[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <637796ED17F7774FB27D6AAE3C6951584A96DC6D@SZXEMA509-MBS.china.huawei.com>
Date: Sat, 18 Mar 2017 04:43:55 +0000
From: Lipengcheng <lpc.li@...ilicon.com>
To: Alan Stern <stern@...land.harvard.edu>
CC: "gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>,
"linux-usb@...r.kernel.org" <linux-usb@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: RE: EHCI
Hi,
> -----Original Message-----
> From: Alan Stern [mailto:stern@...land.harvard.edu]
> Sent: Saturday, March 18, 2017 12:44 AM
> To: Lipengcheng
> Cc: gregkh@...uxfoundation.org; linux-usb@...r.kernel.org; linux-kernel@...r.kernel.org
> Subject: Re: EHCI
>
> On Fri, 17 Mar 2017, Lipengcheng wrote:
>
> > Hi,
> > In the ehci-hub.c,the ehci_handshake(ehci, status_reg, PORT_RESET, 0,
> > 1000) use that the timeout is 1000us. The 1000us timeout can not
> > satisfy all the chips. The EHCI protocol describes:A host controller
> > must terminate the reset and stabilize the state of the port within 2
> > milliseconds of software transitioning this bit from a one to a zero.
> > So I think timeout time is set to 2000us more appropriate. If I set
> > the timeout time is set to 2000us and may it produce side effects?
>
> That is a long time to wait with interrupts disabled. You probably should release the spinlock and enable interrupts during the handshake.
>
> Have you seen any errors with the current 1000 us value? If you haven't, there's no reason to change the code.
>
Yes, I have some error -110(-ETIMEDOUT) with the current 1000us value. When I set the value 2000us, it is ok.
> Alan Stern
Best Regards,
Pengcheng Li
Powered by blists - more mailing lists