[Home]

Summary:DAHLIN-00257: wcb4xxp shows hardware EC in /proc/dahdi/* for B410P although vpmsupport=0
Reporter:Gerald Schnabel (geraldsf)Labels:
Date Opened:2011-09-30 08:49:55Date Closed:2011-11-04 11:55:41
Priority:TrivialRegression?No
Status:Closed/CompleteComponents:wcb4xxp
Versions:2.5.0.1 Frequency of
Occurrence		Constant
Related
Issues:
Environment:Attachments:( 0) 0001-wcb4xxp-Do-not-show-LASVEGAS2-as-echocan-name-if-vpm.patch
Description:If module wcb4xxp is loaded with _vpmsupport=0_ dmesg shows that hardware echo cancellation has been disabled:
{noformat}
wcb4xxp 0000:05:01.0: Identified Wildcard B410P (controller rev 1) at 00011000, IRQ 74
wcb4xxp 0000:05:01.0: VPM 0/1 init: chip ver 33
wcb4xxp 0000:05:01.0: VPM 1/1 init: chip ver 33
wcb4xxp 0000:05:01.0: NOTE: hardware echo cancellation has been disabled
wcb4xxp 0000:05:01.0: Port 1: TE mode
wcb4xxp 0000:05:01.0: Port 2: TE mode
wcb4xxp 0000:05:01.0: Port 3: TE mode
wcb4xxp 0000:05:01.0: Port 4: TE mode
wcb4xxp 0000:05:01.0: Did not do the highestorder stuff
dahdi_transcode: Loaded.
wcb4xxp 0000:05:01.0: new card sync source: port 4
wcb4xxp 0000:05:01.0: new card sync source: port 4
{noformat}

{noformat}
[root@localhost ~]# cat /sys/module/wcb4xxp/parameters/vpmsupport
0
{noformat}

_/proc/dahdi/*_ shows that EC: LASVEGAS2 is used:

{noformat}
[root@localhost ~]# cat /proc/dahdi/*
Span 1: B4/0/1 "B4XXP (PCI) Card 0 Span 1" AMI/CCS RED

          1 B4/0/1/1 Clear RED (EC: LASVEGAS2 - INACTIVE)
          2 B4/0/1/2 Clear RED (EC: LASVEGAS2 - INACTIVE)
          3 B4/0/1/3 Hardware-assisted HDLC RED (EC: LASVEGAS2 - INACTIVE)
Span 2: B4/0/2 "B4XXP (PCI) Card 0 Span 2" AMI/CCS RED

          4 B4/0/2/1 Clear RED (EC: LASVEGAS2 - INACTIVE)
          5 B4/0/2/2 Clear RED (EC: LASVEGAS2 - INACTIVE)
          6 B4/0/2/3 Hardware-assisted HDLC RED (EC: LASVEGAS2 - INACTIVE)
Span 3: B4/0/3 "B4XXP (PCI) Card 0 Span 3" AMI/CCS RED

          7 B4/0/3/1 Clear RED (EC: LASVEGAS2 - INACTIVE)
          8 B4/0/3/2 Clear RED (EC: LASVEGAS2 - INACTIVE)
          9 B4/0/3/3 Hardware-assisted HDLC RED (EC: LASVEGAS2 - INACTIVE)
Span 4: B4/0/4 "B4XXP (PCI) Card 0 Span 4" AMI/CCS RED

         10 B4/0/4/1 Clear RED (EC: LASVEGAS2 - INACTIVE)
         11 B4/0/4/2 Clear RED (EC: LASVEGAS2 - INACTIVE)
         12 B4/0/4/3 Hardware-assisted HDLC RED (EC: LASVEGAS2 - INACTIVE)
{noformat}

I would expect to see that software EC is used in _/proc/dahdi/*_ when hardware EC is disabled:

{noformat}
@@ -2205,7 +2205,7 @@ static const char *b4xxp_echocan_name(co
{
struct b4xxp_span *bspan = container_of(chan->span, struct b4xxp_span,
span);
- if (bspan->parent->card_type == B410P)
+ if (vpmsupport && bspan->parent->card_type == B410P)
return "LASVEGAS2";
return NULL;
}
{noformat}

{noformat}
Span 1: B4/0/1 "B4XXP (PCI) Card 0 Span 1" AMI/CCS RED

          1 B4/0/1/1 Clear RED (EC: MG2 - INACTIVE)
          2 B4/0/1/2 Clear RED (EC: MG2 - INACTIVE)
          3 B4/0/1/3 Hardware-assisted HDLC RED
Span 2: B4/0/2 "B4XXP (PCI) Card 0 Span 2" AMI/CCS RED

          4 B4/0/2/1 Clear RED (EC: MG2 - INACTIVE)
          5 B4/0/2/2 Clear RED (EC: MG2 - INACTIVE)
          6 B4/0/2/3 Hardware-assisted HDLC RED
Span 3: B4/0/3 "B4XXP (PCI) Card 0 Span 3" AMI/CCS RED

          7 B4/0/3/1 Clear RED (EC: MG2 - INACTIVE)
          8 B4/0/3/2 Clear RED (EC: MG2 - INACTIVE)
          9 B4/0/3/3 Hardware-assisted HDLC RED
Span 4: B4/0/4 "B4XXP (PCI) Card 0 Span 4" AMI/CCS RED

         10 B4/0/4/1 Clear RED (EC: MG2 - INACTIVE)
         11 B4/0/4/2 Clear RED (EC: MG2 - INACTIVE)
         12 B4/0/4/3 Hardware-assisted HDLC RED
{noformat}
Comments:By: Shaun Ruffell (sruffell) 2011-09-30 10:38:47.992-0500

I would like to think about this some more before committing to trunk, but I believe [0001-wcb4xxp-Do-not-show-LASVEGAS2-as-echocan-name-if-vpm.patch|https://issues.asterisk.org/jira/secure/attachment/41272/0001-wcb4xxp-Do-not-show-LASVEGAS2-as-echocan-name-if-vpm.patch] is all that is needed.
By: Shaun Ruffell (sruffell) 2011-11-04 11:55:41.888-0500

Committed in [r10251|http://svnview.digium.com/svn/dahdi?view=revision&revision=10251] (with, unfortunately a bad issue number).