experimental support for C-1501

The following experimental patch adds support for the technotrend budget C-1501 dvb-c card. The parameters used to configure the tda10023 demod chip are largely determined experimentally, but works quite for me in my initial tests.

Signed-off-by Sigmund Augdal <[email protected]>

------------------------------------------------------------------------

_______________________________________________
linux-dvb mailing list
[email protected]
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb

Quoted 4 lines · Sigmund Augdal

The following experimental patch adds support for the technotrend budget C-1501 dvb-c card. The parameters used to configure the tda10023 demod chip are largely determined experimentally, but works quite for me in my initial tests.

You finally found correct values :) I see that it uses same clock settings than Anysee. Also deltaf could be correct because I remember from my tests that it cannot set wrong otherwise it will not work at all. How did you find defltaf?

Your patch has at least coding style issues, please run make checkpatch fix errors and resend patch.

regards
Antti

Quoted 10 lines

Signed-off-by Sigmund Augdal <[email protected]>

------------------------------------------------------------------------

_______________________________________________
linux-dvb mailing list
[email protected]
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb

Hello all.
I have a skystar hd2 card and installed the latest mantis driver. Now I am able to watch some dvb-s and dvb-s2 channels from astra 19,2. The LNB for this satalite is connected to the default switch port. When I try to tune to a channel on any other satalite that is on de diseqc switch I get no signal.
I tried this with mythtv 0.21 and the mythtv patch from:
http://svn.mythtv.org/trac/ticket/5403

I have 2 questions.
1. Has anyone used the skystar hd2 or twinhand vp-1041 succesfull with a diseqc switch?

2. What should I do to figure out what the problem is exactly?

Thanks
Joep

ons, 04.06.2008 kl. 16.58 +0300, skrev Antti Palosaari:

Quoted 10 lines

Quoted 4 lines · Sigmund Augdal

The following experimental patch adds support for the technotrend budget C-1501 dvb-c card. The parameters used to configure the tda10023 demod chip are largely determined experimentally, but works quite for me in my initial tests.

You finally found correct values :) I see that it uses same clock settings than Anysee. Also deltaf could be correct because I remember from my tests that it cannot set wrong otherwise it will not work at all. How did you find defltaf?

I guessed the clock settings based on how the same tuner is used by a different demod. The deltaf value was found by trial and error (helped by some scripting). deltaf values slightly off will also work, but tuning will be very slow. I also think the deltaf value will depend on the bandwidth of the transponder tuned. All transponders I've tested with are 8MHz, but I think other values are possible, and these will most likely not work. I submitted the patch anyway in the hope that some broader testing might help uncover any remaining problems.

Quoted 1 line

Quoted 2 lines

Your patch has at least coding style issues, please run make checkpatch fix errors and resend patch.

I was trying to follow the guidelines, but I guess I wasn't doing that good enough. Will try to clean that up and resend soon.

regards
Sigmund

Quoted 16 lines

regards
Antti

Quoted 10 lines

Signed-off-by Sigmund Augdal <[email protected]>

------------------------------------------------------------------------

_______________________________________________
linux-dvb mailing list
[email protected]
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb

_______________________________________________
linux-dvb mailing list
[email protected]
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb

Here is a new version. This one passes checkpatch without warnings. I removed the read_pwm function, as it always uses the fallback path for my card (and frankly I have no idea wether it is actually relevant at all for this kind of card). Furthermore the tda10023 driver doesn't seem to use this value for anything.

Quoted 5 lines · Sigmund Augdal

Here is a new version. This one passes checkpatch without warnings. I removed the read_pwm function, as it always uses the fallback path for my card (and frankly I have no idea wether it is actually relevant at all for this kind of card). Furthermore the tda10023 driver doesn't seem to use this value for anything.

Any issues with this patch? If not I will commit it next weekend.

Quoted 8 lines · Oliver Endriss

Quoted 5 lines · Sigmund Augdal

Here is a new version. This one passes checkpatch without warnings. I removed the read_pwm function, as it always uses the fallback path for my card (and frankly I have no idea wether it is actually relevant at all for this kind of card). Furthermore the tda10023 driver doesn't seem to use this value for anything.

Any issues with this patch? If not I will commit it next weekend.

At the first glance it works fine, thanks Sigmund.

Although for multiproto drivers I slightly modified patches and replaced some files from linux-dvb tree (tda10021.c, tda10023.c, tda1002x.h).

However, I don't know how necessary is tda827x-oops patch with multiproto.

At compile have only one warning:

   CC [M]  /usr/local/src/multiproto-2008-05-14/v4l/budget-av.o

/usr/local/src/multiproto-2008-05-14/v4l/budget-av.c: In function 'frontend_init':
/usr/local/src/multiproto-2008-05-14/v4l/budget-av.c:1306: warning: passing argument 1 of '__a' from incompatible pointer type

Regards,
AK
plain text document attachment (tda827x-oops-multiproto.patch)

drivers/media/dvb/frontends/tda827x.c renamed from drivers/media/dvb/frontends/tda827x.c.old +2 -1

@@ -554,7 +554,7 @@
 	struct tda827x_priv *priv = fe->tuner_priv;
 	unsigned char buf[] = {0x22, 0x01};
 	int arg;
-	struct i2c_msg msg = { .addr = priv->i2c_addr, .flags = 0,
+	struct i2c_msg msg = { .flags = 0,
 			       .buf = buf, .len = sizeof(buf) };
 
 	if (NULL == priv->cfg) {

@@ -562,6 +562,7 @@
 		return;
 	}
 
+	msg.addr = priv->i2c_addr;
 	if (priv->cfg->config) {
 		if (high)
 			dprintk("setting LNA to high gain\n");

plain text document attachment (c-1501-multiproto.patch)

drivers/media/dvb/ttpci/budget-ci.c renamed from drivers/media/dvb/ttpci/budget-ci.c.old +28 -1

@@ -51,6 +51,8 @@
 #include "lnbp21.h"
 #include "bsbe1.h"
 #include "bsru6.h"
+#include "tda1002x.h"
+#include "tda827x.h"
 
 /*
  * Regarding DEBIADDR_IR:

@@ -1337,6 +1339,16 @@
 	.refclock	= 27000000,
 };
 
+static struct tda10023_config tda10023_config = {
+	.demod_address = 0xc,
+	.invert = 0,
+	.xtal = 16000000,
+	.pll_m = 11,
+	.pll_p = 3,
+	.pll_n = 1,
+	.deltaf = 0xA511,
+};
+
 static void frontend_init(struct budget_ci *budget_ci)
 {
 	switch (budget_ci->budget.dev->pci->subsystem_device) {

@@ -1404,7 +1416,20 @@
 				budget_ci->budget.dvb_frontend = NULL;
 			}
 		}
-
+		break;
+	case 0x101a: /* TT Budget-C-1501 (philips tda10023/philips tda8274A) */
+		budget_ci->budget.dvb_frontend =
+			dvb_attach(tda10023_attach, &tda10023_config,
+				   &budget_ci->budget.i2c_adap, 0x48);
+		if (budget_ci->budget.dvb_frontend) {
+			if (dvb_attach(tda827x_attach,
+				       budget_ci->budget.dvb_frontend, 0x61,
+				       &budget_ci->budget.i2c_adap, NULL)
+			    == NULL)
+				printk(KERN_ERR "%s: No tda827x found!\n",
+				       __func__);
+			break;
+		}
 		break;
 
 	case 0x1019:		// TT S2-3200 PCI

@@ -1535,6 +1560,7 @@
 MAKE_BUDGET_INFO(ttbtci, "TT-Budget-T-CI PCI", BUDGET_TT);
 MAKE_BUDGET_INFO(ttbcci, "TT-Budget-C-CI PCI", BUDGET_TT);
 MAKE_BUDGET_INFO(tt3200, "TT-Budget S2-3200 PCI", BUDGET_TT);
+MAKE_BUDGET_INFO(ttc1501, "TT-Budget C-1501 PCI", BUDGET_TT);
 
 static struct pci_device_id pci_tbl[] = {
 	MAKE_EXTENSION_PCI(ttbci, 0x13c2, 0x100c),

@@ -1544,6 +1570,7 @@
 	MAKE_EXTENSION_PCI(ttbtci, 0x13c2, 0x1012),
 	MAKE_EXTENSION_PCI(ttbs2, 0x13c2, 0x1017),
 	MAKE_EXTENSION_PCI(tt3200, 0x13c2, 0x1019),
+	MAKE_EXTENSION_PCI(ttc1501, 0x13c2, 0x101A),
 	{
 	 .vendor = 0,
 	 }

_______________________________________________
linux-dvb mailing list
[email protected]
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb

This i was only to get the patch to apply, because the context had changed slightly (because of the definitions for the s2 cards), right?

This line is, to the best of what I can see not added by my patch, neither in my version, or the version you provided for the multiproto tree.

Quoted 8 lines · Oliver Endriss

Quoted 5 lines · Sigmund Augdal

Here is a new version. This one passes checkpatch without warnings. I removed the read_pwm function, as it always uses the fallback path for my card (and frankly I have no idea wether it is actually relevant at all for this kind of card). Furthermore the tda10023 driver doesn't seem to use this value for anything.

Any issues with this patch? If not I will commit it next weekend.

At the first glance it works fine, thanks Sigmund.
Although for multiproto drivers I slightly modified patches and replaced some
files from linux-dvb tree (tda10021.c, tda10023.c, tda1002x.h).

However, I don't know how necessary is tda827x-oops patch with multiproto.

At compile have only one warning:
CC [M] /usr/local/src/multiproto-2008-05-14/v4l/budget-av.o
/usr/local/src/multiproto-2008-05-14/v4l/budget-av.c: In function 'frontend_init':
/usr/local/src/multiproto-2008-05-14/v4l/budget-av.c:1306: warning:
passing
argument 1 of '__a' from incompatible pointer type

Regards,
AK

--- linux/drivers/media/dvb/frontends/tda827x.c.old 2008-06-06 13:57:57.000000000 +0300
+++ linux/drivers/media/dvb/frontends/tda827x.c 2008-06-06 13:59:52.000000000 +0300

drivers/media/dvb/frontends/tda827x.c renamed from drivers/media/dvb/frontends/tda827x.c.old +1 -1

@@ -554,7 +554,7 @@
 struct tda827x_priv *priv = fe->tuner_priv;
 unsigned char buf[] = {0x22, 0x01};
 int arg;
- struct i2c_msg msg = { .addr = priv->i2c_addr, .flags = 0,
+ struct i2c_msg msg = { .flags = 0,
        .buf = buf, .len = sizeof(buf) };
 
 if (NULL == priv->cfg) {

--- linux/drivers/media/dvb/ttpci/budget-ci.c.old 2008-06-06 13:39:23.000000000 +0300
+++ linux/drivers/media/dvb/ttpci/budget-ci.c 2008-06-06 13:50:18.000000000 +0300

drivers/media/dvb/ttpci/budget-ci.c renamed from drivers/media/dvb/ttpci/budget-ci.c.old +27 -1

@@ -51,6 +51,8 @@
 #include "lnbp21.h"
 #include "bsbe1.h"
 #include "bsru6.h"
+#include "tda1002x.h"
+#include "tda827x.h"
 
 /*
 * Regarding DEBIADDR_IR:

@@ -1337,6 +1339,16 @@
 .refclock = 27000000,
 };
 
+static struct tda10023_config tda10023_config = {
+ .demod_address = 0xc,
+ .invert = 0,
+ .xtal = 16000000,
+ .pll_m = 11,
+ .pll_p = 3,
+ .pll_n = 1,
+ .deltaf = 0xA511,
+};
+
 static void frontend_init(struct budget_ci *budget_ci)
 {
 switch (budget_ci->budget.dev->pci->subsystem_device) {

@@ -1404,7 +1416,20 @@
 budget_ci->budget.dvb_frontend = NULL;
 }
 }
-
+ break;
+ case 0x101a: /* TT Budget-C-1501 (philips tda10023/philips tda8274A) */
+ budget_ci->budget.dvb_frontend =
+ dvb_attach(tda10023_attach, &tda10023_config,
+    &budget_ci->budget.i2c_adap, 0x48);
+ if (budget_ci->budget.dvb_frontend) {
+ if (dvb_attach(tda827x_attach,
+        budget_ci->budget.dvb_frontend, 0x61,
+        &budget_ci->budget.i2c_adap, NULL)
+     == NULL)
+ printk(KERN_ERR "%s: No tda827x found!\n",
+        __func__);
+ break;
+ }
 break;
 
 case 0x1019: // TT S2-3200 PCI

@@ -1535,6 +1560,7 @@
 MAKE_BUDGET_INFO(ttbtci, "TT-Budget-T-CI PCI", BUDGET_TT);
 MAKE_BUDGET_INFO(ttbcci, "TT-Budget-C-CI PCI", BUDGET_TT);
 MAKE_BUDGET_INFO(tt3200, "TT-Budget S2-3200 PCI", BUDGET_TT);
+MAKE_BUDGET_INFO(ttc1501, "TT-Budget C-1501 PCI", BUDGET_TT);
 
 static struct pci_device_id pci_tbl[] = {
 MAKE_EXTENSION_PCI(ttbci, 0x13c2, 0x100c),

_______________________________________________
linux-dvb mailing list
[email protected]
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb

Hi

What am I doing wrong - sorry for the potentially idiot error...

I got what I thought was the latest, which would include these c-1501 patches yes??
hg clone http://linuxtv.org/hg/v4l-dvb

But there's no "linux/drivers/media/dvb/frontends/tda827x.c" and the
"linux/drivers/media/dvb/ttpci/budget-ci.c" is not the same as the patches
references.

[root@freddy v4l-dvb]# find . | grep tda827x.c ./.hg/store/data/linux/drivers/media/common/tuners/tda827x.c.i ./.hg/store/data/linux/drivers/media/dvb/frontends/tda827x.c.i ./linux/drivers/media/common/tuners/tda827x.c

Have I got the wrong repository??

Quoted 22 lines · Sigmund Augdal

ons, 04.06.2008 kl. 16.58 +0300, skrev Antti Palosaari:

Quoted 10 lines

Quoted 4 lines · Sigmund Augdal

The following experimental patch adds support for the technotrend budget C-1501 dvb-c card. The parameters used to configure the tda10023 demod chip are largely determined experimentally, but works quite for me in my initial tests.

You finally found correct values :) I see that it uses same clock settings than Anysee. Also deltaf could be correct because I remember from my tests that it cannot set wrong otherwise it will not work at all. How did you find defltaf?

I guessed the clock settings based on how the same tuner is used by a different demod. The deltaf value was found by trial and error (helped by some scripting). deltaf values slightly off will also work, but tuning will be very slow. I also think the deltaf value will depend on the bandwidth of the transponder tuned. All transponders I've tested with are 8MHz, but I think other values are possible, and these will most likely not work. I submitted the patch anyway in the hope that some broader testing might help uncover any remaining problems.

Quoted 3 lines

Your patch has at least coding style issues, please run make checkpatch fix errors and resend patch.

Quoted 2 lines

I was trying to follow the guidelines, but I guess I wasn't doing that good enough. Will try to clean that up and resend soon.

Here is a new version. This one passes checkpatch without warnings. I removed the read_pwm function, as it always uses the fallback path for my card (and frankly I have no idea wether it is actually relevant at all for this kind of card). Furthermore the tda10023 driver doesn't seem to use this value for anything.