Voyage 0.10+Net5501 -> IRQ issue?

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Voyage 0.10+Net5501 -> IRQ issue?

Muffinman
Hello all,

I've been running Voyage 0.10.0 for a while but unfortunately it doesn't
run very smooth compared to 0.7.5. It's running on a Soekris net5501.
With normal 44.1 files its running fine but with higher resolutions it
stutters/distorts every few seconds or so, quite annoying.

I've loaded 0.7.5 on the CF card of the 5501 and the main differences
are in the load IRQ processes are creating.

5501: The following processes are producing a combined 15-20% cpu load
depending on the resolution currently being played. In 0.7.5. these
processes are not visible, so I'm not sure what that means.
irq/15-ehci_hcd
irq/15-ohci_
irq/15-pata_amd_hcd

However:
- ksoftirq is also slightly elevated at 1.6 when running 0.10.0 compared
to 0 on the 0.7.5.
- The si line in 'top' seems also to be quite high (between 2 and 7)
compared to when running 0.7.5 (between 0 and and occasional 0.3).
- system load is also quite high

It seems to me I have an irq problem, and the output of cat
/proc/interrupts (see below) seems to give some confirmation. The pata
and the usb irq's are shared in 0.10.0. This makes sense since the music
files are locally stored and accessed through a pata interface.

Anyone any ideas on how to fix this? Does anyone know what has changed
between 0.7.5 and 0.10 that might have caused this change?

Kind regards and thanks in advance, Maarten

0.7.5: cat /proc/interrupts
           CPU0      
  0:    2610959    XT-PIC-XT-PIC    timer
  1:          0    XT-PIC-XT-PIC    i8042
  2:          0    XT-PIC-XT-PIC    cascade
  3:          1    XT-PIC-XT-PIC
  4:        273    XT-PIC-XT-PIC    serial
  7:          1    XT-PIC-XT-PIC
 11:      41747    XT-PIC-XT-PIC    eth0
 14:     188673    XT-PIC-XT-PIC    ide0
 15:    3926610    XT-PIC-XT-PIC    ehci_hcd:usb1, ohci_hcd:usb2
NMI:          0   Non-maskable interrupts
LOC:          0   Local timer interrupts
SPU:          0   Spurious interrupts
PMI:          0   Performance monitoring interrupts
IWI:          0   IRQ work interrupts
RES:          0   Rescheduling interrupts
CAL:          0   Function call interrupts
TLB:          0   TLB shootdowns
ERR:          1
MIS:          0

0.10.0: cat /proc/interrupts
           CPU0      
  0:     302642    XT-PIC-XT-PIC    timer
  1:          0    XT-PIC-XT-PIC    i8042
  2:          0    XT-PIC-XT-PIC    cascade
  4:        515    XT-PIC-XT-PIC    serial
  8:          1    XT-PIC-XT-PIC    rtc0
 11:       2473    XT-PIC-XT-PIC    eth0
 14:      11585    XT-PIC-XT-PIC    pata_amd
 15:         38    XT-PIC-XT-PIC    pata_amd, ehci_hcd:usb1, ohci_hcd:usb2
NMI:          0   Non-maskable interrupts
LOC:          0   Local timer interrupts
SPU:          0   Spurious interrupts
PMI:          0   Performance monitoring interrupts
IWI:          0   IRQ work interrupts
RTR:          0   APIC ICR read retries
RES:          0   Rescheduling interrupts
CAL:          0   Function call interrupts
TLB:          0   TLB shootdowns
ERR:          0
MIS:          0



_______________________________________________
Voyage-linux mailing list
[hidden email]
http://list.voyage.hk/mailman/listinfo/voyage-linux
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Voyage 0.10+Net5501 -> IRQ issue?

Muffinman
Hello again,

I've done some experimenting with the issue and I think I have narrowed
the problem down a bit. The issue is partially introduced in Voyage
0.9.2 with the introduction of libata (see /proc/interrupts below). In
0.9.1. hires files play smoothly.

In solving this problem: I suppose there are two possible scenario's:
either fix the problem with libata or reverse to the legacy PATA
drivers. Where do I go to fix libata (code or configuration)? Else, how
do I reverse to the legacy drivers?

I hope anyone has some suggestions on either strategy.

Thanks in advance, Maarten

0.9.1
cat /proc/interrupts
           CPU0      
  0:     290271    XT-PIC-XT-PIC    timer
  1:          0    XT-PIC-XT-PIC    i8042
  2:          0    XT-PIC-XT-PIC    cascade
  4:        556    XT-PIC-XT-PIC    serial
 11:       1828    XT-PIC-XT-PIC    eth0
 14:       6367    XT-PIC-XT-PIC    ide0
 15:         38    XT-PIC-XT-PIC    ehci_hcd:usb1, ohci_hcd:usb2
NMI:          0   Non-maskable interrupts
LOC:          0   Local timer interrupts
SPU:          0   Spurious interrupts
PMI:          0   Performance monitoring interrupts
IWI:          0   IRQ work interrupts
RTR:          0   APIC ICR read retries
RES:          0   Rescheduling interrupts
CAL:          0   Function call interrupts
TLB:          0   TLB shootdowns
ERR:          0
MIS:          0

0.9.2
cat /proc/interrupts
           CPU0      
  0:   16021039    XT-PIC-XT-PIC    timer
  1:          0    XT-PIC-XT-PIC    i8042
  2:          0    XT-PIC-XT-PIC    cascade
  4:        549    XT-PIC-XT-PIC    serial
  7:          3    XT-PIC-XT-PIC
 11:      74406    XT-PIC-XT-PIC    eth0
 14:      74362    XT-PIC-XT-PIC    pata_amd
 15:    9406334    XT-PIC-XT-PIC    pata_amd, ehci_hcd:usb1, ohci_hcd:usb2
NMI:          0   Non-maskable interrupts
LOC:          0   Local timer interrupts
SPU:          0   Spurious interrupts
PMI:          0   Performance monitoring interrupts
IWI:          0   IRQ work interrupts
RTR:          0   APIC ICR read retries
RES:          0   Rescheduling interrupts
CAL:          0   Function call interrupts
TLB:          0   TLB shootdowns
ERR:          3
MIS:          0


On 17/12/15 21:29, Muffinman wrote:

> Hello all,
>
> I've been running Voyage 0.10.0 for a while but unfortunately it doesn't
> run very smooth compared to 0.7.5. It's running on a Soekris net5501.
> With normal 44.1 files its running fine but with higher resolutions it
> stutters/distorts every few seconds or so, quite annoying.
>
> I've loaded 0.7.5 on the CF card of the 5501 and the main differences
> are in the load IRQ processes are creating.
>
> 5501: The following processes are producing a combined 15-20% cpu load
> depending on the resolution currently being played. In 0.7.5. these
> processes are not visible, so I'm not sure what that means.
> irq/15-ehci_hcd
> irq/15-ohci_
> irq/15-pata_amd_hcd
>
> However:
> - ksoftirq is also slightly elevated at 1.6 when running 0.10.0 compared
> to 0 on the 0.7.5.
> - The si line in 'top' seems also to be quite high (between 2 and 7)
> compared to when running 0.7.5 (between 0 and and occasional 0.3).
> - system load is also quite high
>
> It seems to me I have an irq problem, and the output of cat
> /proc/interrupts (see below) seems to give some confirmation. The pata
> and the usb irq's are shared in 0.10.0. This makes sense since the music
> files are locally stored and accessed through a pata interface.
>
> Anyone any ideas on how to fix this? Does anyone know what has changed
> between 0.7.5 and 0.10 that might have caused this change?
>
> Kind regards and thanks in advance, Maarten
>
> 0.7.5: cat /proc/interrupts
>            CPU0      
>   0:    2610959    XT-PIC-XT-PIC    timer
>   1:          0    XT-PIC-XT-PIC    i8042
>   2:          0    XT-PIC-XT-PIC    cascade
>   3:          1    XT-PIC-XT-PIC
>   4:        273    XT-PIC-XT-PIC    serial
>   7:          1    XT-PIC-XT-PIC
>  11:      41747    XT-PIC-XT-PIC    eth0
>  14:     188673    XT-PIC-XT-PIC    ide0
>  15:    3926610    XT-PIC-XT-PIC    ehci_hcd:usb1, ohci_hcd:usb2
> NMI:          0   Non-maskable interrupts
> LOC:          0   Local timer interrupts
> SPU:          0   Spurious interrupts
> PMI:          0   Performance monitoring interrupts
> IWI:          0   IRQ work interrupts
> RES:          0   Rescheduling interrupts
> CAL:          0   Function call interrupts
> TLB:          0   TLB shootdowns
> ERR:          1
> MIS:          0
>
> 0.10.0: cat /proc/interrupts
>            CPU0      
>   0:     302642    XT-PIC-XT-PIC    timer
>   1:          0    XT-PIC-XT-PIC    i8042
>   2:          0    XT-PIC-XT-PIC    cascade
>   4:        515    XT-PIC-XT-PIC    serial
>   8:          1    XT-PIC-XT-PIC    rtc0
>  11:       2473    XT-PIC-XT-PIC    eth0
>  14:      11585    XT-PIC-XT-PIC    pata_amd
>  15:         38    XT-PIC-XT-PIC    pata_amd, ehci_hcd:usb1, ohci_hcd:usb2
> NMI:          0   Non-maskable interrupts
> LOC:          0   Local timer interrupts
> SPU:          0   Spurious interrupts
> PMI:          0   Performance monitoring interrupts
> IWI:          0   IRQ work interrupts
> RTR:          0   APIC ICR read retries
> RES:          0   Rescheduling interrupts
> CAL:          0   Function call interrupts
> TLB:          0   TLB shootdowns
> ERR:          0
> MIS:          0
>
>


_______________________________________________
Voyage-linux mailing list
[hidden email]
http://list.voyage.hk/mailman/listinfo/voyage-linux
Loading...