The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to stable@vger.kernel.org.
To reproduce the conflict and resubmit, you may use the following commands:
git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y git checkout FETCH_HEAD git cherry-pick -x 3c7af52c7616c3aa6dacd2336ec748d4a65df8f4 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to 'stable@vger.kernel.org' --in-reply-to '2024012202-engraving-mute-6fac@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^..
Possible dependencies:
3c7af52c7616 ("usb: dwc3: gadget: Queue PM runtime idle on disconnect event") 8f40fc080813 ("usb: dwc3: gadget: Refactor EP0 forced stall/restart into a separate API") 02435a739b81 ("usb: dwc3: gadget: Stall and restart EP0 if host is unresponsive") 8422b769fa46 ("usb: dwc3: gadget: Submit endxfer command if delayed during disconnect") e1ee843488d5 ("usb: dwc3: gadget: Force sending delayed status during soft disconnect") 9d778f0c5f95 ("usb: dwc3: Fix ep0 handling when getting reset while doing control transfer") f66eef8fb898 ("usb: dwc3: gadget: Delay issuing End Transfer") c96683798e27 ("usb: dwc3: ep0: Don't prepare beyond Setup stage") 8f8034f493b5 ("usb: dwc3: gadget: Don't modify GEVNTCOUNT in pullup()") 861c010a2ee1 ("usb: dwc3: gadget: Refactor pullup()") 0066472de157 ("usb: dwc3: Issue core soft reset before enabling run/stop") 8217f07a5023 ("usb: dwc3: gadget: Avoid starting DWC3 gadget during UDC unbind")
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From 3c7af52c7616c3aa6dacd2336ec748d4a65df8f4 Mon Sep 17 00:00:00 2001 From: Wesley Cheng quic_wcheng@quicinc.com Date: Wed, 3 Jan 2024 13:49:46 -0800 Subject: [PATCH] usb: dwc3: gadget: Queue PM runtime idle on disconnect event
There is a scenario where DWC3 runtime suspend is blocked due to the dwc->connected flag still being true while PM usage_count is zero after DWC3 giveback is completed and the USB gadget session is being terminated. This leads to a case where nothing schedules a PM runtime idle for the device.
The exact condition is seen with the following sequence: 1. USB bus reset is issued by the host 2. Shortly after, or concurrently, a USB PD DR SWAP request is received (sink->source) 3. USB bus reset event handler runs and issues dwc3_stop_active_transfers(), and pending transfer are stopped 4. DWC3 usage_count decremented to 0, and runtime idle occurs while dwc->connected == true, returns -EBUSY 5. DWC3 disconnect event seen, dwc->connected set to false due to DR swap handling 6. No runtime idle after this point
Address this by issuing an asynchronous PM runtime idle call after the disconnect event is completed, as it modifies the dwc->connected flag, which is what blocks the initial runtime idle.
Fixes: fc8bb91bc83e ("usb: dwc3: implement runtime PM") Cc: stable@vger.kernel.org Signed-off-by: Wesley Cheng quic_wcheng@quicinc.com Link: https://lore.kernel.org/r/20240103214946.2596-1-quic_wcheng@quicinc.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org
diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c index c15e965ea95a..019368f8e9c4 100644 --- a/drivers/usb/dwc3/gadget.c +++ b/drivers/usb/dwc3/gadget.c @@ -3989,6 +3989,13 @@ static void dwc3_gadget_disconnect_interrupt(struct dwc3 *dwc) usb_gadget_set_state(dwc->gadget, USB_STATE_NOTATTACHED);
dwc3_ep0_reset_state(dwc); + + /* + * Request PM idle to address condition where usage count is + * already decremented to zero, but waiting for the disconnect + * interrupt to set dwc->connected to FALSE. + */ + pm_request_idle(dwc->dev); }
static void dwc3_gadget_reset_interrupt(struct dwc3 *dwc)