[linux-dvb] Re: TechniSat AirStar2 support

[Javier Marcet <lists-linuxtv@marcet.info>]

* Christopher Pascoe <c.pascoe@itee.uq.edu.au> [040913 01:55]:

Chris THANKS a lot for my first birthday present :)

I'm right now happily watching for the first time terrestrial channels
in full digital quality. It is even better than most SAT channels!
Amazing.

The are still a couple quirks, though. Don't worry, they don't affect
quality or usability.

There are some errors reported.
Doing:
# modprobe dvb-ttpci
# modprobe stv0299
# modprobe skystar2
# modprobe mt352 ### the force_card=2 parameter is in /etc/modules.conf
logs this:

Sep 13 02:34:37 hiroshi saa7146: register extension 'dvb'.
Sep 13 02:34:37 hiroshi ACPI: PCI interrupt 0000:00:0f.0[A] -> GSI 11 (level, low) -> IRQ 11
Sep 13 02:34:37 hiroshi saa7146: found saa7146 @ mem e1ae6000 (revision 1, irq 11) (0x13c2,0x0000).
Sep 13 02:34:37 hiroshi DVB: registering new adapter (Siemens/Technotrend/Hauppauge PCI rev1.3).
Sep 13 02:34:37 hiroshi adapter has MAC addr = 00:d0:5c:03:7a:64
Sep 13 02:34:37 hiroshi gpioirq unknown type=0 len=0
Sep 13 02:34:37 hiroshi DVB: AV7110(0) - firm f0240009, rtsl b0250018, vid 71010068, app 8000261c
Sep 13 02:34:37 hiroshi DVB: AV7110(0) - firmware supports CI link layer interface
Sep 13 02:34:37 hiroshi av7110(0): Crystal audio DAC detected
Sep 13 02:34:37 hiroshi saa7146_vv: saa7146 (0): registered device video1 [v4l2]
Sep 13 02:34:37 hiroshi av7110: found av7110-0.
Sep 13 02:34:39 hiroshi stv0299: try to attach to Siemens/Technotrend/Hauppauge PCI rev1.3
Sep 13 02:34:39 hiroshi stv0299: setup for tuner BSRU6, TDQB-S00x
Sep 13 02:34:39 hiroshi DVB: registering frontend 0 (STV0299/TSA5059/SL1935 based)...
Sep 13 02:34:39 hiroshi ACPI: PCI interrupt 0000:00:0d.0[A] -> GSI 11 (level, low) -> IRQ 11
Sep 13 02:34:39 hiroshi drivers/media/dvb/b2c2/skystar2.c: FlexCopIIB(rev.195) chip found
Sep 13 02:34:39 hiroshi drivers/media/dvb/b2c2/skystar2.c: the chip has 38 hardware filters
Sep 13 02:34:39 hiroshi DVB: registering new adapter (Technisat SkyStar2 driver).
Sep 13 02:34:40 hiroshi master_xfer: write error 0 !
Sep 13 02:34:40 hiroshi message 0: flags=0x0, addr=0x68, buf[0]=0x2, len=2 
Sep 13 02:34:40 hiroshi master_xfer: read error !
Sep 13 02:34:40 hiroshi message 0: flags=0x0, addr=0x68, buf=0x0, len=1 
Sep 13 02:34:40 hiroshi message 1: flags=0x1, addr=0x68, buf=0x0, len=1 
Sep 13 02:34:40 hiroshi mt352_read_register: readreg error (ret == -121)
Sep 13 02:34:40 hiroshi mt352_read_register: readreg error (ret == -121)
Sep 13 02:34:40 hiroshi mt352_read_register: readreg error (ret == -121)
Sep 13 02:34:40 hiroshi mt352_read_register: readreg error (ret == -121)
Sep 13 02:34:40 hiroshi mt352_read_register: readreg error (ret == -38)
Sep 13 02:34:40 hiroshi mt352_read_register: readreg error (ret == -121)
Sep 13 02:34:40 hiroshi mt352_read_register: readreg error (ret == -121)
Sep 13 02:34:40 hiroshi dvbfe_mt352: Setup for Zarlink MT352 + Samsung TDTC9251DH01C DVB-T
Sep 13 02:34:40 hiroshi DVB: registering frontend 1 (Zarlink MT352 + Samsung TDTC9251DH01C DVB-T)...

Up until now I had only used DVB-s cards, and 99% of the channels I
watch are scrambled, hence I only recorded one at a time. I could
combine it with an analog channel.

With the FTA DVB-t channels I wanted to try how many channels I could
record simultaneously without going over my system resources' limits.

When I did the test I could record three DVB-t channels at once, but the
hardware could give more. The limit was within the mt352 or skystar2
driver. Switching to another DVB-t channel I got no errors, but only a
blank screen. However I could switch and also record (while the three
former recording were still taking place) a DVB-s channel.

Do you think this can be improved?

And lastly, if you use -mm, or apply the patch
ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.9-rc1/2.6.9-rc1-mm4/broken-out/remove-unconditional-pci-acpi-irq-routing.patch
you'll have to add 'pci=routeirq' to your kernel's boot parameters, or
else it'll fail.

Shall I report it to the author of the patch as it's said within it, or
is it up to mt352's author (either the original one or you) to fix it?


That's all. I only want to say thanks a lot again :)


--
"Whom are you?" said he, for he had been to night school.
		-- George Ade

Javier Marcet <javier@marcet.info>