[Asterisk-Users] ztdummy inaccuracy on linux-2.6
Tamas
jalsot at gmail.com
Tue Jan 17 07:42:44 MST 2006
Steven Ringwald wrote:
> On Mon, 2006-01-16 at 22:30 +0100, Tamas wrote:
>
>> /var/log/dmesg:
>> ...
>> CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
>> CPU: L2 Cache: 512K (64 bytes/line)
>> mtrr: v2.0 (20020519)
>> CPU: AMD Athlon(tm) 64 Processor 3000+ stepping 02
>> Using IO-APIC 2
>> ..MP-BIOS bug: 8254 timer not connected to IO-APIC
>> works.
>>
> Hm. This looks like something here (above)
>
>
I still didn't find out why do I have such silly message. However many
other people has this, so I guess, this is not really an malfunction.
>> pcie_portdrv_probe->Dev[005d:10de] has invalid IRQ. Check vendor BIOS
>> assign_interrupt_mode Found MSI capability
>> Allocate Port Service[pcie00]
>>
>
>
>
>> Maybe this says somehting...
>>
I revisited the BIOS settings and now it seems to be nicer ;) The
previouse silly messages still appears in logs...
cat /proc/interrupts
CPU0
0: 10698036 IO-APIC-edge timer
1: 8 IO-APIC-edge i8042
8: 10898960 IO-APIC-edge rtc
9: 0 IO-APIC-level acpi
50: 57 IO-APIC-level eth2, HFC PCI
217: 10302 IO-APIC-level 3w-xxxx
225: 1091888 IO-APIC-level eth0
233: 5334 IO-APIC-level eth1
NMI: 10772
LOC: 10696575
ERR: 0
MIS: 0
zttest still shows some crazy values.
Thanks to Tony Mountfield, I won't care much about :) I will test with
real usage (meetme, IAX2 trunk) how it is going in practice...
> Yeah. Also, what does your kernel config look like? Why did you roll
> your own kernel, rather than using the kernel 2.6.15 series that is in
> the updates-testing???
>
I tuned the kernel a bit (vanilla kernel without any external patch),
optimized it (removed tons of unneeded things which are in Ubuntu stock
kernel). Things seems to be fine, just that zttest. Tests will show if
zttest gives any usable information.
> The AMD64 box that I am using is running 2.6.14-1.1656_FC4, btw.
>
> Steve
>
>
Thanks for comments!
Regards,
Tamas
More information about the asterisk-users
mailing list