Re: [Asterisk-Users] PCI front mount chassis?

2004-03-12 Thread Stephen Davies


On Fri, 12 Mar 2004, Brian Capouch wrote:

 I too am running 6 cards in my system, although not in a high traffic 
 capacity load environment.
 
 So far my (limited) high-load simulations have shown no problems.


So - is it apocryphal that the Digium cards (drivers) won't share
interrupts?

If there is a real issue with sharing interrupts then it seems to me
to be a bug that needs fixing.  PCI bus supports shared interrupts,
why doesn't the hardware/driver?

Yours curiously,
Steve


___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] PCI front mount chassis?

2004-03-12 Thread Rich Adamson

  I too am running 6 cards in my system, although not in a high traffic 
  capacity load environment.
  
  So far my (limited) high-load simulations have shown no problems.
 
 
 So - is it apocryphal that the Digium cards (drivers) won't share
 interrupts?
 
 If there is a real issue with sharing interrupts then it seems to me
 to be a bug that needs fixing.  PCI bus supports shared interrupts,
 why doesn't the hardware/driver?

In most cases, sharing an interrupt is not a problem at all. There have
been a few cases where _some_ issue was resolved by moving cards around,
however the majority of those seem to be: a) abrupt system changes with
no effort to seriously identify the root-cause, b) newbie installations
where the condition of the underlying system infrastructure is totally
unknown, or, c) wild recommendations that might have had some basis a
long time ago but no longer apply.

Example: 'cat /proc/interrupts'
  9: 1854652239  XT-PIC  ehci-hcd, eth0, wcfxo, Intel ICH4
works just fine, and I can't imagine a more demanding irq arrangement
where the only nic shares with an x100p, etc.

Obviously there are performance limits and expecting multiple quad T1 
cards or some other _specific_ high-volume configuration to share one 
or two interrupts could create a problem. But, engineering a system for
those conditions is no more difficult then understanding the 
requirements of whatever cards are being used and dealing with them 
appropriately.



___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] PCI front mount chassis?

2004-03-12 Thread Bruno Haas
Wow, 1000 IRQs a second. I'm quite amazed. Does anybody know which 
applications would require such a low latency ? It does seem to me that 
this way of doing things is rather dangerous and prone to problems. 
Anybody can comment ?

Thanks
Bruno
Nicolas Bougues wrote:

On Fri, Mar 12, 2004 at 08:41:36AM -0500, Walt Reed wrote:
 

The voice cards generate an order of magnitude more interrupts than
anything else. This may be why it's not recommended to share
interrupts on voice cards. Don't know if the T1 cards have a similar
issue. I would hope not. The x100p's are a pretty simplistic device.
They probably generate an interrupt for every byte. The x100p's are also
used for timing in things like MOH and MM conferences AFAIK. It seems
like it would be nice to only put one card in timer mode if that is
indeed what is generating all those interrupts. Could someone in the
know enlighten us?
   

Digium boards usually generate 1000 interrupts/sec. This is mostly a
timer interrupt, so that the driver can poll the board for 8 bytes
per channel 1000 times per second. The polling is either slave (PIO),
or busmaster (DMA).
Each board generates 1000 interrupts/second, no matter the kind/number
of ports on the boards. The driver knows if that's a single channel
board, with 8 bytes to fetch, or quad E1, with 128*8 bytes to get on
each interrupt.
There are quite strict timing requirements : if one interrupt is lost,
1/1000th of sound on the line(s) is lost, which can be quite bad for
things like HDLC (on T1/E1) or modem sound (no matter the kind of
channel).
Sharing an IRQ may introduce some latency, because the IRQ has to be
handled by several drivers, and if each of them does not behave
properly, frames can be late/lost.
 

___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
  http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] PCI front mount chassis?

2004-03-12 Thread Rich Adamson
 Looking at my /proc/interrupts:
 
   0:   59709041  XT-PIC  timer
   5:  597050409  XT-PIC  wcfxo
   7:  597211339  XT-PIC  wcfxo
  10:4538876  XT-PIC  eth0
  11:3044608  XT-PIC  aic7xxx, eth1
 
 The voice cards generate an order of magnitude more interrupts than
 anything else. This may be why it's not recommended to share
 interrupts on voice cards. Don't know if the T1 cards have a similar
 issue. I would hope not. The x100p's are a pretty simplistic device.
 They probably generate an interrupt for every byte. The x100p's are also
 used for timing in things like MOH and MM conferences AFAIK. It seems
 like it would be nice to only put one card in timer mode if that is
 indeed what is generating all those interrupts. Could someone in the
 know enlighten us?

The x100p incurres roughly 1000 interrupts per second. Someone on the list
published a short unix command-line that essentially ran the above twice
and calculated the interrupts / sec nicely. It was kind of a handy way
to visually inspect the numbers to diagnose problems/issues. If I recall, 
that was around early/mid 2003 in the archives.

I believe that is the zap timing referred to frequently relative to
do you have a zap card installed.

Anyway, balancing the processing required per card verses how many cards
can share an interrupt is a system engineering task (regardless of
what the cards are doing). However, that engineering task is complicated
by the lack of published _actual_ requirements/expectations for each type
of card that can be installed/supported, and therefore raises questions 
(and some spontaneous off-the-wall responses) that aren't necessarily
based on facts.

Rich


___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] PCI front mount chassis?

2004-03-12 Thread Anton Tinchev
Steven Critchfield wrote:

On Fri, 2004-03-12 at 05:26, Rich Adamson wrote:
 

I too am running 6 cards in my system, although not in a high traffic 
capacity load environment.

So far my (limited) high-load simulations have shown no problems.
   

So - is it apocryphal that the Digium cards (drivers) won't share
interrupts?
If there is a real issue with sharing interrupts then it seems to me
to be a bug that needs fixing.  PCI bus supports shared interrupts,
why doesn't the hardware/driver?
 

In most cases, sharing an interrupt is not a problem at all. There have
been a few cases where _some_ issue was resolved by moving cards around,
however the majority of those seem to be: a) abrupt system changes with
no effort to seriously identify the root-cause, b) newbie installations
where the condition of the underlying system infrastructure is totally
unknown, or, c) wild recommendations that might have had some basis a
long time ago but no longer apply.
   

What I was referring to in this case about sharing an IRQ was the actual
wire trace in the PCI bus. As I understand the PCI spec, there are 4
interrupt lines called A,B,C, and D. In slot 1, They appear in that
order. In slot 2 they shift, in slot 3 they shift and again in slot 4.
By the time you get to slot 5, all interrupts have been in each of the 4
spots and now they have the choice of staying in the same as slot 4 or
shifting again. Most cards use interrupt line A as if they are in slot
1, and therefore if you move them from one slot to another, they would
most likely get their own access to the interrupt line.
This is part of the reason why moving a card around in the chassis
helps. This is also why I would be cautious of trying to run more than 4
high interrupt cards on the PCI bus. 

My next point would be that of, if you need more than 4 cards, you need
more than 1 computer controlling it. That is way too many eggs for one
basket. 
 

Most of the 3.3V Boards have more then 1 pci bus. The boards that i use 
have 3 PCU Busses - 3xXPCI, 3xPCI66, Internal with Adaptec, ide and 
E1000 on board

___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
  http://lists.digium.com/mailman/listinfo/asterisk-users


[Asterisk-Users] PCI front mount chassis?

2004-03-11 Thread Jeff Stohl
I am looking for a good case to house my Digium PCI cards, I was hoping to 
mount them in the front for cleaner access then in the back. Unfortunately I 
haven't found much, does anyone have a good recommendation for chassis to 
use up to six digium cards?

Thanks!

_
FREE pop-up blocking with the new MSN Toolbar – get it now! 
http://clk.atdmt.com/AVE/go/onm00200415ave/direct/01/

___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
  http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] PCI front mount chassis?

2004-03-11 Thread James Sharp
 I am looking for a good case to house my Digium PCI cards, I was hoping to
 mount them in the front for cleaner access then in the back. Unfortunately
 I
 haven't found much, does anyone have a good recommendation for chassis to
 use up to six digium cards?

Probably not cost effective, but i had looked into a few 6U CompactPCI
chassis...stick a CPCI to PCI adapter card in there.

*rummagerummage*  Ah-hah.

http://www.pcisystems.com/PCI-ADAPTER.html

Something like that.  Except that particular one wouldn't work for some
cards because it is PCI 2.1, not 2.2.
___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] PCI front mount chassis?

2004-03-11 Thread Steven Critchfield
On Thu, 2004-03-11 at 23:03, Jeff Stohl wrote:
 I am looking for a good case to house my Digium PCI cards, I was hoping to 
 mount them in the front for cleaner access then in the back. Unfortunately I 
 haven't found much, does anyone have a good recommendation for chassis to 
 use up to six digium cards?

What makes you think you could reliably run 6 cards at once?

The PCI bus only has 4 interrupt lines. Anything over that is shared. 
-- 
Steven Critchfield [EMAIL PROTECTED]

___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] PCI front mount chassis?

2004-03-11 Thread Jeff Stohl
I am running six reliably right now.

Surely I am not the only one doing a large capacity single site?


From: Steven Critchfield [EMAIL PROTECTED]
Reply-To: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Subject: Re: [Asterisk-Users] PCI front mount chassis?
Date: Fri, 12 Mar 2004 00:35:40 -0600
On Thu, 2004-03-11 at 23:03, Jeff Stohl wrote:
 I am looking for a good case to house my Digium PCI cards, I was hoping 
to
 mount them in the front for cleaner access then in the back. 
Unfortunately I
 haven't found much, does anyone have a good recommendation for chassis 
to
 use up to six digium cards?

What makes you think you could reliably run 6 cards at once?

The PCI bus only has 4 interrupt lines. Anything over that is shared.
--
Steven Critchfield [EMAIL PROTECTED]
___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users
_
Fast. Reliable. Get MSN 9 Dial-up - 3 months for the price of 1! 
(Limited-time Offer) http://click.atdmt.com/AVE/go/onm00200361ave/direct/01/

___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
  http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] PCI front mount chassis?

2004-03-11 Thread Brian Capouch
Jeff Stohl wrote:
I am running six reliably right now.

Surely I am not the only one doing a large capacity single site?

I too am running 6 cards in my system, although not in a high traffic 
capacity load environment.

So far my (limited) high-load simulations have shown no problems.

B.
___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
  http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [Asterisk-Users] PCI front mount chassis?

2004-03-11 Thread Nicolas Bougues
On Fri, Mar 12, 2004 at 06:48:41AM +, Jeff Stohl wrote:
 I am running six reliably right now.
 
 Surely I am not the only one doing a large capacity single site?
 

You mean 6 quad span boards ??

In the T1 world, it's 576 channels on a single PC, 720 if E1.

What kind of configuration can handle that load reliably ? What's the
application ?

-- 
Nicolas Bougues
Axialys Interactive
___
Asterisk-Users mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-users
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users