raspberrypi: make tusb_time_millis_api RAM-resident; core1 calls it#11114
Merged
Conversation
Since the TinyUSB update in adafruit#11093, tuh_task_event_ready() calls tusb_time_millis_api() (added upstream in hathach/tinyusb@3a262cb6e to report a due deferred call_after callback). On RP2 boards, usb_host polls tuh_task_event_ready() from core1's PIO-USB frame loop (common-hal/usb_host/Port.c). tuh_task_event_ready() is deliberately placed in RAM by link-rp2*.ld, and everything it calls must be RAM-resident: on RP2350 core1 sets an MPU region that makes flash execute-never, and on RP2040 flash may be unavailable while core0 writes to it. The shared tusb_time_millis_api() implementation (supervisor_ticks_ms32) lives in flash, so core1 died at the first device attach (the first time call_after was pending) and the PIO-USB bus went silent: no USB host device was ever detected. Fixes the regression reported in adafruit#10243. Make the shared implementation MP_WEAK and override it in the raspberrypi port with a PLACE_IN_ITCM function that reads the timer directly. time_us_32() >> 10 yields ~1.024 ms units rather than exact milliseconds, which is fine because TinyUSB only uses this API for relative delay arithmetic and every use goes through the same function. Tested on Feather RP2040 USB Host and Fruit Jam with a USB flash drive: both fail to enumerate on current main and enumerate immediately with this change, with the TinyUSB submodule pristine at 5453ed09f. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Collaborator
Author
|
I know this wasn't directed to me, but yes you can detect this at link time. I'll submit a PR for it. The script picked up two other lives in flash, but needs to be moved to RAM scenarios to run on core1.
|
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
2 participants
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.
Fixes the USB host regression from #11093, reported in #10243. Fixes #10243.
On RP2 boards, core1 drives the PIO-USB host and must only run code that lives in RAM, never flash. The TinyUSB update (hathach/tinyusb@3a262cb6e, from #3566) made core1 ask "what time is it?" via
tusb_time_millis_api()and that function lived in flash. Core1 crashed at the first device attach, so no USB host device was ever detected.Fix: move
tusb_time_millis_api()into RAM for this port (weak shared default,PLACE_IN_ITCMoverride reading the timer directly).Tested on Feather RP2040 USB Host and Fruit Jam with a flash drive: never detected on main, enumerates immediately with this change. TinyUSB submodule unchanged.