From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.10) Gecko/20050720 Fedora/1.0.6-1.1.fc4 Firefox/1.0.6 Description of problem: /usr/share/hdwata/pci{.ids,table} lack entries for the Hauppauge Nova-T model 909 DVB card The driver (cx88-dvb) is present from kernel >= 2.6.12 Version-Release number of selected component (if applicable): hwdata-0.158.1-1 How reproducible: Always Steps to Reproduce: 1. look in the files... 2. 3. Actual Results: no entries for Nova-T... Additional info:
Created attachment 118091 [details] pcitable entries for Nova-T I've submitted the relevant entries for pci.ids to pciids.sf.net
Shouldn't be needed in pcitable ; should be handled fine with modules.pcimap.
It isn't. The modules.pcimap that depmod generates for kernel 2.6.12-1.1398_FC4 doesn't handle the subvendor or something; I don't know how it's supposed to work, but it doesn't load cx88-dvb for this device. ISTR that before I made the change it loaded (uselessly) cx88-blackbird, but I can't revert to check just at the moment.
cx88-dvb 0x000014f1 0x00008802 0xffffffff 0xffffffff 0x00000000 0x00000000 0x0 If it's supposed to match the 0x8800 you mention in your pcitable patch, the driver needs updated. The problem is that both -dvb and -blackbird claim the device, which in FC4 causes one to get loaded at random. We're looking at fixing this for FC5.
The card in question has 0x8800, 0x8802 and 0x8804 devices. Wouldn't the clash be resolved if the subvendor and subdevice ids were filled in? lspci -n -v -s 01:01 01:01.0 Class 0400: 14f1:8800 (rev 05) Subsystem: 0070:9002 Flags: bus master, medium devsel, latency 32, IRQ 10 Memory at fb000000 (32-bit, non-prefetchable) [size=16M] Capabilities: <available only to root> 01:01.2 Class 0480: 14f1:8802 (rev 05) Subsystem: 0070:9002 Flags: bus master, medium devsel, latency 32, IRQ 10 Memory at fc000000 (32-bit, non-prefetchable) [size=16M] Capabilities: <available only to root> 01:01.4 Class 0480: 14f1:8804 (rev 05) Subsystem: 0070:9002 Flags: bus master, medium devsel, latency 32, IRQ 10 Memory at fd000000 (32-bit, non-prefetchable) [size=16M] Capabilities: <available only to root> (as I said, I don't know how it's supposed to work... 0070 is Hauppauge, and I think 9002 only matches the Nova-T DVB model 909) Anyway, with the patch I made it works for me, so I'm happy (and look forward to it just working in FC5), and anyone trying to find out why their Nova-T isn't detected in FC4 will stand a chance of finding this. For the benefit of such searchers, it might be worth mentioning that after patching pcitable and pci.ids, I ran kudzu. I'm not sure if each of these steps is necessary.
*** Bug 168411 has been marked as a duplicate of this bug. ***
This was fixed in mid-December; we now use sysfs & udev to load modules at boot; this should load both modules.