headers
Joel CARNAT | 1 Sep 2008 12:09
Favicon
Gravatar

Re: ACPI S3 and VIA (C3) CPU

Hi,

Le Dim 31 août 2008 23:27, Nicolas Joly a écrit :
> On Sun, Aug 31, 2008 at 11:05:42PM +0200, Joel CARNAT wrote:
>> Le Dim 31 août 2008 22:52, Nicolas Joly a écrit :
>> > On Sun, Aug 31, 2008 at 10:44:29PM +0200, Joel CARNAT wrote:
>> >> Hello,
>> >>
>> >> I tried zzz(8) on my VIA machine (VIA Nehemiah) but got the error:
>> >> acpi0: entering state 3
>> >> acpi0: ACPI S3 not available on this platform
>> >> Using "sudo apm -S" just froze the machine.
> [...]
>> # sysctl hw.acpi
>> hw.acpi.root = 1010208
>> hw.acpi.supported_states = S0 S1 S4 S5
>>

Linux 2.6.24.5 (also) says:
ACPI: (supports S0 S1 S4 S5)

(Don't know why I keep thinking Linux might work better/different...)

>> Is there some hidden option to enable in the kernel ?
>
> No. AFAIK, it won't work without a BIOS supporting it.
>

Would this only be a BIOS issue ?
I mean, if I can find a BIOS upgrade, it may help ?
(Continue reading)

Permalink | Reply | 0 comments |
headers
Brad du Plessis | 1 Sep 2008 13:59
Picon

USB device detection problem

Hi all,

I have a problem on NetBSD 4.0_STABLE with USB ports on an Intel DG33FB 
motherboard not working under certain conditions.

If I follow the following steps the hard-wired rear (mounted on the 
motherboard) ports will fail to detect the insertion of a USB flash drive:

- boot the box cold (a warm reboot never reproduces the problem)
- boot with a USB mouse plugged in on one of the rear ports (with no 
mouse or just a PS/2 mouse the problem cannot be reproduced)

After a matter of seconds (maybe 5-10 seconds?) after having reached the 
login prompt I'll see the log:

uhub2: device problem, disabling port 2

Any device plugged into one of the rear ports will now not get detected. 
I can still plug my USB flash drive into one of the ports plugged into 
the internal USB headers and it works fine. I've reproduced this on 2 
Intel DG33FB's. (Btw, the same procedure does not show any problem when 
booting into Windows XP).

If I change it round a bit and plug the USB mouse into a port connected 
to one of the internal USB headers, the other port on that header will 
not work but the rear ports will.

I'll try and compile with USB_DEBUG and repeat the procedure.

Any help would be much appreciated.
(Continue reading)

Permalink | Reply | 3 comments |
headers
Brad du Plessis | 1 Sep 2008 14:57
Picon

Re: USB device detection problem

Brad du Plessis wrote:
> I'll try and compile with USB_DEBUG and repeat the procedure.

Here are the logs from a USB_DEBUG compile, I take it this:

usbd_new_device: set address 2 failed - trying a port reset

is the problem ?

Thanks,
 Brad

Sep  1 05:45:23 test /netbsd: Copyright (c) 1996, 1997, 1998, 1999, 
2000, 2001, 2002, 2003, 2004, 2005,
Sep  1 05:45:23 test /netbsd: 2006, 2007
Sep  1 05:45:23 test /netbsd: The NetBSD Foundation, Inc.  All rights 
reserved.
Sep  1 05:45:23 test /netbsd: Copyright (c) 1982, 1986, 1989, 1991, 1993
Sep  1 05:45:23 test /netbsd: The Regents of the University of 
California.  All rights reserved.
Sep  1 05:45:23 test /netbsd:
Sep  1 05:45:23 test /netbsd: NetBSD 4.0_STABLE (GENERIC.MP) #0: Mon 
Sep  1 05:27:52 SAST 2008
Sep  1 05:45:23 test /netbsd: 
root <at> test:/usr/data/wd0f/src/sys/arch/i386/compile/GENERIC.MP
Sep  1 05:45:23 test /netbsd: total memory = 996 MB
Sep  1 05:45:23 test /netbsd: rbus: rbus_min_start set to 0x40000000
Sep  1 05:45:23 test /netbsd: avail memory = 968 MB
Sep  1 05:45:23 test /netbsd: timecounter: Timecounters tick every 
10.000 msec
(Continue reading)

Permalink | Reply | 2 comments |
headers
Frederik Sausmikat | 1 Sep 2008 15:32
Picon
Picon

uplcom: can only receive data

Hi,

I'm currently trying to get USB to serial converters to work with
NetBSD-4.0 and -current. When plugging the adapters in they get recognized
by the kernel...

    uplcom0 at uhub0 port 1
    uplcom0: Prolific Technology Inc. USB-Serial Controller D, rev
    1.10/3.00, addr 2
    ucom0 at uplcom0

or

    uplcom0 at uhub3 port 1
    uplcom0: ATEN International Serial adapter, rev 1.10/0.01, addr 2
    ucom0 at uplcom0

...and I am able to send data *from* a regular built-in serial port *to* 
the USB adapter but not the other way round (using cat, cu or minicom). 
This is the case on an Asus EeePC 900 running 4.99.72 GENERIC and my 
home PC running 4.0-STABLE (both i386).

Additionally I tried the adapters with my machine at work, running 
4.0-GENERIC (looping tty00 back to ttyU0 via null modem cable) and there 
they work without any problems (this machine is also i386).

They also work with Debian ("testing", kernel 2.6.26?) on all machines, 
so it doesn't seem to be hardware related.

Now the question is, does anybody have an idea what the problem might be
(Continue reading)

Permalink | Reply | 17 comments |
headers
Martin Husemann | 1 Sep 2008 16:37
Picon

Re: uplcom: can only receive data

On Mon, Sep 01, 2008 at 03:32:07PM +0200, Frederik Sausmikat wrote:
> ...and I am able to send data *from* a regular built-in serial port *to* 
> the USB adapter but not the other way round (using cat, cu or minicom). 
> This is the case on an Asus EeePC 900 running 4.99.72 GENERIC and my 
> home PC running 4.0-STABLE (both i386).

Is the cable full grown, including RTS/CTS/DCD/DTE?

Do you have an entry in /etc/ttys for the usb or the regular serial port?
(not that you need to, just checking if any of the flags available there
interferes with testing)

Martin
Permalink | Reply | 7 comments |
headers
Frederik Sausmikat | 1 Sep 2008 18:22
Picon
Picon

Re: uplcom: can only receive data

Martin Husemann wrote:
> Is the cable full grown, including RTS/CTS/DCD/DTE?
>   
Yes it seems to be, though some of the pins involving one (or two) of 
the above signals are bridged on one end but not the other. But since it 
worked at least on one NetBSD box and under Debian I did not investigate 
into this further.

I'll have another look at the cable and will rewire the pins in question...

> Do you have an entry in /etc/ttys for the usb or the regular serial port?
> (not that you need to, just checking if any of the flags available there
> interferes with testing)
>   
No, on all machines the /etc/ttys file is the one included in NetBSD.

Regards, Freddy
Permalink | Reply | 6 comments |
headers
Frederik Sausmikat | 1 Sep 2008 19:14
Picon
Picon

Re: uplcom: can only receive data

Frederik Sausmikat wrote:
> Martin Husemann wrote:
>> Is the cable full grown, including RTS/CTS/DCD/DTE?
Now it definitely is. Unfortunately it doesn't change a thing, still 
only receiving on the USB side.

Regards, Freddy
Permalink | Reply | 5 comments |
headers
Martin Husemann | 1 Sep 2008 19:46
Picon

Re: uplcom: can only receive data

On Mon, Sep 01, 2008 at 07:14:00PM +0200, Frederik Sausmikat wrote:
> Now it definitely is. Unfortunately it doesn't change a thing, still 
> only receiving on the USB side.

Does "cu -f -l ttyU0" (or whichever tty you are using) work?

I have several different uplcom devices and they all work (bidirectional)
for me. But we have seen subtle difference between them before, so your two
being somehow different wouldn't suprise me.

Martin
Permalink | Reply | 4 comments |
headers
Frederik Sausmikat | 1 Sep 2008 21:05
Picon
Picon

Re: uplcom: can only receive data

Martin Husemann wrote:
> On Mon, Sep 01, 2008 at 07:14:00PM +0200, Frederik Sausmikat wrote:
>> Now it definitely is. Unfortunately it doesn't change a thing, still 
>> only receiving on the USB side.
> 
> Does "cu -f -l ttyU0" (or whichever tty you are using) work?

Already tried that, too. With no success.

> I have several different uplcom devices and they all work (bidirectional)
> for me. But we have seen subtle difference between them before, so your two
> being somehow different wouldn't suprise me.

The strange thing is that there are no problems with my NetBSD box at 
work. And especially interesting to me is why the device is able to 
receive but not to send, since this Prolific adapter is reported to work 
by several people.

Maybe the problem lies somwhere in the upper levels of the USB driver chain?
Permalink | Reply | 3 comments |
headers
Martin Husemann | 1 Sep 2008 21:08
Picon

Re: uplcom: can only receive data

On Mon, Sep 01, 2008 at 09:05:24PM +0200, Frederik Sausmikat wrote:
> Maybe the problem lies somwhere in the upper levels of the USB driver chain?

Unlikely. You don't happen to have a serial line tester?

Martin
Permalink | Reply | 1 comment |

Gmane