ASTERISK-07076: [patch] fxotune either doesn't recognize or doesn't display -l or -m options, causes subsequent tuning failure

[Home]

Summary: ASTERISK-07076: [patch] fxotune either doesn't recognize or doesn't display -l or -m options, causes subsequent tuning failure

Reporter: Barry King (bking) Labels:

Date Opened: 2006-06-01 11:04:36 Date Closed: 2006-07-05 16:18:42

Priority: Major Regression? No

Status: Closed/Complete Components: Core/General

Versions: Frequency of
Occurrence

Related
Issues:

Environment: Attachments: ( 0) fxotune_b.patch

Description: # /usr/src/zaptel/fxotune -l 10 -m 30 -i 99629298 -e 1 -vvvvv
Running with parameters:
doset=0
docalibrate=1
dodump=0
startdev=1
stopdev=1
calibtype=2
waveformtype=2
delaytosilence=2
silencegoodfor=2
dialstr=99629298
debug=5

notice that delaytosilence and silencegoodfor aren't properly set (or displayed?).

I'm using today's SVN.

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

Seems like this is contributing to the below problem, as there is definately no 8 second (or seemingly even a 2 second) pause before the numbers start rolling in.

Tuning module /dev/zap/1
0,0,0,0,0,0,0,0,0,-1,11145.606542,2.358272
10,0,6,1,254,2,255,0,0,-1,2575.600261,0.544965
3,255,255,0,1,0,0,0,0,-1,3164.861937,0.669646
3,1,253,253,2,255,0,0,0,-1,3105.572568,0.657101
9,254,251,255,2,0,1,0,0,-1,3306.597867,0.699635
5,3,251,250,2,254,0,0,255,-1,3270.623773,0.692024
8,253,2,244,255,10,244,3,253,-1,2823.219661,0.597358
10,249,244,8,12,245,252,0,1,-1,3059.400366,0.647331
1,0,0,0,0,0,0,0,0,-1,3341.875194,0.707099
10,252,255,1,255,0,0,0,0,-1,3118.030150,0.659737
7,255,251,251,2,255,255,1,255,-1,3143.439581,0.665113
3,1,251,250,1,254,255,0,255,-1,3070.118443,0.649599
5,252,250,0,0,255,1,0,0,-1,3333.203994,0.705265
5,3,251,250,1,253,0,0,255,-1,3346.334114,0.708043
8,253,2,244,255,10,244,3,253,-1,2891.078823,0.611717
10,249,244,8,12,245,252,0,1,-1,3103.133730,0.656585
2,0,0,0,0,0,0,0,0,-1,3072.097837,0.650018
7,0,0,255,254,0,0,0,0,-1,3159.940594,0.668604
9,0,253,254,2,255,0,0,0,-1,3288.209099,0.695744
5,1,249,254,4,253,1,0,0,-1,3262.483956,0.690301
5,252,250,1,1,254,0,255,0,-1,3264.890036,0.690810
5,3,251,250,2,253,255,255,255,-1,3295.255320,0.697235
8,253,2,244,255,10,244,3,253,-1,2863.640659,0.605911
10,249,244,8,12,245,252,0,1,-1,3108.483296,0.657717
3,0,0,0,0,0,0,0,0,-1,3172.882916,0.671343
7,0,255,254,255,0,255,0,0,-1,3187.736511,0.674486
9,0,253,253,1,255,0,0,0,-1,3338.992730,0.706490
5,1,249,254,3,253,1,0,0,-1,3319.566329,0.702379
5,252,250,1,1,254,0,255,0,-1,3310.293075,0.700417
5,3,251,251,2,253,255,255,255,-1,3314.319906,0.701269
8,253,2,244,255,10,244,3,253,-1,2854.076507,0.603887
10,249,244,8,12,245,252,0,1,-1,3076.868997,0.651027
4,0,0,0,0,0,0,0,0,-1,3011.706852,0.637240
9,255,1,4,0,0,1,255,0,-1,3369.921924,0.713034
9,0,253,0,3,254,0,0,255,-1,3313.629018,0.701123
9,2,250,253,5,253,1,0,255,-1,3310.327917,0.700424
5,252,250,1,2,255,0,255,0,-1,3306.701319,0.699657
5,3,251,250,3,254,255,255,255,-1,3334.169410,0.705469
8,253,2,244,255,10,244,3,253,-1,2893.748941,0.612281
10,249,244,8,12,245,252,0,1,-1,3104.525486,0.656879
5,0,0,0,0,0,0,0,0,-1,3346.589231,0.708097
Could not fill input buffer - got -1 bytes, expected 4000 bytes
Failure!
Unable to tune 1 devices, even though those devices are present

Comments: By: Serge Vecher (serge-v) 2006-06-01 11:27:08

trumpetinc: since you've done GOOD work on fxotune in 7136, could you please comment on this one?
By: Kevin Day (trumpetinc) 2006-06-05 22:03:30

Just uploaded fxotune.patch

The problem described in this bug is caused by an int overflow. The result is that fxotune was not computing elapsed time properly, which means it was not resetting the line when it should have been.

Attached patch properly handles the elpased computation.

FYI - I did not recieve notification on this bug (I was in the process of posting the patch when I noticed I had this bug on my list as well)... Sorry about the delayed response :-)

- K

By: Serge Vecher (serge-v) 2006-06-06 09:07:54

trumpetinc: thank you :)
bking: please test the patch and report your findings. Thanks
By: Kevin Day (trumpetinc) 2006-06-06 12:25:24

Just uploaded fxotune_b.patch - no functional changes, just adjusted comments to use /*... */ instead of //

Darned coding conventions :-)
By: Barry King (bking) 2006-06-06 14:11:19

Output is a lot more verbose now, tests complete without failures, and echo coefficients are successfully set.

Thanks, trumpetinc!
By: Hadley Rich (hads) 2006-06-15 17:35:32

I was having the same issues and this patch works well for me. Cheers.
By: Matt O'Gorman (mogorman) 2006-07-05 16:18:41

Committed revision 1203.