Re: [Astlinux-users] Digium TC400 transcoder card issues

2009-05-28 Thread Michael Keuter
>Interesting...
>
>TC400B - Voice Processing - Data Sheet
>http://docs.digium.com/TC400B/tc400b-datasheet.pdf
>
>It would be cool if of a lower than 120 G.729a Transcodes option was 
>available for a lower cost for embedded applications.
>
>Lonnie
>
>On May 27, 2009, at 10:00 AM, Darrick Hartman wrote:
>
>>  I've not used the transcoder hardware card.  Looks interesting though.
>>  Udev should handle creating the correct devices, but it's likely that
>>  Digium won't officially support the card unless it's on a system with
>>  Dahdi.  We're working towards that, but won't officially support Dahdi
>>  until after the 0.7 release.  It's a major interruption which will 
>>  cause
>>  changes to several parts of the system.
>>
>>  Please let us know what reply you get from Digium.
>>
>>  Regards,
>>
>  > Darrick
>>

In the specs from .pdf above only Zaptel > 1.4.1 is required for use 
with Asterisk 1.4.x.
So DAHDI is NOT required explicitly.

Michael

--
Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT 
is a gathering of tech-side developers & brand creativity professionals. Meet
the minds behind Google Creative Lab, Visual Complexity, Processing, & 
iPhoneDevCamp as they present alongside digital heavyweights like Barbarian 
Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com 
___
Astlinux-users mailing list
Astlinux-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/astlinux-users

Donations to support AstLinux are graciously accepted via PayPal to 
pay...@krisk.org.


Re: [Astlinux-users] Digium TC400 transcoder card issues

2009-05-28 Thread Michael Graves
On Wed, 27 May 2009 10:18:28 -0500, Lonnie Abelbeck wrote:

>Interesting...
>
>TC400B - Voice Processing - Data Sheet
>http://docs.digium.com/TC400B/tc400b-datasheet.pdf
>
>It would be cool if of a lower than 120 G.729a Transcodes option was  
>available for a lower cost for embedded applications.
>
>Lonnie

One wonders at what point you'd be better off just using a moer
powerful host platform over the TC400 card. If you don't need 120
channels then maybe the card is not appropriate. 

I've never actually reached the limit of transcode capability on a 1
GHz Transmeta CPU since I don't have that many G.729 licenses. If the
Net4801 could do two, then I'd guess that the T5700 can do six. Any
reasonable modern CPU even more. So for small embedded installations
maybe a faster host is enough.

Michael


>On May 27, 2009, at 10:00 AM, Darrick Hartman wrote:
>
>> I've not used the transcoder hardware card.  Looks interesting though.
>> Udev should handle creating the correct devices, but it's likely that
>> Digium won't officially support the card unless it's on a system with
>> Dahdi.  We're working towards that, but won't officially support Dahdi
>> until after the 0.7 release.  It's a major interruption which will  
>> cause
>> changes to several parts of the system.
>>
>> Please let us know what reply you get from Digium.
>>
>> Regards,
>>
>> Darrick
>>
>> On 05/27/2009 09:51 AM, Josh Mahonin wrote:
>>> Hi all,
>>>
>>> I've been using the Digium TC400 G729/G711 transcoder card in various
>>> Asterisk installations for a while with great results, but I'd like  
>>> to
>>> start using it on an embedded platform, such as the Soekris net5501.
>>>
>>> I grabbed AstLinux 0.6.5, and with some symlinking (/dev/ 
>>> zaptranscode =>
>>> /dev/zap/transcode) and firmware copying
>>> (/lib/firmware/zaptel-fw-tc400m.bin), got the 'wctc4xxp' transcoder
>>> module loaded with Asterisk recognizing it.  During a call however,  
>>> the
>>> transcoding thread doesn't seem to be able to read from the device,  
>>> and
>>> subsequently no audio is passed between the endpoints.  I've tried  
>>> with
>>> two known working cards, one of which is back in production.  An  
>>> strace
>>> is included the end of the message.
>>>
>>> I'm pretty sure this isn't strictly a problem with AstLinux, because
>>> I've installed Asterisk 1.4.24.1 with Zaptel 1.4.12.1 on a new  
>>> machine a
>>> stock Linux install and had exactly the same results.  Normally I'd  
>>> just
>>> change versions and be done with it, but I'd like to start using
>>> AstLinux for this project, and it doesn't look like upgrading  
>>> versions
>>> is a trivial task.  :)
>>>
>>> I've already put in a ticket with Digium support, but I'm emailing  
>>> the
>>> board because I'm hoping someone else here might have run into a  
>>> similar
>>> problem and might know a workaround.  Unfortunately I haven't found  
>>> any
>>> forum postings or bug reports to date about it.
>>>
>>> Here's a snippet of an strace between a call from a phone in G729
>>> (172.20.33.73) and another Asterisk box in G711 (172.20.32.10).
>>>
>>> 
>>> recvfrom(33,
>>> "\200\0Vu\353\246\250F*\202rC}}~~\377\376\376\376\376}~"..., 8192, 0,
>>> {sa_family=AF_INET, sin_port=htons(25862),
>>> sin_addr=inet_addr("172.20.32.10")}, [16]) = 172
>>> gettimeofday({1243434707, 29870}, NULL) = 0
>>> time(NULL)  = 1243434707
>>> write(37,
>>> "\370\377\360\377\360\377\360\377\370\377\10\0\370\377\370"...,  
>>> 160) = 160
>>> poll([{fd=37, events=POLLOUT, revents=POLLOUT}], 1, 50) = 1
>>> read(37, 0x81e31a0, 8000)   = -1 EAGAIN (Resource
>>> temporarily unavailable)
>>> poll([{fd=29, events=POLLIN|POLLPRI, revents=POLLIN}, {fd=30,
>>> events=POLLIN|POLLPRI}, {fd=31, events=POLLIN|POLLPRI}, {fd=33,
>>> events=POLLIN|POLLPRI}, {fd=34, events=POLLIN|POLLPRI}, {fd=35,
>>> events=POLLIN|POLLPRI}], 6, -1) = 1
>>> fcntl64(31, F_GETFL)= 0x800 (flags O_RDONLY| 
>>> O_NONBLOCK)
>>> read(31, 0xbb7fa53c, 4) = -1 EAGAIN (Resource
>>> temporarily unavailable)
>>> recvfrom(29,
>>> "\200\22\355\310\n3\326\273H\256A\233x\6\36\353\300\272"..., 8192, 0,
>>> {sa_family=AF_INET, sin_port=htons(),
>>> sin_addr=inet_addr("172.20.33.73")}, [16]) = 32
>>> gettimeofday({1243434707, 47654}, NULL) = 0
>>> time(NULL)  = 1243434707
>>> write(38, "x\6\36\353\300\272\316\314\rV\370~ 
>>> \271\270\250\232\326"...,
>>> 20) = 20
>>> poll([{fd=38, events=POLLOUT, revents=POLLOUT}], 1, 50) = 1
>>> read(38, 0x81e5276, 1024)   = -1 EAGAIN (Resource
>>> temporarily unavailable)
>>> poll([{fd=33, events=POLLIN|POLLPRI, revents=POLLIN}, {fd=34,
>>> events=POLLIN|POLLPRI}, {fd=35, events=POLLIN|POLLPRI}, {fd=29,
>>> events=POLLIN|POLLPRI}, {fd=30, events=POLLIN|POLLPRI}, {fd=31,
>>> events=POLLIN|POLLPRI}], 6, -1) = 1
>>> fcntl64(35, F_GETFL)= 0x800 (flags O_RDONLY| 
>>> O_NONBLOCK)
>>> read(35, 0xbb7fa53c, 4)