Re: [LAU] : jack CPU % is going crazy !

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Pieter Palmers <pieterp@...>
Cc: A list for linux audio users <linux-audio-user@...>
Date: Tuesday, April 8, 2008 - 12:27 pm

------=_Part_22029_7345965.1207657628012
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

2008/4/7, Pieter Palmers :

(in
l the
t

Hello,

I ran jackdmp/freebob with -P99 priority but nothing changed. But there is
something Strange. I can't give a -P99 priority directly in Qjackctl. The
max is 89. So I had to run jackdmp in a terminal, then run qjackctl in
"Active" mode.

Well sorry, after some more tests, it is a bit better with -P99 priority. I=
t
is more stable around 3.5% and there are few peaks up to 28% or 53% whereas
with a lower priority (-P80) it is "stable" around 5% and there are many
more peaks up to 13, 22 or 28% and some up to 53%
Perhaps it is a problem with my configurations files

Here are my configuration files :

******************************* rtirq conf script
***************************************

# IRQ thread service names
# (space separated list, from higher to lower priority).
RTIRQ_NAME_LIST=3D"rtc ohci1394 i8042"

# Highest priority.
RTIRQ_PRIO_HIGH=3D99

# Priority decrease step.
RTIRQ_PRIO_DECR=3D5

# Whether to reset all IRQ threads to SCHED_OTHER.
RTIRQ_RESET_ALL=3D0

# On kernel configurations that support it,
# which services should be NOT threaded
# (space separated list).
RTIRQ_NON_THREADED=3D"rtc ohci1394"

# Process names which will be forced to the
# highest realtime priority range (99-91)
# (space separated list, from highest to lower priority).
# RTIRQ_HIGH_LIST=3D"softirq-timer"

************************** /etc/init.d/limits.conf
****************************

@audio - rtprio 99
@audio - nice -15
@audio - memlock 512000

****************************** cat /proc/interrupts
*********************************
CPU0 CPU1
0: 16448905 0 IO-APIC-edge timer
1: 1922 0 IO-APIC-edge i8042
8: 1 0 IO-APIC-edge rtc
9: 1 0 IO-APIC-fasteoi acpi
12: 801 0 IO-APIC-edge i8042
14: 18261 0 IO-APIC-edge ide0
15: 74104 0 IO-APIC-edge ide1
16: 270629 0 IO-APIC-fasteoi nvidia
17: 4022 0 IO-APIC-fasteoi eth0
19: 1564191 0 IO-APIC-fasteoi ohci1394, ohci1394
20: 24 0 IO-APIC-fasteoi ehci_hcd:usb1, uhci_hcd:usb2
21: 113 0 IO-APIC-fasteoi uhci_hcd:usb3, HDA Intel
22: 12548 0 IO-APIC-fasteoi uhci_hcd:usb4
23: 0 0 IO-APIC-fasteoi uhci_hcd:usb5
NMI: 0 0 Non-maskable interrupts
LOC: 138113 8271422 Local timer interrupts
RES: 943079 1904284 Rescheduling interrupts
CAL: 86 54 function call interrupts
TLB: 890 700 TLB shootdowns
TRM: 0 0 Thermal event interrupts
THR: 0 0 Threshold APIC interrupts
SPU: 0 0 Spurious interrupts
ERR: 0

************************************* /etc/init.d/rtirq status
*******************************

* status: started

PID CLS RTPRIO NI PRI %CPU STAT COMMAND
352 FF 95 - 135 0.0 S< IRQ-8 rtc
426 FF 90 - 130 0.3 S< IRQ-19 ohci1394, ohci1394
491 FF 85 - 125 0.0 S< IRQ-1 i8042
490 FF 84 - 124 0.0 S< IRQ-12 i8042
5 FF 50 - 90 0.0 S< softirq-high/0
6 FF 50 - 90 0.7 S< softirq-timer/0
7 FF 50 - 90 0.0 S< softirq-net-tx/
8 FF 50 - 90 0.0 S< softirq-net-rx/
9 FF 50 - 90 0.0 S< softirq-block/0
10 FF 50 - 90 0.0 S< softirq-tasklet
11 FF 50 - 90 0.0 S< softirq-sched/0
12 FF 50 - 90 0.0 S< softirq-hrtimer
13 FF 50 - 90 0.0 S< softirq-rcu/0
17 FF 50 - 90 0.0 S< softirq-high/1
18 FF 50 - 90 0.4 S< softirq-timer/1
19 FF 50 - 90 0.0 S< softirq-net-tx/
20 FF 50 - 90 0.0 S< softirq-net-rx/
21 FF 50 - 90 0.0 S< softirq-block/1
22 FF 50 - 90 0.1 S< softirq-tasklet
23 FF 50 - 90 0.0 S< softirq-sched/1
24 FF 50 - 90 0.0 S< softirq-hrtimer
25 FF 50 - 90 0.0 S< softirq-rcu/1
99 FF 50 - 90 0.0 S< IRQ-9 acpi
395 FF 50 - 90 0.7 S< IRQ-14 ide0
396 FF 50 - 90 0.0 S< IRQ-15 ide1
446 FF 50 - 90 0.0 S< IRQ-20 ehci_hcd:usb1, uhci_hcd:usb=
2
461 FF 50 - 90 0.0 S< IRQ-21 uhci_hcd:usb3, HDA Intel
468 FF 50 - 90 0.0 S< IRQ-22 uhci_hcd:usb4
479 FF 50 - 90 0.0 S< IRQ-23 uhci_hcd:usb5
3452 FF 50 - 90 0.0 S< IRQ-17 eth0
3685 FF 50 - 90 0.1 S< IRQ-16 nvidia

Thank you again for your help.
greets,

Mysth-R

--=20
*
***************************************************************************=
************
* {^_^} Mysth-R {^_^}
* <=3D Aide Auditive =3D>
*
* http://myspace.com/mysthr
* http://myspace.com/aideauditive
* http://mysthr.free.fr/Joomla =3D> Site d=E9di=E9 =E0 l'audio sous
Fedora/PlanetCCRMA.
*
***************************************************************************=
************

------=_Part_22029_7345965.1207657628012
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

2008/4/7, Pieter Palmers <pieterp@joow.be>:
Mysth-R wrote:

2008/4/7, Pieter Palmers <pieterp@joo=
w.be
<mailto:pieterp@joow.be&=
gt;>:

   Mysth-R wrote:

       2008/4/7, Pieter Palmers <pieterp@joow.be
       <mailto:p=
ieterp@joow.be
> <mailto:pieter=
p@joow.be

       <mailto:p=
ieterp@joow.be
>>>:

          Mysth-R wrote:

              I forgot to precise that =
my sound card is a Presonus
       Firebox and
              run with Freebob.
              I got the idea to test wi=
th my U-control Uca202
       (Behringer) Usb
              sound-card, and there is =
no problems. The cpu percent is
       quite
              stable around 2,1%.
              It is a bit strange becau=
se I got the rtirq script
       congiured for
              my firewire card. So it s=
hould not work better than my
       firewire ...

              I am a bit lost !!  =
:O[

          Is the 'real' cpu time (e.g. fr=
om top) also that high? It
       could be
          that there is a bug in the time reporti=
ng of the freebob backend,
          resulting in a bad calculation of the c=
pu time.

          Greets,

          Pieter

       Hi,
       It is very strange. I can't see a real corr=
elation between cpu
       load displayed by htop and the DSP load display=
ed by qjackctl.
       In htop the cpu is quite unstable too, but it d=
oesn'tt change at
       the same time as the dsp load in qjackctl. Perh=
aps it is due to
       a latency. but I can't say if is linked.
       I have made some more test :

       1- runing qjackctl with freebob backend : both =
dsp load
       (qjackctl) and cpu load (htop) are unstable but=
not really linked.
       2- runing qjackctl with alsa backend : dsp load=
is quite stable
       but cpu load is unstable.
       3- running jackdmp in a console with Freebob th=
en alsa : cpu
       load is unstable.
       4- when jackdmp or qjackctl are killed : cpu lo=
ad is normal
       around 2% and stable.

       So :
       Is it normal that the cpu load displayed in hto=
p is unstable ?
       If it is, then as you said pieter, could it be =
a bug with
       freebob ??? I never saw this before, on the sam=
e laptop but with
       32bits OS.

   FreeBoB is guaranteed 110% bug-free.

h=E9h=E9 :D

       Is there a problem in my kernel config ?
       Is there a something wrong with jack or freebob=
with the 64bits
       arch ?

   It could very well be a 64bit issue somewhere. I haven't =
tested
   freebob on 64bit yet.

       Well, if someone as some idea for me to make mo=
re tests ... I
       would be happy ;)

   It might be interesting if you could figure out exactly what =
the
   differences are between your setups. Mainly software versions=
etc...

well, on the same laptop I got a 32bits openSuse system and a gentoo 64bits=
in dual boot.
I have tested on OpenSuse, with both Kde3.5 and Fvwm, just using qjackctl 0=
.3.2 with jackd 0.109 and it works perfectly.
Now on my gentoo 64 bits I tryed on both Kde4.0 and Fvwm-crystal, just usin=
g qjackctl 0.3.2 and jackdmp0.70. I also tryed with qjackctl 0.2.23 (in cas=
e it was due to QT4 library) and jackdmp-svn and jackd 0.109 : in all the c=
ase I got the cpu load problem.

Do you think this is a firewire problem ? perhaps I forgot an important opt=
ion when I compiled my kernel (2.6.24-rt1)
I think I will try with another kernel. Will see if there is new kernel ava=
ilable. otherwise I will try with an older (2.6.22 ...)

Thank you for your answer.

ps : pieter if you haven't tested freebob in 64 bits, perhaps you could=
give me some guidelines, to test it and make a report for you ?

I don't really have a clue about what's going on, so it's prett=
y difficult to do so.

Can you try running jack/freebob with a very high priority (e.g. -P99) to s=
ee if this changes things?Hello,I ran jac=
kdmp/freebob with -P99 priority but nothing changed. But there is something=
Strange. I can't give a -P99 priority directly in Qjackctl. The max is=
89. So I had to run jackdmp in a terminal, then run qjackctl in "Acti=
ve" mode.
Well sorry, after some more tests, it is a bit better with -P99 priorit=
y. It is more stable around 3.5% and there are few peaks up to 28% or 53% w=
hereas with a lower priority (-P80) it is "stable" around 5% and =
there are many more peaks up to 13, 22 or 28% and some up to 53%
Perhaps it is a problem with my configurations filesHere are my=
configuration files : ******************************* rtirq co=
nf script ***************************************# IRQ thread servi=
ce names
# (space separated list, from higher to lower priority).RTIRQ_NAME_LIST=
=3D"rtc ohci1394 i8042"# Highest priority.RTIRQ_PRIO_=
HIGH=3D99# Priority decrease step.RTIRQ_PRIO_DECR=3D5# =
Whether to reset all IRQ threads to SCHED_OTHER.
RTIRQ_RESET_ALL=3D0# On kernel configurations that support it,#=
which services should be NOT threaded# (space separated list).RTIR=
Q_NON_THREADED=3D"rtc ohci1394"# Process names which will=
be forced to the
# highest realtime priority range (99-91)# (space separated list, from =
highest to lower priority).# RTIRQ_HIGH_LIST=3D"softirq-timer&quot=
;************************** /etc/init.d/limits.conf ***********=
*****************
@audio          - &nb=
sp;     rtprio      &nbsp=
;   99@audio        &=
nbsp; -       nice    &nb=
sp;       -15@audio   &nbs=
p;      -       meml=
ock         512000*=
***************************** cat /proc/interrupts ************************=
*********
          CPU0  &nbs=
p;    CPU1         0:=
   16448905         =
0   IO-APIC-edge      timer  1=
:       1922     &nb=
sp;    0   IO-APIC-edge    &nb=
sp; i8042  8:        &nbsp=
; 1          0   IO-=
APIC-edge      rtc  9:   &=
nbsp;      1      &n=
bsp;   0   IO-APIC-fasteoi   acpi
 12:        801   &n=
bsp;      0   IO-APIC-edge  &n=
bsp;   i8042 14:      18261&nbs=
p;         0   IO-APIC-ed=
ge      ide0 15:    &=
nbsp; 74104          0 &n=
bsp; IO-APIC-edge      ide1 16: &nbs=
p;   270629         =
0   IO-APIC-fasteoi   nvidia
 17:       4022    &=
nbsp;     0   IO-APIC-fasteoi   eth=
0 19:    1564191     &nbsp=
;    0   IO-APIC-fasteoi   ohci1394, ohc=
i1394 20:         24 =
         0   IO-APIC-fast=
eoi   ehci_hcd:usb1, uhci_hcd:usb2 21:   =
     113       &nbsp=
;  0   IO-APIC-fasteoi   uhci_hcd:usb3, HDA Intel<=
br>
 22:      12548     =
     0   IO-APIC-fasteoi   uhci_hcd=
:usb4 23:          0&=
nbsp;         0   IO-APIC=
-fasteoi   uhci_hcd:usb5NMI:     &nb=
sp;    0        &nbs=
p; 0   Non-maskable interruptsLOC:     13=
8113    8271422   Local timer interrupts
RES:     943079    1904284   R=
escheduling interruptsCAL:       &nb=
sp; 86         54   funct=
ion call interruptsTLB:        890&n=
bsp;       700   TLB shootdowns=
TRM:          0  &nb=
sp;       0   Thermal event interru=
ptsTHR:          0 &n=
bsp;        0   Threshold APIC=
interrupts
SPU:          0  &nb=
sp;       0   Spurious interruptsERR:          0*=
************************************ /etc/init.d/rtirq status *************=
******************* status:  started  PID CLS RTP=
RIO  NI PRI %CPU STAT COMMAND     
  352 FF      95   - 135  0.0 =
S<   IRQ-8        rtc&n=
bsp; 426 FF      90   - 130  0.3 S&=
lt;   IRQ-19       ohci1394, ohci13=
94  491 FF      85   - 125&nbsp=
; 0.0 S<   IRQ-1        i80=
42  490 FF      84   - 124&nbsp=
; 0.0 S<   IRQ-12       i8042
    5 FF      50   -&nbsp=
; 90  0.0 S<   softirq-high/0    &nbs=
p;      6 FF      50 =
  -  90  0.7 S<   softirq-timer/0  &n=
bsp;       7 FF      =
50   -  90  0.0 S<   softirq-net-tx/ =
         8 FF   &nbsp=
;  50   -  90  0.0 S<   softirq-net-r=
x/     
    9 FF      50   -&nbsp=
; 90  0.0 S<   softirq-block/0    &nb=
sp;    10 FF      50   -&n=
bsp; 90  0.0 S<   softirq-tasklet    =
     11 FF      50   =
-  90  0.0 S<   softirq-sched/0   &nb=
sp;     12 FF      50 &nbs=
p; -  90  0.0 S<   softirq-hrtimer   =
  
   13 FF      50   -  90&=
nbsp; 0.0 S<   softirq-rcu/0     &nbs=
p;     17 FF      50 &nbsp=
; -  90  0.0 S<   softirq-high/1   &n=
bsp;      18 FF      50&nb=
sp;  -  90  0.4 S<   softirq-timer/1 &nbsp=
;       19 FF      50=
   -  90  0.0 S<   softirq-net-tx/ &n=
bsp;   
   20 FF      50   -  90&=
nbsp; 0.0 S<   softirq-net-rx/      <=
br>   21 FF      50   -  =
90  0.0 S<   softirq-block/1    &nbsp=
;    22 FF      50   -&nbs=
p; 90  0.1 S<   softirq-tasklet    &n=
bsp;    23 FF      50   -&=
nbsp; 90  0.0 S<   softirq-sched/1   &nbsp=

   24 FF      50   -  90&=
nbsp; 0.0 S<   softirq-hrtimer      <=
br>   25 FF      50   -  =
90  0.0 S<   softirq-rcu/1     &=
nbsp;     99 FF      50 &n=
bsp; -  90  0.0 S<   IRQ-9    &n=
bsp;   acpi  395 FF      50&nbs=
p;  -  90  0.7 S<   IRQ-14   &nb=
sp;   ide0
  396 FF      50   -  90 =
0.0 S<   IRQ-15       ide1&=
nbsp; 446 FF      50   -  90  =
0.0 S<   IRQ-20       ehci_hcd:u=
sb1, uhci_hcd:usb2  461 FF      50 &=
nbsp; -  90  0.0 S<   IRQ-21    =
   uhci_hcd:usb3, HDA Intel
  468 FF      50   -  90 =
0.0 S<   IRQ-22       uhci_hcd:=
usb4  479 FF      50   -  =
90  0.0 S<   IRQ-23       u=
hci_hcd:usb5 3452 FF      50   =
-  90  0.0 S<   IRQ-17     =
  eth0 3685 FF      50   -=
  90  0.1 S<   IRQ-16     &=
nbsp; nvidia
Thank you again for your help.greets,Mysth-R-- * *******************************=
*********************************************************  {^=
_^} Mysth-R {^_^}
*  <=3D Aide Auditive =3D>** http://myspace.com/mysthr* http://myspace.com/aideauditive* http://mysthr.free.fr/Joomla  =3D&=
gt; Site d=E9di=E9 =E0 l'audio sous Fedora/PlanetCCRMA.
* *************************************************************************=
**************

------=_Part_22029_7345965.1207657628012--

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[LAU] : jack CPU % is going crazy !, Mysth-R, (Sat Apr 5, 5:27 pm)
Re: [LAU] : jack CPU % is going crazy !, Russell Hanaghan, (Sat Apr 5, 8:59 pm)
Re: [LAU] : jack CPU % is going crazy !, Mysth-R, (Sun Apr 6, 1:12 pm)
Re: [LAU] : jack CPU % is going crazy !, Mysth-R, (Sun Apr 6, 1:24 pm)
Re: [LAU] : jack CPU % is going crazy !, Pieter Palmers, (Mon Apr 7, 7:00 am)
Re: [LAU] : jack CPU % is going crazy !, Mysth-R, (Mon Apr 7, 9:43 am)
Re: [LAU] : jack CPU % is going crazy !, Pieter Palmers, (Mon Apr 7, 11:14 am)
Re: [LAU] : jack CPU % is going crazy !, Mysth-R, (Mon Apr 7, 2:41 pm)
Re: [LAU] : jack CPU % is going crazy !, Pieter Palmers, (Mon Apr 7, 3:06 pm)
Re: [LAU] : jack CPU % is going crazy !, Mysth-R, (Tue Apr 8, 12:27 pm)