linux-kernel-mentees.lists.linuxfoundation.org archive mirror
 help / color / mirror / Atom feed
From: hverkuil at xs4all.nl (Hans Verkuil)
Subject: [Linux-kernel-mentees] [PATCH v2 1/2] cec-follower: add tuner step increment/decrement
Date: Wed, 25 Sep 2019 09:12:26 +0200	[thread overview]
Message-ID: <c56379c6-51e5-0199-1bfc-b1be359ca9a1@xs4all.nl> (raw)
In-Reply-To: <fb849c63-6bac-5d40-f78d-f0ac57250a95@xs4all.nl>

On 9/25/19 8:54 AM, Hans Verkuil wrote:
> Hi Jiunn,
> 
> On 9/24/19 9:24 PM, Jiunn Chang wrote:
>> Tuner step increment/decrement will select the next highest or next
>> lowest service frequency.  There are a total of three possible
>> frequencies from analog_freqs_khz given a broadcast type and system.
>>
>> Opcodes implemented:
>>   - <Tuner Step Increment>
>>   - <Tuner Step Decrement>
>>
>> Signed-off-by: Jiunn Chang <c0d1n61at3 at gmail.com>
>> ---
>>  utils/cec-follower/cec-tuner.cpp | 47 ++++++++++++++++++++++++++++++--
>>  1 file changed, 44 insertions(+), 3 deletions(-)
>>
>> diff --git a/utils/cec-follower/cec-tuner.cpp b/utils/cec-follower/cec-tuner.cpp
>> index acc3fd00..1a9b9d90 100644
>> --- a/utils/cec-follower/cec-tuner.cpp
>> +++ b/utils/cec-follower/cec-tuner.cpp
>> @@ -135,6 +135,22 @@ static bool analog_set_tuner_dev_info(struct node *node, struct cec_msg *msg)
>>  	return false;
>>  }
>>  
>> +static int analog_find_freq_index(struct cec_op_tuner_device_info *info)
>> +{
>> +	if (info->analog.ana_freq == 0)
>> +		return -1;
>> +
>> +	int ana_freq_khz = (info->analog.ana_freq * 625) / 10;
>> +
>> +	for (int i = 0; i < NUM_ANALOG_FREQS; i++) {
>> +		int freq = analog_freqs_khz[info->analog.ana_bcast_type][info->analog.bcast_system][i];
>> +
>> +		if (ana_freq_khz == freq)
>> +			return i;
>> +	}
>> +	return -1;
>> +}
> 
> Hmm, I don't like this code. I think that it would be cleaner to just add:
> 
> static unsigned int analog_freq_idx[3][9];
> 
> Which contains the index of the current frequency for each bcast_type/system
> combination.
> 
> This ensures that 1) there always is a current frequency and 2) tuner stepping
> is now trivial since all you need to do is to update the corresponding index.

I've been reading up on this a bit more, and section CEC 13.8.2 states that the
Tuner Step commands cycle through a set of services defined by the tuner device.
But normally it will cycle through all services.

That means that rather than cycling through the frequencies of just the current
bcast_type/system, the follower should cycle through ALL channels, both analog
and digital.

So for analog we really have 3 * 9 * 3 channels, and this will be more once the
digital channels are added.

In any case, this means that there is just a single 'static unsigned int freq_idx;'
and not an array.

Regards,

	Hans

> 
> Now, that said, I was thinking that it would be nice to have one bcast_type/system
> combo with no channels at all, since that's a nice test as well.
> 
> I think that setting secam-lq for satellite to { 0, 0, 0 } to indicate no
> defined channels would make a nice test case.
> 
> Something for a follow-up patch, though.
> 
>> +
>>  void process_tuner_record_timer_msgs(struct node *node, struct cec_msg &msg, unsigned me)
>>  {
>>  	bool is_bcast = cec_msg_is_broadcast(&msg);
>> @@ -178,12 +194,37 @@ void process_tuner_record_timer_msgs(struct node *node, struct cec_msg &msg, uns
>>  		return;
>>  
>>  	case CEC_MSG_SELECT_DIGITAL_SERVICE:
>> -	case CEC_MSG_TUNER_STEP_DECREMENT:
>> -	case CEC_MSG_TUNER_STEP_INCREMENT:
>> -		if (!cec_has_tuner(1 << me))
>> +	case CEC_MSG_TUNER_STEP_DECREMENT: {
>> +		if (!cec_has_tuner(1 << me) && !cec_has_tv(1 << me))
>>  			break;
>> +
>> +		struct cec_op_tuner_device_info *info = &node->state.tuner_dev_info;
>> +		int freq_idx = analog_find_freq_index(info);
>> +		int idx = (freq_idx == 0) ? NUM_ANALOG_FREQS : freq_idx;
>> +
>> +		if (info->is_analog) {
>> +			int freq = analog_freqs_khz[info->analog.ana_bcast_type][info->analog.bcast_system][--idx];
>> +
>> +			info->analog.ana_freq = (freq * 10) / 625;
>> +		}
>>  		return;
>> +	}
>> +
>> +	case CEC_MSG_TUNER_STEP_INCREMENT: {
>> +		if (!cec_has_tuner(1 << me) && !cec_has_tv(1 << me))
>> +			break;
>>  
>> +		struct cec_op_tuner_device_info *info = &node->state.tuner_dev_info;
>> +		int freq_idx = analog_find_freq_index(info);
>> +		int idx = (freq_idx == NUM_ANALOG_FREQS - 1) ? -1 : freq_idx;
>> +
>> +		if (info->is_analog) {
>> +			int freq = analog_freqs_khz[info->analog.ana_bcast_type][info->analog.bcast_system][++idx];
>> +
>> +			info->analog.ana_freq = (freq * 10) / 625;
>> +		}
>> +		return;
>> +	}
>>  
>>  		/*
>>  		  One Touch Record
>>
> 
> Regards,
> 
> 	Hans
> 

WARNING: multiple messages have this Message-ID (diff)
From: hverkuil@xs4all.nl (Hans Verkuil)
Subject: [Linux-kernel-mentees] [PATCH v2 1/2] cec-follower: add tuner step increment/decrement
Date: Wed, 25 Sep 2019 09:12:26 +0200	[thread overview]
Message-ID: <c56379c6-51e5-0199-1bfc-b1be359ca9a1@xs4all.nl> (raw)
Message-ID: <20190925071226.8ygYgaln9mSXno4N7FMel0pqLWAqDubsLKI2SBJGaic@z> (raw)
In-Reply-To: <fb849c63-6bac-5d40-f78d-f0ac57250a95@xs4all.nl>

On 9/25/19 8:54 AM, Hans Verkuil wrote:
> Hi Jiunn,
> 
> On 9/24/19 9:24 PM, Jiunn Chang wrote:
>> Tuner step increment/decrement will select the next highest or next
>> lowest service frequency.  There are a total of three possible
>> frequencies from analog_freqs_khz given a broadcast type and system.
>>
>> Opcodes implemented:
>>   - <Tuner Step Increment>
>>   - <Tuner Step Decrement>
>>
>> Signed-off-by: Jiunn Chang <c0d1n61at3 at gmail.com>
>> ---
>>  utils/cec-follower/cec-tuner.cpp | 47 ++++++++++++++++++++++++++++++--
>>  1 file changed, 44 insertions(+), 3 deletions(-)
>>
>> diff --git a/utils/cec-follower/cec-tuner.cpp b/utils/cec-follower/cec-tuner.cpp
>> index acc3fd00..1a9b9d90 100644
>> --- a/utils/cec-follower/cec-tuner.cpp
>> +++ b/utils/cec-follower/cec-tuner.cpp
>> @@ -135,6 +135,22 @@ static bool analog_set_tuner_dev_info(struct node *node, struct cec_msg *msg)
>>  	return false;
>>  }
>>  
>> +static int analog_find_freq_index(struct cec_op_tuner_device_info *info)
>> +{
>> +	if (info->analog.ana_freq == 0)
>> +		return -1;
>> +
>> +	int ana_freq_khz = (info->analog.ana_freq * 625) / 10;
>> +
>> +	for (int i = 0; i < NUM_ANALOG_FREQS; i++) {
>> +		int freq = analog_freqs_khz[info->analog.ana_bcast_type][info->analog.bcast_system][i];
>> +
>> +		if (ana_freq_khz == freq)
>> +			return i;
>> +	}
>> +	return -1;
>> +}
> 
> Hmm, I don't like this code. I think that it would be cleaner to just add:
> 
> static unsigned int analog_freq_idx[3][9];
> 
> Which contains the index of the current frequency for each bcast_type/system
> combination.
> 
> This ensures that 1) there always is a current frequency and 2) tuner stepping
> is now trivial since all you need to do is to update the corresponding index.

I've been reading up on this a bit more, and section CEC 13.8.2 states that the
Tuner Step commands cycle through a set of services defined by the tuner device.
But normally it will cycle through all services.

That means that rather than cycling through the frequencies of just the current
bcast_type/system, the follower should cycle through ALL channels, both analog
and digital.

So for analog we really have 3 * 9 * 3 channels, and this will be more once the
digital channels are added.

In any case, this means that there is just a single 'static unsigned int freq_idx;'
and not an array.

Regards,

	Hans

> 
> Now, that said, I was thinking that it would be nice to have one bcast_type/system
> combo with no channels at all, since that's a nice test as well.
> 
> I think that setting secam-lq for satellite to { 0, 0, 0 } to indicate no
> defined channels would make a nice test case.
> 
> Something for a follow-up patch, though.
> 
>> +
>>  void process_tuner_record_timer_msgs(struct node *node, struct cec_msg &msg, unsigned me)
>>  {
>>  	bool is_bcast = cec_msg_is_broadcast(&msg);
>> @@ -178,12 +194,37 @@ void process_tuner_record_timer_msgs(struct node *node, struct cec_msg &msg, uns
>>  		return;
>>  
>>  	case CEC_MSG_SELECT_DIGITAL_SERVICE:
>> -	case CEC_MSG_TUNER_STEP_DECREMENT:
>> -	case CEC_MSG_TUNER_STEP_INCREMENT:
>> -		if (!cec_has_tuner(1 << me))
>> +	case CEC_MSG_TUNER_STEP_DECREMENT: {
>> +		if (!cec_has_tuner(1 << me) && !cec_has_tv(1 << me))
>>  			break;
>> +
>> +		struct cec_op_tuner_device_info *info = &node->state.tuner_dev_info;
>> +		int freq_idx = analog_find_freq_index(info);
>> +		int idx = (freq_idx == 0) ? NUM_ANALOG_FREQS : freq_idx;
>> +
>> +		if (info->is_analog) {
>> +			int freq = analog_freqs_khz[info->analog.ana_bcast_type][info->analog.bcast_system][--idx];
>> +
>> +			info->analog.ana_freq = (freq * 10) / 625;
>> +		}
>>  		return;
>> +	}
>> +
>> +	case CEC_MSG_TUNER_STEP_INCREMENT: {
>> +		if (!cec_has_tuner(1 << me) && !cec_has_tv(1 << me))
>> +			break;
>>  
>> +		struct cec_op_tuner_device_info *info = &node->state.tuner_dev_info;
>> +		int freq_idx = analog_find_freq_index(info);
>> +		int idx = (freq_idx == NUM_ANALOG_FREQS - 1) ? -1 : freq_idx;
>> +
>> +		if (info->is_analog) {
>> +			int freq = analog_freqs_khz[info->analog.ana_bcast_type][info->analog.bcast_system][++idx];
>> +
>> +			info->analog.ana_freq = (freq * 10) / 625;
>> +		}
>> +		return;
>> +	}
>>  
>>  		/*
>>  		  One Touch Record
>>
> 
> Regards,
> 
> 	Hans
> 

  parent reply	other threads:[~2019-09-25  7:12 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-09-24 14:17 [Linux-kernel-mentees] [PATCH 0/2] cec-compliance: tuner control c0d1n61at3
2019-09-24 14:17 ` Jiunn Chang
2019-09-24 14:17 ` [Linux-kernel-mentees] [PATCH 1/2] cec-follower: add tuner step increment/decrement c0d1n61at3
2019-09-24 14:17   ` Jiunn Chang
2019-09-24 15:20   ` hverkuil
2019-09-24 15:20     ` Hans Verkuil
2019-09-24 14:17 ` [Linux-kernel-mentees] [PATCH 2/2] cec-compliance: add/refactor tuner control tests c0d1n61at3
2019-09-24 14:17   ` Jiunn Chang
2019-09-24 19:24 ` [Linux-kernel-mentees] [PATCH v2 0/2] cec-compliance: tuner control c0d1n61at3
2019-09-24 19:24   ` Jiunn Chang
2019-09-30  4:30   ` [Linux-kernel-mentees] [PATCH v3 0/3] " c0d1n61at3
2019-09-30  4:30     ` Jiunn Chang
2019-10-01  3:18     ` [Linux-kernel-mentees] [PATCH v4 " c0d1n61at3
2019-10-01  3:18       ` Jiunn Chang
2019-10-03  3:18       ` [Linux-kernel-mentees] [PATCH v5 0/2] " c0d1n61at3
2019-10-03  3:18         ` Jiunn Chang
2019-10-04  4:05         ` [Linux-kernel-mentees] [PATCH v6 0/1] " c0d1n61at3
2019-10-04  4:05           ` Jiunn Chang
2019-10-04  4:05         ` [Linux-kernel-mentees] [PATCH v6 1/1] Add test for new features in cec-follower c0d1n61at3
2019-10-04  4:05           ` Jiunn Chang
2019-10-03  3:18       ` [Linux-kernel-mentees] [PATCH v5 1/2] cec-follower: add tuner step increment/decrement c0d1n61at3
2019-10-03  3:18         ` Jiunn Chang
2019-10-03  3:18       ` [Linux-kernel-mentees] [PATCH v5 2/2] Add test for new features in cec-follower c0d1n61at3
2019-10-03  3:18         ` Jiunn Chang
2019-10-03  6:51         ` hverkuil
2019-10-03  6:51           ` Hans Verkuil
2019-10-01  3:18     ` [Linux-kernel-mentees] [PATCH v4 1/3] cec-follower: fix bugs for tuner emulation c0d1n61at3
2019-10-01  3:18       ` Jiunn Chang
2019-10-01  3:18     ` [Linux-kernel-mentees] [PATCH v4 2/3] cec-follower: add tuner step increment/decrement c0d1n61at3
2019-10-01  3:18       ` Jiunn Chang
2019-10-01  7:12       ` hverkuil
2019-10-01  7:12         ` Hans Verkuil
2019-10-01  3:18     ` [Linux-kernel-mentees] [PATCH v4 3/3] cec-compliance: add tuner control test c0d1n61at3
2019-10-01  3:18       ` Jiunn Chang
2019-10-01  7:51       ` hverkuil
2019-10-01  7:51         ` Hans Verkuil
2019-09-30  4:30   ` [Linux-kernel-mentees] [PATCH v3 1/3] cec-follower: fix bugs for tuner emulation c0d1n61at3
2019-09-30  4:30     ` Jiunn Chang
2019-09-30  9:12     ` hverkuil
2019-09-30  9:12       ` Hans Verkuil
2019-09-30  4:30   ` [Linux-kernel-mentees] [PATCH v3 2/3] cec-follower: add tuner step increment/decrement c0d1n61at3
2019-09-30  4:30     ` Jiunn Chang
2019-09-30  9:35     ` hverkuil
2019-09-30  9:35       ` Hans Verkuil
2019-09-30  4:30   ` [Linux-kernel-mentees] [PATCH v3 3/3] cec-compliance: refactor tuner control tests c0d1n61at3
2019-09-30  4:30     ` Jiunn Chang
2019-09-30  9:43     ` hverkuil
2019-09-30  9:43       ` Hans Verkuil
2019-09-24 19:24 ` [Linux-kernel-mentees] [PATCH v2 1/2] cec-follower: add tuner step increment/decrement c0d1n61at3
2019-09-24 19:24   ` Jiunn Chang
2019-09-24 19:31   ` skhan
2019-09-24 19:31     ` Shuah Khan
2019-09-25  6:54   ` hverkuil
2019-09-25  6:54     ` Hans Verkuil
2019-09-25  7:12     ` hverkuil [this message]
2019-09-25  7:12       ` Hans Verkuil
2019-09-24 19:24 ` [Linux-kernel-mentees] [PATCH v2 2/2] cec-compliance: add/refactor tuner control tests c0d1n61at3
2019-09-24 19:24   ` Jiunn Chang
2019-09-24 19:44   ` skhan
2019-09-24 19:44     ` Shuah Khan
2019-09-25  7:22   ` hverkuil
2019-09-25  7:22     ` Hans Verkuil

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=c56379c6-51e5-0199-1bfc-b1be359ca9a1@xs4all.nl \
    --to=linux-kernel-mentees@lists.linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).