT completions. It also updates HIL board configuration and includes small formatting/comment-only cleanup in unrelated areas.
Changes:
- DWC2: emit device-mode SETUP events on
SETUP_DONEand add a v3.10a-specific guard to drop spurious EP0RX_COMPLETEwords flagged bySTPKTRX/STSPHSRX. - HIL: add
stm32f407discoandstm32l476discoentries totest/hil/tinyusb.json. - Minor cleanups: brace-style tweak in MTP class code; comment-out CMake target folder properties; add a commented JLink option line for stm32f769disco.
Reviewed changes
Copilot reviewed 5 out of 5 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
test/hil/tinyusb.json |
Adds STM32F407 Disco and STM32L476 Disco HIL configurations. |
src/portable/synopsys/dwc2/dcd_dwc2.c |
Moves SETUP notification to SETUP_DONE and applies a v3.10a RX_COMPLETE quirk filter; small comment tweaks. |
src/class/mtp/mtp_device.c |
Brace/comment clarification for OUT completion logic in the MTP transfer callback. |
hw/bsp/stm32f7/boards/stm32f769disco/board.cmake |
Adds a commented example JLINK_OPTION line. |
hw/bsp/family_support.cmake |
Comments out unused CMake target folder properties. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Comment on lines
+346
to
+349
| { | ||
| "name": "stm32f407disco", | ||
| "uid": "30001A000647313332353735", | ||
| "tests": { |
| case GRXSTS_PKTSTS_SETUP_DONE: | ||
| // Setup packet done: | ||
| case GRXSTS_PKTSTS_SETUP_DONE: { | ||
| // Pop this word cause Setup interrupt |
|
|
||
| case GRXSTS_PKTSTS_RX_COMPLETE: { | ||
| // Out packet done | ||
| // Pop this word cause xfer complete interrupt |
| if (dir == TUSB_DIR_IN) { | ||
| handle_epin_slave(rhport, epnum, intr.diepint_bm); | ||
| } else { | ||
| // epout is handled in rxflv |
TinyUSB Average Code Size Metrics
Input files
|
|
| target | .text | .rodata | .data | .bss | total | % diff |
|---|---|---|---|---|---|---|
| xmc4500_relax/hid_generic_inout | 12,900 → 12,964 (+64) | — | — | — | 25,852 → 25,980 (+128) | +0.5% |
| xmc4500_relax/hid_multiple_interface | 13,712 → 13,776 (+64) | — | — | — | 27,476 → 27,604 (+128) | +0.5% |
| sipeed_longan_nano/dfu_runtime | 12,556 → 12,616 (+60) | 590 → 594 (+4) | — | — | 13,814 → 13,878 (+64) | +0.5% |
| xmc4500_relax/printer_to_cdc | 14,968 → 15,032 (+64) | — | — | — | 29,984 → 30,112 (+128) | +0.4% |
| xmc4500_relax/audio_test | 15,060 → 15,124 (+64) | — | — | — | 30,192 → 30,320 (+128) | +0.4% |
| sipeed_longan_nano/hid_composite | 14,692 → 14,756 (+64) | — | — | — | 16,102 → 16,166 (+64) | +0.4% |
| xmc4500_relax/cdc_msc_throughput | 17,096 → 17,160 (+64) | — | — | — | 34,240 → 34,368 (+128) | +0.4% |
| sipeed_longan_nano/audio_test | 16,412 → 16,476 (+64) | — | — | — | 17,566 → 17,630 (+64) | +0.4% |
| sipeed_longan_nano/audio_test_multi_rate | 16,732 → 16,796 (+64) | — | — | — | 17,866 → 17,930 (+64) | +0.4% |
| sipeed_longan_nano/midi_test | 15,130 → 15,192 (+62) | 510 → 506 (-4) | — | — | 16,308 → 16,366 (+58) | +0.4% |
… dedup) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Collaborator
|
How did you find out DOEPINT STUP need to cleared to allow RXSTSP SETUP_DONE fire again ? I read back and forth the guide but didn't find it. |
Owner
Author
It is undocumented, I just guess and try it out. The L476 is troublesome, even though I fixed it once, I kind of forgot/mis-interprete it this time again and take quite a bit of trial. :) |
hathach
commented
Jun 4, 2026
hathach
left a comment
hathach
left a comment
Owner
Author
There was a problem hiding this comment.
f723 host cdc failed is pre-PR issue. The hil test need a few re-run possibly due to over current to usb when multiple boards are active. I saw this happen when a couple of imxrt running especially 1064. Will try to add external power to make it more reliable.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
follow up to #3634
Primary change: move the device-mode control SETUP event from the
SETUP_RXRX-FIFO status toSETUP_DONEfor all DWC2 cores.SETUP_DONEis the correct hand-off point — the core assertsDOEPINT.SetUpright after the Setup-Stage-Done word is popped, so the 8-byte packet is final (no more back-to-back SETUPs).This exposed a DWC2 core rev 3.10a (e.g. STM32L476) quirk: the core emits an extra EP0
RX_COMPLETE(RXFLVLPKTSTS0x3) that is not a real OUT data completion, in two cases flagged onDOEPINT:STPKTRX(Setup Packet Received) — pushed betweenSETUP_RXandSETUP_DONEof every control transfer.STSPHSRX(Status Phase Received, control write) — pushed after the OUT data stage when the host starts the IN status phase.Acting on these double-counted control-OUT completions (e.g.
SET_LINE_CODING) and tripped ausbd_edpt_xferassert, breaking enumeration.Fix
Guarded by
gsnpsid == DWC2_CORE_REV_3_10a, in theGRXSTS_PKTSTS_RX_COMPLETEcase: drop the event whenSTPKTRXorSTSPHSRXis set, and W1C-clear the flag so a latchedSTSPHSRXcannot block the core from emitting the nextSETUP_DONE.usbdstill handles the real OUT data and queues the IN status ZLP itself (the core does not auto-complete the control-write status). The guard makes the workaround inert on all non-3.10a cores.HIL validation — full device suite,
Total failed: 0Also bundled in this branch
test/hil/tinyusb.json: add STM32F407 / STM32L476 disco HIL board configs.src/class/mtp/mtp_device.c: brace-style cleanup.hw/bsp/family_support.cmake: comment out unused target folder properties.🤖 Generated with Claude Code