[linux-dvb] [EXPERIMENTAL] cx88+xc3028 - tests are required - was: Re: When xc3028/xc2028 will be supported?

Mauro Carvalho Chehab mchehab at infradead.org
Tue Feb 19 19:05:14 CET 2008


On Tue, 19 Feb 2008 14:43:56 -0300
Dâniel Fraga <fragabr at gmail.com> wrote:

> On Tue, 19 Feb 2008 13:38:45 -0300
> Mauro Carvalho Chehab <mchehab at infradead.org> wrote:
> 
> > I've did some changesets fixing both drivers at:
> > 	http://linuxtv.org/hg/~mchehab/cx88-xc2028/
> 
> > Dâniel Fraga,
> > 
> > Please test if the gpio's from Daniel Gimpelvinch works for you, against the
> > newer tree. I suspect it will work. However, is it not uncommon to have two
> > cards with the same brand name, but needing different gpio settings.
> 
> 	Some good news Mauro:

Good news!
 
> cx88[0]: subsystem: 14f1:ea3d, board: PowerColor Real Angel 330
> [card=62,autodetected] cx88[0]: TV tuner type 71, Radio tuner type 0
> tuner' 0-0061: chip found @ 0xc2 (cx88[0])
> xc2028 0-0061: type set to XCeive xc2028/xc3028 tuner
> xc2028 0-0061: xc2028/3028 firmware name not set!
> cx88[0]: Asking xc2028/3028 to load firmware xc3028-v27.fw
> cx88[0]/0: found at 0000:01:08.0, rev: 5, irq: 18, latency: 32, mmio:
> 0xfc000000 cx88[0]/0: registered device video0 [v4l2]
> cx88[0]/0: registered device vbi0
> cx88[0]/0: registered device radio0
> xc2028 0-0061: Loading 80 firmware images from xc3028-v27.fw, type:
> xc2028 firmware, ver 2.7 cx88[0]: Calling XC2028/3028 callback
> xc2028 0-0061: Loading firmware for type=BASE MTS (5), id
> 0000000000000000. cx88[0]: Calling XC2028/3028 callback
> xc2028 0-0061: Loading firmware for type=MTS (4), id 000000000000b700.
> xc2028 0-0061: Device is Xceive 3028 version 1.0, firmware version 2.7
> cx88[0]: Calling XC2028/3028 callback
> xc2028 0-0061: Device is Xceive 3028 version 1.0, firmware version 2.7
> cx88[0]: Calling XC2028/3028 callback
> 
> 	Notice that it's loading the firmware correctly now ;). I tried
> xawtv and gqradio (and radio from xawtv too). I hear just white noise.

The above logs seems OK to my eyes.

> I tried to change the gpios, but without success...
> 
> 	Maybe Daniel Gimpelevich could try on his Powercolor to confirm
> this.... I think we're almost there... 

It seems so. The bug were trivial: just call request_module("tuner") before
sending commands to tuner ;)

> xc2028 0-0061: Loading firmware for type=BASE FM (401), id 0000000000000000.
> cx88[0]: Calling XC2028/3028 callback
> xc2028 0-0061: Loading firmware for type=FM (400), id 0000000000000000.
The above logs are OK.

> xc2028 0-0061: Loading SCODE for type=SCODE HAS_IF_3280 (60000000), id 0000000000000000. 
		 ^^^^^^^^^^^^^
		 This is wrong.

There's a tuner-xc2028 patch missing on this tree that fixes the above bug.
I've imported the fix patch from master tree. This will avoid to wrongly load an
scode table for FM.

Please update from the tree and try again.

I guess that your device uses input2. However, if I'm wrong, you'll need this
patch for FM:

diff -r 968d9b6b657d linux/drivers/media/video/cx88/cx88-cards.c
--- a/linux/drivers/media/video/cx88/cx88-cards.c	Wed Feb 13 23:52:48 2008 -0200
+++ b/linux/drivers/media/video/cx88/cx88-cards.c	Tue Feb 19 14:59:44 2008 -0300
@@ -2430,6 +2430,7 @@ static void cx88_card_setup(struct cx88_
 
 		ctl.fname   = XC2028_DEFAULT_FIRMWARE;
 		ctl.max_len = 64;
+		ctl.input1 = 1;
 
 		switch (core->boardnr) {
 		case CX88_BOARD_DVICO_FUSIONHDTV_DVB_T_PRO:

Could you also try if TV is properly working?

Cheers,
Mauro



More information about the linux-dvb mailing list