falo

Looks like this has been an issue for newer versions of these cards for a couple of years - found this git issue on Hauppauge's repo. As far as their other documentation is concerned, the card should be supported but it feels like they haven't kept up with hardware revisions and slightly shoddy code paths. Noted on future hardware choices!

I get you about the Hauppauge cards, and their site is ... unhelpful. Unfortunately it was the only accessible combo DVB-T and DVB-S card I could find (or indeed any good DVB-S card). card=59 doesn't load - fails on the same error from the m88ds3103 driver (and doesn't even try to load the DVB-T adapter). Going back into the source for that driver, I found this snippet - I think this is where it dies and fails the frontend setup (this is here) as if it's not one of those two hardcoded IDs (70 or 74 from the header file) then it throws an error. I'd try slipping 71 into there to see if it passes (maybe try to load it as the RT6000 as that's what the 5525 allegedly uses according to LinuxTV), but I have no idea how to build / deploy the driver if I did. How do firmware updates work with Hauppauge cards? Can't seem to find any documentation on it - do I need to pop it into a Windows box and do it via their software?

Oh yeah, definitely giving it the full /dev/dvb - I feel like if I've decided to split individual adapters across different dockers then I should just be rethinking my setup! The LinuxTV wiki suggests there should be two (the Si 2168/57 for DVB-T/DVB-C) and the Montage for the DVB-S). I'm sure in my desperate searching last night I found a picture of both listed side by side in a TVHeadend install but can't for the life of me find it now. It may have been an oddly specific dream. No rush though, it's one of many projects I'm playing with. Wasn't expecting you to look at this so soon, so thanks! [Edit] I'm not going mad, which is a good start. There was one in this very thread (but I'm assuming it's one of the earlier card revisions). [Edit 2] Just saw your syslinux note and have amended accordingly. Feels much better than having it probe, fail, I unload, then reprobe so ta!

Just changed it to device to check - both ways around headend sees /dev/dvb/adapter0, which is the DVB-T2 tuner in the card and the only adapter listed under /dev/dvb from an unraid shell following the probe. The expectation is that there should be one additional adapter to represent the DVB-S tuner from the card

Diagnostics attached. Was aiming to use it on TVHeadend, but do have Plex up and running as well. Headend does see the DVB-T2 adapter after the forced modprobe. servermonkey3-diagnostics-20230630-1657.zip

I was wondering if you could steer me in the right direction for getting my WinTV-HVR5525 working. It's a brand new board, direct from Hauppauge. Initially it wouldn't detect at all (syslog showed cx23885 not recognising it) so I forced detection with rmmod cx23885 modprobe cx23885 card=52 and get this output in the logs: Now I can see the DVB-T tuner (single adapter in /dev/dvb), but the DVB-S tuner is not present. It looks like it's still not properly detecting the card, and so not loading the correct part for that second tuner the card has. Looking at the Linux TV wiki for this card , the card I have appears to be a newer revision (C6I6) than the previous C316. I've found one other thread (with no answers) on Stackoverflow that suggests there's also a C5I6 floating around with similar model numbering. Key differences that stand out in my logs vs the output in the linuxtv wiki: Expecting subsystem 0070:f038 but the newer cards have 0070:f039 Expecting model 150329 where it is now 150339 (though I don't think this makes a difference looking at the kernel source), just throws a warning. The chip ID of the DVB-S tuner seems to be different (71 vs 64) so this could be the real issue if it's using a new chip. Trying to connect some dots from source, it looks like the driver used for the DVB-S tuner in the older card is M88DS3103 and it uses the chip_id to switch between either the M88DS3103 or the M88RS6000 (the latter being what the 5525 uses). These are defined in linux/drivers/media/dvb-frontends/m88ds3103_priv.h (M88RS6000_CHIP_ID etc.). As it doesn't have a definition for 0x71, it just waves goodbye and moves on to the next tuner. This is where I run dry on knowledge. Is this something I can patch locally for just dvb on unraid, or something I'm going to need to raise at the kernel level to hopefully get a change/fix implemented? Is there any way I can fool it into trying against a different chip_id to see if it's still compatible with the existing driver (there's no distinction in versions on WinTV's site, so I'd hope it would still be).

Just started using Unraid while also sorting out my network setup and ran into this exact issue. Something like this would be of considerable help for routing.