[mythtv-users] Firewire troubles after upgrade. Is test-mepg2 mightier than MythTV?

Jarod Wilson jarod at wilsonet.com
Wed Jan 27 06:08:04 UTC 2010


On Wed, Jan 27, 2010 at 1:00 AM, Jå§òÑ M@¢þhè®$Øñ
<jason.macpherson at gmail.com> wrote:
>> >
>> > I'm currently recompiling my Kernel with the the one-line patch you
>> > mentioned.
>> > http://patchwork.kernel.org/patch/75234/
>>
>> Ah! That has a very nice summary from Stefan Richter (upstream
>> firewire driver maintainer) of the details I was only vaguely
>> recalling. So it *was* the AB22/A that was previously blacklisted, and
>> its any address over 2GB, and he's confirmed the AB23 has the same
>> problem. So yeah, this is highly relevant for Deyan's setup.
>>
>> The patch I had in mind that everyone ought to be trying out his this though:
>>
>> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=090699c0530ae5380a9b8511d76f656cc437bb6e
>>
>> It universally disables dual-buffer mode by default for all controllers.
>>
>
>
> I'm looking over that patch now...
> Does use_dualbuffer ever get initialized to false?
> Maybe there is a memset call I'm not seeing.

The fw_ohci struct that use_dualbuffer is a member of is kzalloc'd, so
its 0 from the get go.

static int __devinit pci_probe(struct pci_dev *dev,
			       const struct pci_device_id *ent)
{
	struct fw_ohci *ohci;
	u32 bus_options, max_receive, link_speed, version;
	u64 guid;
	int err;
	size_t size;

	ohci = kzalloc(sizeof(*ohci), GFP_KERNEL);
...

-- 
Jarod Wilson
jarod at wilsonet.com


More information about the mythtv-users mailing list