linux-kernel - [RFC][PATCH 0/2] Avoiding DWC3 transfer stalls/hangs when using adb over f

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20200122222645.38805-1-john.stultz@linaro.org>
Date:   Wed, 22 Jan 2020 22:26:43 +0000
From:   John Stultz <john.stultz@...aro.org>
To:     lkml <linux-kernel@...r.kernel.org>
Cc:     John Stultz <john.stultz@...aro.org>,
        Felipe Balbi <felipe.balbi@...ux.intel.com>,
        Yang Fei <fei.yang@...el.com>,
        Thinh Nguyen <thinhn@...opsys.com>,
        Tejas Joglekar <tejas.joglekar@...opsys.com>,
        Andrzej Pietrasiewicz <andrzej.p@...labora.com>,
        Jack Pham <jackp@...eaurora.org>, Todd Kjos <tkjos@...gle.com>,
        Greg KH <gregkh@...uxfoundation.org>,
        Linux USB List <linux-usb@...r.kernel.org>,
        stable <stable@...r.kernel.org>
Subject: [RFC][PATCH 0/2] Avoiding DWC3 transfer stalls/hangs when using adb over f_fs

Hey all,
  I wanted to send these out for comment and thoughts.

Since ~4.20, when the functionfs gadget enabled scatter-gather
support, we have seen problems with adb connections stalling and
stopping to function on hardware with dwc3 usb controllers.
Specifically, HiKey960, Dragonboard 845c, and Pixel3 devices.

Initally the workaround we used was to simply disable scatter
gather support on the dwc3 by commenting out the
"dwc->gadget.sg_supported = true;" line.

After working with Fei Yang, who was seeing similar trouble on
Intel dwc3 based hardare, Thinh Nguyen mentioned that a fix had
already been found and pointed me to one of Anurag's patches.

This solved the issue on HiKey960 and I sent it out to the list
but didn't get any feedback.

Additional testing with the Dragonboard 845c found that that
first fix was not sufficient, and so I've sat on the fix
thinking something deeper was amiss and went back to the hack
of disabling sg_supported on all dwc3 platforms. 

In the following months Fei's continued and repeated efforts
didn't seem to get enough review to result in a fix, and they've
since moved on to other work.

Recently, I found that folks at qcom have seen similer issues
and pointed me to the second patch in this series, which does
seem to resolve the issue on the Dragonboard 845c, but not the
HiKey960 on its own.

So I wanted to send these patches out for comment. There's
clearly a number of folks seeing broken behavior for ahwile on
dwc3 hardware, and we're all seeemingly working around it in our
own ways, so either those individual fixes need to get upstream
or we need to figure out some deeper solution to the issue.

So I wanted to send these two out for review and feedback.

thanks
-john

Cc: Felipe Balbi <felipe.balbi@...ux.intel.com>
Cc: Yang Fei <fei.yang@...el.com>
Cc: Thinh Nguyen <thinhn@...opsys.com>
Cc: Tejas Joglekar <tejas.joglekar@...opsys.com>
Cc: Andrzej Pietrasiewicz <andrzej.p@...labora.com>
Cc: Jack Pham <jackp@...eaurora.org>
Cc: Todd Kjos <tkjos@...gle.com>
Cc: Greg KH <gregkh@...uxfoundation.org>
Cc: Linux USB List <linux-usb@...r.kernel.org>
Cc: stable <stable@...r.kernel.org>

Anurag Kumar Vulisha (2):
  usb: dwc3: gadget: Check for IOC/LST bit in both event->status and
    TRB->ctrl fields
  usb: dwc3: gadget: Correct the logic for finding last SG entry

 drivers/usb/dwc3/gadget.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

-- 
2.17.1