DiBxxxx: fixes and improvements

GIT PULL Patrick Boettcher 4 participants 5 messages Apr 22, 2013 – Jun 2, 2013 Media (V4L/DVB)

lore

Share & tools

lore

Patrick Boettcher

author

2013-04-22 08:12 UTC #0 lore raw

Hi Mauro,

These patches contains some fixes and changes for the DiBcom demods and SIPs.

Please merge for 3.10 if possible.

The following changes since commit 60d509fa6a9c4653a86ad830e4c4b30360b23f0e:

Linux 3.9-rc8 (2013-04-21 14:38:45 -0700)

are available in the git repository at:

git://git.linuxtv.org/pb/media_tree.git/ master

for you to fetch changes up to 7e39d1958b186e5af259b9fde1a006853b4663ab:

[media] dib8000: do not freeze AGCs by default (2013-04-22 10:06:48 +0200)

----------------------------------------------------------------
Olivier Grenie (6):
[media] dib8000: enhancement
[media] dib7000p: enhancement
[media] dib0090: enhancement
[media] dib8096: enhancement
[media] dib7090p: remove the support for the dib7090E [media] dib7090p: improve the support of the dib7090 and dib7790

Patrick Boettcher (1):
[media] dib8000: do not freeze AGCs by default

 drivers/media/dvb-core/dvb-usb-ids.h         |    3 +-
 drivers/media/dvb-frontends/dib0090.c        |  438 ++---
 drivers/media/dvb-frontends/dib7000p.c       |   17 +-
 drivers/media/dvb-frontends/dib7000p.h       |    7 +
 drivers/media/dvb-frontends/dib8000.c        | 2239 +++++++++++++---------
 drivers/media/dvb-frontends/dib8000.h        |    6 +-
 drivers/media/dvb-frontends/dibx000_common.h |    3 +-
 drivers/media/usb/dvb-usb/dib0700_devices.c  |  465 +++--
 8 files changed, 1780 insertions(+), 1398 deletions(-)

regards,
--
Patrick

Mauro Carvalho Chehab

2013-04-27 14:28 UTC #1 lore raw

Hi Patrick,

Em Mon, 22 Apr 2013 10:12:34 +0200
Patrick Boettcher <[email protected]> escreveu:

Quoted 8 lines

Hi Mauro,
These patches contains some fixes and changes for the DiBcom demods and SIPs.
Please merge for 3.10 if possible.

Quoted 7 lines

The following changes since commit 60d509fa6a9c4653a86ad830e4c4b30360b23f0e:
Linux 3.9-rc8 (2013-04-21 14:38:45 -0700)
are available in the git repository at:
git://git.linuxtv.org/pb/media_tree.git/ master

Hmm... I suspect that there's something wrong with those changes.

Testing it with a dib8076 usb stick seems that the code is worse than before, as it is now harder to get a lock here.

With the previous code:

INFO Scanning frequency #1 725142857
Carrier(0x03) Signal= 67.46% C/N= 0.00% UCB= 0 postBER= 0 Viterbi(0x05) Signal= 67.08% C/N= 0.00% UCB= 0 postBER= 2097151 Viterbi(0x07) Signal= 67.54% C/N= 0.25% UCB= 165 postBER= 0
Sync (0x0f) Signal= 67.06% C/N= 0.23% UCB= 151 postBER= 0
Lock (0x1f) Signal= 67.58% C/N= 0.24% UCB= 160 postBER= 338688
Service #0 (60320) BAND HD channel 57.1.0
Service #1 (60345) BAND 1SEG channel 57.1.1

With the new code:

INFO Scanning frequency #1 725142857

       (0x00) Signal= 68.80% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.78% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.69% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.82% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.29% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.27% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.27% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.55% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.50% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.65% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.75% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.29% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.25% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.46% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.90% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.50% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.41% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.41% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.96% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.42% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.24% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.25% C/N= 0.00% UCB= 0 postBER= 0

So, it seems that the changes broke something.

Regards,
Mauro

Olivier GRENIE

2013-04-28 20:28 UTC #2 lore raw

Hello Mauro,
I will have a look tomorrow and gives a patch. Sorry for this mistake.

regards,
Olivier
________________________________________
From: Mauro Carvalho Chehab [[email protected]]
Sent: Saturday, April 27, 2013 4:28 PM
To: Patrick Boettcher
Cc: [email protected]; Olivier GRENIE; Patrick BOETTCHER
Subject: Re: [GIT PULL FOR 3.10] DiBxxxx: fixes and improvements

Hi Patrick,

Em Mon, 22 Apr 2013 10:12:34 +0200
Patrick Boettcher <[email protected]> escreveu:

Quoted 8 lines

Hi Mauro,
These patches contains some fixes and changes for the DiBcom demods and SIPs.
Please merge for 3.10 if possible.

Quoted 7 lines

The following changes since commit 60d509fa6a9c4653a86ad830e4c4b30360b23f0e:
Linux 3.9-rc8 (2013-04-21 14:38:45 -0700)
are available in the git repository at:
git://git.linuxtv.org/pb/media_tree.git/ master

Hmm... I suspect that there's something wrong with those changes.

Testing it with a dib8076 usb stick seems that the code is worse than before, as it is now harder to get a lock here.

With the previous code:

With the new code:

INFO Scanning frequency #1 725142857

       (0x00) Signal= 68.80% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.78% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.69% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.82% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.29% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.27% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.27% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.55% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.50% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.65% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.75% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.29% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.25% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.46% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.90% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.50% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.41% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.41% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.96% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.42% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.24% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.25% C/N= 0.00% UCB= 0 postBER= 0

So, it seems that the changes broke something.

Regards,
Mauro

Olivier GRENIE

2013-05-04 04:08 UTC #3 lore raw

In reply to Mauro Carvalho Chehab

Hello Mauro,
can you apply the attached patch. This patch correct the proposed patch by Patrick for the dib807x. Sorry to not have seen it before.

regards,
Olivier

________________________________________
From: Mauro Carvalho Chehab [[email protected]]
Sent: Saturday, April 27, 2013 4:28 PM
To: Patrick Boettcher
Cc: [email protected]; Olivier GRENIE; Patrick BOETTCHER
Subject: Re: [GIT PULL FOR 3.10] DiBxxxx: fixes and improvements

Hi Patrick,

Em Mon, 22 Apr 2013 10:12:34 +0200
Patrick Boettcher <[email protected]> escreveu:

Quoted 8 lines

Hi Mauro,
These patches contains some fixes and changes for the DiBcom demods and SIPs.
Please merge for 3.10 if possible.

Quoted 7 lines

The following changes since commit 60d509fa6a9c4653a86ad830e4c4b30360b23f0e:
Linux 3.9-rc8 (2013-04-21 14:38:45 -0700)
are available in the git repository at:
git://git.linuxtv.org/pb/media_tree.git/ master

Hmm... I suspect that there's something wrong with those changes.

Testing it with a dib8076 usb stick seems that the code is worse than before, as it is now harder to get a lock here.

With the previous code:

With the new code:

INFO Scanning frequency #1 725142857

       (0x00) Signal= 68.80% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.78% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.69% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.82% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.29% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.27% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.27% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.55% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.50% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.65% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.75% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.29% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.25% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.46% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.90% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.50% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.41% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.41% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.96% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.42% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.24% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.25% C/N= 0.00% UCB= 0 postBER= 0

So, it seems that the changes broke something.

Regards,
Mauro

From 6d5679558aa99d9763fb2d638de771a8f52c94fa Mon Sep 17 00:00:00 2001
From: Olivier Grenie <[email protected]>
Date: Thu, 2 May 2013 16:11:14 +0200
Subject: [PATCH] [media] dib8000: correct previous commit The intend of this
patch is to correct a previous commit. This commit corrects
the behavior of the dib807x.

Signed-off-by Olivier Grenie <[email protected]>

---
 drivers/media/dvb-frontends/dib8000.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

drivers/media/dvb-frontends/dib8000.c +3 -3

index 4ba7c90..a57926b 100644
@@ -2439,7 +2439,7 @@ static int dib8000_autosearch_start(struct dvb_frontend *fe)
 	if (state->revision == 0x8090)
 		internal = dib8000_read32(state, 23) / 1000;
 
-	if (state->autosearch_state == AS_SEARCHING_FFT) {
+	if ((state->revision >= 0x8002) && (state->autosearch_state == AS_SEARCHING_FFT)) {
 		dib8000_write_word(state,  37, 0x0065); /* P_ctrl_pha_off_max default values */
 		dib8000_write_word(state, 116, 0x0000); /* P_ana_gain to 0 */

@@ -2475,7 +2475,7 @@ static int dib8000_autosearch_start(struct dvb_frontend *fe)
 		dib8000_write_word(state, 770, (dib8000_read_word(state, 770) & 0xdfff) | (1 << 13)); /* P_restart_ccg = 1 */
 		dib8000_write_word(state, 770, (dib8000_read_word(state, 770) & 0xdfff) | (0 << 13)); /* P_restart_ccg = 0 */
 		dib8000_write_word(state, 0, (dib8000_read_word(state, 0) & 0x7ff) | (0 << 15) | (1 << 13)); /* P_restart_search = 0; */
-	} else if (state->autosearch_state == AS_SEARCHING_GUARD) {
+	} else if ((state->revision >= 0x8002) && (state->autosearch_state == AS_SEARCHING_GUARD)) {
 		state->fe[0]->dtv_property_cache.transmission_mode = TRANSMISSION_MODE_8K;
 		state->fe[0]->dtv_property_cache.guard_interval = GUARD_INTERVAL_1_8;
 		state->fe[0]->dtv_property_cache.inversion = 0;

@@ -2577,7 +2577,7 @@ static int dib8000_autosearch_irq(struct dvb_frontend *fe)
 	struct dib8000_state *state = fe->demodulator_priv;
 	u16 irq_pending = dib8000_read_word(state, 1284);
 
-	if (state->autosearch_state == AS_SEARCHING_FFT) {
+	if ((state->revision >= 0x8002) && (state->autosearch_state == AS_SEARCHING_FFT)) {
 		if (irq_pending & 0x1) {
 			dprintk("dib8000_autosearch_irq: max correlation result available");
 			return 3;

Signature · 1 line

1.7.10.4

Michael Krufky

2013-06-02 17:03 UTC #4 lore raw

Olivier,

I have regenerated your patch against the most recent codebase, as your patch no longer applies. (There were changes merged on the 29th of April that broke it.)

The new patch is attached, but I am reluctant to merge it, as I do not have any description of the patch, and it lacks a sign-off.

Please test this patch and confirm that it produces the desired effect, then resubmit with a short description and your sign-off.

Best regards,

Mike Krufky

Quoted 86 lines · Olivier GRENIE

Hello Mauro,
can you apply the attached patch. This patch correct the proposed patch by Patrick for the dib807x. Sorry to not have seen it before.
regards,
Olivier
________________________________________
From: Mauro Carvalho Chehab [[email protected]]
Sent: Saturday, April 27, 2013 4:28 PM
To: Patrick Boettcher
Cc: [email protected]; Olivier GRENIE; Patrick BOETTCHER
Subject: Re: [GIT PULL FOR 3.10] DiBxxxx: fixes and improvements
Hi Patrick,
Em Mon, 22 Apr 2013 10:12:34 +0200
Patrick Boettcher <[email protected]> escreveu:
Quoted 8 lines
Hi Mauro,
These patches contains some fixes and changes for the DiBcom demods and SIPs.
Please merge for 3.10 if possible.
Quoted 7 lines
The following changes since commit 60d509fa6a9c4653a86ad830e4c4b30360b23f0e:
Linux 3.9-rc8 (2013-04-21 14:38:45 -0700)
are available in the git repository at:
git://git.linuxtv.org/pb/media_tree.git/ master
Hmm... I suspect that there's something wrong with those changes.
Testing it with a dib8076 usb stick seems that the code is worse than before, as it is now harder to get a lock here.
With the previous code:
INFO Scanning frequency #1 725142857
Carrier(0x03) Signal= 67.46% C/N= 0.00% UCB= 0 postBER= 0 Viterbi(0x05) Signal= 67.08% C/N= 0.00% UCB= 0 postBER= 2097151 Viterbi(0x07) Signal= 67.54% C/N= 0.25% UCB= 165 postBER= 0
Sync (0x0f) Signal= 67.06% C/N= 0.23% UCB= 151 postBER= 0
Lock (0x1f) Signal= 67.58% C/N= 0.24% UCB= 160 postBER= 338688
Service #0 (60320) BAND HD channel 57.1.0
Service #1 (60345) BAND 1SEG channel 57.1.1
With the new code:
INFO Scanning frequency #1 725142857
       (0x00) Signal= 68.80% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.78% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.69% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.82% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.29% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.27% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.27% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.55% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.50% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.65% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.75% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.29% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.25% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.46% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.90% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.50% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.28% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.43% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.41% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.41% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 68.96% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.42% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.24% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.22% C/N= 0.00% UCB= 0 postBER= 0
RF     (0x01) Signal= 69.25% C/N= 0.00% UCB= 0 postBER= 0

Quoted 4 lines

So, it seems that the changes broke something.
Regards,
Mauro

drivers/media/dvb-frontends/dib8000.c +6 -3

index 9053614..487c253 100644
@@ -2445,7 +2445,8 @@ static int dib8000_autosearch_start(struct dvb_frontend *fe)
 	if (state->revision == 0x8090)
 		internal = dib8000_read32(state, 23) / 1000;
 
-	if (state->autosearch_state == AS_SEARCHING_FFT) {
+	if ((state->revision >= 0x8002) &&
+	    (state->autosearch_state == AS_SEARCHING_FFT)) {
 		dib8000_write_word(state,  37, 0x0065); /* P_ctrl_pha_off_max default values */
 		dib8000_write_word(state, 116, 0x0000); /* P_ana_gain to 0 */

@@ -2481,7 +2482,8 @@ static int dib8000_autosearch_start(struct dvb_frontend *fe)
 		dib8000_write_word(state, 770, (dib8000_read_word(state, 770) & 0xdfff) | (1 << 13)); /* P_restart_ccg = 1 */
 		dib8000_write_word(state, 770, (dib8000_read_word(state, 770) & 0xdfff) | (0 << 13)); /* P_restart_ccg = 0 */
 		dib8000_write_word(state, 0, (dib8000_read_word(state, 0) & 0x7ff) | (0 << 15) | (1 << 13)); /* P_restart_search = 0; */
-	} else if (state->autosearch_state == AS_SEARCHING_GUARD) {
+	} else if ((state->revision >= 0x8002) &&
+		   (state->autosearch_state == AS_SEARCHING_GUARD)) {
 		c->transmission_mode = TRANSMISSION_MODE_8K;
 		c->guard_interval = GUARD_INTERVAL_1_8;
 		c->inversion = 0;

@@ -2583,7 +2585,8 @@ static int dib8000_autosearch_irq(struct dvb_frontend *fe)
 	struct dib8000_state *state = fe->demodulator_priv;
 	u16 irq_pending = dib8000_read_word(state, 1284);
 
-	if (state->autosearch_state == AS_SEARCHING_FFT) {
+	if ((state->revision >= 0x8002) &&
+	    (state->autosearch_state == AS_SEARCHING_FFT)) {
 		if (irq_pending & 0x1) {
 			dprintk("dib8000_autosearch_irq: max correlation result available");
 			return 3;