Mailing List Archive: Sparc + Firewire problem with Sony DV Cam

Sparc + Firewire problem with Sony DV Cam

Sep 5, 2004, 4:21 AM

Post #1 of 3 (2184 views)

Morning All,

Sorry to cross post, but I'm not sure who's best placed to help here.

I have a Sony DCR-TRV-255 digital 8 camcorder, which I'm trying to
connect to my Sun Ultra 5 running Gentoo.

I've tried with both 2.4.25 and 2.6.7 kernels with similar
non-functioning results.

The card is a PCI ohci1394 compliant one with a Texas Instruments
chipset. It uses the ohci1394 driver, not the pcilynx one.

lspci gives:

0000:02:03.0 FireWire (IEEE 1394): Texas Instruments TSB12LV26 IEEE-1394
Controller (Link)

Success is reported on linux1394.org with this chipset. Is it the
camera?

On 2.4.25, I get the following in /proc/bus/ieee1394/devices:

Node[0-00:1023] GUID[0030ffa6000042da]:
Vendor ID: `Linux OHCI-1394' [0x000000]
Capabilities: 0x0083c0
Bus Options:
IRMC(1) CMC(1) ISC(1) BMC(0) PMC(0) GEN(0)
LSPD(2) MAX_REC(2048) CYC_CLK_ACC(0)
Host Node Status:
Host Driver : ohci1394
Nodes connected : 1
Nodes active : 1
SelfIDs received: 1
Irm ID : [0-00:1023]
BusMgr ID : [0-63:1023]
In Bus Reset : no
Root : yes
Cycle Master : yes
IRM : yes
Bus Manager : no

...but /proc/bus/ieee1394 doesn't exist in 2.6.7 although everything
appears to be OK and the ohci1394 driver loads.

When I connect the camera, the weird stuff starts (this is on 2.4.25,
but the results on 2.6.7 are identical).

Sep 5 10:09:42 ultra5 ohci1394_0: SelfID received, but NodeID invalid
(probably new bus reset occurred): 0800FFC0
Sep 5 10:09:42 ultra5 ohci1394_0: SelfID is inconsistent
[0x807e7f80/0xc88062a7]
Sep 5 10:09:42 ultra5 ieee1394: impossible ack_complete from node 65535
(tcode 4)
Sep 5 10:09:42 ultra5 ieee1394: Current remote IRM is not 1394a-2000
compliant, resetting...
Sep 5 10:09:42 ultra5 ohci1394_0: Get PHY Reg timeout
[0x013f017f/0x00000000/100]
Sep 5 10:09:42 ultra5 ohci1394_0: SelfID is inconsistent
[0x80707f80/0xc880691f]
Sep 5 10:09:42 ultra5 ieee1394: impossible ack_complete from node 65535
(tcode 4)
Sep 5 10:09:42 ultra5 ieee1394: Current remote IRM is not 1394a-2000
compliant, resetting...
Sep 5 10:09:42 ultra5 ohci1394_0: SelfID is inconsistent
[0x807c7f80/0xc8806dc2]
Sep 5 10:09:42 ultra5 ieee1394: impossible ack_complete from node 65535
(tcode 4)
Sep 5 10:09:42 ultra5 ieee1394: Current remote IRM is not 1394a-2000
compliant, resetting...
Sep 5 10:09:43 ultra5 ohci1394_0: Get PHY Reg timeout
[0x01bf01ff/0x00000000/100]
Sep 5 10:09:43 ultra5 ohci1394_0: SelfID is inconsistent
[0x807f7f80/0xc8807447]
Sep 5 10:09:43 ultra5 ieee1394: impossible ack_complete from node 65535
(tcode 4)
Sep 5 10:09:43 ultra5 ieee1394: Current remote IRM is not 1394a-2000
compliant, resetting...
Sep 5 10:09:43 ultra5 ohci1394_0: Get PHY Reg timeout
[0x01bf01ff/0x00000000/100]
Sep 5 10:09:43 ultra5 ohci1394_0: SelfID is inconsistent
[0x807c7f80/0xc8807ad1]
Sep 5 10:09:43 ultra5 ieee1394: impossible ack_complete from node 65535
(tcode 4)
Sep 5 10:09:43 ultra5 ieee1394: Current remote IRM is not 1394a-2000
compliant, resetting...
Sep 5 10:09:43 ultra5 ohci1394_0: SelfID is inconsistent
[0x80707f80/0xc8808034]
Sep 5 10:09:43 ultra5 ieee1394: impossible ack_complete from node 65535
(tcode 4)
Sep 5 10:09:43 ultra5 ieee1394: Current remote IRM is not 1394a-2000
compliant, resetting...
Sep 5 10:09:43 ultra5 ieee1394: Stopping reset loop for IRM sanity
Sep 5 10:09:43 ultra5 ieee1394: Node removed: ID:BUS[0-00:1023]
GUID[0030ffa6000042da]
Sep 5 10:09:44 ultra5 ohci1394_0: SelfID is inconsistent
[0x81627e93/0xc4808c88]
Sep 5 10:09:44 ultra5 ieee1394: SelfIDs failed root check
Sep 5 10:09:44 ultra5 ieee1394: Error in SelfID stage, resetting
Sep 5 10:09:44 ultra5 ohci1394_0: SelfID is inconsistent
[0x81627e93/0xc48098e6]

... then the last three entries are repeated until:
Sep 5 10:09:47 ultra5 ieee1394: SelfIDs failed root check
Sep 5 10:09:47 ultra5 ieee1394: Stopping out-of-control reset loop
Sep 5 10:09:47 ultra5 ieee1394: Warning - topology map and speed map
will not be valid
Sep 5 10:09:47 ultra5 ieee1394: Host added: ID:BUS[0-00:1023]
GUID[0030ffa6000042da]
Sep 5 10:09:48 ultra5 ohci1394_0: SelfID is inconsistent
[0x816e7e93/0xc480d876]

Then the three line loop above starts again.

At no point does the camera show up in /proc/bus/ieee1394/devices. The
system slows considerably during the above activity.

Has anybody got any ideas?

Regards,

James

--
James Andrew Holden, Leeds, UK (james at jamesholden dot net)
GPG Key: 1024D/8358863A *Please encrypt mail where possible!*
Fingerprint: 32C9 A76F 3CFE A06C 1B00 5AAB 9877 4742 8358 863A
jamesholden.net ICQ: 11290827 >Buy Linux CDs from fastdiscs.com<

Re: Sparc + Firewire problem with Sony DV Cam [ In reply to ]

gustavoz at gentoo

Sep 5, 2004, 8:08 PM

Post #2 of 3 (2086 views)

Permalink

James Holden wrote:

> I have a Sony DCR-TRV-255 digital 8 camcorder, which I'm trying to
> connect to my Sun Ultra 5 running Gentoo.
>
> I've tried with both 2.4.25 and 2.6.7 kernels with similar
> non-functioning results.
>
> The card is a PCI ohci1394 compliant one with a Texas Instruments
> chipset. It uses the ohci1394 driver, not the pcilynx one.
>
> lspci gives:
>
> 0000:02:03.0 FireWire (IEEE 1394): Texas Instruments TSB12LV26 IEEE-1394
> Controller (Link)
>
> Success is reported on linux1394.org with this chipset. Is it the
> camera?

Hi.
I've been testing firewire video for sparc lately.
Basically i have the same firewire controller as you do on my u5, using
ohci1394. My cam is a Canon mv300i (mini-dv) and i've been successful in
controlling it, capturing (raw mode only since others take up too much
cpu, they crash and burn in the u5 soon enough) and recording to it.
I think last time i played with it i was using 2.4.26, but it may have
been 2.4.25.
So yes, it mostly looks like the camera isn't liking something.

--
Gustavo Zacarias
Gentoo/SPARC & HPPA monkey

--
gentoo-sparc@gentoo.org mailing list

Re: Sparc + Firewire problem with Sony DV Cam [ In reply to ]

gentoo at jamesholden

Sep 6, 2004, 4:24 AM

Post #3 of 3 (2081 views)

Permalink

On Mon, 2004-09-06 at 04:08, Gustavo Zacarias wrote:
...snip...
> Hi.
> I've been testing firewire video for sparc lately.
> Basically i have the same firewire controller as you do on my u5, using
> ohci1394. My cam is a Canon mv300i (mini-dv) and i've been successful in
> controlling it, capturing (raw mode only since others take up too much
> cpu, they crash and burn in the u5 soon enough) and recording to it.
> I think last time i played with it i was using 2.4.26, but it may have
> been 2.4.25.
> So yes, it mostly looks like the camera isn't liking something.

Hi!

I had a feeling that might be the case.

I've read some stuff in the linux1394-users list archives about the way
that <red-herring type="possible"> the ohci1394 handles checksums in
certain types of packets where there's no payload</red-herring>.

I'll try contacting the guy who put the entry on the ieee1394 hardware
database for a very similar camera (a DCR-TRV250E instead of my 255E).
Maybe he did something special.

Cheers,

James

--
James Andrew Holden, Leeds, UK (james at jamesholden dot net)
GPG Key: 1024D/8358863A *Please encrypt mail where possible!*
Fingerprint: 32C9 A76F 3CFE A06C 1B00 5AAB 9877 4742 8358 863A
jamesholden.net ICQ: 11290827 >Buy Linux CDs from fastdiscs.com<

Mailing List Archive

Attached Files:

Attached Files: