1394 ohci driver causes issues in win 7 64

NTDEV
1

hi all,

we have a wdm whql qualified driver, which runs on the root bus. the driver
works well for all oses and bit depths. However, in win 7 64 bit RC (and
RTM), when the 1394 OHCI compliant host controller is present (driver
version 6.1.7201.0) the driver never lets the machine go into S3. 1394 card
details:

VID = 1217 ,DID = 10F7

when we Disabled 1394?s ASPM function everything worked well. System will
show ?black screen? after _*PTS*, but don?t enter S3.

is this a known issue with the concerned 1394 card?

2

Hi Albert,

We did see similar issues with PCI\VEN_1217&DEV_00F7 device but were never able to find root cause. We think it’s a hw/fw issue but haven’t been able to get confirmation on that.

Thanks,

-Don[MS]

3

we are seeing it on a ‘non-released’ intel platform. It is highly possible
that the BIOS code to allocate pci space is messed up, but we can’t confirm
yet, as it is extremely difficult to debug. also, the moment we connrect teh
windbg, the issue vanishes, as the 1394 driver never gets loaded :frowning:

Don, can you tell us which vendor and device that device ID is associated
with, I do not have the platform with me right now, and so I cant double
check on that part very fast.

thanks

ap

On Thu, Aug 13, 2009 at 4:08 AM, wrote:

> Hi Albert,
>
> We did see similar issues with PCI\VEN_1217&DEV_00F7 device but were never
> able to find root cause. We think it’s a hw/fw issue but haven’t been able
> to get confirmation on that.
>
> Thanks,
>
> -Don[MS]
>
> —
> NTDEV is sponsored by OSR
>
> For our schedule of WDF, WDM, debugging and other seminars visit:
> http://www.osr.com/seminars
>
> To unsubscribe, visit the List Server section of OSR Online at
> http://www.osronline.com/page.cfm?name=ListServer
>

4

I have an ASUS M4A78-E motherboard that has a newer VIA OHCI chipset that did not work
with the new stack - I got a yellowbang on it. I figured I had something wrong, but maybe
this is the HW that you mention.

Works great with the Vista stack though. In fact, we ususally have trouble with non-TI chipsets,
but this one from VIA works great, even pushing it hard in isoch.

I’ll reenable the Win7 stack and see what error code is thrown - actually, I hope I haven’t clobbered
the “setupapi.log” (can’t remember what it’s called in Win7) - I can post a snippet here.

Greg Havenga
Sr. System Architect / Sr. Software Engineer

Photometrics, Inc.
3440 E Britannia Dr.
Tucson, AZ 85749

(520) 547-2566
http://www.photomet.com

From: xxxxx@lists.osr.com [xxxxx@lists.osr.com] On Behalf Of A P [xxxxx@gmail.com]
Sent: Wednesday, August 12, 2009 10:06 PM
To: Windows System Software Devs Interest List
Subject: Re: [ntdev] 1394 ohci driver causes issues in win 7 64

we are seeing it on a ‘non-released’ intel platform. It is highly possible that the BIOS code to allocate pci space is messed up, but we can’t confirm yet, as it is extremely difficult to debug. also, the moment we connrect teh windbg, the issue vanishes, as the 1394 driver never gets loaded :frowning:

Don, can you tell us which vendor and device that device ID is associated with, I do not have the platform with me right now, and so I cant double check on that part very fast.

thanks

ap

On Thu, Aug 13, 2009 at 4:08 AM, > wrote:
Hi Albert,

We did see similar issues with PCI\VEN_1217&DEV_00F7 device but were never able to find root cause. We think it’s a hw/fw issue but haven’t been able to get confirmation on that.

Thanks,

-Don[MS]

—
NTDEV is sponsored by OSR

For our schedule of WDF, WDM, debugging and other seminars visit:
http://www.osr.com/seminars

To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

— NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

5

>we are seeing it on a ‘non-released’ intel platform. It is highly possible

that the BIOS code to allocate pci space is messed up, but we can’t confirm
yet, as it is extremely difficult to debug.

Is it one of those Intel CRB/blue boards? If so, I’ve only dealt with two of them, but based on my experience, what you’re describing wouldn’t exactly be unprecedented, especially if the board in question features a newly released feature.

In any case, do you have an Arium?

mm

6

no it is a production board.

we have been able to isolate it to some extent. it never happens when we
connect a debugger to 1394 (obvious, as the driver for it doesnt load at
all).

My question is, how do I debug it.

We have aurium and ITP but we cannot connect it as it is not a CRB.

It doesnt have a COM port either.

The only way would be to connect another 1394 card to it through pcmcia.

Myquestions to u experts wuld be, once I have two 1394 ports on the debugee,
how does windbg decide which one to pick as the debuging channel? letting
the other one load normally?

is there a way?

thanks

ap

On Thu, Aug 13, 2009 at 7:21 PM, wrote:

> >we are seeing it on a ‘non-released’ intel platform. It is highly possible
> >that the BIOS code to allocate pci space is messed up, but we can’t
> confirm
> >yet, as it is extremely difficult to debug.
>
> Is it one of those Intel CRB/blue boards? If so, I’ve only dealt with two
> of them, but based on my experience, what you’re describing wouldn’t exactly
> be unprecedented, especially if the board in question features a newly
> released feature.
>
> In any case, do you have an Arium?
>
>
> mm
>
> —
> NTDEV is sponsored by OSR
>
> For our schedule of WDF, WDM, debugging and other seminars visit:
> http://www.osr.com/seminars
>
> To unsubscribe, visit the List Server section of OSR Online at
> http://www.osronline.com/page.cfm?name=ListServer
>