[SOLVED!!!] Dead Firewire problem after sleep/wake up.

[powerpcg5]

UPDATE:
Guys, great news!

Problem is SOLVED!!!!

Buy this item from Amazon
http://www.amazon.com/gp/product/B002S53IG8
$25

It comes with a small bracket, but there's a longer bracket also included in the box. So just swap the brackets. It's got a *golden* TI chipset.... not some lame-o VIA firewire chipset.

Install card into a PCI Express slot.
Go to your bios and DISABLE the on-board Gigabyte FW port.
Reboot to OSX.

In your console log, you'll see a new message
NEW: FireWire (OHCI) TI ID 823f PCI now active, GUID 785634120000000b; max speed s800.

instead of the built-in Gigabyte Firewire port (remember to disable in BIOS onboard FW)
OLD: FireWire (OHCI) VendorID 1106 ID 3044 PCI now active, GUID 00656f1cda78cb00; max speed s400.

Notice, the new TI card also gives you 800Mbit speed.

Moment of truth test...
Played itunes. MOTU audio works.
Go to sleep.
Wake up.
Check Console Log. NO MORE ERRORS (Firewire unknown fffffff GUID)!
Play itunes again....
YES!... audio resumes where it left off.

... so if you'll be buying a Gigabyte motherboard, save your dollars and buy one without the onboard FW port (because they're crap anyway). Instead buy a TI-chipset Firewire card.

YAY!!!!

See also this thread
viewtopic.php?f=22&t=45039

ORIGINAL POST:
Here's the problem.... upon fresh boot, my firewire port works. (Giga Z68x UD4 mobo).
I have a MOTU audio interface connected to the FW port.
It works.

But If I go to Sleep, and then wake up, the MOTU audio interface goes dead. I've traced this to the Firewire port not coming back alive.
Checking console, it says:
FW3.png

Here's the mystery.
Upon fresh boot, according to Console Log....
FW1.png
First, there's that error message "runtime power conservation disabled"
Second, it shows Vendor ID 1106, ID 3044 PCI and a GUID of 00656f1..... cb00

Who or what is that mystery vendor ID 1106? Who knows? That's not mine.

Because upon fresh boot, connecting my MOTU audio interface and Sony DV camera, I only have these 2 Vendor and GUIDs that are listed below, and I know are valid.
FW2.png

There is a 3rd "Unknown" entry in About this Mac... my bet this is that mystery vendor ID 1106.

So to recap, Firewire functionality goes to crap upon waking up from reboot.
None of your FW ports will work anymore.
Console reports Firewire GUID ffffffffffffff is invalid after waking up.
Whatever valid Firewire GUID/device you have working before sleep, will disappear from About this Mac upon wake up.
You'll be left with that "Unknown device".
If you reboot your computer again to regain functionality of Firewire, Console status message reports some Bogus Firewire Vendor ID and Device ID, and GUID... which most probably, is a device you don't have.

So what's the real problem here?
The Firewire chipset of the Gigabyte motherboard?
A firewire kext?
The DSDT's firewire portion is wrong? (hardcoded with that bogus GUID?)

I hope we can all find a fix to this Firewire problem.

I've ordered a Firewire card with TI chipset... will be here in a couple of days. I'll report if that fixes the firewire problem.

** But I still want to know where that bogus Vendor ID, device ID, and GUID that Console is reporting coming from.

 

[powerpcg5]

Re: Firewire problem caused by Bogus Firewire GUID in DSDT?

Update:
Checking Console Log, and without doing anything, the Gigabyte Firewire port came back alive again.... AFTER 38 MINUTES.

FW4.png

It took 38 minutes for firewire to come back and work again!

Dome some research, 1106 is the Vendor ID for VIA Technologies
and device 3044 is VIA Fire II 1394a OHCI Link Layer Ctrlr1

So now the question is... why does it take 38 minutes for Firewire to come alive again after Wake up from Sleep?

 

[kryz]

Re: Firewire problem caused by Bogus Firewire GUID in DSDT?

Hmm, I too have a motu fw card and was almost going to get the same mobo but with that info I think I might be better off with the UD5. That seems to be the only one with TI FW (also 3 ports) and hdmi on board.

 

[powerpcg5]

Re: Firewire problem caused by Bogus Firewire GUID in DSDT?

I'll let you know if a FW card with TI chipset will solve the MOTU post-sleep problem.

Don't get me wrong... I'm still happy with the UD4 board, especially getting the ability to sleep again and overall mobo performance.

If I can just fix this post-sleep dead Firewire issue, it will be perfect.

 

[kryz]

Re: Firewire problem caused by Bogus Firewire GUID in DSDT?

Yeah, thanks for that. I already ordered the Z68XP-UD5 today, though, but I could still change that around. I hope you get the problem solved soon.

 

[powerpcg5]

Re: Firewire problem caused by Bogus Firewire GUID in DSDT?

Tried some DSDT changes (as discussed at InsanelyMac) re: giving HPET (4) IRQs based on data gathered from IORegistryExplorer.

Still no go... no change.

I can verify the problem isn't with just the MOTU firewire... FW external HD is also affected by this problem. i.e. FW drives go dead after wake from sleep.

Also, can't select FW drives to be boot devices during startup.....
So.... this is giving me the feeling that Gigabyte on-board firewire isn't 100% compatible with Lion... it must be a VIA chipset/hardware problem.

 

[kryz]

Re: Firewire problem caused by Bogus Firewire GUID in DSDT?

That sucks. Now if I anybody could confirm that this is a via-chip-only issue and the ti built in fw works better...

 

[akimov]

Re: Firewire problem caused by Bogus Firewire GUID in DSDT?

Hi,

I've got the same problem with a TC electronic Konnekt 24D firewire card.
It works great but I can't wake up the firewire port. (sleep doesn't work properly anyway).

I can't mount firewire drive too.

 

[powerpcg5]

Re: Firewire problem caused by Bogus Firewire GUID in DSDT?

I can't mount firewire drive too.

I can mount my FW drive, but only when on the Lion desktop. I can access its contents, etc...

What I can't do is use the FW drive as a boot drive because it's invisible to Chimera. (whether this is a Chimera problem or a VIA chipset problem, I don't know.)

Also, after sleep, I lose access to the FW drive. Finder still shows the drive(s), (i.e. not disconnected), but I can't access it's contents anymore... Then if I try to EJECT the FW drive from Finder, then the computer hangs.

Are we the only guys using Firewire? I can't believe I didn't come across these problems during my previous research.

-- I guess the quick-fix solution to this is don't go to sleep... (everything is fine before sleep) FW only craps out after wake up.

 

[kryz]

Re: Firewire problem caused by Bogus Firewire GUID in DSDT?

I remember having had that problem with a fw drive on my macbook pro. I think it had something to do with the time machine backup. Unfortunately I forgot the details how I solved that. So this might not be a problem coming from that particular board. Maybe you can find some helpful info in the apple knowledge base?