Hi, Baolin Wang <baolin.wang@xxxxxxxxxx> writes: <trim> >> Also note that the usb_endpoint_xfer_isoc() call on line 2067 of >> gadget.c (as in my testing/next from today) won't even get executed, so >> we're safe there. > > Never will be executed? then we can remove the > usb_endpoint_xfer_isoc() (line 2025) at risk? > > 2023 clean_busy = dwc3_cleanup_done_reqs(dwc, dep, event, status); > 2024 if (clean_busy && (is_xfer_complete || > 2025 > usb_endpoint_xfer_isoc(dep->endpoint.desc))) > 2026 dep->flags &= ~DWC3_EP_BUSY; hmm, now that I look at this again, in case of XferInProgress, we could still have a problem. I'll fix it up in that commit I pointed you to. -- balbi
Attachment:
signature.asc
Description: PGP signature