[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230328160756.30520-1-quic_kriskura@quicinc.com>
Date: Tue, 28 Mar 2023 21:37:54 +0530
From: Krishna Kurapati <quic_kriskura@...cinc.com>
To: Thinh Nguyen <Thinh.Nguyen@...opsys.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Alan Stern <stern@...land.harvard.edu>,
"Geert Uytterhoeven" <geert+renesas@...der.be>,
Colin Ian King <colin.i.king@...il.com>,
Jiantao Zhang <water.zhangjiantao@...wei.com>,
"Rafael J . Wysocki" <rafael@...nel.org>
CC: <linux-usb@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
<quic_ppratap@...cinc.com>, <quic_wcheng@...cinc.com>,
<quic_jackp@...cinc.com>, <quic_ugoswami@...cinc.com>,
Krishna Kurapati <quic_kriskura@...cinc.com>
Subject: [PATCH v2 0/2] Handle core soft reset failure in pullup
When core soft reset timeout happens, pullup doesn't check for the
return value and proceeds setting up of event buffers and starts the
controller.
In this scneario, it is observed sometimes that the GEVTADDR LO/HI
registers read zero while we are setting the run stop bit and we end
up accessing address 0x00 leading to a crash. This series tries to
address this issue by handling the timeout and return back appropriate
error code to configfs for it to retry enumeration if it chooses to.
Link to v1: https://lore.kernel.org/all/20230322092740.28491-1-quic_kriskura@quicinc.com/
changes in v2:
Fixed comments addressing incomplete error handling in udc core
Krishna Kurapati (2):
usb: dwc3: gadget: Bail out in pullup if soft reset timeout happens
usb: gadget: udc: Handle gadget_connect failure during bind operation
drivers/usb/dwc3/gadget.c | 5 ++++-
drivers/usb/gadget/udc/core.c | 20 ++++++++++++++++----
2 files changed, 20 insertions(+), 5 deletions(-)
--
2.40.0
Powered by blists - more mailing lists