Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Thinh Nguyen <Thinh.Nguyen@synopsys.com>
To:     Anurag Kumar Vulisha <anurag.kumar.vulisha@xilinx.com>,
        "balbi@kernel.org" <balbi@kernel.org>,
        "gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>
CC:     "linux-usb@vger.kernel.org" <linux-usb@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "Thinh.Nguyen@synopsys.com" <Thinh.Nguyen@synopsys.com>,
        "v.anuragkumar@gmail.com" <v.anuragkumar@gmail.com>
Subject: Re: [PATCH v3 4/8] usb: dwc3: implement stream transfer timeout
Thread-Topic: [PATCH v3 4/8] usb: dwc3: implement stream transfer timeout
Thread-Index: AQHURqy4/mMoK+M5zEyImU/x5ug6Dg==
Date:   Fri, 7 Sep 2018 23:28:07 +0000
Message-ID: <30102591E157244384E984126FC3CB4F544AD89B@us01wembx1.internal.synopsys.com>
References: <1536326029-16327-1-git-send-email-anurag.kumar.vulisha@xilinx.com>
 <1536326029-16327-5-git-send-email-anurag.kumar.vulisha@xilinx.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 9/7/2018 6:14 AM, Anurag Kumar Vulisha wrote:=0A=
> According to dwc3 databook when streams are used, it may be possible=0A=
> for the host and device become out of sync, where device may wait for=0A=
> host to issue prime transcation and host may wait for device to issue=0A=
> erdy. To avoid such deadlock, timeout needs to be implemented. After=0A=
> timeout occurs, device will first stop transfer and restart the transfer=
=0A=
> again. This patch does the same.=0A=
>=0A=
> Signed-off-by: Anurag Kumar Vulisha <anurag.kumar.vulisha@xilinx.com>=0A=
> Reviewed-by: Thinh Nguyen <thinhn@synopsys.com>=0A=
> ---=0A=
>  Changes in v3:=0A=
> 	1. Added the changes suggested by "Thinh Nguyen"=0A=
>=0A=
>  Changes in v2:=0A=
> 	1. Changed STREAM_TIMEOUT to STREAM_TIMEOUT_MS as suggested by=0A=
> 	   "Andy Shevchenko"=0A=
> ---=0A=
>  drivers/usb/dwc3/core.h   |  7 +++++++=0A=
>  drivers/usb/dwc3/gadget.c | 36 ++++++++++++++++++++++++++++++++++++=0A=
>  2 files changed, 43 insertions(+)=0A=
>=0A=
> diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h=0A=
> index 5bfb625..0b255e97 100644=0A=
> --- a/drivers/usb/dwc3/core.h=0A=
> +++ b/drivers/usb/dwc3/core.h=0A=
> @@ -633,6 +633,11 @@ struct dwc3_event_buffer {=0A=
>  =0A=
>  #define DWC3_TRB_NUM		256=0A=
>  =0A=
> +/*=0A=
> + * Timeout value in msecs used by stream_timeout_timer when streams are =
enabled=0A=
> + */=0A=
> +#define STREAM_TIMEOUT_MS	50=0A=
> +=0A=
>  /**=0A=
>   * struct dwc3_ep - device side endpoint representation=0A=
>   * @endpoint: usb endpoint=0A=
> @@ -656,6 +661,7 @@ struct dwc3_event_buffer {=0A=
>   * @name: a human readable name e.g. ep1out-bulk=0A=
>   * @direction: true for TX, false for RX=0A=
>   * @stream_capable: true when streams are enabled=0A=
> + * @stream_timeout_timer: timer used to aviod deadlock when streams are =
used=0A=
=0A=
Change aviod -> avoid. Also, the timer is being used when there's a=0A=
deadlock rather than to avoid it. You probably need to explain what=0A=
deadlock it is if you mentioned it here, but I think it's ok to simply=0A=
say timeout timer for streams (unless Felipe has any objection).=0A=
=0A=
Thinh=0A=
=0A=