[Xenomai-help] High Latency on PXA

All of lore.kernel.org
 help / color / mirror / Atom feed

* [Xenomai-help] High Latency on PXA
@ 2009-09-08 19:59 Felipe Castro
  2009-09-08 20:19 ` Gilles Chanteperdrix
  0 siblings, 1 reply; 13+ messages in thread
From: Felipe Castro @ 2009-09-08 19:59 UTC (permalink / raw)
  To: xenomai

[-- Attachment #1: Type: text/plain, Size: 2791 bytes --]

I'm running Xenomai 2.4.8 with kernel 2.6.27 on a Voipac development kit
with PXA270 processor.

Once applied a patch (Thanks Giles) everythink looks to run ok , but running
the latency test program i got:

root@domain.hid$ ./latency -p 3000
== Sampling period: 3000 us
== Test mode: periodic user-mode task
== All results in microseconds
warming up...
RTT|  00:00:01  (periodic user-mode task, 3000 us period, priority 99)
RTH|-----lat min|-----lat avg|-----lat max|-overrun|----lat best|---lat
worst
RTD|      14.153|      14.769|      39.999|       0|      14.153|
39.999
RTD|      14.153|      17.538|     209.538|       0|      14.153|
209.538
RTD|      14.153|      56.923|   13829.846|       4|      14.153|
13829.846
RTD|      13.846|      57.538|   13769.538|       8|      13.846|
13829.846
RTD|      14.153|      56.923|   13816.615|      12|      13.846|
13829.846
RTD|      14.153|      57.846|   13833.846|      16|      13.846|
13833.846
RTD|      14.153|      57.230|   13827.692|      20|      13.846|
13833.846
RTD|      14.153|      57.538|   13812.615|      24|      13.846|
13833.846
RTD|      13.846|      56.923|   13823.999|      28|      13.846|
13833.846
RTD|      14.153|      57.538|   13812.307|      32|      13.846|
13833.846
RTD|      14.153|     200.923|   13833.538|      36|      13.846|
13833.846
RTD|      13.846|     201.846|   14044.923|      40|      13.846|
14044.923
RTD|      13.846|     201.538|   14044.307|      44|      13.846|
14044.923
RTD|      13.846|     201.538|   14043.384|      48|      13.846|
14044.923

There are overrun and max. latency is to high ! Is there something to fix
it.

Running with tasks with more than 10ms there are overrun , and the results
looks better ...

root@domain.hid$ ./latency -p 10000
== Sampling period: 10000 us
== Test mode: periodic user-mode task
== All results in microseconds
warming up...
RTT|  00:00:01  (periodic user-mode task, 10000 us period, priority 99)
RTH|-----lat min|-----lat avg|-----lat max|-overrun|----lat best|---lat
worst
RTD|      13.538|      14.769|      19.999|       0|      13.538|
19.999
RTD|      13.538|      18.769|      81.538|       0|      13.538|
81.538
RTD|      13.230|      85.846|    6833.538|       0|      13.230|
6833.538
RTD|      13.538|      87.076|    6807.999|       0|      13.230|
6833.538
RTD|      13.230|      85.230|    6827.999|       0|      13.230|
6833.538
RTD|      13.538|      87.384|    6804.307|       0|      13.230|
6833.538
RTD|      13.230|      85.846|    6841.846|       0|      13.230|
6841.846
RTD|      13.230|      87.999|    6801.538|       0|      13.230|
6841.846
RTD|      13.230|      84.923|    6820.615|       0|      13.230|
6841.846
RTD|      13.230|      88.615|    6795.692|       0|      13.230|
6841.846



Thanks !x

[-- Attachment #2: Type: text/html, Size: 3061 bytes --]

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-08 19:59 [Xenomai-help] High Latency on PXA Felipe Castro
@ 2009-09-08 20:19 ` Gilles Chanteperdrix
  2009-09-09 15:03   ` Felipe Castro
  0 siblings, 1 reply; 13+ messages in thread
From: Gilles Chanteperdrix @ 2009-09-08 20:19 UTC (permalink / raw)
  To: Felipe Castro; +Cc: xenomai

Felipe Castro wrote:
> Spam detection software, running on the system "xenomai.org", has
> identified this incoming email as possible spam.  The original message
> has been attached to this so you can view it (if it isn't spam) or label
> similar future email.  If you have any questions, see
> the administrator of that system for details.

> 
> Content preview:  I'm running Xenomai 2.4.8 with kernel 2.6.27 on a Voipac development
>    kit with PXA270 processor. Once applied a patch (Thanks Giles) everythink
>    looks to run ok , but running the latency test program i got: [...] 
> 
> Content analysis details:   (4.8 points, 4.5 required)
> 
>  pts rule name              description
> ---- ---------------------- --------------------------------------------------
>  1.1 URIBL_RHS_DOB          Contains an URI of a new domain (Day Old Bread)
>                             [URIs: gna.org]
>  2.0 HTML_MESSAGE           BODY: HTML included in message
>  1.8 MIME_BASE64_TEXT       RAW: Message text disguised using base64 encoding
> 
> The original message was not completely plain text, and may be unsafe to
> open with some email clients; in particular, it may contain a virus,
> or confirm that your address can receive spam.  If you wish to view
> it, it may be safer to save it to a file and open it with an editor.
> 

Please avoid sending messages which look like spam, you will get faster
responses since I do read my spam box only once a day.

> I'm running Xenomai 2.4.8 with kernel 2.6.27 on a Voipac development kit
> with PXA270 processor.

Please switch to Xenomai 2.4.9.1.

> 
> Once applied a patch (Thanks Giles) everythink looks to run ok , but running
> the latency test program i got:

All patches have been included in 2.4.9.1. What is your status with
regard to floating points? Are you compiling with -msoft-float? It is
mandatory on PXA.


-- 
					    Gilles.


^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-08 20:19 ` Gilles Chanteperdrix
@ 2009-09-09 15:03   ` Felipe Castro
  2009-09-09 15:08     ` Gilles Chanteperdrix
  0 siblings, 1 reply; 13+ messages in thread
From: Felipe Castro @ 2009-09-09 15:03 UTC (permalink / raw)
  To: xenomai

[-- Attachment #1: Type: text/plain, Size: 4373 bytes --]

Yes , i compiled with msoft-float , anyway  I downloaded xenomai 2.4.9.1 and
recompile my kernel,

During the startup it was ok ,

[    1.340935] NetWinder Floating Point Emulator V0.97 (double precision)
[    1.344875] I-pipe: Domain Xenomai registered.
[    1.344970] Xenomai: hal/arm started.
[    1.375581] I-pipe: Domain IShield registered.
[    1.377746] Xenomai: real-time nucleus v2.4.9.1 (Big Bad Moon) loaded.

and running xeno-info i got:

root@domain.hid$ ./xeno-info
If some fields are empty or look unusual you may have an old version.
Compare to the current minimal requirements in Documentation/Changes.

BusyBox v1.6.1 (2009-07-15 08:44:38 BRT) multi-call binary
Linux vpac270 2.6.27-vpac1 #55 PREEMPT Wed Sep 9 06:07:03 BRT 2009 armv5tel
unknown

module-init-tools      found
Modules Loaded         xeno_native

But running latency test program:

root@domain.hid$ ./latency -p 3000
== Sampling period: 3000 us
== Test mode: periodic user-mode task
== All results in microseconds
[  360.030571] Xenomai: fatal: bug at kernel/xenomai/nucleus/shadow.c:1128
(!irqs_disabled_hw())
[  360.030608]  CPU  PID    PRI      TIMEOUT  STAT      NAME
[  360.030631]    0  0       -1      0        00500088  ROOT
[  360.030653] >  0  363      0      0        00300180  display-361
[  360.030678] Master time base: clock=1170102652
[  360.030697]
[  360.123559] [<c0021dec>] (show_stack+0x0/0x48) from [<c007e0a8>]
(xnshadow_harden+0x1b8/0x210)
[  360.149304] [<c007def0>] (xnshadow_harden+0x0/0x210) from [<c007e5c8>]
(xnshadow_wait_barrier+0xf8/0x108)
[  360.177811]  r8:00000018 r7:00000000 r6:c787bfb0 r5:c7a80810 r4:c787a000
[  360.197738] [<c007e4d0>] (xnshadow_wait_barrier+0x0/0x108) from
[<c007fb18>] (xnshadow_sys_barrier+0x14/0x18)
[  360.227281]  r6:c787a000 r5:c787bfb0 r4:00000001
[  360.241003] [<c007fb04>] (xnshadow_sys_barrier+0x0/0x18) from
[<c007feac>] (losyscall_event+0xbc/0x1ac)
[  360.268996] [<c007fdf0>] (losyscall_event+0x0/0x1ac) from [<c0066094>]
(__ipipe_dispatch_event+0xd0/0x1c8)
[  360.297788] [<c0065fc4>] (__ipipe_dispatch_event+0x0/0x1c8) from
[<c0022c8c>] (__ipipe_syscall_root+0x7c/0x110)
[  360.327847] [<c0022c10>] (__ipipe_syscall_root+0x0/0x110) from
[<c001e028>] (vector_swi+0x68/0xa8)

any sugestion ?

Thanks!!

On Tue, Sep 8, 2009 at 5:19 PM, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> Felipe Castro wrote:
> > Spam detection software, running on the system "xenomai.org", has
> > identified this incoming email as possible spam.  The original message
> > has been attached to this so you can view it (if it isn't spam) or label
> > similar future email.  If you have any questions, see
> > the administrator of that system for details.
>
> >
> > Content preview:  I'm running Xenomai 2.4.8 with kernel 2.6.27 on a
> Voipac development
> >    kit with PXA270 processor. Once applied a patch (Thanks Giles)
> everythink
> >    looks to run ok , but running the latency test program i got: [...]
> >
> > Content analysis details:   (4.8 points, 4.5 required)
> >
> >  pts rule name              description
> > ---- ----------------------
> --------------------------------------------------
> >  1.1 URIBL_RHS_DOB          Contains an URI of a new domain (Day Old
> Bread)
> >                             [URIs: gna.org]
> >  2.0 HTML_MESSAGE           BODY: HTML included in message
> >  1.8 MIME_BASE64_TEXT       RAW: Message text disguised using base64
> encoding
> >
> > The original message was not completely plain text, and may be unsafe to
> > open with some email clients; in particular, it may contain a virus,
> > or confirm that your address can receive spam.  If you wish to view
> > it, it may be safer to save it to a file and open it with an editor.
> >
>
> Please avoid sending messages which look like spam, you will get faster
> responses since I do read my spam box only once a day.
>
> > I'm running Xenomai 2.4.8 with kernel 2.6.27 on a Voipac development kit
> > with PXA270 processor.
>
> Please switch to Xenomai 2.4.9.1.
>
> >
> > Once applied a patch (Thanks Giles) everythink looks to run ok , but
> running
> > the latency test program i got:
>
> All patches have been included in 2.4.9.1. What is your status with
> regard to floating points? Are you compiling with -msoft-float? It is
> mandatory on PXA.
>
>
> --
>                                             Gilles.
>

[-- Attachment #2: Type: text/html, Size: 5326 bytes --]

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-09 15:03   ` Felipe Castro
@ 2009-09-09 15:08     ` Gilles Chanteperdrix
  2009-09-09 19:06       ` Felipe Castro
  0 siblings, 1 reply; 13+ messages in thread
From: Gilles Chanteperdrix @ 2009-09-09 15:08 UTC (permalink / raw)
  To: Felipe Castro; +Cc: xenomai

Felipe Castro wrote:
> Yes , i compiled with msoft-float , anyway  I downloaded xenomai 2.4.9.1
> and recompile my kernel,
> 
> During the startup it was ok ,
> 
> [    1.340935] NetWinder Floating Point Emulator V0.97 (double precision)
> [    1.344875] I-pipe: Domain Xenomai registered.
> [    1.344970] Xenomai: hal/arm started.
> [    1.375581] I-pipe: Domain IShield registered.
> [    1.377746] Xenomai: real-time nucleus v2.4.9.1 (Big Bad Moon) loaded.
> 
> and running xeno-info i got:
> 
> root@domain.hid$ ./xeno-info
> If some fields are empty or look unusual you may have an old version.
> Compare to the current minimal requirements in Documentation/Changes.
>  
> BusyBox v1.6.1 (2009-07-15 08:44:38 BRT) multi-call binary
> Linux vpac270 2.6.27-vpac1 #55 PREEMPT Wed Sep 9 06:07:03 BRT 2009
> armv5tel unknown
>  
> module-init-tools      found
> Modules Loaded         xeno_native
> 
> But running latency test program:
> 
> root@domain.hid$ ./latency -p 3000
> == Sampling period: 3000 us
> == Test mode: periodic user-mode task
> == All results in microseconds
> [  360.030571] Xenomai: fatal: bug at
> kernel/xenomai/nucleus/shadow.c:1128 (!irqs_disabled_hw())
> [  360.030608]  CPU  PID    PRI      TIMEOUT  STAT      NAME
> [  360.030631]    0  0       -1      0        00500088  ROOT
> [  360.030653] >  0  363      0      0        00300180  display-361
> [  360.030678] Master time base: clock=1170102652
> [  360.030697]
> [  360.123559] [<c0021dec>] (show_stack+0x0/0x48) from [<c007e0a8>]
> (xnshadow_harden+0x1b8/0x210)
> [  360.149304] [<c007def0>] (xnshadow_harden+0x0/0x210) from
> [<c007e5c8>] (xnshadow_wait_barrier+0xf8/0x108)
> [  360.177811]  r8:00000018 r7:00000000 r6:c787bfb0 r5:c7a80810 r4:c787a000
> [  360.197738] [<c007e4d0>] (xnshadow_wait_barrier+0x0/0x108) from
> [<c007fb18>] (xnshadow_sys_barrier+0x14/0x18)
> [  360.227281]  r6:c787a000 r5:c787bfb0 r4:00000001
> [  360.241003] [<c007fb04>] (xnshadow_sys_barrier+0x0/0x18) from
> [<c007feac>] (losyscall_event+0xbc/0x1ac)
> [  360.268996] [<c007fdf0>] (losyscall_event+0x0/0x1ac) from
> [<c0066094>] (__ipipe_dispatch_event+0xd0/0x1c8)
> [  360.297788] [<c0065fc4>] (__ipipe_dispatch_event+0x0/0x1c8) from
> [<c0022c8c>] (__ipipe_syscall_root+0x7c/0x110)
> [  360.327847] [<c0022c10>] (__ipipe_syscall_root+0x0/0x110) from
> [<c001e028>] (vector_swi+0x68/0xa8)
> 
> any sugestion ?

You are probably not using the I-pipe patch which comes with Xenomai
2.4.9.1,
type:
cat /proc/ipipe/version

-- 
                                          Gilles



^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-09 15:08     ` Gilles Chanteperdrix
@ 2009-09-09 19:06       ` Felipe Castro
  2009-09-09 19:23         ` Gilles Chanteperdrix
  2009-09-09 21:40         ` Gilles Chanteperdrix
  0 siblings, 2 replies; 13+ messages in thread
From: Felipe Castro @ 2009-09-09 19:06 UTC (permalink / raw)
  To: xenomai

[-- Attachment #1: Type: text/plain, Size: 2951 bytes --]

i applied the adeos path that comes with xenomai 2.4.9.1 overriding the old
one that I was using, is that a problem ?

the result is :

root@domain.hid:~$ cat /proc/ipipe/version
1.12-02


On Wed, Sep 9, 2009 at 12:08 PM, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> Felipe Castro wrote:
> > Yes , i compiled with msoft-float , anyway  I downloaded xenomai 2.4.9.1
> > and recompile my kernel,
> >
> > During the startup it was ok ,
> >
> > [    1.340935] NetWinder Floating Point Emulator V0.97 (double precision)
> > [    1.344875] I-pipe: Domain Xenomai registered.
> > [    1.344970] Xenomai: hal/arm started.
> > [    1.375581] I-pipe: Domain IShield registered.
> > [    1.377746] Xenomai: real-time nucleus v2.4.9.1 (Big Bad Moon) loaded.
> >
> > and running xeno-info i got:
> >
> > root@domain.hid$ ./xeno-info
> > If some fields are empty or look unusual you may have an old version.
> > Compare to the current minimal requirements in Documentation/Changes.
> >
> > BusyBox v1.6.1 (2009-07-15 08:44:38 BRT) multi-call binary
> > Linux vpac270 2.6.27-vpac1 #55 PREEMPT Wed Sep 9 06:07:03 BRT 2009
> > armv5tel unknown
> >
> > module-init-tools      found
> > Modules Loaded         xeno_native
> >
> > But running latency test program:
> >
> > root@domain.hid$ ./latency -p 3000
> > == Sampling period: 3000 us
> > == Test mode: periodic user-mode task
> > == All results in microseconds
> > [  360.030571] Xenomai: fatal: bug at
> > kernel/xenomai/nucleus/shadow.c:1128 (!irqs_disabled_hw())
> > [  360.030608]  CPU  PID    PRI      TIMEOUT  STAT      NAME
> > [  360.030631]    0  0       -1      0        00500088  ROOT
> > [  360.030653] >  0  363      0      0        00300180  display-361
> > [  360.030678] Master time base: clock=1170102652
> > [  360.030697]
> > [  360.123559] [<c0021dec>] (show_stack+0x0/0x48) from [<c007e0a8>]
> > (xnshadow_harden+0x1b8/0x210)
> > [  360.149304] [<c007def0>] (xnshadow_harden+0x0/0x210) from
> > [<c007e5c8>] (xnshadow_wait_barrier+0xf8/0x108)
> > [  360.177811]  r8:00000018 r7:00000000 r6:c787bfb0 r5:c7a80810
> r4:c787a000
> > [  360.197738] [<c007e4d0>] (xnshadow_wait_barrier+0x0/0x108) from
> > [<c007fb18>] (xnshadow_sys_barrier+0x14/0x18)
> > [  360.227281]  r6:c787a000 r5:c787bfb0 r4:00000001
> > [  360.241003] [<c007fb04>] (xnshadow_sys_barrier+0x0/0x18) from
> > [<c007feac>] (losyscall_event+0xbc/0x1ac)
> > [  360.268996] [<c007fdf0>] (losyscall_event+0x0/0x1ac) from
> > [<c0066094>] (__ipipe_dispatch_event+0xd0/0x1c8)
> > [  360.297788] [<c0065fc4>] (__ipipe_dispatch_event+0x0/0x1c8) from
> > [<c0022c8c>] (__ipipe_syscall_root+0x7c/0x110)
> > [  360.327847] [<c0022c10>] (__ipipe_syscall_root+0x0/0x110) from
> > [<c001e028>] (vector_swi+0x68/0xa8)
> >
> > any sugestion ?
>
> You are probably not using the I-pipe patch which comes with Xenomai
> 2.4.9.1,
> type:
> cat /proc/ipipe/version
>
> --
>                                           Gilles
>
>

[-- Attachment #2: Type: text/html, Size: 3772 bytes --]

^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-09 19:06       ` Felipe Castro
@ 2009-09-09 19:23         ` Gilles Chanteperdrix
  2009-09-09 21:40         ` Gilles Chanteperdrix
  1 sibling, 0 replies; 13+ messages in thread
From: Gilles Chanteperdrix @ 2009-09-09 19:23 UTC (permalink / raw)
  To: Felipe Castro; +Cc: xenomai

Felipe Castro wrote:
> i applied the adeos path that comes with xenomai 2.4.9.1 overriding the old
> one that I was using, is that a problem ?

Yes, please start from scratch. And I suspect you may not be running the
kernel you think you are running since it is a 2.6.27, and there is no
I-pipe patch for Linux 2.6.27 in xenomai 2.4.9.1.

> 
> the result is :
> 
> root@domain.hid:~$ cat /proc/ipipe/version
> 1.12-02

xenomai 2.4.9.1 ships with the following I-pipe patches for ARM:
adeos-ipipe-2.6.28-arm-1.12-05
adeos-ipipe-2.6.29-arm-1.13-03
adeos-ipipe-2.6.30-arm-1.14-02

-- 
					    Gilles.


^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-09 19:06       ` Felipe Castro
  2009-09-09 19:23         ` Gilles Chanteperdrix
@ 2009-09-09 21:40         ` Gilles Chanteperdrix
  2009-09-10 17:28           ` Felipe Castro
  1 sibling, 1 reply; 13+ messages in thread
From: Gilles Chanteperdrix @ 2009-09-09 21:40 UTC (permalink / raw)
  To: Felipe Castro; +Cc: xenomai

Felipe Castro wrote:
> i applied the adeos path that comes with xenomai 2.4.9.1 overriding the old
> one that I was using, is that a problem ?
> 
> the result is :
> 
> root@domain.hid:~$ cat /proc/ipipe/version
> 1.12-02
> 
> 
> On Wed, Sep 9, 2009 at 12:08 PM, Gilles Chanteperdrix <
> gilles.chanteperdrix@xenomai.org> wrote:
> 
>> Felipe Castro wrote:
>>> Yes , i compiled with msoft-float , anyway  I downloaded xenomai 2.4.9.1
>>> and recompile my kernel,
>>>
>>> During the startup it was ok ,
>>>
>>> [    1.340935] NetWinder Floating Point Emulator V0.97 (double precision)

By the way, you should disable the floating point emulator, this way you
would be sure that the user-space code is actually using soft-floats.

-- 
					    Gilles.


^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-09 21:40         ` Gilles Chanteperdrix
@ 2009-09-10 17:28           ` Felipe Castro
  2009-09-10 17:45             ` Gilles Chanteperdrix
  0 siblings, 1 reply; 13+ messages in thread
From: Felipe Castro @ 2009-09-10 17:28 UTC (permalink / raw)
  To: xenomai


[-- Attachment #1.1: Type: text/plain, Size: 1489 bytes --]

I'm a little bit lost about this issues cause when running the xenomai 2.4.8
with the patch that you send me it worked, but with new version it doesn't
work.

Acctually i using the kernel 2.6.27 and i understand what you told me , in
version 2.4.9.1 there aren't a patch to this kernel version, neither xenomai
2.4.8.  :S

The problem is, I'm using a Voipac development kit, and the have only the
kernel 2.6.27 BSP.

So , i tried to disable the float point emulation but the kernel didn't load
OK .

I attached a log file, I hope it can help .

Thank you

On Wed, Sep 9, 2009 at 6:40 PM, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> Felipe Castro wrote:
> > i applied the adeos path that comes with xenomai 2.4.9.1 overriding the
> old
> > one that I was using, is that a problem ?
> >
> > the result is :
> >
> > root@domain.hid:~$ cat /proc/ipipe/version
> > 1.12-02
> >
> >
> > On Wed, Sep 9, 2009 at 12:08 PM, Gilles Chanteperdrix <
> > gilles.chanteperdrix@xenomai.org> wrote:
> >
> >> Felipe Castro wrote:
> >>> Yes , i compiled with msoft-float , anyway  I downloaded xenomai
> 2.4.9.1
> >>> and recompile my kernel,
> >>>
> >>> During the startup it was ok ,
> >>>
> >>> [    1.340935] NetWinder Floating Point Emulator V0.97 (double
> precision)
>
> By the way, you should disable the floating point emulator, this way you
> would be sure that the user-space code is actually using soft-floats.
>
> --
>                                             Gilles.
>

[-- Attachment #1.2: Type: text/html, Size: 2093 bytes --]

[-- Attachment #2: log --]
[-- Type: application/octet-stream, Size: 23901 bytes --]

//--------------------------------------------------------------------------------------//
//											//
// KERNEL STARTING									//
//											//
// with NetWinder Floating Point Emulator V0.97 (extended precision)			//
//--------------------------------------------------------------------------------------//

U-Boot code: 5C010000 -> 5C01B6A0  BSS: -> 5C01EAF0
RAM Configuration:
Bank #0: a0000000 128 MB
Bank #1: 80000000  0 kB
Flash: 32 MB
Hit any key to stop autoboot:  0 
Writing TAGS to DRAM at 0xa0000100
## Starting application at 0x00020000 ...
Uncompressing Linux...................................................................................................... done, booting the kernel.
[    0.000000] Linux version 2.6.27-vpac1 (root@debian) (gcc version 3.4.1) #62 Thu Sep 10 10:31:40 BRT 2009
[    0.000000] CPU: XScale-PXA270 [69054117] revision 7 (ARMv5TE), cr=0000397f
[    0.000000] Machine: Voipac PXA270 Module
[    0.000000] Memory policy: ECC disabled, Data cache writeback
[    0.000000] Run Mode clock: 208.00MHz (*16)
[    0.000000] Turbo Mode clock: 312.00MHz (*1.5, active)
[    0.000000] Memory clock: 208.00MHz (/2)
[    0.000000] System bus clock: 208.00MHz 
[    0.000000] CPU0: D VIVT undefined 5 cache
[    0.000000] CPU0: I cache: 32768 bytes, associativity 32, 32 byte lines, 32 sets
[    0.000000] CPU0: D cache: 32768 bytes, associativity 32, 32 byte lines, 32 sets
[    0.000000] Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 32512
[    0.000000] Kernel command line: root=/dev/mtdblock2 rootfstype=jffs2 console=ttyS0,38400
[    0.000000] PID hash table entries: 512 (order: 9, 2048 bytes)
[    0.960989] I-pipe 1.12-02: pipeline enabled.
[    0.961244] Console: colour dummy device 80x30
[    0.961573] Dentry cache hash table entries: 16384 (order: 4, 65536 bytes)
[    0.962340] Inode-cache hash table entries: 8192 (order: 3, 32768 bytes)
[    0.979420] Memory: 128MB = 128MB total
[    0.979514] Memory: 126320KB available (2932K code, 408K data, 88K init)
[    0.981615] Calibrating delay loop... 311.29 BogoMIPS (lpj=1556480)
[    1.192212] Mount-cache hash table entries: 512
[    1.194133] CPU: Testing write buffer coherency: ok
[    1.215653] net_namespace: 288 bytes
[    1.227451] NET: Registered protocol family 16
[    1.496057] SCSI subsystem initialized
[    1.530238] I2C: i2c-1: PXA I2C adapter
[    1.542991] I2C: i2c-0: PXA I2C adapter
[    1.632625] NET: Registered protocol family 2
[    1.733876] IP route cache hash table entries: 1024 (order: 0, 4096 bytes)
[    1.740862] TCP established hash table entries: 4096 (order: 3, 32768 bytes)
[    1.741949] TCP bind hash table entries: 4096 (order: 2, 16384 bytes)
[    1.742454] TCP: Hash tables configured (established 4096 bind 4096)
[    1.742809] TCP reno registered
[    1.774448] NET: Registered protocol family 1
[    1.783436] NetWinder Floating Point Emulator V0.97 (extended precision)
[    1.803933] I-pipe: Domain Xenomai registered.
[    1.804467] Xenomai: hal/arm started.
[    1.959616] I-pipe: Domain IShield registered.
[    1.972511] Xenomai: real-time nucleus v2.4.9.1 (Big Bad Moon) loaded.
[    1.996556] Xenomai: starting native API services.
[    1.996926] Xenomai: starting POSIX services.
[    2.003003] Xenomai: starting RTDM services.
[    2.019515] JFFS2 version 2.2. (NAND) © 2001-2006 Red Hat, Inc.
[    2.035190] msgmni has been set to 247
[    2.036762] io scheduler noop registered
[    2.037089] io scheduler deadline registered (default)
[    2.045322] pxa2xx-fb pxa2xx-fb: machine LCCR0 setting contains illegal bits: 00300878
[    2.073681] Console: switching to colour frame buffer device 80x30
[    2.458029] pxa2xx-uart.0: ttyS0 at MMIO 0x40100000 (irq = 22) is a FFUART
[    2.458595] console [ttyS0] enabled
[    3.282818] pxa2xx-uart.1: ttyS1 at MMIO 0x40200000 (irq = 21) is a BTUART
[    3.313044] pxa2xx-uart.2: ttyS2 at MMIO 0x40700000 (irq = 20) is a STUART
[    3.347432] dm9000 Ethernet Driver, V1.31
[    3.481591] dm9000 dm9000.0: eth%d: Invalid ethernet MAC address. Please set using ifconfig
[    3.518814] eth0: dm9000e at c8a1e300,c8a32304 IRQ 178 MAC: 00:00:00:00:00:00 (chip)
[    3.552678] Driver 'sd' needs updating - please use bus_type methods
[    3.589383] Probing vpac270dimm at physical address 0x00000000 (16-bit bankwidth)
[    3.613352] vpac270dimm: Found 1 x16 devices at 0x0 in 16-bit bank
[    3.633148]  Intel/Sharp Extended Query Table at 0x010A
[    3.649528]  Intel/Sharp Extended Query Table at 0x010A
[    3.665876]  Intel/Sharp Extended Query Table at 0x010A
[    3.682545]  Intel/Sharp Extended Query Table at 0x010A
[    3.699039]  Intel/Sharp Extended Query Table at 0x010A
[    3.715397] Using buffer write method
[    3.726982] Using auto-unlock on power-up/resume
[    3.741448] cfi_cmdset_0001: Erase suspend on write enabled
[    3.765259] RedBoot partition parsing not available
[    3.785770] cmdlinepart partition parsing not available
[    3.802196] Creating 3 MTD partitions on "vpac270dimm":
[    3.818935] 0x00000000-0x00020000 : "Bootloader"
[    3.856775] 0x00020000-0x00220000 : "Kernel"
[    3.895392] 0x00220000-0x02000000 : "Filesystem"
[    3.963921] mice: PS/2 mouse device common for all mice
[    4.040664] rtc-ds1307 0-0068: rtc core: registered ds1339 as rtc0
[    4.073396] sa1100-rtc sa1100-rtc: rtc core: registered sa1100-rtc as rtc1
[    4.102104] i2c /dev entries driver
[    4.142153] Registered led device: vpac270:orange
[    4.163565] TCP cubic registered
[    4.174612] NET: Registered protocol family 17
[    4.188869] ieee80211: 802.11 data/management/control stack, git-1.1.13
[    4.209525] ieee80211: Copyright (C) 2004-2005 Intel Corporation <jketreno@linux.intel.com>
[    4.236092] XScale iWMMXt coprocessor detected.
[    4.269197] rtc-ds1307 0-0068: setting system clock to 2000-01-15 22:22:44 UTC (947974964)
[    5.001592] VFS: Mounted root (jffs2 filesystem).
[    5.018550] Freeing init memory: 88K
init started: BusyBox v1.6.1 (2009-07-15 08:44:38 BRT) multi-call binary
mount: mounting usbfs on /proc/bus/usb failed
Starting syslogd/klogd: start-stop-daemon: lseek: Invalid argument
start-stop-daemon: lseek: Invalid argument
done
Setting up IP spoofing protection: rp_filter.
Configuring network interfaces... ip: either "dev" is duplicate, or "ether" is garbage
ip: RTNETLINK answers: Network is unreachable
done.
starting Busybox Telnet: telnetd... start-stop-daemon: lseek: Invalid argument
done.

//--------------------------------------------------------------------------------------//
//											//
// Running Latency									//
//											//
//--------------------------------------------------------------------------------------//


root@vpac270:/usr/xenomai$ ls
bin      include  lib      sbin     share
root@vpac270:/usr/xenomai$ cd bin
root@vpac270:/usr/xenomai/bin$ ls
clocktest    klatency     rtcansend    xeno-config  xeno-test
cyclictest   latency      switchbench  xeno-info
irqloop      rtcanrecv    switchtest   xeno-load
root@vpac270:/usr/xenomai/bin$ ./latency -p 3000
== Sampling period: 3000 us
== Test mode: periodic user-mode task
== All results in microseconds
[   33.003266] Xenomai: fatal: bug at kernel/xenomai/nucleus/shadow.c:1128 (!irqs_disabled_hw())
[   33.003311]  CPU  PID    PRI      TIMEOUT  STAT      NAME
[   33.003338]    0  0       -1      0        00500088  ROOT
[   33.003365] >  0  314      0      0        00300180  display-312
[   33.003394] Master time base: clock=107260786
[   33.003417] 
[   33.096027] [<c0022ef0>] (show_stack+0x0/0x4c) from [<c0087f44>] (xnshadow_harden+0x1fc/0x258)
[   33.121800] [<c0087d48>] (xnshadow_harden+0x0/0x258) from [<c0088558>] (xnshadow_wait_barrier+0x110/0x120)
[   33.150565]  r8:00000018 r7:00000000 r6:c7c33fb0 r5:c8986c10 r4:c7c32000
[   33.170491] [<c0088448>] (xnshadow_wait_barrier+0x0/0x120) from [<c0089e78>] (xnshadow_sys_barrier+0x18/0x1c)
[   33.200036]  r6:c7c32000 r5:c7c33fb0 r4:00000001
[   33.213758] [<c0089e60>] (xnshadow_sys_barrier+0x0/0x1c) from [<c008a308>] (losyscall_event+0xc0/0x1b0)
[   33.241751] [<c008a248>] (losyscall_event+0x0/0x1b0) from [<c006a878>] (__ipipe_dispatch_event+0xdc/0x1e8)
[   33.270568] [<c006a79c>] (__ipipe_dispatch_event+0x0/0x1e8) from [<c0023e4c>] (__ipipe_syscall_root+0x80/0x120)
[   33.300628] [<c0023dcc>] (__ipipe_syscall_root+0x0/0x120) from [<c001f010>] (vector_swi+0x70/0xb0)
[   33.327329] I-pipe tracer log (100 points):
[   33.339756]    #  func                    0 ipipe_trace_panic_freeze+0x10 (xnshadow_harden+0x184)
[   33.366196]  | #  end     0x80000000     -4 __sched_text_start+0x5d0 (xnshadow_harden+0x90)
[   33.391085]  | #  [    1] init    -1    -21 xnpod_schedule+0xb4 (xnpod_schedule_handler+0x54)
[   33.416518]  | #  func                  -24 xnpod_schedule+0x14 (xnpod_schedule_handler+0x54)
[   33.441924]  | #  func                  -26 xnpod_schedule_handler+0x10 (__ipipe_sync_stage+0x1b4)
[   33.468623]  | +  func                  -30 __ipipe_sync_stage+0x10 (ipipe_suspend_domain+0xa4)
[   33.494546]  | +  func                  -33 ipipe_suspend_domain+0x10 (__ipipe_walk_pipeline+0x74)
[   33.521245]  |   +func                  -36 __ipipe_walk_pipeline+0x10 (__ipipe_restore_pipeline_head+0xc4)
[   33.550270]  | * +func                  -38 __ipipe_restore_pipeline_head+0x10 (gatekeeper_thread+0x180)
[   33.578521]  | * +func                  -42 __ipipe_set_irq_pending+0x10 (__ipipe_dispatch_wired+0x7c)
[   33.606254]  | * +func                  -46 __ipipe_dispatch_wired+0x10 (__ipipe_handle_irq+0xa4)
[   33.632694]  | * +func                  -50 __ipipe_handle_irq+0x10 (ipipe_trigger_irq+0x7c)
[   33.657842]  | * +func                  -54 ipipe_trigger_irq+0x10 (xnpod_schedule+0x34)
[   33.681957]  | * +func                  -57 xnpod_schedule+0x14 (gatekeeper_thread+0x164)
[   33.706330]  | * +func                  -59 disengage_irq_shield+0x10 (gatekeeper_thread+0x160)
[   33.732253]  | * +[  314] display- 0    -66 xnpod_resume_thread+0x70 (gatekeeper_thread+0x15c)
[   33.757919]  | * +func                  -69 xnpod_resume_thread+0x14 (gatekeeper_thread+0x15c)
[   33.783583]  |   +begin   0x80000000    -71 gatekeeper_thread+0x140 (kthread+0x58)
[   33.806172]  |   +end     0x80000000    -74 __ipipe_restore_pipeline_head+0xd8 (rpi_pop+0x374)
[   33.831838]  | * +func                  -76 __ipipe_restore_pipeline_head+0x10 (rpi_pop+0x374)
[   33.857502]  |   +begin   0x80000000    -79 rpi_pop+0x3c (gatekeeper_thread+0x120)
[   33.880067]      +func                  -82 rpi_pop+0x14 (gatekeeper_thread+0x120)
[   33.902630]      +func                  -86 kthread_should_stop+0x10 (gatekeeper_thread+0xf8)
[   33.928037]      +func                  -89 finish_task_switch+0x10 (__sched_text_start+0x644)
[   33.953703]  |   +end     0x80000000    -92 __sched_text_start+0x5d0 (gatekeeper_thread+0xf4)
[   33.979109]  |   +func                  -97 iwmmxt_do+0x10 (notifier_call_chain+0x4c)
[   34.002474]  |   +func                  -99 nwfpe_notify+0x10 (notifier_call_chain+0x4c)
[   34.026614]  |   +func                 -101 notifier_call_chain+0x10 (__atomic_notifier_call_chain+0x20)
[   34.054864]  |   +func                 -103 __atomic_notifier_call_chain+0x14 (atomic_notifier_call_chain+0x24)
[   34.084923]  |   +func                 -105 atomic_notifier_call_chain+0x14 (__switch_to+0x30)
[   34.110588]  |   +begin   0x80000000   -109 __sched_text_start+0x3cc (xnshadow_harden+0x90)
[   34.135477]  |   +end     0x80000001   -111 __ipipe_dispatch_event+0x1c8 (__sched_text_start+0x3ac)
[   34.162436]  |   +begin   0x80000001   -114 __ipipe_dispatch_event+0x104 (__sched_text_start+0x3ac)
[   34.189393]  |   +end     0x80000000   -116 __ipipe_restore_pipeline_head+0xd8 (schedule_event+0xa24)
[   34.216868]  | * +func                 -119 __ipipe_restore_pipeline_head+0x10 (schedule_event+0xa24)
[   34.244341]  |   +begin   0x80000000   -122 schedule_event+0x64c (__ipipe_dispatch_event+0xdc)
[   34.270007]      +func                 -126 schedule_event+0x14 (__ipipe_dispatch_event+0xdc)
[   34.295413]  |   +end     0x80000001   -128 __ipipe_dispatch_event+0xc4 (__sched_text_start+0x3ac)
[   34.322113]  |   +begin   0x80000001   -130 __ipipe_dispatch_event+0x40 (__sched_text_start+0x3ac)
[   34.348812]      +func                 -133 __ipipe_dispatch_event+0x14 (__sched_text_start+0x3ac)
[   34.375510]  |   +end     0x80000000   -135 __ipipe_unstall_root+0x64 (__sched_text_start+0x354)
[   34.401693]  |   #begin   0x80000000   -137 __ipipe_unstall_root+0x34 (__sched_text_start+0x354)
[   34.427874]      #func                 -139 __ipipe_unstall_root+0x10 (__sched_text_start+0x354)
[   34.454056]      #func                 -144 pick_next_task_rt+0x10 (__sched_text_start+0x1e8)
[   34.479464]      #func                 -147 check_spread+0x10 (put_prev_task_fair+0x3c)
[   34.503319]      #func                 -149 put_prev_task_fair+0x10 (__sched_text_start+0x1b0)
[   34.528985]      #func                 -154 calc_delta_mine+0x14 (update_curr+0xcc)
[   34.551806]      #func                 -156 update_curr+0x14 (dequeue_task_fair+0x2c)
[   34.575146]      #func                 -159 dequeue_task_fair+0x10 (dequeue_task+0x124)
[   34.599001]      #func                 -165 dequeue_task+0x14 (deactivate_task+0x2c)
[   34.622082]      #func                 -167 deactivate_task+0x10 (__sched_text_start+0x190)
[   34.646972]  |   #end     0x80000001   -170 __sched_text_start+0x124 (xnshadow_harden+0x90)
[   34.671861]  |   #begin   0x80000001   -172 __sched_text_start+0x10c (xnshadow_harden+0x90)
[   34.696752]      #func                 -175 sched_clock+0x10 (sched_clock_cpu+0x30)
[   34.719574]      #func                 -177 sched_clock_cpu+0x10 (__sched_text_start+0xe0)
[   34.744204]  |   #end     0x80000001   -179 __sched_text_start+0xd4 (xnshadow_harden+0x90)
[   34.768836]  |   +begin   0x80000001   -182 __sched_text_start+0xb4 (xnshadow_harden+0x90)
[   34.793468]      +func                 -185 ipipe_check_context+0x10 (__sched_text_start+0x3c)
[   34.819132]      +func                 -187 __sched_text_start+0x14 (xnshadow_harden+0x90)
[   34.843763]  |   +end     0x80000000   -189 __ipipe_unstall_root+0x64 (__ipipe_restore_root+0x64)
[   34.870206]  |   #begin   0x80000000   -191 __ipipe_unstall_root+0x34 (__ipipe_restore_root+0x64)
[   34.896645]      #func                 -194 __ipipe_unstall_root+0x10 (__ipipe_restore_root+0x64)
[   34.923085]      #func                 -196 __ipipe_restore_root+0x10 (__wake_up_sync+0xb4)
[   34.947976]  |   #end     0x80000001   -198 __ipipe_restore_root+0x58 (try_to_wake_up+0x118)
[   34.973124]  |   #begin   0x80000001   -201 __ipipe_restore_root+0x38 (try_to_wake_up+0x118)
[   34.998272]      #func                 -204 __ipipe_restore_root+0x10 (try_to_wake_up+0x118)
[   35.023420]  |   #end     0x80000001   -206 resched_task+0x4c (check_preempt_wakeup+0x178)
[   35.048051]  |   #begin   0x80000001   -208 resched_task+0x34 (check_preempt_wakeup+0x178)
[   35.072682]      #func                 -210 resched_task+0x10 (check_preempt_wakeup+0x178)
[   35.097314]      #func                 -213 calc_delta_mine+0x14 (update_curr+0xcc)
[   35.120136]      #func                 -216 update_curr+0x14 (check_preempt_wakeup+0x4c)
[   35.144250]      #func                 -219 sched_clock+0x10 (sched_clock_cpu+0x30)
[   35.167072]      #func                 -221 sched_clock_cpu+0x10 (check_preempt_wakeup+0x3c)
[   35.192221]      #func                 -224 check_preempt_wakeup+0x14 (try_to_wake_up+0xe8)
[   35.217111]      #func                 -226 inc_nr_running+0x10 (activate_task+0x34)
[   35.240191]  |   #end     0x80000001   -228 __ipipe_restore_root+0x58 (hrtimer_start+0x130)
[   35.265082]  |   #begin   0x80000001   -231 __ipipe_restore_root+0x38 (hrtimer_start+0x130)
[   35.289970]      #func                 -233 __ipipe_restore_root+0x10 (hrtimer_start+0x130)
[   35.314861]      #func                 -236 rb_insert_color+0x10 (enqueue_hrtimer+0x148)
[   35.339000]      #func                 -239 enqueue_hrtimer+0x10 (hrtimer_start+0x118)
[   35.362598]  |   #end     0x80000001   -243 hrtimer_start+0x68 (__enqueue_rt_entity+0x104)
[   35.387229]  |   #begin   0x80000001   -247 hrtimer_start+0x48 (__enqueue_rt_entity+0x104)
[   35.411861]      #func                 -249 hrtimer_start+0x14 (__enqueue_rt_entity+0x104)
[   35.436491]      #func                 -253 ktime_add_safe+0x18 (hrtimer_forward+0x14c)
[   35.460348]      #func                 -258 ktime_add_ns+0x1c (hrtimer_forward+0xf8)
[   35.483428]      #func                 -264 ktime_divns+0x10 (hrtimer_forward+0xd0)
[   35.506251]      #func                 -268 hrtimer_forward+0x18 (__enqueue_rt_entity+0xf0)
[   35.531139]      #func                 -272 set_normalized_timespec+0x10 (ktime_get_ts+0x58)
[   35.556289]      #func                 -278 pxa_read_oscr+0x10 (getnstimeofday+0x50)
[   35.579369]      #func                 -282 getnstimeofday+0x14 (ktime_get_ts+0x30)
[   35.602191]      #func                 -285 ktime_get_ts+0x10 (ktime_get+0x28)
[   35.623720]      #func                 -288 ktime_get+0x18 (__enqueue_rt_entity+0xd4)
[   35.647060]      #func                 -293 __enqueue_rt_entity+0x14 (enqueue_task_rt+0x3c)
[   35.671949]      #func                 -297 dequeue_rt_stack+0x10 (enqueue_task_rt+0x30)
[   35.696064]      #func                 -301 enqueue_task_rt+0x10 (enqueue_task+0x3c)
[   35.719145]      #func                 -304 enqueue_task+0x10 (activate_task+0x2c)
[   35.741709]      #func                 -306 activate_task+0x10 (try_to_wake_up+0xcc)
[   35.764789]      #func                 -310 sched_clock+0x10 (sched_clock_cpu+0x30)
[   35.787610]      #func                 -312 sched_clock_cpu+0x10 (try_to_wake_up+0xb8)
[   35.811209]  |   #end     0x80000001   -315 task_rq_lock+0x54 (try_to_wake_up+0x3c)


//--------------------------------------------------------------------------------------//
//											//
// KERNEL STARTING									//
//											//
// without NetWinder Floating Point Emulator V0.97 (extended precision)			//
//--------------------------------------------------------------------------------------//


U-Boot 1.1.2 (Jun 20 2007 - 14:11:13)

U-Boot code: 5C010000 -> 5C01B6A0  BSS: -> 5C01EAF0
RAM Configuration:
Bank #0: a0000000 128 MB
Bank #1: 80000000  0 kB
Flash: 32 MB
Hit any key to stop autoboot:  0 
Writing TAGS to DRAM at 0xa0000100
## Starting application at 0x00020000 ...
Uncompressing Linux.................................................................................................... done, booting the kernel.
[    0.000000] Linux version 2.6.27-vpac1 (root@debian) (gcc version 3.4.1) #63 Thu Sep 10 11:04:45 BRT 2009
[    0.000000] CPU: XScale-PXA270 [69054117] revision 7 (ARMv5TE), cr=0000397f
[    0.000000] Machine: Voipac PXA270 Module
[    0.000000] Memory policy: ECC disabled, Data cache writeback
[    0.000000] Run Mode clock: 208.00MHz (*16)
[    0.000000] Turbo Mode clock: 312.00MHz (*1.5, active)
[    0.000000] Memory clock: 208.00MHz (/2)
[    0.000000] System bus clock: 208.00MHz 
[    0.000000] CPU0: D VIVT undefined 5 cache
[    0.000000] CPU0: I cache: 32768 bytes, associativity 32, 32 byte lines, 32 sets
[    0.000000] CPU0: D cache: 32768 bytes, associativity 32, 32 byte lines, 32 sets
[    0.000000] Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 32512
[    0.000000] Kernel command line: root=/dev/mtdblock2 rootfstype=jffs2 console=ttyS0,38400
[    0.000000] PID hash table entries: 512 (order: 9, 2048 bytes)
[    0.945947] I-pipe 1.12-02: pipeline enabled.
[    0.946200] Console: colour dummy device 80x30
[    0.946531] Dentry cache hash table entries: 16384 (order: 4, 65536 bytes)
[    0.947298] Inode-cache hash table entries: 8192 (order: 3, 32768 bytes)
[    0.964229] Memory: 128MB = 128MB total
[    0.964321] Memory: 126392KB available (2876K code, 407K data, 88K init)
[    0.966408] Calibrating delay loop... 311.29 BogoMIPS (lpj=1556480)
[    1.177162] Mount-cache hash table entries: 512
[    1.179089] CPU: Testing write buffer coherency: ok
[    1.200627] net_namespace: 288 bytes
[    1.212405] NET: Registered protocol family 16
[    1.480735] SCSI subsystem initialized
[    1.514911] I2C: i2c-1: PXA I2C adapter
[    1.527629] I2C: i2c-0: PXA I2C adapter
[    1.617195] NET: Registered protocol family 2
[    1.718522] IP route cache hash table entries: 1024 (order: 0, 4096 bytes)
[    1.725481] TCP established hash table entries: 4096 (order: 3, 32768 bytes)
[    1.726567] TCP bind hash table entries: 4096 (order: 2, 16384 bytes)
[    1.727071] TCP: Hash tables configured (established 4096 bind 4096)
[    1.727427] TCP reno registered
[    1.759088] NET: Registered protocol family 1
[    1.788016] I-pipe: Domain Xenomai registered.
[    1.788540] Xenomai: hal/arm started.
[    1.942892] I-pipe: Domain IShield registered.
[    1.955141] Xenomai: real-time nucleus v2.4.9.1 (Big Bad Moon) loaded.
[    1.979878] Xenomai: starting native API services.
[    1.980247] Xenomai: starting POSIX services.
[    1.986299] Xenomai: starting RTDM services.
[    2.002735] JFFS2 version 2.2. (NAND) © 2001-2006 Red Hat, Inc.
[    2.018343] msgmni has been set to 247
[    2.019921] io scheduler noop registered
[    2.020245] io scheduler deadline registered (default)
[    2.028306] pxa2xx-fb pxa2xx-fb: machine LCCR0 setting contains illegal bits: 00300878
[    2.056984] Console: switching to colour frame buffer device 80x30
[    2.439728] pxa2xx-uart.0: ttyS0 at MMIO 0x40100000 (irq = 22) is a FFUART
[    2.440293] console [ttyS0] enabled
[    3.244674] pxa2xx-uart.1: ttyS1 at MMIO 0x40200000 (irq = 21) is a BTUART
[    3.273470] pxa2xx-uart.2: ttyS2 at MMIO 0x40700000 (irq = 20) is a STUART
[    3.308459] dm9000 Ethernet Driver, V1.31
[    3.447636] eth0: dm9000e at c8a1e300,c8a32304 IRQ 178 MAC: 00:01:02:03:04:05 (chip)
[    3.481327] Driver 'sd' needs updating - please use bus_type methods
[    3.519052] Probing vpac270dimm at physical address 0x00000000 (16-bit bankwidth)
[    3.542403] vpac270dimm: Found 1 x16 devices at 0x0 in 16-bit bank
[    3.562132]  Intel/Sharp Extended Query Table at 0x010A
[    3.578541]  Intel/Sharp Extended Query Table at 0x010A
[    3.594877]  Intel/Sharp Extended Query Table at 0x010A
[    3.611299]  Intel/Sharp Extended Query Table at 0x010A
[    3.627956]  Intel/Sharp Extended Query Table at 0x010A
[    3.644404] Using buffer write method
[    3.656061] Using auto-unlock on power-up/resume
[    3.670502] cfi_cmdset_0001: Erase suspend on write enabled
[    3.693433] RedBoot partition parsing not available
[    3.713172] cmdlinepart partition parsing not available
[    3.730124] Creating 3 MTD partitions on "vpac270dimm":
[    3.746441] 0x00000000-0x00020000 : "Bootloader"
[    3.784435] 0x00020000-0x00220000 : "Kernel"
[    3.821591] 0x00220000-0x02000000 : "Filesystem"
[    3.890696] mice: PS/2 mouse device common for all mice
[    3.967888] rtc-ds1307 0-0068: rtc core: registered ds1339 as rtc0
[    4.000356] sa1100-rtc sa1100-rtc: rtc core: registered sa1100-rtc as rtc1
[    4.027985] i2c /dev entries driver
[    4.068546] Registered led device: vpac270:orange
[    4.089921] TCP cubic registered
[    4.100338] NET: Registered protocol family 17
[    4.115049] ieee80211: 802.11 data/management/control stack, git-1.1.13
[    4.135929] ieee80211: Copyright (C) 2004-2005 Intel Corporation <jketreno@linux.intel.com>
[    4.162467] XScale iWMMXt coprocessor detected.
[    4.194945] rtc-ds1307 0-0068: setting system clock to 2000-01-15 22:30:53 UTC (947975453)
[    4.923883] VFS: Mounted root (jffs2 filesystem).
[    4.940871] Freeing init memory: 88K
[    5.259645] Kernel panic - not syncing: Attempted to kill init!






^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-10 17:28           ` Felipe Castro
@ 2009-09-10 17:45             ` Gilles Chanteperdrix
  2009-09-15 12:49               ` Felipe Castro
  0 siblings, 1 reply; 13+ messages in thread
From: Gilles Chanteperdrix @ 2009-09-10 17:45 UTC (permalink / raw)
  To: Felipe Castro; +Cc: xenomai

Felipe Castro wrote:
> I'm a little bit lost about this issues cause when running the xenomai
> 2.4.8 with the patch that you send me it worked, but with new version it
> doesn't work.
> Acctually i using the kernel 2.6.27 and i understand what you told me ,
> in version 2.4.9.1 there aren't a patch to this kernel version, neither
> xenomai 2.4.8.  :S

2.4.8 appeared to work, but it had a bug.
See https://mail.gna.org/public/xenomai-core/2009-08/msg00021.html

> 
> The problem is, I'm using a Voipac development kit, and the have only
> the kernel 2.6.27 BSP.
> 
> So , i tried to disable the float point emulation but the kernel didn't
> load OK .

It proves that your user-space support is compiled for hardware float,
but not PXA hardware floats. This is a pure waste of CPU power. See:

http://www.arm.linux.org.uk/mailinglists/faq.php#f3

Note that Xenomai applications compiled that way will not work either.

-- 
                                          Gilles



^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-10 17:45             ` Gilles Chanteperdrix
@ 2009-09-15 12:49               ` Felipe Castro
  2009-09-15 13:15                 ` Henri Roosen
  2009-09-15 13:18                 ` Philippe Gerum
  0 siblings, 2 replies; 13+ messages in thread
From: Felipe Castro @ 2009-09-15 12:49 UTC (permalink / raw)
  To: xenomai


[-- Attachment #1.1: Type: text/plain, Size: 1603 bytes --]

Hi,

I guess I compiled all with msoft-float OK , i disabled the float point
emulation in the kernel , did a new filesystem with busybox compiled with
soft float and compiled the xenomai user space to but running the latency
program i still having a problem.

During the boot everything seems to be ok ....

Attached follow a log file, with the kernel load , and ipipe trace. what i
supposed to do to get it running ok ?

As you asked me , I'm running version 2.4.9.1 of xenomai .

On Thu, Sep 10, 2009 at 2:45 PM, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> Felipe Castro wrote:
> > I'm a little bit lost about this issues cause when running the xenomai
> > 2.4.8 with the patch that you send me it worked, but with new version it
> > doesn't work.
> > Acctually i using the kernel 2.6.27 and i understand what you told me ,
> > in version 2.4.9.1 there aren't a patch to this kernel version, neither
> > xenomai 2.4.8.  :S
>
> 2.4.8 appeared to work, but it had a bug.
> See https://mail.gna.org/public/xenomai-core/2009-08/msg00021.html
>
> >
> > The problem is, I'm using a Voipac development kit, and the have only
> > the kernel 2.6.27 BSP.
> >
> > So , i tried to disable the float point emulation but the kernel didn't
> > load OK .
>
> It proves that your user-space support is compiled for hardware float,
> but not PXA hardware floats. This is a pure waste of CPU power. See:
>
> http://www.arm.linux.org.uk/mailinglists/faq.php#f3
>
> Note that Xenomai applications compiled that way will not work either.
>
> --
>                                           Gilles
>
>

[-- Attachment #1.2: Type: text/html, Size: 2308 bytes --]

[-- Attachment #2: Log --]
[-- Type: application/octet-stream, Size: 17434 bytes --]

// Kernel boot
//-----------------------------------------------------

U-Boot 1.1.2 (Jun 20 2007 - 14:11:13)

U-Boot code: 5C010000 -> 5C01B6A0  BSS: -> 5C01EAF0
RAM Configuration:
Bank #0: a0000000 128 MB
Bank #1: 80000000  0 kB
Flash: 32 MB
Hit any key to stop autoboot:  0 
Writing TAGS to DRAM at 0xa0000100
## Starting application at 0x00020000 ...
Uncompressing Linux...................................................................................................... done, booting the kernel.
[    0.000000] Linux version 2.6.27-vpac1 (root@debian) (gcc version 4.3.3 (Sourcery G++ 4.3-234) ) #80 PREEMPT Mon Sep 14 11:49:59 BRT 2009
[    0.000000] CPU: XScale-PXA270 [69054117] revision 7 (ARMv5TE), cr=0000397f
[    0.000000] Machine: Voipac PXA270 Module
[    0.000000] Memory policy: ECC disabled, Data cache writeback
[    0.000000] Run Mode clock: 208.00MHz (*16)
[    0.000000] Turbo Mode clock: 312.00MHz (*1.5, active)
[    0.000000] Memory clock: 208.00MHz (/2)
[    0.000000] System bus clock: 208.00MHz 
[    0.000000] CPU0: D VIVT undefined 5 cache
[    0.000000] CPU0: I cache: 32768 bytes, associativity 32, 32 byte lines, 32 sets
[    0.000000] CPU0: D cache: 32768 bytes, associativity 32, 32 byte lines, 32 sets
[    0.000000] Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 32512
[    0.000000] Kernel command line: root=/dev/mtdblock2 rootfstype=jffs2 console=ttyS0,38400
[    0.000000] PID hash table entries: 512 (order: 9, 2048 bytes)
[    1.102694] I-pipe 1.12-02: pipeline enabled.
[    1.102955] Console: colour dummy device 80x30
[    1.103282] Dentry cache hash table entries: 16384 (order: 4, 65536 bytes)
[    1.104095] Inode-cache hash table entries: 8192 (order: 3, 32768 bytes)
[    1.120841] Memory: 128MB = 128MB total
[    1.120930] Memory: 126336KB available (2960K code, 377K data, 80K init)
[    1.123114] Calibrating delay loop... 311.29 BogoMIPS (lpj=1556480)
[    1.333935] Mount-cache hash table entries: 512
[    1.335972] CPU: Testing write buffer coherency: ok
[    1.357282] net_namespace: 288 bytes
[    1.362664] NET: Registered protocol family 16
[    1.500866] SCSI subsystem initialized
[    1.519429] I2C: i2c-1: PXA I2C adapter
[    1.525551] I2C: i2c-0: PXA I2C adapter
[    1.579125] NET: Registered protocol family 2
[    1.666690] IP route cache hash table entries: 1024 (order: 0, 4096 bytes)
[    1.669009] TCP established hash table entries: 4096 (order: 3, 32768 bytes)
[    1.669403] TCP bind hash table entries: 4096 (order: 2, 16384 bytes)
[    1.669688] TCP: Hash tables configured (established 4096 bind 4096)
[    1.669795] TCP reno registered
[    1.697108] NET: Registered protocol family 1
[    1.711078] I-pipe: Domain Xenomai registered.
[    1.711317] Xenomai: hal/arm started.
[    1.790781] I-pipe: Domain IShield registered.
[    1.796655] Xenomai: real-time nucleus v2.4.9.1 (Big Bad Moon) loaded.
[    1.815310] Xenomai: starting native API services.
[    1.815468] Xenomai: starting POSIX services.
[    1.819523] Xenomai: starting RTDM services.
[    1.825071] JFFS2 version 2.2. (NAND) © 2001-2006 Red Hat, Inc.
[    1.829718] fuse init (API version 7.9)
[    1.833112] msgmni has been set to 247
[    1.833774] io scheduler noop registered
[    1.833889] io scheduler deadline registered (default)
[    1.838610] pxa2xx-fb pxa2xx-fb: machine LCCR0 setting contains illegal bits: 00300878
[    1.858856] Console: switching to colour frame buffer device 80x30
[    2.053265] pxa2xx-uart.0: ttyS0 at MMIO 0x40100000 (irq = 22) is a FFUART
[    2.053494] console [ttyS0] enabled
[    2.870506] pxa2xx-uart.1: ttyS1 at MMIO 0x40200000 (irq = 21) is a BTUART
[    2.896513] pxa2xx-uart.2: ttyS2 at MMIO 0x40700000 (irq = 20) is a STUART
[    2.923979] dm9000 Ethernet Driver, V1.31
[    3.055906] dm9000 dm9000.0: eth%d: Invalid ethernet MAC address. Please set using ifconfig
[    3.086301] eth0: dm9000e at c8a1c300,c8a1e304 IRQ 178 MAC: 00:00:00:00:00:00 (chip)
[    3.114616] Driver 'sd' needs updating - please use bus_type methods
[    3.144197] Probing vpac270dimm at physical address 0x00000000 (16-bit bankwidth)
[    3.167062] vpac270dimm: Found 1 x16 devices at 0x0 in 16-bit bank
[    3.186473]  Intel/Sharp Extended Query Table at 0x010A
[    3.202451]  Intel/Sharp Extended Query Table at 0x010A
[    3.218312]  Intel/Sharp Extended Query Table at 0x010A
[    3.234156]  Intel/Sharp Extended Query Table at 0x010A
[    3.249992]  Intel/Sharp Extended Query Table at 0x010A
[    3.265832] Using buffer write method
[    3.277073] Using auto-unlock on power-up/resume
[    3.291075] cfi_cmdset_0001: Erase suspend on write enabled
[    3.311476] RedBoot partition parsing not available
[    3.328875] cmdlinepart partition parsing not available
[    3.345124] Creating 3 MTD partitions on "vpac270dimm":
[    3.360971] 0x00000000-0x00020000 : "Bootloader"
[    3.386566] 0x00020000-0x00220000 : "Kernel"
[    3.411350] 0x00220000-0x02000000 : "Filesystem"
[    3.451762] mice: PS/2 mouse device common for all mice
[    3.500097] rtc-ds1307 0-0068: rtc core: registered ds1339 as rtc0
[    3.524774] sa1100-rtc sa1100-rtc: rtc core: registered sa1100-rtc as rtc1
[    3.549794] i2c /dev entries driver
[    3.574805] Registered led device: vpac270:orange
[    3.592077] TCP cubic registered
[    3.602600] NET: Registered protocol family 17
[    3.616284] ieee80211: 802.11 data/management/control stack, git-1.1.13
[    3.636322] ieee80211: Copyright (C) 2004-2005 Intel Corporation <jketreno@linux.intel.com>
[    3.661776] XScale iWMMXt coprocessor detected.
[    3.685972] rtc-ds1307 0-0068: setting system clock to 2000-01-20 17:55:52 UTC (948390952)
[    4.034614] VFS: Mounted root (jffs2 filesystem).
[    4.050443] Freeing init memory: 80K
init started: BusyBox v1.6.1 (2009-09-14 10:24:40 BRT) multi-call binary
mount: mounting usbfs on /proc/bus/usb failed
Starting syslogd/klogd: start-stop-daemon: lseek: Invalid argument
start-stop-daemon: lseek: Invalid argument
done
Setting up IP spoofing protection: rp_filter.
Configuring network interfaces... ifconfig: SIOCSIFHWADDR: Cannot assign requested address
ifconfig: SIOCSIFFLAGS: Cannot assign requested address
route: SIOC[ADD|DEL]RT: Network is unreachable
done.
starting Busybox Telnet: telnetd... start-stop-daemon: lseek: Invalid argument
done.

vpac270 login: 

//--------------------
// Running latency

root@vpac270:/usr$ cd xenomai/
root@vpac270:/usr/xenomai$ ls
bin      include  lib      sbin     share
root@vpac270:/usr/xenomai$ cd bin
root@vpac270:/usr/xenomai/bin$ ./latency -p 3000
== Sampling period: 3000 us
== Test mode: periodic user-mode task
== All results in microseconds
[   94.974190] Xenomai: fatal: bug at kernel/xenomai/nucleus/shadow.c:1128 (!irqs_disabled_hw())
[   94.974225]  CPU  PID    PRI      TIMEOUT  STAT      NAME
[   94.974246]    0  0       -1      0        00500088  ROOT
[   94.974267] >  0  311      0      0        00300180  display-310
[   94.974290] Master time base: clock=308666876
[   94.974308] 
[   95.066916] [<c00217bc>] (show_stack+0x0/0x50) from [<c00837ac>] (xnshadow_harden+0x208/0x270)
[   95.092711] [<c00835a4>] (xnshadow_harden+0x0/0x270) from [<c0083e80>] (xnshadow_wait_barrier+0x130/0x148)
[   95.121501]  r7:00000000 r6:c7a42000 r5:c7a43fb0 r4:c8986c10
[   95.138378] [<c0083d50>] (xnshadow_wait_barrier+0x0/0x148) from [<c0083eac>] (xnshadow_sys_barrier+0x14/0x18)
[   95.167947]  r5:c7a43fb0 r4:00000001
[   95.178595] [<c0083e98>] (xnshadow_sys_barrier+0x0/0x18) from [<c0083c3c>] (losyscall_event+0xd8/0x1ec)
[   95.206612] [<c0083b64>] (losyscall_event+0x0/0x1ec) from [<c006221c>] (__ipipe_dispatch_event+0xd8/0x1dc)
[   95.235430] [<c0062144>] (__ipipe_dispatch_event+0x0/0x1dc) from [<c0023418>] (__ipipe_syscall_root+0x80/0x14c)
[   95.265515] [<c0023398>] (__ipipe_syscall_root+0x0/0x14c) from [<c001d0b4>] (vector_swi+0x74/0xb4)
[   95.292240]  r7:000f0042 r6:000000e0 r5:00010000 r4:00010228
[   95.309070] I-pipe tracer log (100 points):
[   95.321496]  | #  end     0x80000000      0 schedule+0x654 (xnshadow_harden+0x98)
[   95.343877]  | #  [    1] init    -1    -24 xnpod_schedule+0xb0 (xnpod_schedule_handler+0x54)
[   95.369310]  | * +[  311] display- 0    -48 xnpod_resume_thread+0x6c (gatekeeper_thread+0x170)
[   95.395001]  |   +begin   0x80000000    -51 gatekeeper_thread+0x150 (kthread+0x54)
[   95.417565]  |   +end     0x80000000    -53 __ipipe_restore_pipeline_head+0x108 (rpi_pop+0x534)
[   95.443488]  |   +begin   0x80000000    -57 rpi_pop+0x34 (gatekeeper_thread+0x130)
[   95.466052]  |   +end     0x80000000    -64 schedule+0x654 (gatekeeper_thread+0x104)
[   95.489133]  |   +begin   0x80000000    -69 schedule+0x434 (xnshadow_harden+0x98)
[   95.511438]  |   +end     0x80000001    -71 __ipipe_dispatch_event+0x1c4 (schedule+0x414)
[   95.535810]  |   +begin   0x80000001    -74 __ipipe_dispatch_event+0x100 (schedule+0x414)
[   95.560184]  |   +end     0x80000000    -77 __ipipe_restore_pipeline_head+0x108 (schedule_event+0xc6c)
[   95.587916]  |   +begin   0x80000000    -79 schedule_event+0x704 (__ipipe_dispatch_event+0xd8)
[   95.613582]  |   +end     0x80000001    -82 __ipipe_dispatch_event+0xc4 (schedule+0x414)
[   95.637696]  |   +begin   0x80000001    -87 __ipipe_dispatch_event+0x38 (schedule+0x414)
[   95.661810]  |   +end     0x80000000    -90 __ipipe_unstall_root+0x64 (schedule+0x39c)
[   95.685407]  |   #begin   0x80000000    -93 __ipipe_unstall_root+0x2c (schedule+0x39c)
[   95.709005]  |   #end     0x80000001   -104 schedule+0x188 (xnshadow_harden+0x98)
[   95.731310]  |   #begin   0x80000001   -107 schedule+0x16c (xnshadow_harden+0x98)
[   95.753615]  |   #end     0x80000001   -110 schedule+0x128 (xnshadow_harden+0x98)
[   95.775920]  |   +begin   0x80000001   -113 schedule+0x108 (xnshadow_harden+0x98)
[   95.798227]  |   +end     0x80000000   -117 __ipipe_unstall_root+0x64 (__ipipe_restore_root+0x60)
[   95.824667]  |   #begin   0x80000000   -119 __ipipe_unstall_root+0x2c (__ipipe_restore_root+0x60)
[   95.851107]  |   #end     0x80000001   -123 __ipipe_restore_root+0x54 (try_to_wake_up+0x12c)
[   95.876256]  |   #begin   0x80000001   -125 __ipipe_restore_root+0x34 (try_to_wake_up+0x12c)
[   95.901403]  |   #end     0x80000001   -127 resched_task+0x4c (check_preempt_wakeup+0x134)
[   95.926062]  |   #begin   0x80000001   -130 resched_task+0x30 (check_preempt_wakeup+0x134)
[   95.950692]  |   #end     0x80000001   -137 __ipipe_restore_root+0x54 (hrtimer_start+0x120)
[   95.975581]  |   #begin   0x80000001   -139 __ipipe_restore_root+0x34 (hrtimer_start+0x120)
[   96.000472]  |   #end     0x80000001   -147 hrtimer_start+0x60 (enqueue_task_rt+0x13c)
[   96.024069]  |   #begin   0x80000001   -151 hrtimer_start+0x40 (enqueue_task_rt+0x13c)
[   96.047665]  |   #end     0x80000001   -185 task_rq_lock+0x50 (try_to_wake_up+0x34)
[   96.070489]  |   #begin   0x80000001   -188 task_rq_lock+0x30 (try_to_wake_up+0x34)
[   96.093311]  |   #end     0x80000001   -194 __wake_up_sync+0x60 (xnshadow_harden+0x94)
[   96.116907]  |   +begin   0x80000001   -197 __wake_up_sync+0x40 (xnshadow_harden+0x94)
[   96.140505]  |   +end     0x80000000   -201 __ipipe_unstall_root+0x64 (__ipipe_restore_root+0x60)
[   96.166946]  |   #begin   0x80000000   -204 __ipipe_unstall_root+0x2c (__ipipe_restore_root+0x60)
[   96.193386]  |   #end     0x80000001   -207 down_interruptible+0x50 (xnshadow_harden+0x58)
[   96.218018]  |   +begin   0x80000001   -209 down_interruptible+0x30 (xnshadow_harden+0x58)
[   96.242650]  |   +end     0x80000000   -219 schedule+0x654 (xnshadow_wait_barrier+0xc0)
[   96.266505]  |   +begin   0x80000000   -225 schedule+0x434 (ret_slow_syscall+0x0)
[   96.288811]  |   +end     0x80000001   -227 __ipipe_dispatch_event+0x1c4 (schedule+0x414)
[   96.313183]  |   +begin   0x80000001   -230 __ipipe_dispatch_event+0x100 (schedule+0x414)
[   96.337557]  |   +end     0x80000000   -232 __ipipe_restore_pipeline_head+0x108 (schedule_event+0xc6c)
[   96.365289]  |   +begin   0x80000000   -236 schedule_event+0x704 (__ipipe_dispatch_event+0xd8)
[   96.390955]  |   +end     0x80000001   -238 disengage_irq_shield+0x184 (schedule_event+0x2e4)
[   96.416361]  |   +begin   0x80000001   -241 disengage_irq_shield+0x2c (schedule_event+0x2e4)
[   96.441509]  |   +end     0x80000001   -244 __ipipe_dispatch_event+0xc4 (schedule+0x414)
[   96.465624]  |   +begin   0x80000001   -247 __ipipe_dispatch_event+0x38 (schedule+0x414)
[   96.489737]  |   +end     0x80000000   -251 __ipipe_unstall_root+0x64 (schedule+0x39c)
[   96.513335]  |   #begin   0x80000000   -253 __ipipe_unstall_root+0x2c (schedule+0x39c)
[   96.536933]  |   #end     0x80000001   -265 schedule+0x188 (ret_slow_syscall+0x0)
[   96.559237]  |   #begin   0x80000001   -268 schedule+0x16c (ret_slow_syscall+0x0)
[   96.581543]  |   #end     0x80000001   -271 schedule+0x128 (ret_slow_syscall+0x0)
[   96.603848]  |   +begin   0x80000001   -273 schedule+0x108 (ret_slow_syscall+0x0)
[   96.626154]  |   +end     0x80000001   -277 __ipipe_syscall_root+0x110 (vector_swi+0x74)
[   96.650269]  |   #begin   0x80000000   -280 __ipipe_sync_stage+0x220 (__ipipe_syscall_root+0x100)
[   96.676709]  |   #end     0x80000001   -288 __ipipe_restore_root+0x54 (try_to_wake_up+0x12c)
[   96.701857]  |   #begin   0x80000001   -291 __ipipe_restore_root+0x34 (try_to_wake_up+0x12c)
[   96.727006]  |   #end     0x80000001   -294 resched_task+0x4c (check_preempt_wakeup+0x134)
[   96.751636]  |   #begin   0x80000001   -297 resched_task+0x30 (check_preempt_wakeup+0x134)
[   96.776268]  |   #end     0x80000001   -319 task_rq_lock+0x50 (try_to_wake_up+0x34)
[   96.799089]  |   #begin   0x80000001   -322 task_rq_lock+0x30 (try_to_wake_up+0x34)
[   96.821912]  |   #end     0x80000000   -332 __ipipe_sync_stage+0x154 (__ipipe_syscall_root+0x100)
[   96.848353]  |   +begin   0x80000001   -339 __ipipe_syscall_root+0xe8 (vector_swi+0x74)
[   96.872208]  |   +end     0x80000001   -342 __ipipe_dispatch_event+0x1c4 (__ipipe_syscall_root+0x80)
[   96.899425]  | +  begin   0x80000001   -345 __ipipe_dispatch_event+0x100 (__ipipe_syscall_root+0x80)
[   96.926640]  | +  end     0x80000000   -350 __ipipe_restore_pipeline_head+0x108 (rt_task_start+0xd8)
[   96.953856]  | #  [    1] init    -1   -355 xnpod_schedule+0xb0 (xnpod_start_thread+0x21c)
[   96.978512]  | #  [  311] display- 0   -374 xnpod_resume_thread+0x6c (xnshadow_start+0x28)
[   97.003145]  | +  begin   0x80000000   -388 rt_task_start+0x58 (__rt_task_start+0x64)
[   97.026484]  | +  end     0x80000000   -392 __ipipe_restore_pipeline_head+0x108 (xnregistry_fetch+0xc8)
[   97.054476]  | +  begin   0x80000000   -396 xnregistry_fetch+0x30 (__rt_task_start+0x4c)
[   97.078589]  | +  end     0x80000001   -403 __ipipe_dispatch_event+0xc4 (__ipipe_syscall_root+0x80)
[   97.105546]  |   +begin   0x80000001   -407 __ipipe_dispatch_event+0x38 (__ipipe_syscall_root+0x80)
[   97.132504]  |   +end     0x80000001   -459 __ipipe_syscall_root+0x110 (vector_swi+0x74)
[   97.156619]  |   +begin   0x80000001   -461 __ipipe_syscall_root+0xe8 (vector_swi+0x74)
[   97.180474]  |   +end     0x80000001   -464 __ipipe_dispatch_event+0x1c4 (__ipipe_syscall_root+0x80)
[   97.207690]  |   +begin   0x80000001   -466 __ipipe_dispatch_event+0x100 (__ipipe_syscall_root+0x80)
[   97.234907]  |   +end     0x80000000   -470 __ipipe_unstall_root+0x64 (__ipipe_restore_root+0x60)
[   97.261347]  |   #begin   0x80000000   -472 __ipipe_unstall_root+0x2c (__ipipe_restore_root+0x60)
[   97.287786]  |   #end     0x80000001   -474 up+0x50 (xnshadow_sys_completion+0xb8)
[   97.310351]  |   +begin   0x80000001   -477 up+0x30 (xnshadow_sys_completion+0xb8)
[   97.332915]  |   +end     0x80000000   -481 __ipipe_unstall_root+0x64 (__ipipe_restore_root+0x60)
[   97.359355]  |   #begin   0x80000000   -483 __ipipe_unstall_root+0x2c (__ipipe_restore_root+0x60)
[   97.385796]  |   #end     0x80000001   -485 down+0x50 (xnshadow_sys_completion+0x2c)
[   97.408876]  |   +begin   0x80000001   -488 down+0x30 (xnshadow_sys_completion+0x2c)
[   97.431956]  |   +end     0x80000001   -491 __ipipe_dispatch_event+0xc4 (__ipipe_syscall_root+0x80)
[   97.458915]  | +  begin   0x80000001   -494 __ipipe_dispatch_event+0x100 (__ipipe_syscall_root+0x80)
[   97.486130]  | +  end     0x80000001   -497 __ipipe_dispatch_event+0xc4 (__ipipe_syscall_root+0x80)
[   97.513089]  |   +begin   0x80000001   -500 __ipipe_dispatch_event+0x38 (__ipipe_syscall_root+0x80)
[   97.540046]  |   +end     0x80000000   -512 schedule+0x654 (preempt_schedule+0x40)
[   97.562609]  |   +begin   0x80000000   -524 schedule+0x434 (xnshadow_wait_barrier+0xc0)
[   97.586464]  |   +end     0x80000001   -526 __ipipe_dispatch_event+0x1c4 (schedule+0x414)
[   97.610838]  |   +begin   0x80000001   -529 __ipipe_dispatch_event+0x100 (schedule+0x414)
[   97.635210]  |   +end     0x80000000   -532 __ipipe_restore_pipeline_head+0x108 (schedule_event+0xc6c)
[   97.662943]  |   +begin   0x80000000   -536 schedule_event+0x704 (__ipipe_dispatch_event+0xd8)
[   97.688608]  |   +end     0x80000000   -539 __ipipe_restore_pipeline_head+0x108 (schedule_event+0x6c4)
[   97.716341]  |   +begin   0x80000000   -542 schedule_event+0x364 (__ipipe_dispatch_event+0xd8)
[   97.742006]  |   +end     0x80000001   -546 disengage_irq_shield+0x184 (schedule_event+0x2e4)
[   97.767413]  |   +begin   0x80000001   -550 disengage_irq_shield+0x2c (schedule_event+0x2e4)




^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-15 12:49               ` Felipe Castro
@ 2009-09-15 13:15                 ` Henri Roosen
  2009-09-15 13:18                 ` Philippe Gerum
  1 sibling, 0 replies; 13+ messages in thread
From: Henri Roosen @ 2009-09-15 13:15 UTC (permalink / raw)
  To: Felipe Castro; +Cc: xenomai

Felipe, your kernel still runs ipipe version 1.12-02. That version of
the ipipe for ARM has a known bug which very likely is causing the
problem you see.

If you cannot use a later version of the kernel (which would be highly
recommended), then you should at least apply the patch as provided by
Gilles in his Xenomai 2.4.9.1 announcement:

https://mail.gna.org/public/xenomai-help/2009-08/msg00158.html

Hope that helps!

    Henri

On Tue, Sep 15, 2009 at 2:49 PM, Felipe Castro <felipecpc1@domain.hid> wrote:
> Hi,
>
> I guess I compiled all with msoft-float OK , i disabled the float point
> emulation in the kernel , did a new filesystem with busybox compiled with
> soft float and compiled the xenomai user space to but running the latency
> program i still having a problem.
>
> During the boot everything seems to be ok ....
>
> Attached follow a log file, with the kernel load , and ipipe trace. what i
> supposed to do to get it running ok ?
>
> As you asked me , I'm running version 2.4.9.1 of xenomai .
>
> On Thu, Sep 10, 2009 at 2:45 PM, Gilles Chanteperdrix
> <gilles.chanteperdrix@xenomai.org> wrote:
>>
>> Felipe Castro wrote:
>> > I'm a little bit lost about this issues cause when running the xenomai
>> > 2.4.8 with the patch that you send me it worked, but with new version it
>> > doesn't work.
>> > Acctually i using the kernel 2.6.27 and i understand what you told me ,
>> > in version 2.4.9.1 there aren't a patch to this kernel version, neither
>> > xenomai 2.4.8.  :S
>>
>> 2.4.8 appeared to work, but it had a bug.
>> See https://mail.gna.org/public/xenomai-core/2009-08/msg00021.html
>>
>> >
>> > The problem is, I'm using a Voipac development kit, and the have only
>> > the kernel 2.6.27 BSP.
>> >
>> > So , i tried to disable the float point emulation but the kernel didn't
>> > load OK .
>>
>> It proves that your user-space support is compiled for hardware float,
>> but not PXA hardware floats. This is a pure waste of CPU power. See:
>>
>> http://www.arm.linux.org.uk/mailinglists/faq.php#f3
>>
>> Note that Xenomai applications compiled that way will not work either.
>>
>> --
>>                                          Gilles
>>
>
>
> _______________________________________________
> Xenomai-help mailing list
> Xenomai-help@domain.hid
> https://mail.gna.org/listinfo/xenomai-help
>
>


^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-15 12:49               ` Felipe Castro
  2009-09-15 13:15                 ` Henri Roosen
@ 2009-09-15 13:18                 ` Philippe Gerum
  2009-09-15 14:52                   ` Felipe Castro
  1 sibling, 1 reply; 13+ messages in thread
From: Philippe Gerum @ 2009-09-15 13:18 UTC (permalink / raw)
  To: Felipe Castro; +Cc: xenomai

On Tue, 2009-09-15 at 09:49 -0300, Felipe Castro wrote:
> Hi, 
> 
> I guess I compiled all with msoft-float OK , i disabled the float
> point emulation in the kernel , did a new filesystem with busybox
> compiled with soft float and compiled the xenomai user space to but
> running the latency program i still having a problem.
> 
> During the boot everything seems to be ok ....
> 
> Attached follow a log file, with the kernel load , and ipipe trace.
> what i supposed to do to get it running ok ?
> 

As Gilles pointed out already, you are supposed to use a pipeline patch
which does not exhibit the issue reported by Xenomai, which raises the
debug assertion reported in the log.

This patch only exists starting from 2.6.28, since the fix was not
backported to earlier kernel releases. Since you want to stick with a
non-mainline kernel, you will have to backport the fix yourself, and see
how it goes.

http://git.denx.de/?p=ipipe-2.6.git;a=commit;h=912dcfb63c3315d8a20b8d467f72e443dad3852b
 

> As you asked me , I'm running version 2.4.9.1 of xenomai .
> 
> On Thu, Sep 10, 2009 at 2:45 PM, Gilles Chanteperdrix
> <gilles.chanteperdrix@xenomai.org> wrote:
>         Felipe Castro wrote:
>         > I'm a little bit lost about this issues cause when running
>         the xenomai
>         > 2.4.8 with the patch that you send me it worked, but with
>         new version it
>         > doesn't work.
>         > Acctually i using the kernel 2.6.27 and i understand what
>         you told me ,
>         > in version 2.4.9.1 there aren't a patch to this kernel
>         version, neither
>         > xenomai 2.4.8.  :S
>         
>         
>         2.4.8 appeared to work, but it had a bug.
>         See
>         https://mail.gna.org/public/xenomai-core/2009-08/msg00021.html
>         
>         >
>         > The problem is, I'm using a Voipac development kit, and the
>         have only
>         > the kernel 2.6.27 BSP.
>         >
>         > So , i tried to disable the float point emulation but the
>         kernel didn't
>         > load OK .
>         
>         
>         It proves that your user-space support is compiled for
>         hardware float,
>         but not PXA hardware floats. This is a pure waste of CPU
>         power. See:
>         
>         http://www.arm.linux.org.uk/mailinglists/faq.php#f3
>         
>         Note that Xenomai applications compiled that way will not work
>         either.
>         
>         --
>                                                  Gilles
>         
> 
> _______________________________________________
> Xenomai-help mailing list
> Xenomai-help@domain.hid
> https://mail.gna.org/listinfo/xenomai-help
-- 
Philippe.




^ permalink raw reply	[flat|nested] 13+ messages in thread

* Re: [Xenomai-help] High Latency on PXA
  2009-09-15 13:18                 ` Philippe Gerum
@ 2009-09-15 14:52                   ` Felipe Castro
  0 siblings, 0 replies; 13+ messages in thread
From: Felipe Castro @ 2009-09-15 14:52 UTC (permalink / raw)
  To: xenomai

[-- Attachment #1: Type: text/plain, Size: 3967 bytes --]

So, i applied the patch sent by gilles that Henri told and it works .....
At afternoon I'll do more tests , but for now i'm happy with the results
.......

by the way thanks for attention !!!

here the latency test running

root@domain.hid$ ./latency -p 3000
== Sampling period: 3000 us
== Test mode: periodic user-mode task
== All results in microseconds
[   69.267317] Xenomai: registered exported object dispsem-309 (semaphores)
warming up...
RTT|  00:00:01  (periodic user-mode task, 3000 us period, priority 99)
RTH|-----lat min|-----lat avg|-----lat max|-overrun|----lat best|---lat
worst
RTD|      40.923|      47.692|     127.384|       0|      40.923|
127.384
RTD|      40.923|      47.384|     116.307|       0|      40.923|
127.384
RTD|      40.923|      47.999|     137.230|       0|      40.923|
137.230
RTD|      40.923|      46.769|     123.692|       0|      40.923|
137.230
RTD|      40.923|      47.999|     139.076|       0|      40.923|
139.076
RTD|      40.923|      47.076|     127.692|       0|      40.923|
139.076
RTD|      40.923|      47.384|     122.769|       0|      40.923|
139.076

Felipe

On Tue, Sep 15, 2009 at 10:18 AM, Philippe Gerum <rpm@xenomai.org> wrote:

> On Tue, 2009-09-15 at 09:49 -0300, Felipe Castro wrote:
> > Hi,
> >
> > I guess I compiled all with msoft-float OK , i disabled the float
> > point emulation in the kernel , did a new filesystem with busybox
> > compiled with soft float and compiled the xenomai user space to but
> > running the latency program i still having a problem.
> >
> > During the boot everything seems to be ok ....
> >
> > Attached follow a log file, with the kernel load , and ipipe trace.
> > what i supposed to do to get it running ok ?
> >
>
> As Gilles pointed out already, you are supposed to use a pipeline patch
> which does not exhibit the issue reported by Xenomai, which raises the
> debug assertion reported in the log.
>
> This patch only exists starting from 2.6.28, since the fix was not
> backported to earlier kernel releases. Since you want to stick with a
> non-mainline kernel, you will have to backport the fix yourself, and see
> how it goes.
>
>
> http://git.denx.de/?p=ipipe-2.6.git;a=commit;h=912dcfb63c3315d8a20b8d467f72e443dad3852b
>
>
> > As you asked me , I'm running version 2.4.9.1 of xenomai .
> >
> > On Thu, Sep 10, 2009 at 2:45 PM, Gilles Chanteperdrix
> > <gilles.chanteperdrix@xenomai.org> wrote:
> >         Felipe Castro wrote:
> >         > I'm a little bit lost about this issues cause when running
> >         the xenomai
> >         > 2.4.8 with the patch that you send me it worked, but with
> >         new version it
> >         > doesn't work.
> >         > Acctually i using the kernel 2.6.27 and i understand what
> >         you told me ,
> >         > in version 2.4.9.1 there aren't a patch to this kernel
> >         version, neither
> >         > xenomai 2.4.8.  :S
> >
> >
> >         2.4.8 appeared to work, but it had a bug.
> >         See
> >         https://mail.gna.org/public/xenomai-core/2009-08/msg00021.html
> >
> >         >
> >         > The problem is, I'm using a Voipac development kit, and the
> >         have only
> >         > the kernel 2.6.27 BSP.
> >         >
> >         > So , i tried to disable the float point emulation but the
> >         kernel didn't
> >         > load OK .
> >
> >
> >         It proves that your user-space support is compiled for
> >         hardware float,
> >         but not PXA hardware floats. This is a pure waste of CPU
> >         power. See:
> >
> >         http://www.arm.linux.org.uk/mailinglists/faq.php#f3
> >
> >         Note that Xenomai applications compiled that way will not work
> >         either.
> >
> >         --
> >                                                  Gilles
> >
> >
> > _______________________________________________
> > Xenomai-help mailing list
> > Xenomai-help@domain.hid
> > https://mail.gna.org/listinfo/xenomai-help
> --
> Philippe.
>
>
>

[-- Attachment #2: Type: text/html, Size: 5409 bytes --]

^ permalink raw reply	[flat|nested] 13+ messages in thread

end of thread, other threads:[~2009-09-15 14:52 UTC | newest]

Thread overview: 13+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-09-08 19:59 [Xenomai-help] High Latency on PXA Felipe Castro
2009-09-08 20:19 ` Gilles Chanteperdrix
2009-09-09 15:03   ` Felipe Castro
2009-09-09 15:08     ` Gilles Chanteperdrix
2009-09-09 19:06       ` Felipe Castro
2009-09-09 19:23         ` Gilles Chanteperdrix
2009-09-09 21:40         ` Gilles Chanteperdrix
2009-09-10 17:28           ` Felipe Castro
2009-09-10 17:45             ` Gilles Chanteperdrix
2009-09-15 12:49               ` Felipe Castro
2009-09-15 13:15                 ` Henri Roosen
2009-09-15 13:18                 ` Philippe Gerum
2009-09-15 14:52                   ` Felipe Castro

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.