[Asterisk-Users] Re: hfc-s card, brii-stuff.0.1.0-RC4a, zaphfc: sync lost, pci performance too low. you might have some cpu throtteling enabled.

[Alexander Szlezak]

HI Hartmut,

I do have the same problem as you decribed earlier. The Billion HFC 
Cards (two of them) work flawlessly in my old Pentium II, but in my more 
powerful Athlon XP 2400+ (Via KT400 Chipset), I allways get the "pci 
performance too low" message and syslog kills the system.

I'll try it now with commenting out the message. Do you know anything 
more in the meantime. Did Mr. Junghanns have any statement about it? I 
wonder if chan_mISDN is a better choice by now?

Thanks for any advice you might have!

yours,
Alexander

Hartmut Wahl wrote:
> Hello, 
> 
> I have investigated the issue a bit further, I was not able to find the
> root cause, maybe it is the KT133 Chipset of my ASUS A7V. However I
> found a bad hack to make it work under some circumstances. I commented
> out the line:
> 
> printk(KERN_CRIT "zaphfc: sync lost, pci performance too low. you might
> have some cpu throtteling enabled.\n");
> 
> in zaphfc/zaphfc.c. Since when this situation happened once, the syslog
> started and this caused the situation again -> endless loop with high
> sysload. Now the sync lost happens probably every now and then but it
> does not go into an endless loop.
> 
> I also recognized that I must not run setiathome since then the audio
> quality of connections via the hfc-card will suffer (crackling).
> Amazingly cpuburn (takes every cpu-Time it gets as well) does not have
> this effect. I have no idea what strange things setiathome does to cause
> this but it reminds me of a sound card problem on this board. My SB-Live 
> had crackling when I ran setiathome but I think with a newer driver 
> (and ALSA I think) this problem was gone.
> 
> Things that did not help:
> - Trying to change the latency (is fixed to 16 and cannot be changed)
> - Trying to change the latency of my other pci-devices (much higher and
>   much lower).
> - Playing around with BIOS Options like delayed transaction, etc.
> - Changing PCI-slots
> - Making sure that the ISDN-card did not share the IRQ 
> - Having only the ISDN-card and the Video-card in the System
> 
> Hope that helps others who encounter this problem as well
>  Hartmut
> 
> On Sat, Sep 11, 2004 at 12:55:21PM +0200, Hartmut Wahl wrote:
> 
>>Hi!
>>
>>On Sat, Sep 11, 2004 at 10:38:48AM +0200, Maurizio Marini wrote:
>>
>>>my machine did hangup as growing logs fullfilled partition
>>
>>hmm I see, mine is 8G, but it has gronwn from 0.5G to 2.0G
>> 
>>
>>>it does apply to asterisk, not to zaphfc :(
>>>it was a misleading suggestion, so
>>>i solved it installing in an other more powerful machine:
>>>
>>>processor       : 0
>>>vendor_id       : GenuineIntel
>>>cpu family      : 6
>>>model           : 8
>>>model name      : Pentium III (Coppermine)
>>>stepping        : 10
>>>cpu MHz         : 999.556
>>>cache size      : 256 KB
>>>fdiv_bug        : no
>>>hlt_bug         : no
>>>f00f_bug        : no
>>>coma_bug        : no
>>>fpu             : yes
>>>fpu_exception   : yes
>>>cpuid level     : 2
>>>wp              : yes
>>>flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca 
>>>cmov pat pse36 mmx fxsr sse
>>>bogomips        : 1957.88
>>>
>>>with this hw i've no issues at all; even strange messages i complained about 
>>>in my previous posts like:
>>
>>ok we are getting closer, although speed shouldn't be a problem:
>>
>>vendor_id       : AuthenticAMD
>>model name      : AMD Athlon(tm) Processor
>>stepping        : 4
>>cpu MHz         : 1059.618
>>cache size      : 256 KB
>>bogomips        : 2097.15
>>
>>it is a 1,4GHz underclocked, since my ASUS A7V with KT133 does only
>>support 100MHz FSB, but it requires only slow and quite fans.
>>
>>I think I'll try different kernels maybe something in the Debian
>>k7-kernel-image interferes.
>>
>>Greetings
>>  Hartmut
>>
> 
>