Re: cvs commit: src/sys/dev/usb ohci.c ohci_pci.c ohcivar.h

From: ie9t-sbgk <>
Date: Tue, 23 Dec 2003 11:14:48 +0900
Hi Nate-san.

On Mon, 22 Dec 2003 09:46:54 -0800 (PST), Nate Lawson <> wrote:

> On Mon, 22 Dec 2003, Takeshi Shibagaki wrote:
>>   Modified files:
>>     sys/dev/usb          ohci.c ohci_pci.c ohcivar.h
>>   Log:
>>   methods for USB devices in the same way of uhci driver. But this change
>>   is not complete because some ohci controlers are not initialized completely.
>>   So "kernel: usb0: 1 scheduling overruns" interrupt will generate many times.
>>   This change will be same one in PR kern/60099.
>>   Discussed on [bsd-nomads:16737] - [bsd-nomads:16746].
>>   Revision  Changes    Path
>>   1.140     +2 -12     src/sys/dev/usb/ohci.c
>>   1.38      +40 -0     src/sys/dev/usb/ohci_pci.c
>>   1.36      +3 -0      src/sys/dev/usb/ohcivar.h
> Also,
> Lots of input from:	Anish Mistry <>
> Now any chance of getting you to look at uhci(4)?  It has similar
> suspend/resume problems.

I don't understand you well, but do you ask me whether looking at
"man 4 uhci" or uhci driver source(uchci.c...)?

If you are talking about uchi driver source, I have been look, and
I'm using VAIO Z505D,Z505V/BP, there is no problem. these machine
have uhci(Intel 82371AB/EB (PIIX4) USB controller).

Please tell me an example in the case of uhci, ohci problem.

> Perhaps Linux has some clues to the best
> approach.

I have researched linux 2.6.0 drivers. Maybe my Libretto L1 don't
work well at resume process in linux, also Victor InterLink MP-XP7210
('s machine). In japanese some laptop machines,
OHCI registers are clear almostly at suspend/resume process.
I had caught up re-init process added OHCI register write code step by
step, and I know it is necessary about 50% OHCI register init codes of
ohci_init() at suspend/resume. So I think it is better holding in common
between boot init codes and resume init one. Is it wrong?

Takeshi Shibagaki
Received on Mon Dec 22 2003 - 18:08:19 UTC