[Home]

Summary:ASTERISK-02368: Call progress detection for Costa Rica is broken
Reporter:hades (hades)Labels:
Date Opened:2004-09-08 09:14:55Date Closed:2011-06-07 14:10:20
Priority:MinorRegression?No
Status:Closed/CompleteComponents:Core/General
Versions:Frequency of
Occurrence
Related
Issues:
Environment:Attachments:( 0) cp_dsp_patch.diff
Description:The call progress detection algorithm is initializing hz[HZ_425] as 0 (zero) always, thus making the progress detection to fail.

****** ADDITIONAL INFORMATION ******

I just fooled Asterisk and replaced the code:

static struct progress {
 int size;
 int freqs[7];
} modes[] = {
 { GSAMP_SIZE_NA, { 425, 440, 480, 620, 950, 1400, 1800 }
                    ^^^
}, /* North America */
 { GSAMP_SIZE_CR, { 425 } },
};

...

     case PROG_MODE_NA:
       if (hz[HZ_425] > TONE_MIN_THRESH * TONE_THRESH) {
         newstate = TONE_STATE_RINGING;
       } else if (dsp->genergy > TONE_MIN_THRESH * TONE_THRESH) {
         newstate = TONE_STATE_TALKING;
       } else
         newstate = TONE_STATE_SILENCE;
       break;
Comments:By: Brian West (bkw918) 2004-09-08 09:38:48

This is not a MAJOR bug .. a major bug is one that stops the normal usage of the software.  This isn't something that works very well on Analog lines anyway.
By: hades (hades) 2004-09-08 15:09:09

Sorry about that. Yes, it doesn't work very well on analog lines. I'm using it on an E100P E1 card and it's working just fine.
By: Mark Spencer (markster) 2004-09-08 21:05:51

Easiest if you find me on IRC so I can debug it on your system.
By: alric (alric) 2004-09-25 08:22:34

Reminder sent to hades

Have you made any attempt to contact markster (kram) on IRC?  What is the status of this bug report?
By: Mark Spencer (markster) 2004-10-04 10:45:06

User seems to have lost interest in this bug.
By: hades (hades) 2004-10-04 13:26:50

Sorry, I was out for Astricon and then I got back and had a huge pile of problems to solve and I forgot about the bug. I'll post the ugly "patch" that I used to solve my problem in my case. This week I'll have more time to work on the problem.

edited on: 10-04-04 13:51
By: hades (hades) 2004-10-04 13:51:03

Note: I have to use progzone=us in zapata.conf for that to work. It seems that at some point in dsp.c things are not initialized as they should when progzone=cr. Maybe the hz[] initialization is the problem (the array for PROG_MODE_CR has only one element, but 7 are actually initialized each time). As I mentioned before, hz[HZ_425] was always zero'ed.
By: twisted (twisted) 2004-10-27 16:34:55

/me gently taps kram on the shoulder

what's goin on here?  Still something that is being worked on?
By: twisted (twisted) 2004-11-14 21:17:10.000-0600

While i feel this may still be a bug, there is a definate lack of response.  I'm going to suspend this one, in hopes of someone waking it back up.