[Xenomai] Xenomai tests

[Xenomai] Xenomai tests

[vibnwis]

Having managed to compile Xenomai apps and libs, I went to install them on
target. The way I installed is by creating a xenomai directory in /usr and
set PATH to /usr/xenomai/bin, /usr/xenomai/lib etc.
However, when I ran latency I got this error message
  0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory

and when I ran xeno latency -T 25, I got these error messages,

Xenomai: /dev/rtheap is missing
(chardev, major=10 minor=254)

In the installed directory, it does not have dev direcotry but bin, etc,
sbin, lib, share, demo, include. The target does not have rtheap dev as
well. Where and how it is created?

The dmesg | grep -i xenomai produced

[    0.666100] [Xenomai] scheduling class idle registered.
[    0.666112] [Xenomai] scheduling class rt registered.
[    0.666309] I-pipe: head domain Xenomai registered.
[    0.668400] [Xenomai] Cobalt v3.0-rc7 (Exact Zero)


Appreciate any comments, please.

thank you

Re: [Xenomai] Xenomai tests

[Lennart Sorensen]

On Tue, Sep 22, 2015 at 12:15:31AM +1200, vibnwis wrote:
> Having managed to compile Xenomai apps and libs, I went to install them on
> target. The way I installed is by creating a xenomai directory in /usr and
> set PATH to /usr/xenomai/bin, /usr/xenomai/lib etc.
> However, when I ran latency I got this error message
>   0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
> 
> and when I ran xeno latency -T 25, I got these error messages,
> 
> Xenomai: /dev/rtheap is missing
> (chardev, major=10 minor=254)
> 
> In the installed directory, it does not have dev direcotry but bin, etc,
> sbin, lib, share, demo, include. The target does not have rtheap dev as
> well. Where and how it is created?

Usually by udev or some other tool that creates devices for modules
(or built in code) when loaded.

> The dmesg | grep -i xenomai produced
> 
> [    0.666100] [Xenomai] scheduling class idle registered.
> [    0.666112] [Xenomai] scheduling class rt registered.
> [    0.666309] I-pipe: head domain Xenomai registered.
> [    0.668400] [Xenomai] Cobalt v3.0-rc7 (Exact Zero)
> 
> Appreciate any comments, please.

If you run with a static /dev directory, you will have to create required
device nodes yourself somehow.  I get the impression this is not really
how anyone does things anymore.

-- 
Len Sorensen

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Mon, Sep 21, 2015 at 02:54:25PM -0400, Lennart Sorensen wrote:
> On Tue, Sep 22, 2015 at 12:15:31AM +1200, vibnwis wrote:
> > Having managed to compile Xenomai apps and libs, I went to install them on
> > target. The way I installed is by creating a xenomai directory in /usr and
> > set PATH to /usr/xenomai/bin, /usr/xenomai/lib etc.
> > However, when I ran latency I got this error message
> >   0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
> > 
> > and when I ran xeno latency -T 25, I got these error messages,
> > 
> > Xenomai: /dev/rtheap is missing
> > (chardev, major=10 minor=254)
> > 
> > In the installed directory, it does not have dev direcotry but bin, etc,
> > sbin, lib, share, demo, include. The target does not have rtheap dev as
> > well. Where and how it is created?
> 
> Usually by udev or some other tool that creates devices for modules
> (or built in code) when loaded.

The alternatives are:
- the devtmpfs filesystem, which requires no further tools installed
- the mdev tool on systems based on busybox.

> 
> > The dmesg | grep -i xenomai produced
> > 
> > [    0.666100] [Xenomai] scheduling class idle registered.
> > [    0.666112] [Xenomai] scheduling class rt registered.
> > [    0.666309] I-pipe: head domain Xenomai registered.
> > [    0.668400] [Xenomai] Cobalt v3.0-rc7 (Exact Zero)
> > 
> > Appreciate any comments, please.
> 
> If you run with a static /dev directory, you will have to create required
> device nodes yourself somehow.  I get the impression this is not really
> how anyone does things anymore.

Well, in fact, with devtmpfs, I do not even see any reason to have
udev.

-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[Lennart Sorensen]

On Mon, Sep 21, 2015 at 09:41:51PM +0200, Gilles Chanteperdrix wrote:
> The alternatives are:
> - the devtmpfs filesystem, which requires no further tools installed
> - the mdev tool on systems based on busybox.
> 
> Well, in fact, with devtmpfs, I do not even see any reason to have
> udev.

Isn't devtmpfs the predecesor to udev that died?  Or was that devfs or
something else.  I can't remember, it was years ago.

Hmm, I must be thinking of devfs.  Yeah devtmpfs is even appears to be
required by udev these days and does much of the work already it seems.

Not much left for udev to do with devtmpfs around it seems.

-- 
Len Sorensen

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Mon, Sep 21, 2015 at 06:00:04PM -0400, Lennart Sorensen wrote:
> On Mon, Sep 21, 2015 at 09:41:51PM +0200, Gilles Chanteperdrix wrote:
> > The alternatives are:
> > - the devtmpfs filesystem, which requires no further tools installed
> > - the mdev tool on systems based on busybox.
> > 
> > Well, in fact, with devtmpfs, I do not even see any reason to have
> > udev.
> 
> Isn't devtmpfs the predecesor to udev that died?  Or was that devfs or
> something else.  I can't remember, it was years ago.
> 
> Hmm, I must be thinking of devfs.  Yeah devtmpfs is even appears to be
> required by udev these days and does much of the work already it seems.
> 
> Not much left for udev to do with devtmpfs around it seems.

udev still allows executing actions such as running scripts,
changing modes. Like auto-mounting USB keys.

-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[Philippe Gerum]

On 09/21/2015 02:15 PM, vibnwis wrote:
> Having managed to compile Xenomai apps and libs, I went to install them on
> target. The way I installed is by creating a xenomai directory in /usr and
> set PATH to /usr/xenomai/bin, /usr/xenomai/lib etc.
> However, when I ran latency I got this error message
>   0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
> 
> and when I ran xeno latency -T 25, I got these error messages,
> 
> Xenomai: /dev/rtheap is missing
> (chardev, major=10 minor=254)
>

/dev/rtheap is specific to Xenomai 2.x, it is irrelevant to Xenomai 3.x.
Looking at the log messages, you are definitely mixing 3.x kernel
support with 2.x libraries and executables. You may want to start from a
fresh and clean installation.

> In the installed directory, it does not have dev direcotry but bin, etc,
> sbin, lib, share, demo, include. The target does not have rtheap dev as
> well. Where and how it is created?
> 
> The dmesg | grep -i xenomai produced
> 
> [    0.666100] [Xenomai] scheduling class idle registered.
> [    0.666112] [Xenomai] scheduling class rt registered.
> [    0.666309] I-pipe: head domain Xenomai registered.
> [    0.668400] [Xenomai] Cobalt v3.0-rc7 (Exact Zero)
> 
> 
> Appreciate any comments, please.
> 
> thank you
> _______________________________________________
> Xenomai mailing list
> Xenomai@xenomai.org
> http://xenomai.org/mailman/listinfo/xenomai
> 


-- 
Philippe.

Re: [Xenomai] Xenomai tests

[vibnwis]

By regenerating ./configure using autoreconf -fi, I managed to compile
xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in their own
place by make install. However, when running latency, it produces

> latency
0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory

> xeno-config --info
Xenomai version: Xenomai/cobalt v3.0-rc7 --
Linux arm 3.18.20-ipipe #1 SMP Tue Sep 22 21:04:49 NZST 2015 armv7l armv7l
armv7l GNU/Linux
Kernel parameters: console=ttyO2,115200n8 root=/dev/mmcblk0p1 ro
rootfstype=ext4 rootwait fixrtc
I-pipe release #5 detected
Cobalt core disabled on this system

> dmesg | grep -i xenomai
<nothing came back>

> dmesg | grep -i ipipe
[    0.000000] Linux version 3.18.20-ipipe (qws@qws-MacBookPro) (gcc
version 4.9.2 (Ubuntu/Linaro 4.9.2-10ubuntu10) ) #1 SMP Tue Sep 22 21:04:49
NZST 2015
[    0.623763] 3.18.20-ipipe #1 Not tainted
[    0.623994]   [<c00d3094>] __ipipe_do_sync_stage+0x288/0x2d0
[    0.624011]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
[    0.624321] CPU: 1 PID: 22 Comm: kblockd Not tainted 3.18.20-ipipe #1
[    0.624479] [<c00843b4>] (trace_hardirqs_on_caller) from [<c05da7f4>]
(__ipipe_fast_svc_irq_exit+0x4/0x18)
[    0.624504] [<c05da7f4>] (__ipipe_fast_svc_irq_exit) from [<c001aacc>]
(ipip_test_and_stall_root+0x2c/0x3c)
[    0.624524] [<c001aacc>] (ipipe_test_and_stall_root) from [<c0086e58>]
(lock_acquire+0x44/0x108)
[    0.728394] Switched to clocksource ipipe_tsc
[    2.464800] usb usb1: Manufacturer: Linux 3.18.20-ipipe ehci_hcd
[    6.364516] usb usb2: Manufacturer: Linux 3.18.20-ipipe musb-hcd


I have googled, but those are beyond my comprehension. Would you somebody
help please?

Is Cobalt not enabled? if yes, what do those return messages from dmesg |
grep -i ipipe? I presume if Cobalt is enabled, the memory allocation issues
would be cleared, wouldn't it?

Appreciate any comments please.

Thank you




On 22 September 2015 at 18:44, Philippe Gerum <rpm@xenomai.org> wrote:

> On 09/21/2015 02:15 PM, vibnwis wrote:
> > Having managed to compile Xenomai apps and libs, I went to install them
> on
> > target. The way I installed is by creating a xenomai directory in /usr
> and
> > set PATH to /usr/xenomai/bin, /usr/xenomai/lib etc.
> > However, when I ran latency I got this error message
> >   0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
> >
> > and when I ran xeno latency -T 25, I got these error messages,
> >
> > Xenomai: /dev/rtheap is missing
> > (chardev, major=10 minor=254)
> >
>
> /dev/rtheap is specific to Xenomai 2.x, it is irrelevant to Xenomai 3.x.
> Looking at the log messages, you are definitely mixing 3.x kernel
> support with 2.x libraries and executables. You may want to start from a
> fresh and clean installation.
>
> > In the installed directory, it does not have dev direcotry but bin, etc,
> > sbin, lib, share, demo, include. The target does not have rtheap dev as
> > well. Where and how it is created?
> >
> > The dmesg | grep -i xenomai produced
> >
> > [    0.666100] [Xenomai] scheduling class idle registered.
> > [    0.666112] [Xenomai] scheduling class rt registered.
> > [    0.666309] I-pipe: head domain Xenomai registered.
> > [    0.668400] [Xenomai] Cobalt v3.0-rc7 (Exact Zero)
> >
> >
> > Appreciate any comments, please.
> >
> > thank you
> > _______________________________________________
> > Xenomai mailing list
> > Xenomai@xenomai.org
> > http://xenomai.org/mailman/listinfo/xenomai
> >
>
>
> --
> Philippe.
>

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> By regenerating ./configure using autoreconf -fi, I managed to compile
> xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in their own
> place by make install. However, when running latency, it produces
> 
> > latency
> 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory

You should either run the latency test as root, or have a look at
this page:
http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/


> > dmesg | grep -i ipipe
> [    0.000000] Linux version 3.18.20-ipipe (qws@qws-MacBookPro) (gcc
> version 4.9.2 (Ubuntu/Linaro 4.9.2-10ubuntu10) ) #1 SMP Tue Sep 22 21:04:49
> NZST 2015
> [    0.623763] 3.18.20-ipipe #1 Not tainted
> [    0.623994]   [<c00d3094>] __ipipe_do_sync_stage+0x288/0x2d0
> [    0.624011]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
> [    0.624321] CPU: 1 PID: 22 Comm: kblockd Not tainted 3.18.20-ipipe #1
> [    0.624479] [<c00843b4>] (trace_hardirqs_on_caller) from [<c05da7f4>]
> (__ipipe_fast_svc_irq_exit+0x4/0x18)
> [    0.624504] [<c05da7f4>] (__ipipe_fast_svc_irq_exit) from [<c001aacc>]
> (ipip_test_and_stall_root+0x2c/0x3c)
> [    0.624524] [<c001aacc>] (ipipe_test_and_stall_root) from [<c0086e58>]
> (lock_acquire+0x44/0x108)
> [    0.728394] Switched to clocksource ipipe_tsc
> [    2.464800] usb usb1: Manufacturer: Linux 3.18.20-ipipe ehci_hcd
> [    6.364516] usb usb2: Manufacturer: Linux 3.18.20-ipipe musb-hcd

Please post the full kernel logs. You likely enabled a kernel
configuration option which has an issue with CONFIG_IPIPE

-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[vibnwis]

On 23 September 2015 at 05:44, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > By regenerating ./configure using autoreconf -fi, I managed to compile
> > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in their own
> > place by make install. However, when running latency, it produces
> >
> > > latency
> > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
>
> You should either run the latency test as root, or have a look at
> this page:
> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
>
> By root I get this.
root@arm:/usr/xenomai# latency
   0"000.000| BUG in low_init(): [main] binding failed: Function not
implemented

However, the PATH has set the Lib

root@arm:/usr/xenomai# echo $PATH
/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/usr/xenomai/bin:/usr/xenomai/lib:/usr/xenomai/include


>
> > > dmesg | grep -i ipipe
> > [    0.000000] Linux version 3.18.20-ipipe (qws@qws-MacBookPro) (gcc
> > version 4.9.2 (Ubuntu/Linaro 4.9.2-10ubuntu10) ) #1 SMP Tue Sep 22
> 21:04:49
> > NZST 2015
> > [    0.623763] 3.18.20-ipipe #1 Not tainted
> > [    0.623994]   [<c00d3094>] __ipipe_do_sync_stage+0x288/0x2d0
> > [    0.624011]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
> > [    0.624321] CPU: 1 PID: 22 Comm: kblockd Not tainted 3.18.20-ipipe #1
> > [    0.624479] [<c00843b4>] (trace_hardirqs_on_caller) from [<c05da7f4>]
> > (__ipipe_fast_svc_irq_exit+0x4/0x18)
> > [    0.624504] [<c05da7f4>] (__ipipe_fast_svc_irq_exit) from [<c001aacc>]
> > (ipip_test_and_stall_root+0x2c/0x3c)
> > [    0.624524] [<c001aacc>] (ipipe_test_and_stall_root) from [<c0086e58>]
> > (lock_acquire+0x44/0x108)
> > [    0.728394] Switched to clocksource ipipe_tsc
> > [    2.464800] usb usb1: Manufacturer: Linux 3.18.20-ipipe ehci_hcd
> > [    6.364516] usb usb2: Manufacturer: Linux 3.18.20-ipipe musb-hcd
>
> Please post the full kernel logs. You likely enabled a kernel
> configuration option which has an issue with CONFIG_IPIPE
>
Please find it in attachment. Many thanks (very grateful indeed)

>
> --
>                                             Gilles.
> https://click-hack.org
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: kern.log
Type: text/x-log
Size: 242697 bytes
Desc: not available
URL: <http://xenomai.org/pipermail/xenomai/attachments/20150923/f1729a84/attachment.bin>

Re: [Xenomai] Xenomai tests

[vibnwis]

On 23 September 2015 at 12:41, vibnwis <vibnwis@gmail.com> wrote:

>
>
> On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> gilles.chanteperdrix@xenomai.org> wrote:
>
>> On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
>> > By regenerating ./configure using autoreconf -fi, I managed to compile
>> > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in their
>> own
>> > place by make install. However, when running latency, it produces
>> >
>> > > latency
>> > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
>>
>> You should either run the latency test as root, or have a look at
>> this page:
>>
>> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
>
>
Referring to the link, I have the following outcomes.
/dev/rtpipe <- not exist

/sys/module/xenomai/parameters/allowed_group <- not exist as well

tried with the following commands:

ubuntu@arm:/usr/xenomai/bin$ sudo find /sys/module -type f -name
"allowed_group" -print
[sudo] password for ubuntu:
<nothing came back>

ubuntu@arm:/usr/xenomai/bin$ sudo find /sys/module -type f -name
"parameters" -print
<nothing came back>

Any idea what/where went wrong?


>
>>
>> By root I get this.
> root@arm:/usr/xenomai# latency
>    0"000.000| BUG in low_init(): [main] binding failed: Function not
> implemented
>
> However, the PATH has set the Lib
>
> root@arm:/usr/xenomai# echo $PATH
>
> /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/usr/xenomai/bin:/usr/xenomai/lib:/usr/xenomai/include
>
>
>>
>> > > dmesg | grep -i ipipe
>> > [    0.000000] Linux version 3.18.20-ipipe (qws@qws-MacBookPro) (gcc
>> > version 4.9.2 (Ubuntu/Linaro 4.9.2-10ubuntu10) ) #1 SMP Tue Sep 22
>> 21:04:49
>> > NZST 2015
>> > [    0.623763] 3.18.20-ipipe #1 Not tainted
>> > [    0.623994]   [<c00d3094>] __ipipe_do_sync_stage+0x288/0x2d0
>> > [    0.624011]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
>> > [    0.624321] CPU: 1 PID: 22 Comm: kblockd Not tainted 3.18.20-ipipe #1
>> > [    0.624479] [<c00843b4>] (trace_hardirqs_on_caller) from [<c05da7f4>]
>> > (__ipipe_fast_svc_irq_exit+0x4/0x18)
>> > [    0.624504] [<c05da7f4>] (__ipipe_fast_svc_irq_exit) from
>> [<c001aacc>]
>> > (ipip_test_and_stall_root+0x2c/0x3c)
>> > [    0.624524] [<c001aacc>] (ipipe_test_and_stall_root) from
>> [<c0086e58>]
>> > (lock_acquire+0x44/0x108)
>> > [    0.728394] Switched to clocksource ipipe_tsc
>> > [    2.464800] usb usb1: Manufacturer: Linux 3.18.20-ipipe ehci_hcd
>> > [    6.364516] usb usb2: Manufacturer: Linux 3.18.20-ipipe musb-hcd
>>
>> Please post the full kernel logs. You likely enabled a kernel
>> configuration option which has an issue with CONFIG_IPIPE
>>
> Please find it in attachment. Many thanks (very grateful indeed)
>
>>
>> --
>>                                             Gilles.
>> https://click-hack.org
>>
>
>

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Wed, Sep 23, 2015 at 12:41:38PM +1200, vibnwis wrote:
> On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> gilles.chanteperdrix@xenomai.org> wrote:
> 
> > On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > By regenerating ./configure using autoreconf -fi, I managed to compile
> > > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in their own
> > > place by make install. However, when running latency, it produces
> > >
> > > > latency
> > > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
> >
> > You should either run the latency test as root, or have a look at
> > this page:
> > http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> >
> > By root I get this.
> root@arm:/usr/xenomai# latency
>    0"000.000| BUG in low_init(): [main] binding failed: Function not
> implemented
> 
> However, the PATH has set the Lib

Running as root was necessary to avoid the mlockall issue. The issue
you have now is different: the logs you sent show that the last
kernel you booted is not patched with Xenomai. The kernel is patched
with the I-pipe patch, but this is not sufficient, it is not
prepared with prepare-kernel.sh

> 
> root@arm:/usr/xenomai# echo $PATH
> /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/usr/xenomai/bin:/usr/xenomai/lib:/usr/xenomai/include
> 
> 
> >
> > > > dmesg | grep -i ipipe
> > > [    0.000000] Linux version 3.18.20-ipipe (qws@qws-MacBookPro) (gcc
> > > version 4.9.2 (Ubuntu/Linaro 4.9.2-10ubuntu10) ) #1 SMP Tue Sep 22
> > 21:04:49
> > > NZST 2015
> > > [    0.623763] 3.18.20-ipipe #1 Not tainted
> > > [    0.623994]   [<c00d3094>] __ipipe_do_sync_stage+0x288/0x2d0
> > > [    0.624011]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
> > > [    0.624321] CPU: 1 PID: 22 Comm: kblockd Not tainted 3.18.20-ipipe #1
> > > [    0.624479] [<c00843b4>] (trace_hardirqs_on_caller) from [<c05da7f4>]
> > > (__ipipe_fast_svc_irq_exit+0x4/0x18)
> > > [    0.624504] [<c05da7f4>] (__ipipe_fast_svc_irq_exit) from [<c001aacc>]
> > > (ipip_test_and_stall_root+0x2c/0x3c)
> > > [    0.624524] [<c001aacc>] (ipipe_test_and_stall_root) from [<c0086e58>]
> > > (lock_acquire+0x44/0x108)
> > > [    0.728394] Switched to clocksource ipipe_tsc
> > > [    2.464800] usb usb1: Manufacturer: Linux 3.18.20-ipipe ehci_hcd
> > > [    6.364516] usb usb2: Manufacturer: Linux 3.18.20-ipipe musb-hcd
> >
> > Please post the full kernel logs. You likely enabled a kernel
> > configuration option which has an issue with CONFIG_IPIPE
> >
> Please find it in attachment. Many thanks (very grateful indeed)

Ok, when I say "please post the full kernel logs", I mean only the
logs of the kernel currently running, not the logs of previous
kernel runs, these are irrelevant. Anyway, I think you have
CONFG_CONTEX_TRACKING enabled, which does not work with
CONFIG_IPIPE, at least on ARM. Could you post you kernel
configuration to be sure?

-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Wed, Sep 23, 2015 at 01:21:55PM +1200, vibnwis wrote:
> On 23 September 2015 at 12:41, vibnwis <vibnwis@gmail.com> wrote:
> 
> >
> >
> > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > gilles.chanteperdrix@xenomai.org> wrote:
> >
> >> On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> >> > By regenerating ./configure using autoreconf -fi, I managed to compile
> >> > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in their
> >> own
> >> > place by make install. However, when running latency, it produces
> >> >
> >> > > latency
> >> > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
> >>
> >> You should either run the latency test as root, or have a look at
> >> this page:
> >>
> >> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> >
> >
> Referring to the link, I have the following outcomes.
> /dev/rtpipe <- not exist
> 
> /sys/module/xenomai/parameters/allowed_group <- not exist as well
> 
> tried with the following commands:
> 
> ubuntu@arm:/usr/xenomai/bin$ sudo find /sys/module -type f -name
> "allowed_group" -print
> [sudo] password for ubuntu:
> <nothing came back>
> 
> ubuntu@arm:/usr/xenomai/bin$ sudo find /sys/module -type f -name
> "parameters" -print
> <nothing came back>
> 
> Any idea what/where went wrong?

Same as previous question: it seems you booted a kernel without
Xenomai support.

As explained here:
http://xenomai.org/installing-xenomai-3-x/#Booting_the_Cobalt_kernel

When you boot a kernel with Xenomai support, your kernel logs (the
result of running the "dmesg" command, not the persistent ones saved
in /var/log/kern.log), should contain a message about Cobalt being
enabled.

-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Wed, Sep 23, 2015 at 07:43:19AM +0200, Gilles Chanteperdrix wrote:
> On Wed, Sep 23, 2015 at 01:21:55PM +1200, vibnwis wrote:
> > On 23 September 2015 at 12:41, vibnwis <vibnwis@gmail.com> wrote:
> > 
> > >
> > >
> > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > gilles.chanteperdrix@xenomai.org> wrote:
> > >
> > >> On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > >> > By regenerating ./configure using autoreconf -fi, I managed to compile
> > >> > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in their
> > >> own
> > >> > place by make install. However, when running latency, it produces
> > >> >
> > >> > > latency
> > >> > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
> > >>
> > >> You should either run the latency test as root, or have a look at
> > >> this page:
> > >>
> > >> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > >
> > >
> > Referring to the link, I have the following outcomes.
> > /dev/rtpipe <- not exist
> > 
> > /sys/module/xenomai/parameters/allowed_group <- not exist as
> well

So, this is probably because you are not running a kernel with
xenomai support.

> > 
> > tried with the following commands:
> > 
> > ubuntu@arm:/usr/xenomai/bin$ sudo find /sys/module -type f -name
> > "allowed_group" -print
> > [sudo] password for ubuntu:
> > <nothing came back>
> > 
> > ubuntu@arm:/usr/xenomai/bin$ sudo find /sys/module -type f -name
> > "parameters" -print
> > <nothing came back>

But there should be a few files named "parameters" under
/sys/module. Maybe you will see them if you remove "type f" from
find command line ?

-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[vibnwis]

On 23 September 2015 at 17:38, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> On Wed, Sep 23, 2015 at 12:41:38PM +1200, vibnwis wrote:
> > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > gilles.chanteperdrix@xenomai.org> wrote:
> >
> > > On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > By regenerating ./configure using autoreconf -fi, I managed to
> compile
> > > > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in
> their own
> > > > place by make install. However, when running latency, it produces
> > > >
> > > > > latency
> > > > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
> > >
> > > You should either run the latency test as root, or have a look at
> > > this page:
> > >
> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > >
> > > By root I get this.
> > root@arm:/usr/xenomai# latency
> >    0"000.000| BUG in low_init(): [main] binding failed: Function not
> > implemented
> >
> > However, the PATH has set the Lib
>
> Running as root was necessary to avoid the mlockall issue. The issue
> you have now is different: the logs you sent show that the last
> kernel you booted is not patched with Xenomai. The kernel is patched
> with the I-pipe patch, but this is not sufficient, it is not
> prepared with prepare-kernel.sh
>

I had thought of that too. However, the patch namely
ipipe-core-3.18.20-arm-5.patch I used is meant for kernel 3.18.20. And
there is no such patch available in the latest xenomai-3-3.0-rc7. ie under
the kernel/Cobalt/arch/arm/patches/, there is such patch available for
3.18.20. With the observation, I concluded that
ipipe-core-3.18.20-arm-5.patch would patch all is needed for xenomai to
work. In addition, I checked in its make menuconfig, I did not see Cobalt
feature, but I did see ipipe was integrated. Obviously now I know it was
wrong. However, I still yet found a xenomai which has patches for 3.18.20
kernel. Any comments?

>
> >
> > root@arm:/usr/xenomai# echo $PATH
> >
> /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/usr/xenomai/bin:/usr/xenomai/lib:/usr/xenomai/include
> >
> >
> > >
> > > > > dmesg | grep -i ipipe
> > > > [    0.000000] Linux version 3.18.20-ipipe (qws@qws-MacBookPro) (gcc
> > > > version 4.9.2 (Ubuntu/Linaro 4.9.2-10ubuntu10) ) #1 SMP Tue Sep 22
> > > 21:04:49
> > > > NZST 2015
> > > > [    0.623763] 3.18.20-ipipe #1 Not tainted
> > > > [    0.623994]   [<c00d3094>] __ipipe_do_sync_stage+0x288/0x2d0
> > > > [    0.624011]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
> > > > [    0.624321] CPU: 1 PID: 22 Comm: kblockd Not tainted
> 3.18.20-ipipe #1
> > > > [    0.624479] [<c00843b4>] (trace_hardirqs_on_caller) from
> [<c05da7f4>]
> > > > (__ipipe_fast_svc_irq_exit+0x4/0x18)
> > > > [    0.624504] [<c05da7f4>] (__ipipe_fast_svc_irq_exit) from
> [<c001aacc>]
> > > > (ipip_test_and_stall_root+0x2c/0x3c)
> > > > [    0.624524] [<c001aacc>] (ipipe_test_and_stall_root) from
> [<c0086e58>]
> > > > (lock_acquire+0x44/0x108)
> > > > [    0.728394] Switched to clocksource ipipe_tsc
> > > > [    2.464800] usb usb1: Manufacturer: Linux 3.18.20-ipipe ehci_hcd
> > > > [    6.364516] usb usb2: Manufacturer: Linux 3.18.20-ipipe musb-hcd
> > >
> > > Please post the full kernel logs. You likely enabled a kernel
> > > configuration option which has an issue with CONFIG_IPIPE
> > >
> > Please find it in attachment. Many thanks (very grateful indeed)
>
> Ok, when I say "please post the full kernel logs", I mean only the
> logs of the kernel currently running, not the logs of previous
> kernel runs, these are irrelevant. Anyway, I think you have
> CONFG_CONTEX_TRACKING enabled, which does not work with
> CONFIG_IPIPE, at least on ARM. Could you post you kernel
> configuration to be sure?
>

My apologies, there were two kern.log and kern.log.1. I had taken the wrong
perhaps. To make sure I got it right, I deleted it and reboot the system.
Attached is the newly created kern.log


>
> --
>                                             Gilles.
> https://click-hack.org
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: kern.log
Type: text/x-log
Size: 40986 bytes
Desc: not available
URL: <http://xenomai.org/pipermail/xenomai/attachments/20150923/91da454f/attachment.bin>

Re: [Xenomai] Xenomai tests

[vibnwis]

On 23 September 2015 at 17:54, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> On Wed, Sep 23, 2015 at 07:43:19AM +0200, Gilles Chanteperdrix wrote:
> > On Wed, Sep 23, 2015 at 01:21:55PM +1200, vibnwis wrote:
> > > On 23 September 2015 at 12:41, vibnwis <vibnwis@gmail.com> wrote:
> > >
> > > >
> > > >
> > > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > > gilles.chanteperdrix@xenomai.org> wrote:
> > > >
> > > >> On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > >> > By regenerating ./configure using autoreconf -fi, I managed to
> compile
> > > >> > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in
> their
> > > >> own
> > > >> > place by make install. However, when running latency, it produces
> > > >> >
> > > >> > > latency
> > > >> > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate
> memory
> > > >>
> > > >> You should either run the latency test as root, or have a look at
> > > >> this page:
> > > >>
> > > >>
> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > > >
> > > >
> > > Referring to the link, I have the following outcomes.
> > > /dev/rtpipe <- not exist
> > >
> > > /sys/module/xenomai/parameters/allowed_group <- not exist as
> > well
>
> So, this is probably because you are not running a kernel with
> xenomai support.
>
Since there isn't any 3.18.20 patch, should I use 3.18.12. It is still
3.8.x. Should be alright ? I will give it a go.

>
> > >
> > > tried with the following commands:
> > >
> > > ubuntu@arm:/usr/xenomai/bin$ sudo find /sys/module -type f -name
> > > "allowed_group" -print
> > > [sudo] password for ubuntu:
> > > <nothing came back>
> > >
> > > ubuntu@arm:/usr/xenomai/bin$ sudo find /sys/module -type f -name
> > > "parameters" -print
> > > <nothing came back>
>
> But there should be a few files named "parameters" under
> /sys/module. Maybe you will see them if you remove "type f" from
> find command line ?
>
> --
>                                             Gilles.
> https://click-hack.org
>

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Wed, Sep 23, 2015 at 06:29:25PM +1200, vibnwis wrote:
> On 23 September 2015 at 17:38, Gilles Chanteperdrix <
> gilles.chanteperdrix@xenomai.org> wrote:
> 
> > On Wed, Sep 23, 2015 at 12:41:38PM +1200, vibnwis wrote:
> > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > gilles.chanteperdrix@xenomai.org> wrote:
> > >
> > > > On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > > By regenerating ./configure using autoreconf -fi, I managed to
> > compile
> > > > > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in
> > their own
> > > > > place by make install. However, when running latency, it produces
> > > > >
> > > > > > latency
> > > > > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
> > > >
> > > > You should either run the latency test as root, or have a look at
> > > > this page:
> > > >
> > http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > > >
> > > > By root I get this.
> > > root@arm:/usr/xenomai# latency
> > >    0"000.000| BUG in low_init(): [main] binding failed: Function not
> > > implemented
> > >
> > > However, the PATH has set the Lib
> >
> > Running as root was necessary to avoid the mlockall issue. The issue
> > you have now is different: the logs you sent show that the last
> > kernel you booted is not patched with Xenomai. The kernel is patched
> > with the I-pipe patch, but this is not sufficient, it is not
> > prepared with prepare-kernel.sh
> >
> 
> I had thought of that too. However, the patch namely
> ipipe-core-3.18.20-arm-5.patch I used is meant for kernel 3.18.20. And
> there is no such patch available in the latest xenomai-3-3.0-rc7. ie under
> the kernel/Cobalt/arch/arm/patches/, there is such patch available for
> 3.18.20. With the observation, I concluded that
> ipipe-core-3.18.20-arm-5.patch would patch all is needed for xenomai to
> work. In addition, I checked in its make menuconfig, I did not see Cobalt
> feature, but I did see ipipe was integrated. Obviously now I know it was
> wrong. However, I still yet found a xenomai which has patches for 3.18.20
> kernel. Any comments?

You can use Xenomai 3.0-rc7 with the I-pipe patch for 3.18.20.
Simply pass the path to the patch to the prepare-kernel.sh --adeos
option.

> > Ok, when I say "please post the full kernel logs", I mean only the
> > logs of the kernel currently running, not the logs of previous
> > kernel runs, these are irrelevant. Anyway, I think you have
> > CONFG_CONTEX_TRACKING enabled, which does not work with
> > CONFIG_IPIPE, at least on ARM. Could you post you kernel
> > configuration to be sure?
> >
> 
> My apologies, there were two kern.log and kern.log.1. I had taken the wrong
> perhaps. To make sure I got it right, I deleted it and reboot the system.
> Attached is the newly created kern.log

It is simpler to send the output of the "dmesg" command. Anyway, now
that we know what the problem is, you do not need to send the logs
again.

-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Wed, Sep 23, 2015 at 06:33:57PM +1200, vibnwis wrote:
> On 23 September 2015 at 17:54, Gilles Chanteperdrix <
> gilles.chanteperdrix@xenomai.org> wrote:
> 
> > On Wed, Sep 23, 2015 at 07:43:19AM +0200, Gilles Chanteperdrix wrote:
> > > On Wed, Sep 23, 2015 at 01:21:55PM +1200, vibnwis wrote:
> > > > On 23 September 2015 at 12:41, vibnwis <vibnwis@gmail.com> wrote:
> > > >
> > > > >
> > > > >
> > > > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > > > gilles.chanteperdrix@xenomai.org> wrote:
> > > > >
> > > > >> On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > >> > By regenerating ./configure using autoreconf -fi, I managed to
> > compile
> > > > >> > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in
> > their
> > > > >> own
> > > > >> > place by make install. However, when running latency, it produces
> > > > >> >
> > > > >> > > latency
> > > > >> > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate
> > memory
> > > > >>
> > > > >> You should either run the latency test as root, or have a look at
> > > > >> this page:
> > > > >>
> > > > >>
> > http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > > > >
> > > > >
> > > > Referring to the link, I have the following outcomes.
> > > > /dev/rtpipe <- not exist
> > > >
> > > > /sys/module/xenomai/parameters/allowed_group <- not exist as
> > > well
> >
> > So, this is probably because you are not running a kernel with
> > xenomai support.
> >
> Since there isn't any 3.18.20 patch, should I use 3.18.12. It is still
> 3.8.x. Should be alright ? I will give it a go.

Please see my answer to your previous mail. You can use the patch
for 3.18.20 with xenomai 3.0-rc7.

-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Wed, Sep 23, 2015 at 07:38:43AM +0200, Gilles Chanteperdrix wrote:
> On Wed, Sep 23, 2015 at 12:41:38PM +1200, vibnwis wrote:
> > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > gilles.chanteperdrix@xenomai.org> wrote:
> > 
> > > On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > By regenerating ./configure using autoreconf -fi, I managed to compile
> > > > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in their own
> > > > place by make install. However, when running latency, it produces
> > > >
> > > > > latency
> > > > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate memory
> > >
> > > You should either run the latency test as root, or have a look at
> > > this page:
> > > http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > >
> > > By root I get this.
> > root@arm:/usr/xenomai# latency
> >    0"000.000| BUG in low_init(): [main] binding failed: Function not
> > implemented
> > 
> > However, the PATH has set the Lib
> 
> Running as root was necessary to avoid the mlockall issue. The issue
> you have now is different: the logs you sent show that the last
> kernel you booted is not patched with Xenomai. The kernel is patched
> with the I-pipe patch, but this is not sufficient, it is not
> prepared with prepare-kernel.sh
> 
> > 
> > root@arm:/usr/xenomai# echo $PATH
> > /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/usr/xenomai/bin:/usr/xenomai/lib:/usr/xenomai/include
> > 
> > 
> > >
> > > > > dmesg | grep -i ipipe
> > > > [    0.000000] Linux version 3.18.20-ipipe (qws@qws-MacBookPro) (gcc
> > > > version 4.9.2 (Ubuntu/Linaro 4.9.2-10ubuntu10) ) #1 SMP Tue Sep 22
> > > 21:04:49
> > > > NZST 2015
> > > > [    0.623763] 3.18.20-ipipe #1 Not tainted
> > > > [    0.623994]   [<c00d3094>] __ipipe_do_sync_stage+0x288/0x2d0
> > > > [    0.624011]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
> > > > [    0.624321] CPU: 1 PID: 22 Comm: kblockd Not tainted 3.18.20-ipipe #1
> > > > [    0.624479] [<c00843b4>] (trace_hardirqs_on_caller) from [<c05da7f4>]
> > > > (__ipipe_fast_svc_irq_exit+0x4/0x18)
> > > > [    0.624504] [<c05da7f4>] (__ipipe_fast_svc_irq_exit) from [<c001aacc>]
> > > > (ipip_test_and_stall_root+0x2c/0x3c)
> > > > [    0.624524] [<c001aacc>] (ipipe_test_and_stall_root) from [<c0086e58>]
> > > > (lock_acquire+0x44/0x108)
> > > > [    0.728394] Switched to clocksource ipipe_tsc
> > > > [    2.464800] usb usb1: Manufacturer: Linux 3.18.20-ipipe ehci_hcd
> > > > [    6.364516] usb usb2: Manufacturer: Linux 3.18.20-ipipe musb-hcd
> > >
> > > Please post the full kernel logs. You likely enabled a kernel
> > > configuration option which has an issue with CONFIG_IPIPE
> > >
> > Please find it in attachment. Many thanks (very grateful indeed)
> 
> Ok, when I say "please post the full kernel logs", I mean only the
> logs of the kernel currently running, not the logs of previous
> kernel runs, these are irrelevant. Anyway, I think you have
> CONFG_CONTEX_TRACKING enabled, which does not work with
> CONFIG_IPIPE, at least on ARM. Could you post you kernel
> configuration to be sure?

The problematic option is CONFIG_TRACE_IRQFLAGS, not
CONFIG_CONTEXT_TRACKING, and you have it enabled, since without it,
you would not be able to enable CONFIG_LOCKDEP. Please disable
CONFIG_TRACE_IRQFLAGS. You will need to disable all the options
depending on that option, such as CONFIG_LOCKDEP.

-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[vibnwis]

On 23 September 2015 at 18:56, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> On Wed, Sep 23, 2015 at 06:33:57PM +1200, vibnwis wrote:
> > On 23 September 2015 at 17:54, Gilles Chanteperdrix <
> > gilles.chanteperdrix@xenomai.org> wrote:
> >
> > > On Wed, Sep 23, 2015 at 07:43:19AM +0200, Gilles Chanteperdrix wrote:
> > > > On Wed, Sep 23, 2015 at 01:21:55PM +1200, vibnwis wrote:
> > > > > On 23 September 2015 at 12:41, vibnwis <vibnwis@gmail.com> wrote:
> > > > >
> > > > > >
> > > > > >
> > > > > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > > > > gilles.chanteperdrix@xenomai.org> wrote:
> > > > > >
> > > > > >> On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > > >> > By regenerating ./configure using autoreconf -fi, I managed to
> > > compile
> > > > > >> > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put
> in
> > > their
> > > > > >> own
> > > > > >> > place by make install. However, when running latency, it
> produces
> > > > > >> >
> > > > > >> > > latency
> > > > > >> > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate
> > > memory
> > > > > >>
> > > > > >> You should either run the latency test as root, or have a look
> at
> > > > > >> this page:
> > > > > >>
> > > > > >>
> > >
> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > > > > >
> > > > > >
> > > > > Referring to the link, I have the following outcomes.
> > > > > /dev/rtpipe <- not exist
> > > > >
> > > > > /sys/module/xenomai/parameters/allowed_group <- not exist as
> > > > well
> > >
> > > So, this is probably because you are not running a kernel with
> > > xenomai support.
> > >
> > Since there isn't any 3.18.20 patch, should I use 3.18.12. It is still
> > 3.8.x. Should be alright ? I will give it a go.
>
> Please see my answer to your previous mail. You can use the patch
> for 3.18.20 with xenomai 3.0-rc7. I will revert back to 3.18.12.
>

Tried it, but failed. See below.

./scripts/prepare-kernel.sh --arch=arm --linux=../linux-3.18.20
--ipipe=./kernel/cobalt/arch/arm/patches/ipipe-core-3.18.12-arm-4.patch

 Hunk #1 succeeded at 1804 (offset -3 lines).
Hunk #2 succeeded at 1864 (offset -3 lines).
Hunk #3 succeeded at 1934 (offset -3 lines).
checking file kernel/sched/core.c
Hunk #15 FAILED at 3650.
Hunk #16 succeeded at 4773 (offset 4 lines).
Hunk #17 succeeded at 8298 (offset 4 lines).
1 out of 17 hunks FAILED
checking file kernel/sched/wait.c
checking file kernel/signal.c
checking file kernel/time/clockevents.c
checking file kernel/time/clocksource.c
checking file kernel/time/tick-common.c
checking file kernel/time/tick-sched.c
checking file kernel/time/timekeeping.c
checking file kernel/time/timer.c
checking file kernel/timer.c
checking file kernel/trace/Kconfig
checking file kernel/trace/ftrace.c
checking file kernel/trace/ring_buffer.c
Hunk #1 FAILED at 2684.
Hunk #2 FAILED at 2697.
2 out of 2 hunks FAILED
checking file kernel/trace/trace.c
checking file kernel/trace/trace_clock.c
checking file kernel/trace/trace_functions.c
checking file kernel/trace/trace_functions_graph.c
checking file lib/Kconfig.debug
checking file lib/atomic64.c
checking file lib/bust_spinlocks.c
checking file lib/ioremap.c
checking file lib/smp_processor_id.c
checking file mm/Kconfig
checking file mm/memory.c
checking file mm/mlock.c
checking file mm/mmap.c
checking file mm/mmu_context.c
checking file mm/mprotect.c
checking file mm/vmalloc.c
prepare-kernel.sh: Unable to patch kernel 3.18.20 with
ipipe-core-3.18.12-arm-4.patch.

>
> --
>                                             Gilles.
> https://click-hack.org
>

Re: [Xenomai] Xenomai tests

[vibnwis]

On 23 September 2015 at 19:06, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> On Wed, Sep 23, 2015 at 07:38:43AM +0200, Gilles Chanteperdrix wrote:
> > On Wed, Sep 23, 2015 at 12:41:38PM +1200, vibnwis wrote:
> > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > gilles.chanteperdrix@xenomai.org> wrote:
> > >
> > > > On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > > By regenerating ./configure using autoreconf -fi, I managed to
> compile
> > > > > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in
> their own
> > > > > place by make install. However, when running latency, it produces
> > > > >
> > > > > > latency
> > > > > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate
> memory
> > > >
> > > > You should either run the latency test as root, or have a look at
> > > > this page:
> > > >
> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > > >
> > > > By root I get this.
> > > root@arm:/usr/xenomai# latency
> > >    0"000.000| BUG in low_init(): [main] binding failed: Function not
> > > implemented
> > >
> > > However, the PATH has set the Lib
> >
> > Running as root was necessary to avoid the mlockall issue. The issue
> > you have now is different: the logs you sent show that the last
> > kernel you booted is not patched with Xenomai. The kernel is patched
> > with the I-pipe patch, but this is not sufficient, it is not
> > prepared with prepare-kernel.sh
> >
> > >
> > > root@arm:/usr/xenomai# echo $PATH
> > >
> /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/usr/xenomai/bin:/usr/xenomai/lib:/usr/xenomai/include
> > >
> > >
> > > >
> > > > > > dmesg | grep -i ipipe
> > > > > [    0.000000] Linux version 3.18.20-ipipe (qws@qws-MacBookPro)
> (gcc
> > > > > version 4.9.2 (Ubuntu/Linaro 4.9.2-10ubuntu10) ) #1 SMP Tue Sep 22
> > > > 21:04:49
> > > > > NZST 2015
> > > > > [    0.623763] 3.18.20-ipipe #1 Not tainted
> > > > > [    0.623994]   [<c00d3094>] __ipipe_do_sync_stage+0x288/0x2d0
> > > > > [    0.624011]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
> > > > > [    0.624321] CPU: 1 PID: 22 Comm: kblockd Not tainted
> 3.18.20-ipipe #1
> > > > > [    0.624479] [<c00843b4>] (trace_hardirqs_on_caller) from
> [<c05da7f4>]
> > > > > (__ipipe_fast_svc_irq_exit+0x4/0x18)
> > > > > [    0.624504] [<c05da7f4>] (__ipipe_fast_svc_irq_exit) from
> [<c001aacc>]
> > > > > (ipip_test_and_stall_root+0x2c/0x3c)
> > > > > [    0.624524] [<c001aacc>] (ipipe_test_and_stall_root) from
> [<c0086e58>]
> > > > > (lock_acquire+0x44/0x108)
> > > > > [    0.728394] Switched to clocksource ipipe_tsc
> > > > > [    2.464800] usb usb1: Manufacturer: Linux 3.18.20-ipipe ehci_hcd
> > > > > [    6.364516] usb usb2: Manufacturer: Linux 3.18.20-ipipe musb-hcd
> > > >
> > > > Please post the full kernel logs. You likely enabled a kernel
> > > > configuration option which has an issue with CONFIG_IPIPE
> > > >
> > > Please find it in attachment. Many thanks (very grateful indeed)
> >
> > Ok, when I say "please post the full kernel logs", I mean only the
> > logs of the kernel currently running, not the logs of previous
> > kernel runs, these are irrelevant. Anyway, I think you have
> > CONFG_CONTEX_TRACKING enabled, which does not work with
> > CONFIG_IPIPE, at least on ARM. Could you post you kernel
> > configuration to be sure?
>
> The problematic option is CONFIG_TRACE_IRQFLAGS, not
> CONFIG_CONTEXT_TRACKING, and you have it enabled, since without it,
> you would not be able to enable CONFIG_LOCKDEP. Please disable
> CONFIG_TRACE_IRQFLAGS. You will need to disable all the options
> depending on that option, such as CONFIG_LOCKDEP.
>

Spent hours looking for the item in kernel config (make menuconfig), but
could find any. Would you please share where that is, please?

I googled it and it looks like under Kernel Hacking but, tried one by one
of those options, nothing came up would unset CONFIG_TRACE_IRQFLAGS_SUPPORT

#
# Kernel hacking
#
CONFIG_TRACE_IRQFLAGS_SUPPORT=y
# CONFIG_PRINTK_TIME is not set
CONFIG_ENABLE_MUST_CHECK=y



>
> --
>                                             Gilles.
> https://click-hack.org
>

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Wed, Sep 23, 2015 at 10:38:05PM +1200, vibnwis wrote:
> On 23 September 2015 at 18:56, Gilles Chanteperdrix <
> gilles.chanteperdrix@xenomai.org> wrote:
> 
> > On Wed, Sep 23, 2015 at 06:33:57PM +1200, vibnwis wrote:
> > > On 23 September 2015 at 17:54, Gilles Chanteperdrix <
> > > gilles.chanteperdrix@xenomai.org> wrote:
> > >
> > > > On Wed, Sep 23, 2015 at 07:43:19AM +0200, Gilles Chanteperdrix wrote:
> > > > > On Wed, Sep 23, 2015 at 01:21:55PM +1200, vibnwis wrote:
> > > > > > On 23 September 2015 at 12:41, vibnwis <vibnwis@gmail.com> wrote:
> > > > > >
> > > > > > >
> > > > > > >
> > > > > > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > > > > > gilles.chanteperdrix@xenomai.org> wrote:
> > > > > > >
> > > > > > >> On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > > > >> > By regenerating ./configure using autoreconf -fi, I managed to
> > > > compile
> > > > > > >> > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put
> > in
> > > > their
> > > > > > >> own
> > > > > > >> > place by make install. However, when running latency, it
> > produces
> > > > > > >> >
> > > > > > >> > > latency
> > > > > > >> > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate
> > > > memory
> > > > > > >>
> > > > > > >> You should either run the latency test as root, or have a look
> > at
> > > > > > >> this page:
> > > > > > >>
> > > > > > >>
> > > >
> > http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > > > > > >
> > > > > > >
> > > > > > Referring to the link, I have the following outcomes.
> > > > > > /dev/rtpipe <- not exist
> > > > > >
> > > > > > /sys/module/xenomai/parameters/allowed_group <- not exist as
> > > > > well
> > > >
> > > > So, this is probably because you are not running a kernel with
> > > > xenomai support.
> > > >
> > > Since there isn't any 3.18.20 patch, should I use 3.18.12. It is still
> > > 3.8.x. Should be alright ? I will give it a go.
> >
> > Please see my answer to your previous mail. You can use the patch
> > for 3.18.20 with xenomai 3.0-rc7. I will revert back to 3.18.12.
> >
> 
> Tried it, but failed. See below.
> 
> ./scripts/prepare-kernel.sh --arch=arm --linux=../linux-3.18.20
> --ipipe=./kernel/cobalt/arch/arm/patches/ipipe-core-3.18.12-arm-4.patch

ipipe-core-3.18.12* is for the Linux 3.18.12 kernel. The I-pipe
patch for Linux 3.18.20 is named ipipe-core-3.18.20something
-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Wed, Sep 23, 2015 at 11:29:21PM +1200, vibnwis wrote:
> On 23 September 2015 at 19:06, Gilles Chanteperdrix <
> gilles.chanteperdrix@xenomai.org> wrote:
> 
> > On Wed, Sep 23, 2015 at 07:38:43AM +0200, Gilles Chanteperdrix wrote:
> > > On Wed, Sep 23, 2015 at 12:41:38PM +1200, vibnwis wrote:
> > > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > > gilles.chanteperdrix@xenomai.org> wrote:
> > > >
> > > > > On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > > > By regenerating ./configure using autoreconf -fi, I managed to
> > compile
> > > > > > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in
> > their own
> > > > > > place by make install. However, when running latency, it produces
> > > > > >
> > > > > > > latency
> > > > > > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate
> > memory
> > > > >
> > > > > You should either run the latency test as root, or have a look at
> > > > > this page:
> > > > >
> > http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > > > >
> > > > > By root I get this.
> > > > root@arm:/usr/xenomai# latency
> > > >    0"000.000| BUG in low_init(): [main] binding failed: Function not
> > > > implemented
> > > >
> > > > However, the PATH has set the Lib
> > >
> > > Running as root was necessary to avoid the mlockall issue. The issue
> > > you have now is different: the logs you sent show that the last
> > > kernel you booted is not patched with Xenomai. The kernel is patched
> > > with the I-pipe patch, but this is not sufficient, it is not
> > > prepared with prepare-kernel.sh
> > >
> > > >
> > > > root@arm:/usr/xenomai# echo $PATH
> > > >
> > /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/usr/xenomai/bin:/usr/xenomai/lib:/usr/xenomai/include
> > > >
> > > >
> > > > >
> > > > > > > dmesg | grep -i ipipe
> > > > > > [    0.000000] Linux version 3.18.20-ipipe (qws@qws-MacBookPro)
> > (gcc
> > > > > > version 4.9.2 (Ubuntu/Linaro 4.9.2-10ubuntu10) ) #1 SMP Tue Sep 22
> > > > > 21:04:49
> > > > > > NZST 2015
> > > > > > [    0.623763] 3.18.20-ipipe #1 Not tainted
> > > > > > [    0.623994]   [<c00d3094>] __ipipe_do_sync_stage+0x288/0x2d0
> > > > > > [    0.624011]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
> > > > > > [    0.624321] CPU: 1 PID: 22 Comm: kblockd Not tainted
> > 3.18.20-ipipe #1
> > > > > > [    0.624479] [<c00843b4>] (trace_hardirqs_on_caller) from
> > [<c05da7f4>]
> > > > > > (__ipipe_fast_svc_irq_exit+0x4/0x18)
> > > > > > [    0.624504] [<c05da7f4>] (__ipipe_fast_svc_irq_exit) from
> > [<c001aacc>]
> > > > > > (ipip_test_and_stall_root+0x2c/0x3c)
> > > > > > [    0.624524] [<c001aacc>] (ipipe_test_and_stall_root) from
> > [<c0086e58>]
> > > > > > (lock_acquire+0x44/0x108)
> > > > > > [    0.728394] Switched to clocksource ipipe_tsc
> > > > > > [    2.464800] usb usb1: Manufacturer: Linux 3.18.20-ipipe ehci_hcd
> > > > > > [    6.364516] usb usb2: Manufacturer: Linux 3.18.20-ipipe musb-hcd
> > > > >
> > > > > Please post the full kernel logs. You likely enabled a kernel
> > > > > configuration option which has an issue with CONFIG_IPIPE
> > > > >
> > > > Please find it in attachment. Many thanks (very grateful indeed)
> > >
> > > Ok, when I say "please post the full kernel logs", I mean only the
> > > logs of the kernel currently running, not the logs of previous
> > > kernel runs, these are irrelevant. Anyway, I think you have
> > > CONFG_CONTEX_TRACKING enabled, which does not work with
> > > CONFIG_IPIPE, at least on ARM. Could you post you kernel
> > > configuration to be sure?
> >
> > The problematic option is CONFIG_TRACE_IRQFLAGS, not
> > CONFIG_CONTEXT_TRACKING, and you have it enabled, since without it,
> > you would not be able to enable CONFIG_LOCKDEP. Please disable
> > CONFIG_TRACE_IRQFLAGS. You will need to disable all the options
> > depending on that option, such as CONFIG_LOCKDEP.
> >
> 
> Spent hours looking for the item in kernel config (make menuconfig), but
> could find any. Would you please share where that is, please?

In make menuconfig type the / key, then type "lockdep".

-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[vibnwis]

On 23 September 2015 at 23:59, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> On Wed, Sep 23, 2015 at 11:29:21PM +1200, vibnwis wrote:
> > On 23 September 2015 at 19:06, Gilles Chanteperdrix <
> > gilles.chanteperdrix@xenomai.org> wrote:
> >
> > > On Wed, Sep 23, 2015 at 07:38:43AM +0200, Gilles Chanteperdrix wrote:
> > > > On Wed, Sep 23, 2015 at 12:41:38PM +1200, vibnwis wrote:
> > > > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > > > gilles.chanteperdrix@xenomai.org> wrote:
> > > > >
> > > > > > On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > > > > By regenerating ./configure using autoreconf -fi, I managed to
> > > compile
> > > > > > > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in
> > > their own
> > > > > > > place by make install. However, when running latency, it
> produces
> > > > > > >
> > > > > > > > latency
> > > > > > > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate
> > > memory
> > > > > >
> > > > > > You should either run the latency test as root, or have a look at
> > > > > > this page:
> > > > > >
> > >
> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > > > > >
> > > > > > By root I get this.
> > > > > root@arm:/usr/xenomai# latency
> > > > >    0"000.000| BUG in low_init(): [main] binding failed: Function
> not
> > > > > implemented
> > > > >
> > > > > However, the PATH has set the Lib
> > > >
> > > > Running as root was necessary to avoid the mlockall issue. The issue
> > > > you have now is different: the logs you sent show that the last
> > > > kernel you booted is not patched with Xenomai. The kernel is patched
> > > > with the I-pipe patch, but this is not sufficient, it is not
> > > > prepared with prepare-kernel.sh
> > > >
> > > > >
> > > > > root@arm:/usr/xenomai# echo $PATH
> > > > >
> > >
> /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/usr/xenomai/bin:/usr/xenomai/lib:/usr/xenomai/include
> > > > >
> > > > >
> > > > > >
> > > > > > > > dmesg | grep -i ipipe
> > > > > > > [    0.000000] Linux version 3.18.20-ipipe (qws@qws-MacBookPro
> )
> > > (gcc
> > > > > > > version 4.9.2 (Ubuntu/Linaro 4.9.2-10ubuntu10) ) #1 SMP Tue
> Sep 22
> > > > > > 21:04:49
> > > > > > > NZST 2015
> > > > > > > [    0.623763] 3.18.20-ipipe #1 Not tainted
> > > > > > > [    0.623994]   [<c00d3094>] __ipipe_do_sync_stage+0x288/0x2d0
> > > > > > > [    0.624011]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
> > > > > > > [    0.624321] CPU: 1 PID: 22 Comm: kblockd Not tainted
> > > 3.18.20-ipipe #1
> > > > > > > [    0.624479] [<c00843b4>] (trace_hardirqs_on_caller) from
> > > [<c05da7f4>]
> > > > > > > (__ipipe_fast_svc_irq_exit+0x4/0x18)
> > > > > > > [    0.624504] [<c05da7f4>] (__ipipe_fast_svc_irq_exit) from
> > > [<c001aacc>]
> > > > > > > (ipip_test_and_stall_root+0x2c/0x3c)
> > > > > > > [    0.624524] [<c001aacc>] (ipipe_test_and_stall_root) from
> > > [<c0086e58>]
> > > > > > > (lock_acquire+0x44/0x108)
> > > > > > > [    0.728394] Switched to clocksource ipipe_tsc
> > > > > > > [    2.464800] usb usb1: Manufacturer: Linux 3.18.20-ipipe
> ehci_hcd
> > > > > > > [    6.364516] usb usb2: Manufacturer: Linux 3.18.20-ipipe
> musb-hcd
> > > > > >
> > > > > > Please post the full kernel logs. You likely enabled a kernel
> > > > > > configuration option which has an issue with CONFIG_IPIPE
> > > > > >
> > > > > Please find it in attachment. Many thanks (very grateful indeed)
> > > >
> > > > Ok, when I say "please post the full kernel logs", I mean only the
> > > > logs of the kernel currently running, not the logs of previous
> > > > kernel runs, these are irrelevant. Anyway, I think you have
> > > > CONFG_CONTEX_TRACKING enabled, which does not work with
> > > > CONFIG_IPIPE, at least on ARM. Could you post you kernel
> > > > configuration to be sure?
> > >
> > > The problematic option is CONFIG_TRACE_IRQFLAGS, not
> > > CONFIG_CONTEXT_TRACKING, and you have it enabled, since without it,
> > > you would not be able to enable CONFIG_LOCKDEP. Please disable
> > > CONFIG_TRACE_IRQFLAGS. You will need to disable all the options
> > > depending on that option, such as CONFIG_LOCKDEP.
> > >
> >
> > Spent hours looking for the item in kernel config (make menuconfig), but
> > could find any. Would you please share where that is, please?
>
> In make menuconfig type the / key, then type "lockdep".
>

Many thanks. Under kernel hacking, I disable everything in there, yet the
CONFIG_TRACE_IRQFLAGS and CONFIG_LOCKDEP remain unchanged. FYI, I used
omap2plus_defconfig. Would it be the cause of that?

>
> --
>                                             Gilles.
> https://click-hack.org
>

Re: [Xenomai] Xenomai tests

[vibnwis]

On 23 September 2015 at 18:55, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> On Wed, Sep 23, 2015 at 06:29:25PM +1200, vibnwis wrote:
> > On 23 September 2015 at 17:38, Gilles Chanteperdrix <
> > gilles.chanteperdrix@xenomai.org> wrote:
> >
> > > On Wed, Sep 23, 2015 at 12:41:38PM +1200, vibnwis wrote:
> > > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > > gilles.chanteperdrix@xenomai.org> wrote:
> > > >
> > > > > On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > > > By regenerating ./configure using autoreconf -fi, I managed to
> > > compile
> > > > > > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in
> > > their own
> > > > > > place by make install. However, when running latency, it produces
> > > > > >
> > > > > > > latency
> > > > > > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate
> memory
> > > > >
> > > > > You should either run the latency test as root, or have a look at
> > > > > this page:
> > > > >
> > >
> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > > > >
> > > > > By root I get this.
> > > > root@arm:/usr/xenomai# latency
> > > >    0"000.000| BUG in low_init(): [main] binding failed: Function not
> > > > implemented
> > > >
> > > > However, the PATH has set the Lib
> > >
> > > Running as root was necessary to avoid the mlockall issue. The issue
> > > you have now is different: the logs you sent show that the last
> > > kernel you booted is not patched with Xenomai. The kernel is patched
> > > with the I-pipe patch, but this is not sufficient, it is not
> > > prepared with prepare-kernel.sh
> > >
> >
> > I had thought of that too. However, the patch namely
> > ipipe-core-3.18.20-arm-5.patch I used is meant for kernel 3.18.20. And
> > there is no such patch available in the latest xenomai-3-3.0-rc7. ie
> under
> > the kernel/Cobalt/arch/arm/patches/, there is such patch available for
> > 3.18.20. With the observation, I concluded that
> > ipipe-core-3.18.20-arm-5.patch would patch all is needed for xenomai to
> > work. In addition, I checked in its make menuconfig, I did not see Cobalt
> > feature, but I did see ipipe was integrated. Obviously now I know it was
> > wrong. However, I still yet found a xenomai which has patches for 3.18.20
> > kernel. Any comments?
>
> You can use Xenomai 3.0-rc7 with the I-pipe patch for 3.18.20.
> Simply pass the path to the patch to the prepare-kernel.sh --adeos
> option.
>
If I am not wrong, prepare-kernel.sh is for preparing kernel by patching
the patch files under the ..arm/patches/ipipe-core...3.18.12.patch, and in
my case I patch kernel 3.18.12 with ipipe-core-3.18.12.patch which I did
not use --adeos but --ipipe.

The other issue I pointed out above, Xenomai-3.0.rc7 does not have 3.18.20
patch. Hope this clarifies things. I am happy to use kernel 3.18.12 anyway.

>
> > > Ok, when I say "please post the full kernel logs", I mean only the
> > > logs of the kernel currently running, not the logs of previous
> > > kernel runs, these are irrelevant. Anyway, I think you have
> > > CONFG_CONTEX_TRACKING enabled, which does not work with
> > > CONFIG_IPIPE, at least on ARM. Could you post you kernel
> > > configuration to be sure?
> > >
> >
> > My apologies, there were two kern.log and kern.log.1. I had taken the
> wrong
> > perhaps. To make sure I got it right, I deleted it and reboot the system.
> > Attached is the newly created kern.log
>
> It is simpler to send the output of the "dmesg" command. Anyway, now
> that we know what the problem is, you do not need to send the logs
> again.
>
> --
>                                             Gilles.
> https://click-hack.org
>

Re: [Xenomai] Xenomai tests

[vibnwis]

On 24 September 2015 at 00:42, vibnwis <vibnwis@gmail.com> wrote:

>
>
> On 23 September 2015 at 23:59, Gilles Chanteperdrix <
> gilles.chanteperdrix@xenomai.org> wrote:
>
>> On Wed, Sep 23, 2015 at 11:29:21PM +1200, vibnwis wrote:
>> > On 23 September 2015 at 19:06, Gilles Chanteperdrix <
>> > gilles.chanteperdrix@xenomai.org> wrote:
>> >
>> > > On Wed, Sep 23, 2015 at 07:38:43AM +0200, Gilles Chanteperdrix wrote:
>> > > > On Wed, Sep 23, 2015 at 12:41:38PM +1200, vibnwis wrote:
>> > > > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
>> > > > > gilles.chanteperdrix@xenomai.org> wrote:
>> > > > >
>> > > > > > On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
>> > > > > > > By regenerating ./configure using autoreconf -fi, I managed to
>> > > compile
>> > > > > > > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put
>> in
>> > > their own
>> > > > > > > place by make install. However, when running latency, it
>> produces
>> > > > > > >
>> > > > > > > > latency
>> > > > > > > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate
>> > > memory
>> > > > > >
>> > > > > > You should either run the latency test as root, or have a look
>> at
>> > > > > > this page:
>> > > > > >
>> > >
>> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
>> > > > > >
>> > > > > > By root I get this.
>> > > > > root@arm:/usr/xenomai# latency
>> > > > >    0"000.000| BUG in low_init(): [main] binding failed: Function
>> not
>> > > > > implemented
>> > > > >
>> > > > > However, the PATH has set the Lib
>> > > >
>> > > > Running as root was necessary to avoid the mlockall issue. The issue
>> > > > you have now is different: the logs you sent show that the last
>> > > > kernel you booted is not patched with Xenomai. The kernel is patched
>> > > > with the I-pipe patch, but this is not sufficient, it is not
>> > > > prepared with prepare-kernel.sh
>> > > >
>> > > > >
>> > > > > root@arm:/usr/xenomai# echo $PATH
>> > > > >
>> > >
>> /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/usr/xenomai/bin:/usr/xenomai/lib:/usr/xenomai/include
>> > > > >
>> > > > >
>> > > > > >
>> > > > > > > > dmesg | grep -i ipipe
>> > > > > > > [    0.000000] Linux version 3.18.20-ipipe (qws@qws-MacBookPro
>> )
>> > > (gcc
>> > > > > > > version 4.9.2 (Ubuntu/Linaro 4.9.2-10ubuntu10) ) #1 SMP Tue
>> Sep 22
>> > > > > > 21:04:49
>> > > > > > > NZST 2015
>> > > > > > > [    0.623763] 3.18.20-ipipe #1 Not tainted
>> > > > > > > [    0.623994]   [<c00d3094>]
>> __ipipe_do_sync_stage+0x288/0x2d0
>> > > > > > > [    0.624011]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
>> > > > > > > [    0.624321] CPU: 1 PID: 22 Comm: kblockd Not tainted
>> > > 3.18.20-ipipe #1
>> > > > > > > [    0.624479] [<c00843b4>] (trace_hardirqs_on_caller) from
>> > > [<c05da7f4>]
>> > > > > > > (__ipipe_fast_svc_irq_exit+0x4/0x18)
>> > > > > > > [    0.624504] [<c05da7f4>] (__ipipe_fast_svc_irq_exit) from
>> > > [<c001aacc>]
>> > > > > > > (ipip_test_and_stall_root+0x2c/0x3c)
>> > > > > > > [    0.624524] [<c001aacc>] (ipipe_test_and_stall_root) from
>> > > [<c0086e58>]
>> > > > > > > (lock_acquire+0x44/0x108)
>> > > > > > > [    0.728394] Switched to clocksource ipipe_tsc
>> > > > > > > [    2.464800] usb usb1: Manufacturer: Linux 3.18.20-ipipe
>> ehci_hcd
>> > > > > > > [    6.364516] usb usb2: Manufacturer: Linux 3.18.20-ipipe
>> musb-hcd
>> > > > > >
>> > > > > > Please post the full kernel logs. You likely enabled a kernel
>> > > > > > configuration option which has an issue with CONFIG_IPIPE
>> > > > > >
>> > > > > Please find it in attachment. Many thanks (very grateful indeed)
>> > > >
>> > > > Ok, when I say "please post the full kernel logs", I mean only the
>> > > > logs of the kernel currently running, not the logs of previous
>> > > > kernel runs, these are irrelevant. Anyway, I think you have
>> > > > CONFG_CONTEX_TRACKING enabled, which does not work with
>> > > > CONFIG_IPIPE, at least on ARM. Could you post you kernel
>> > > > configuration to be sure?
>> > >
>> > > The problematic option is CONFIG_TRACE_IRQFLAGS, not
>> > > CONFIG_CONTEXT_TRACKING, and you have it enabled, since without it,
>> > > you would not be able to enable CONFIG_LOCKDEP. Please disable
>> > > CONFIG_TRACE_IRQFLAGS. You will need to disable all the options
>> > > depending on that option, such as CONFIG_LOCKDEP.
>> > >
>> >
>> > Spent hours looking for the item in kernel config (make menuconfig), but
>> > could find any. Would you please share where that is, please?
>>
>> In make menuconfig type the / key, then type "lockdep".
>>
>
> Many thanks. Under kernel hacking, I disable everything in there, yet the
> CONFIG_TRACE_IRQFLAGS and CONFIG_LOCKDEP remain unchanged. FYI, I used
> omap2plus_defconfig. Would it be the cause of that?
>

To do a quick test if these CONFIG_LOCKDEP_SUPPORT
CONFIG_TRACE_IRQFLAGS_SUPPORT ever able to be changed. I simply turn off
everything in make menuconfig. I mean everything and these two flags are
still there unchanged.
FYI, I had the .config opened in a editor. Whenever changes were made, it
would prompt in the editor to reload the contents.

>
>> --
>>                                             Gilles.
>> https://click-hack.org
>>
>
>

Re: [Xenomai] Xenomai tests

[Lennart Sorensen]

On Thu, Sep 24, 2015 at 01:26:31AM +1200, vibnwis wrote:
> To do a quick test if these CONFIG_LOCKDEP_SUPPORT
> CONFIG_TRACE_IRQFLAGS_SUPPORT ever able to be changed. I simply turn off
> everything in make menuconfig. I mean everything and these two flags are
> still there unchanged.
> FYI, I had the .config opened in a editor. Whenever changes were made, it
> would prompt in the editor to reload the contents.

Certainly looking at lib/Kconfig.debug, nowhere else can turn on
CONFIG_LOCKDEP except stuff in there.  Doing a search in make menuconfig
for LOCKDEP should show you that symbol and even explain (at least
usually it seems to) why the setting is what it is.

Here is what I see on 3.14 where I do have LOCKDEP disabled and
ipipe/xenomai in use:

 .config - Linux/arm 3.14.51 Kernel Configuration
 > Search (LOCKDEP) -----------------------------------------------------------
  +---------------------------- Search Results -----------------------------+
  | Symbol: LOCKDEP [=n]                                                    |  
  | Type  : boolean                                                         |  
  |   Defined at lib/Kconfig.debug:928                                      |  
  |   Depends on: DEBUG_KERNEL [=y] && TRACE_IRQFLAGS_SUPPORT [=y] &&       |
  |               STACKTRACE_SUPPORT [=y] && LOCKDEP_SUPPORT [=y]           |                                                           
  |   Selects: STACKTRACE [=y] && FRAME_POINTER [=n] && KALLSYMS [=y] &&    |
  |            KALLSYMS_ALL [=y]                                            |                                          
  |   Selected by: DEBUG_LOCK_ALLOC [=n] && DEBUG_KERNEL [=y] &&            |
  |                TRACE_IRQFLAGS_SUPPORT [=y] && STACKTRACE_SUPPORT [=y] &&| 
  |                LOCKDEP_SUPPORT [=y] || PROVE_LOCKING [=n] &&            |
  |                DEBUG_KERNEL [=y] && TRACE_IRQFLAGS_SUPPORT [=y] &&      | 
  |                STACKTRACE_SUPPORT [=y] && LOCKDEP_SUPPORT [=y           |
  |                                                                         |
  |                                                                         |  
  | Symbol: DEBUG_LOCKDEP [=n]                                              |  
  | Type  : boolean                                                         |  
  | Prompt: Lock dependency engine debugging                                |  
  |   Location:                                                             |  
  |     -> Kernel hacking                                                   |
  | (1)   -> Lock Debugging (spinlocks, mutexes, etc...)                    |  
  |   Defined at lib/Kconfig.debug:957                                      |  
  |   Depends on: DEBUG_KERNEL [=y] && LOCKDEP [=n]                         |  
  |                                                                         |  
  |                                                                         |  
  | Symbol: LOCKDEP_SUPPORT [=y]                                            |  
  | Type  : boolean                                                         |  
  |   Defined at arch/arm/Kconfig:169                                       |  
  |                                                                         |
  +-----------------------------------------------------------------(100%)--+
  |                                < Exit >                                 |  
  +-------------------------------------------------------------------------+

So in this case since the 'Selected by' says DEBUG_LOCK_ALLOC is off,
then the first block (up to ||) is false, and PROVE_LOCKING is off,
so the next block is false, which means all selectors of LOCKDEP are
false and it is itself false.

-- 
Len Sorensen

Re: [Xenomai] Xenomai tests

[vibnwis]

On 23 September 2015 at 23:58, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> On Wed, Sep 23, 2015 at 10:38:05PM +1200, vibnwis wrote:
> > On 23 September 2015 at 18:56, Gilles Chanteperdrix <
> > gilles.chanteperdrix@xenomai.org> wrote:
> >
> > > On Wed, Sep 23, 2015 at 06:33:57PM +1200, vibnwis wrote:
> > > > On 23 September 2015 at 17:54, Gilles Chanteperdrix <
> > > > gilles.chanteperdrix@xenomai.org> wrote:
> > > >
> > > > > On Wed, Sep 23, 2015 at 07:43:19AM +0200, Gilles Chanteperdrix
> wrote:
> > > > > > On Wed, Sep 23, 2015 at 01:21:55PM +1200, vibnwis wrote:
> > > > > > > On 23 September 2015 at 12:41, vibnwis <vibnwis@gmail.com>
> wrote:
> > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > > > > > > gilles.chanteperdrix@xenomai.org> wrote:
> > > > > > > >
> > > > > > > >> On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > > > > >> > By regenerating ./configure using autoreconf -fi, I
> managed to
> > > > > compile
> > > > > > > >> > xenomai-3-3.0-rc7 natively on PandaBoard. All things are
> put
> > > in
> > > > > their
> > > > > > > >> own
> > > > > > > >> > place by make install. However, when running latency, it
> > > produces
> > > > > > > >> >
> > > > > > > >> > > latency
> > > > > > > >> > 0"000.000| BUG in low_init(): [main] mlockall: Cannot
> allocate
> > > > > memory
> > > > > > > >>
> > > > > > > >> You should either run the latency test as root, or have a
> look
> > > at
> > > > > > > >> this page:
> > > > > > > >>
> > > > > > > >>
> > > > >
> > >
> http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > > > > > > >
> > > > > > > >
> > > > > > > Referring to the link, I have the following outcomes.
> > > > > > > /dev/rtpipe <- not exist
> > > > > > >
> > > > > > > /sys/module/xenomai/parameters/allowed_group <- not exist as
> > > > > > well
> > > > >
> > > > > So, this is probably because you are not running a kernel with
> > > > > xenomai support.
> > > > >
> > > > Since there isn't any 3.18.20 patch, should I use 3.18.12. It is
> still
> > > > 3.8.x. Should be alright ? I will give it a go.
> > >
> > > Please see my answer to your previous mail. You can use the patch
> > > for 3.18.20 with xenomai 3.0-rc7. I will revert back to 3.18.12.
> > >
> >
> > Tried it, but failed. See below.
> >
> > ./scripts/prepare-kernel.sh --arch=arm --linux=../linux-3.18.20
> > --ipipe=./kernel/cobalt/arch/arm/patches/ipipe-core-3.18.12-arm-4.patch
>
> ipipe-core-3.18.12* is for the Linux 3.18.12 kernel. The I-pipe
> patch for Linux 3.18.20 is named ipipe-core-3.18.20something
> --
>                                             Gilles.
> https://click-hack.org
>

While not able to disable these items CONFIG_TRACE_IRQFLAGS CONFIG_LOCKDEP,
I tried my luck on the patched ipipe-ipipe-core-3.18.12-arm-4. Obviously,
it still suffers the problem.

[    0.000000] Booting Linux on physical CPU 0x0
[    0.000000] Initializing cgroup subsys cpuset
[    0.000000] Initializing cgroup subsys cpu
[    0.000000] Initializing cgroup subsys cpuacct
[    0.000000] Linux version 3.18.12-ipipe (qws@qws-MacBookPro) (gcc
version 4.5
[    0.000000] CPU: ARMv7 Processor [412fc09a] revision 10 (ARMv7),
cr=10c5387d
[    0.000000] CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing
instructie
[    0.000000] Machine model: TI OMAP4 PandaBoard-ES
[    0.000000] cma: Reserved 16 MiB at 0xbe800000
[    0.000000] Memory policy: Data cache writealloc
[    0.000000] On node 0 totalpages: 261888
[    0.000000] free_area_init_node: node 0, pgdat c0982c40, node_mem_map
eeff900
[    0.000000]   Normal zone: 1520 pages used for memmap
[    0.000000]   Normal zone: 0 pages reserved
[    0.000000]   Normal zone: 194560 pages, LIFO batch:31
[    0.000000]   HighMem zone: 526 pages used for memmap
[    0.000000]   HighMem zone: 67328 pages, LIFO batch:15
[    0.000000] OMAP4460 ES1.1
[    0.000000] L2C: platform modifies aux control register: 0x0e070000 ->
0x3e40
[    0.000000] L2C: DT/platform modifies aux control register: 0x0e070000
-> 0x0
[    0.000000] L2C-310 enabling early BRESP for Cortex-A9
[    0.000000] L2C-310 ID prefetch enabled, offset 1 lines
[    0.000000] OMAP L2C310: ROM does not support power control setting
[    0.000000] L2C-310 dynamic clock gating disabled, standby mode disabled
[    0.000000] L2C-310 cache controller enabled, 16 ways, 1024 kB
[    0.000000] L2C-310: CACHE_ID 0x410000c7, AUX_CTRL 0x7ec70000
[    0.000000] PERCPU: Embedded 14 pages/cpu @eef95000 s25344 r8192 d23808
u5734
[    0.000000] pcpu-alloc: s25344 r8192 d23808 u57344 alloc=14*4096
[    0.000000] pcpu-alloc: [0] 0 [0] 1
[    0.000000] Built 1 zonelists in Zone order, mobility grouping on.
Total pa8
[    0.000000] Kernel command line: console=ttyO2,115200n8
root=/dev/mmcblk0p1 c
[    0.000000] PID hash table entries: 4096 (order: 2, 16384 bytes)
[    0.000000] Dentry cache hash table entries: 131072 (order: 7, 524288
bytes)
[    0.000000] Inode-cache hash table entries: 65536 (order: 6, 262144
bytes)
[    0.000000] allocated 2095104 bytes of page_cgroup
[    0.000000] please try 'cgroup_disable=memory' option if you don't want
memos
[    0.000000] Memory: 1001420K/1047552K available (6180K kernel code, 692K
rwd)
[    0.000000] Virtual kernel memory layout:
[    0.000000]     vector  : 0xffff0000 - 0xffff1000   (   4 kB)
[    0.000000]     fixmap  : 0xffc00000 - 0xffe00000   (2048 kB)
[    0.000000]     vmalloc : 0xf0000000 - 0xff000000   ( 240 MB)
[    0.000000]     lowmem  : 0xc0000000 - 0xef800000   ( 760 MB)
[    0.000000]     pkmap   : 0xbfe00000 - 0xc0000000   (   2 MB)
[    0.000000]     modules : 0xbf000000 - 0xbfe00000   (  14 MB)
[    0.000000]       .text : 0xc0008000 - 0xc085e2d8   (8537 kB)
[    0.000000]       .init : 0xc085f000 - 0xc08dc000   ( 500 kB)
[    0.000000]       .data : 0xc08dc000 - 0xc09892d0   ( 693 kB)
[    0.000000]        .bss : 0xc09892d0 - 0xc11d4430   (8493 kB)
[    0.000000] Hierarchical RCU implementation.
[    0.000000] NR_IRQS:16 nr_irqs:16 16
[    0.000000] ti_dt_clocks_register: failed to lookup clock node
bandgap_fclk
[    0.000000] OMAP clockevent source: timer1 at 32768 Hz
[    0.000000] sched_clock: 32 bits at 32kHz, resolution 30517ns, wraps
every 6s
[    0.000030] OMAP clocksource: 32k_counter at 32768 Hz
[    0.000183] I-pipe, 349.988 MHz timer
[    0.000313] sched_clock: 64 bits at 349MHz, resolution 2ns, wraps every
3141s
[    0.000331] I-pipe, 349.988 MHz clocksource, wrap in 12271 ms
[    0.001699] Interrupt pipeline (release #4)
[    0.002692] Console: colour dummy device 80x30
[    0.002744] Lock dependency validator: Copyright (c) 2006 Red Hat, Inc.,
Ingr
[    0.002755] ... MAX_LOCKDEP_SUBCLASSES:  8
[    0.002764] ... MAX_LOCK_DEPTH:          48
[    0.002773] ... MAX_LOCKDEP_KEYS:        8191
[    0.002782] ... CLASSHASH_SIZE:          4096
[    0.002791] ... MAX_LOCKDEP_ENTRIES:     32768
[    0.002800] ... MAX_LOCKDEP_CHAINS:      65536
[    0.002808] ... CHAINHASH_SIZE:          32768
[    0.002817]  memory used by lock dependency info: 5167 kB
[    0.002826]  per task-struct memory footprint: 1152 bytes
[    0.002856] Calibrating delay loop... 1391.00 BogoMIPS (lpj=6955008)
[    0.080229] pid_max: default: 32768 minimum: 301
[    0.080690] Security Framework initialized
[    0.080896] Mount-cache hash table entries: 2048 (order: 1, 8192 bytes)
[    0.080917] Mountpoint-cache hash table entries: 2048 (order: 1, 8192
bytes)
[    0.083879] Initializing cgroup subsys memory
[    0.083950] Initializing cgroup subsys devices
[    0.084042] Initializing cgroup subsys freezer
[    0.084127] Initializing cgroup subsys blkio
[    0.084214] Initializing cgroup subsys perf_event
[    0.084300] CPU: Testing write buffer coherency: ok
[    0.085562] CPU0: thread -1, cpu 0, socket 0, mpidr 80000000
[    0.085653] Setting up static identity map for 0x805e3180 - 0x805e31f0
[    0.092103] CPU1: Booted secondary processor
[    0.092127] I-pipe, 349.988 MHz timer
[    0.180228] CPU1: thread -1, cpu 1, socket 0, mpidr 80000001
[    0.180765] Brought up 2 CPUs
[    0.180794] SMP: Total of 2 processors activated (2788.55 BogoMIPS).
[    0.180804] CPU: All CPU(s) started in SVC mode.
[    0.183974] devtmpfs: initialized
[    0.197472] VFP support v0.3: implementor 41 architecture 3 part 30
variant 4
[    0.203561] omap_hwmod: l3_main_3 using broken dt data from ocp
[    0.209291] omap_hwmod: l3_main_2 using broken dt data from ocp
[    0.426195] pinctrl core: initialized pinctrl subsystem
[    0.428946] regulator-dummy: no parameters
[    0.495204] NET: Registered protocol family 16
[    0.501218] DMA: preallocated 256 KiB pool for atomic coherent
allocations
[    0.516431] gpiochip_add: registered GPIOs 0 to 31 on device: gpio
[    0.517486] OMAP GPIO hardware version 0.1
[    0.518818] gpiochip_add: registered GPIOs 32 to 63 on device: gpio
[    0.521062] gpiochip_add: registered GPIOs 64 to 95 on device: gpio
[    0.523238] gpiochip_add: registered GPIOs 96 to 127 on device: gpio
[    0.525357] gpiochip_add: registered GPIOs 128 to 159 on device: gpio
[    0.527551] gpiochip_add: registered GPIOs 160 to 191 on device: gpio
[    0.529296] omap-gpmc 50000000.gpmc: GPMC revision 6.0
[    0.530389] irq: no irq domain found for /ocp/pinmux@4a100040 !
[    0.531046] irq: no irq domain found for /ocp/pinmux@4a100040 !
[    0.531697] irq: no irq domain found for /ocp/pinmux@4a100040 !
[    0.555409] platform 4b501000.aes: Cannot lookup hwmod 'aes'
[    0.556010] platform 480a5000.des: Cannot lookup hwmod 'des'
[    0.566634] No ATAGs?
[    0.566709] hw-breakpoint: found 5 (+1 reserved) breakpoint and 1
watchpoint.
[    0.566722] hw-breakpoint: maximum watchpoint size is 4 bytes.
[    0.569875] OMAP DMA hardware revision 0.0
[    0.623730]
[    0.623744] =================================
[    0.623753] [ INFO: inconsistent lock state ]
[    0.623764] 3.18.12-ipipe #1 Not tainted
[    0.623773] ---------------------------------
[    0.623783] inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage.
[    0.623795] kblockd/22 [HC0[0]:SC0[0]:HE1:SE1] takes:
[    0.623804]  (std_spinlock_raw(&rq->lock)){?.-...}, at: [<c05db640>]
__sched4
[    0.623844] {IN-HARDIRQ-W} state was registered at:
[    0.623852]   [<c05e0538>] _raw_spin_lock+0x28/0x38
[    0.623869]   [<c006797c>] scheduler_tick+0x38/0xd4
[    0.623886]   [<c00a0820>] update_process_times+0x54/0x60
[    0.623905]   [<c00ae070>] tick_periodic+0x30/0xec
[    0.623923]   [<c00ae2f0>] tick_handle_periodic+0x28/0x88
[    0.623938]   [<c0014f9c>] twd_handler+0x38/0x44
[    0.623955]   [<c0094250>] handle_percpu_devid_irq+0x78/0x178
[    0.623971]   [<c0090eb4>] generic_handle_irq+0x28/0x3c
[    0.623993]   [<c00911a4>] __handle_domain_irq+0x64/0xc4
[    0.624008]   [<c00d3880>] __ipipe_do_sync_stage+0x288/0x2d0
[    0.624025]   [<c0008634>] __ipipe_grab_irq+0x5c/0x70
[    0.624039]   [<c0008948>] gic_handle_irq+0x28/0x60
[    0.624053]   [<eefb3980>] 0xeefb3980
[    0.624065]   [<c05e1124>] __irq_svc+0x44/0x50
[    0.624081]   [<c00095a0>] calibrate_delay+0x3a4/0x5b8
[    0.624096]   [<c00095a0>] calibrate_delay+0x3a4/0x5b8
[    0.624109]   [<c085fbc4>] start_kernel+0x314/0x3d4
[    0.624124]   [<80008074>] 0x80008074
[    0.624137] irq event stamp: 14
[    0.624145] hardirqs last  enabled at (13): [<c05e07a8>]
_raw_spin_unlock_irc
[    0.624161] hardirqs last disabled at (14): [<c05e06c0>]
_raw_spin_lock_irq+4
[    0.624175] softirqs last  enabled at (0): [<c003ffbc>]
copy_process.part.554
[    0.624195] softirqs last disabled at (0): [<  (null)>]   (null)
[    0.624207]
[    0.624207] other info that might help us debug this:
[    0.624217]  Possible unsafe locking scenario:
[    0.624217]
[    0.624227]        CPU0
[    0.624235]        ----
[    0.624242]   lock(std_spinlock_raw(&rq->lock));
[    0.624256]   <Interrupt>
[    0.624263]     lock(std_spinlock_raw(&rq->lock));
[    0.624277]
[    0.624277]  *** DEADLOCK ***
[    0.624277]
[    0.624291] 1 lock held by kblockd/22:
[    0.624299]  #0:  (std_spinlock_raw(&rq->lock)){?.-...}, at:
[<c05db640>] __4
[    0.624331]
[    0.624331] stack backtrace:
[    0.624347] CPU: 1 PID: 22 Comm: kblockd Not tainted 3.18.12-ipipe #1
[    0.624388] [<c0015edc>] (unwind_backtrace) from [<c001260c>]
(show_stack+0x)
[    0.624410] [<c001260c>] (show_stack) from [<c05da8e8>]
(dump_stack+0x84/0x9)
[    0.624438] [<c05da8e8>] (dump_stack) from [<c0084480>]
(print_usage_bug+0x1)
[    0.624457] [<c0084480>] (print_usage_bug) from [<c0084764>]
(mark_lock+0x1e)
[    0.624476] [<c0084764>] (mark_lock) from [<c0084cd4>]
(mark_held_locks+0x70)
[    0.624495] [<c0084cd4>] (mark_held_locks) from [<c0084d78>]
(trace_hardirqs)
[    0.624514] [<c0084d78>] (trace_hardirqs_on_caller) from [<c05e1134>]
(__ipi)
[    0.624537] [<c05e1134>] (__ipipe_fast_svc_irq_exit) from [<c0058870>]
(wq_w)
[    0.624556] [<c0058870>] (wq_worker_sleeping) from [<c05dba8c>]
(__schedule+)
[    0.624575] [<c05dba8c>] (__schedule) from [<c0057cc0>]
(rescuer_thread+0x1b)
[    0.624596] [<c0057cc0>] (rescuer_thread) from [<c005c810>]
(kthread+0xc8/0x)
[    0.624614] [<c005c810>] (kthread) from [<c000eb00>]
(ret_from_fork+0x18/0x3)
[    0.648145] omap-dma-engine 4a056000.dma-controller: OMAP DMA engine
driver
[    0.648852] of_get_named_gpiod_flags: parsed 'gpio' property of node
'/hsusb)
[    0.720241] hsusb1_vbus: 3300 mV
[    0.720458] reg-fixed-voltage wl12xx_vmmc: could not find pctldev for
node /e
[    0.720484] platform wl12xx_vmmc: Driver reg-fixed-voltage requests
probe del
[    0.724086] SCSI subsystem initialized
[    0.724613] usbcore: registered new interface driver usbfs
[    0.724731] usbcore: registered new interface driver hub
[    0.724844] usbcore: registered new device driver usb
[    0.725167] of_get_named_gpiod_flags: parsed 'reset-gpios' property of
node )
[    0.725855] omap_i2c 48070000.i2c: could not find pctldev for node
/ocp/pinme
[    0.725881] platform 48070000.i2c: Driver omap_i2c requests probe
deferral
[    0.725935] omap_i2c 48072000.i2c: could not find pctldev for node
/ocp/pinme
[    0.725958] platform 48072000.i2c: Driver omap_i2c requests probe
deferral
[    0.726007] omap_i2c 48060000.i2c: could not find pctldev for node
/ocp/pinme
[    0.726029] platform 48060000.i2c: Driver omap_i2c requests probe
deferral
[    0.726077] omap_i2c 48350000.i2c: could not find pctldev for node
/ocp/pinme
[    0.726097] platform 48350000.i2c: Driver omap_i2c requests probe
deferral
[    0.727915] Switched to clocksource ipipe_tsc
[    0.822234] NET: Registered protocol family 2
[    0.823173] TCP established hash table entries: 8192 (order: 3, 32768
bytes)
[    0.823352] TCP bind hash table entries: 8192 (order: 6, 294912 bytes)
[    0.824242] TCP: Hash tables configured (established 8192 bind 8192)
[    0.824326] TCP: reno registered
[    0.824351] UDP hash table entries: 512 (order: 3, 40960 bytes)
[    0.824493] UDP-Lite hash table entries: 512 (order: 3, 40960 bytes)
[    0.825010] NET: Registered protocol family 1
[    0.825532] RPC: Registered named UNIX socket transport module.
[    0.825546] RPC: Registered udp transport module.
[    0.825557] RPC: Registered tcp transport module.
[    0.825567] RPC: Registered tcp NFSv4.1 backchannel transport module.
[    0.826669] hw perfevents: enabled with armv7_cortex_a9 PMU driver, 7
countee
[    0.829999] futex hash table entries: 512 (order: 3, 32768 bytes)
[    0.830169] audit: initializing netlink subsys (disabled)
[    0.830227] audit: type=2000 audit(0.820:1): initialized
[    0.831084] [Xenomai] scheduling class idle registered.
[    0.831097] [Xenomai] scheduling class rt registered.
[    0.831298] I-pipe: head domain Xenomai registered.
[    0.833495] [Xenomai] Cobalt v3.0-rc7 (Exact Zero)
[    0.834750] VFS: Disk quotas dquot_6.5.2
[    0.834871] Dquot-cache hash table entries: 1024 (order 0, 4096 bytes)
[    0.836029] NFS: Registering the id_resolver key type
[    0.836075] Key type id_resolver registered
[    0.836087] Key type id_legacy registered
[    0.836184] jffs2: version 2.2. (NAND) (SUMMARY)
\xffffffc2\xffffffa9\xffff.
[    0.836534] msgmni has been set to 1461
[    0.838483] bounce: pool size: 64 pages
[    0.838509] io scheduler noop registered
[    0.838530] io scheduler deadline registered
[    0.838606] io scheduler cfq registered (default)
[    0.841863] pinctrl-single 4a100040.pinmux: 203 pins at pa fc100040 size
406
[    0.842192] pinctrl-single 4a31e040.pinmux: 28 pins at pa fc31e040 size
56
[    0.844191] pbias_mmc_omap4: 1800 <--> 3000 mV at 3000 mV
[    0.845187] abb_mpu: 1025 <--> 1389 mV
[    0.845689] abb_iva: 950 <--> 1376 mV
[    0.846750] Serial: 8250/16550 driver, 4 ports, IRQ sharing enabled
[    0.849967] omap_uart 4806a000.serial: no wakeirq for uart0
[    0.849990] of_get_named_gpiod_flags: can't parse 'rts-gpio' property of
nod'
[    0.850159] 4806a000.serial: ttyO0 at MMIO 0x4806a000 (irq = 104,
base_baud 0
[    0.850838] of_get_named_gpiod_flags: can't parse 'rts-gpio' property of
nod'
[    0.850990] 4806c000.serial: ttyO1 at MMIO 0x4806c000 (irq = 105,
base_baud 1
[    0.851632] of_get_named_gpiod_flags: can't parse 'rts-gpio' property of
nod'
[    0.851790] 48020000.serial: ttyO2 at MMIO 0x48020000 (irq = 106,
base_baud 2
[    2.267692] console [ttyO2] enabled
[    2.272047] of_get_named_gpiod_flags: can't parse 'rts-gpio' property of
nod'
[    2.282045] 4806e000.serial: ttyO3 at MMIO 0x4806e000 (irq = 102,
base_baud 3
[    2.306247] brd: module loaded
[    2.316318] loop: module loaded
[    2.321577] mtdoops: mtd device (mtddev=name/number) must be supplied
[    2.332955] usbcore: registered new interface driver asix
[    2.338742] usbcore: registered new interface driver ax88179_178a
[    2.345203] usbcore: registered new interface driver cdc_ether
[    2.351450] usbcore: registered new interface driver smsc95xx
[    2.357541] usbcore: registered new interface driver net1080
[    2.363581] usbcore: registered new interface driver cdc_subset
[    2.369887] usbcore: registered new interface driver zaurus
[    2.375856] usbcore: registered new interface driver cdc_ncm
[    2.382346] usbcore: registered new interface driver cdc_wdm
[    2.388431] usbcore: registered new interface driver usb-storage
[    2.394827] usbcore: registered new interface driver usbtest
[    2.401972] mousedev: PS/2 mouse device common for all mice
[    2.409529] i2c /dev entries driver
[    2.413224] Driver for 1-wire Dallas network protocol.
[    2.419284] of_get_named_gpiod_flags: parsed 'gpios' property of node
'/ocp/)
[    2.430691] omap_wdt: OMAP Watchdog Timer Rev 0x01: initial timeout 60
sec
[    2.439049] of_get_named_gpiod_flags: can't parse 'cd-gpios' property of
nod'
[    2.448619] of_get_named_gpiod_flags: can't parse 'wp-gpios' property of
nod'
[    2.458530] omap_hsmmc 4809c000.mmc: unable to get vmmc regulator -517
[    2.465431] platform 4809c000.mmc: Driver omap_hsmmc requests probe
deferral
[    2.472954] of_get_named_gpiod_flags: can't parse 'cd-gpios' property of
nod'
[    2.482519] of_get_named_gpiod_flags: can't parse 'wp-gpios' property of
nod'
[    2.492370] omap_hsmmc 480d5000.mmc: unable to get vmmc regulator -517
[    2.499296] platform 480d5000.mmc: Driver omap_hsmmc requests probe
deferral
[    2.507279] of_get_named_gpiod_flags: parsed 'gpios' property of node
'/leds)
[    2.517135] of_get_named_gpiod_flags: parsed 'gpios' property of node
'/leds)
[    2.526452] of_get_named_gpiod_flags: parsed 'gpios' property of node
'/leds)
[    2.536530] of_get_named_gpiod_flags: parsed 'gpios' property of node
'/leds)
[    2.546194] ledtrig-cpu: registered to indicate activity on CPUs
[    2.552840] usbcore: registered new interface driver usbhid
[    2.558682] usbhid: USB HID core driver
[    2.564809] oprofile: using arm/armv7-ca9
[    2.569350] TCP: cubic registered
[    2.572840] Initializing XFRM netlink socket
[    2.577347] NET: Registered protocol family 17
[    2.582076] NET: Registered protocol family 15
[    2.586834] Key type dns_resolver registered
[    2.591484] twl: not initialized
[    2.594859] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000
Vs 0
[    2.603244] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000
Vs 0
[    2.611650] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000
Vs 0
[    2.620034] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000
Vs 0
[    2.628433] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000
Vs 0
[    2.636791] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1375000
Vs 0
[    2.645176] twl6030_uv_to_vsel:OUT OF RANGE! non mapped vsel for 1410000
Vs 0
[    2.653572] omap2_set_init_voltage: unable to find boot up OPP for
vdd_mpu
[    2.660774] omap2_set_init_voltage: unable to set vdd_mpu
[    2.666424] omap2_set_init_voltage: unable to find boot up OPP for
vdd_core
[    2.673724] omap2_set_init_voltage: unable to set vdd_core
[    2.679487] omap2_set_init_voltage: unable to find boot up OPP for
vdd_iva
[    2.686665] omap2_set_init_voltage: unable to set vdd_iva
[    2.693601] Power Management for TI OMAP4+ devices.
[    2.698733] OMAP4 PM: u-boot >= v2012.07 is required for full PM support
[    2.705902] ThumbEE CPU extension supported.
[    2.710441] Registering SWP/SWPB emulation handler
[    2.715461] SmartReflex Class3 initialized
[    2.722189] of_get_named_gpiod_flags: parsed 'gpio' property of node
'/wl12x)
[    2.731955] vwl1271: 1800 mV
[    2.738244] Skipping twl internal clock init and using bootloader value
(unk)
[    2.747506] twl 0-0048: PIH (irq 39) nested IRQs
[    2.753340] twl_rtc 48070000.i2c:twl@48:rtc: Power up reset detected.
[    2.760937] twl_rtc 48070000.i2c:twl@48:rtc: Enabling TWL-RTC
[    2.770049] twl_rtc 48070000.i2c:twl@48:rtc: rtc core: registered
48070000.i0
[    2.779984] VAUX1_6030: 1000 <--> 3000 mV at 2800 mV
[    2.786424] VAUX2_6030: 1200 <--> 2800 mV at 1800 mV
[    2.792925] VAUX3_6030: 1000 <--> 3000 mV at 1200 mV
[    2.799412] VMMC: 1200 <--> 3000 mV at 3000 mV
[    2.805308] VPP: 1800 <--> 2500 mV at 1900 mV
[    2.811166] VUSIM: 1200 <--> 2900 mV at 1800 mV
[    2.816919] VDAC: 1800 mV
[    2.820511] VANA: 2100 mV
[    2.824459] VCXIO: 1800 mV
[    2.828180] VUSB: 3300 mV
[    2.832133] V1V8: 1800 mV
[    2.836054] V2V1: 2100 mV
[    2.841773] twl6040 0-004b: clk32k is not handled
[    2.848193] of_get_named_gpiod_flags: parsed 'ti,audpwron-gpio' property
of )
[    2.861369] omap_i2c 48070000.i2c: bus 0 rev0.11 at 400 kHz
[    2.868303] omap_i2c 48072000.i2c: bus 1 rev0.11 at 400 kHz
[    2.875701] omap_i2c 48060000.i2c: bus 2 rev0.11 at 100 kHz
[    2.882569] omap_i2c 48350000.i2c: bus 3 rev0.11 at 400 kHz
[    2.888745] of_get_named_gpiod_flags: can't parse 'cd-gpios' property of
nod'
[    2.898370] of_get_named_gpiod_flags: can't parse 'wp-gpios' property of
nod'
[    2.948397] of_get_named_gpiod_flags: can't parse 'cd-gpios' property of
nod'
[    2.958044] of_get_named_gpiod_flags: can't parse 'wp-gpios' property of
nod'
[    3.090248] twl_rtc 48070000.i2c:twl@48:rtc: setting system clock to
2000-01)
[    3.100222] sr_init: No PMIC hook to init smartreflex
[    3.105763] smartreflex smartreflex.0: omap_sr_probe: SmartReflex driver
inid
[    3.114274] smartreflex smartreflex.1: omap_sr_probe: SmartReflex driver
inid
[    3.122781] smartreflex smartreflex.2: omap_sr_probe: SmartReflex driver
inid
[    3.157412] VUSB: disabling
[    3.161004] VANA: disabling
[    3.164521] VDAC: disabling
[    3.170776] Waiting for root device /dev/mmcblk0p1...
[    3.205076] mmc0: host does not support reading read-only switch,
assuming we
[    3.215326] mmc0: new high speed SDHC card at address 0001
[    3.221791] mmcblk0: mmc0:0001 00000 3.73 GiB
[    3.228402]  mmcblk0: p1
[    3.236636] omap_hsmmc 480d5000.mmc: card claims to support voltages
below de
[    3.256445] mmc1: queuing unknown CIS tuple 0x91 (3 bytes)
[    3.264678] mmc1: new SDIO card at address 0001
[    3.281726] EXT4-fs (mmcblk0p1): INFO: recovery required on readonly
filesysm
[    3.289434] EXT4-fs (mmcblk0p1): write access will be enabled during
recovery
[    5.220723] EXT4-fs (mmcblk0p1): recovery complete
[    5.346414] EXT4-fs (mmcblk0p1): mounted filesystem with ordered data
mode. )
[    5.355010] VFS: Mounted root (ext4 filesystem) readonly on device 179:1.
[    5.363453] devtmpfs: mounted
[    5.367154] Freeing unused kernel memory: 500K (c085f000 - c08dc000)
[    5.804425] random: init urandom read with 25 bits of entropy available
[    6.038568] init: plymouth-upstart-bridge main process (64) terminated
with 1
[    6.047275] init: plymouth-upstart-bridge main process ended, respawning
[    6.109836] init: plymouth-upstart-bridge main process (74) terminated
with 1
[    6.118101] init: plymouth-upstart-bridge main process ended, respawning
[    6.151333] init: plymouth-upstart-bridge main process (77) terminated
with 1
[    6.164019] init: plymouth-upstart-bridge main process ended, respawning
[    6.212263] init: plymouth-upstart-bridge main process (79) terminated
with 1
[    6.220638] init: plymouth-upstart-bridge main process ended, respawning
[    6.271148] init: ureadahead main process (67) terminated with status 5
[    7.412464] systemd-udevd[191]: starting version 204
[    8.605289] OMAP DSS rev 4.0
[    8.661348] of_get_named_gpiod_flags: parsed 'gpios' property of node
'/enco)
[    8.661389] of_get_named_gpiod_flags: parsed 'gpios' property of node
'/enco)
[    8.661425] of_get_named_gpiod_flags: parsed 'gpios' property of node
'/enco)
[    8.683465] connector-dvi connector@0: failed to find video source
[    8.690053] platform connector@0: Driver connector-dvi requests probe
deferrl
[    8.691548] of_get_named_gpiod_flags: can't parse 'hpd-gpios' property
of no'
[    8.692526] connector-dvi connector@0: failed to find video source
[    8.699103] platform connector@0: Driver connector-dvi requests probe
deferrl
[    8.700439] of_get_named_gpiod_flags: parsed 'powerdown-gpios' property
of n)
[    8.823853] Console: switching to colour frame buffer device 240x67
[    8.929143] omapfb omapfb: using display 'dvi' mode 1920x1080
[    9.179119] EXT4-fs (mmcblk0p1): re-mounted. Opts: errors=remount-ro
[    9.785174] cfg80211: Calling CRDA to update world regulatory domain
[    9.822683] omap-abe-twl6040 sound: twl6040-legacy <-> 40132000.mcpdm
mappink
[   10.623772] random: nonblocking pool is initialized
[   11.308232] wlcore: loaded
[   11.528889] init: failsafe main process (404) killed by TERM signal
[   12.621987] init: avahi-cups-reload main process (513) terminated with
statu1
[   13.054446] NET: Registered protocol family 10
[   13.258077] init: plymouth-upstart-bridge main process ended, respawning
[   13.806691] init: lxdm main process (605) terminated with status 127
[   16.915965] init: plymouth-stop pre-start process (825) terminated with
stat1

Re: [Xenomai] Xenomai tests

[vibnwis]

On 24 September 2015 at 03:34, Lennart Sorensen <
lsorense@csclub.uwaterloo.ca> wrote:

> On Thu, Sep 24, 2015 at 01:26:31AM +1200, vibnwis wrote:
> > To do a quick test if these CONFIG_LOCKDEP_SUPPORT
> > CONFIG_TRACE_IRQFLAGS_SUPPORT ever able to be changed. I simply turn off
> > everything in make menuconfig. I mean everything and these two flags are
> > still there unchanged.
> > FYI, I had the .config opened in a editor. Whenever changes were made, it
> > would prompt in the editor to reload the contents.
>
> Certainly looking at lib/Kconfig.debug, nowhere else can turn on
> CONFIG_LOCKDEP except stuff in there.  Doing a search in make menuconfig
> for LOCKDEP should show you that symbol and even explain (at least
> usually it seems to) why the setting is what it is.
>
> Here is what I see on 3.14 where I do have LOCKDEP disabled and
> ipipe/xenomai in use:
>
>  .config - Linux/arm 3.14.51 Kernel Configuration
>  > Search (LOCKDEP)
> -----------------------------------------------------------
>   +---------------------------- Search Results
> -----------------------------+
>   | Symbol: LOCKDEP [=n]
>   |
>   | Type  : boolean
>  |
>   |   Defined at lib/Kconfig.debug:928
>   |
>   |   Depends on: DEBUG_KERNEL [=y] && TRACE_IRQFLAGS_SUPPORT [=y] &&
>  |
>   |               STACKTRACE_SUPPORT [=y] && LOCKDEP_SUPPORT [=y]
>  |
>   |   Selects: STACKTRACE [=y] && FRAME_POINTER [=n] && KALLSYMS [=y] &&
>   |
>   |            KALLSYMS_ALL [=y]
>   |
>   |   Selected by: DEBUG_LOCK_ALLOC [=n] && DEBUG_KERNEL [=y] &&
>   |
>   |                TRACE_IRQFLAGS_SUPPORT [=y] && STACKTRACE_SUPPORT [=y]
> &&|
>   |                LOCKDEP_SUPPORT [=y] || PROVE_LOCKING [=n] &&
>   |
>   |                DEBUG_KERNEL [=y] && TRACE_IRQFLAGS_SUPPORT [=y] &&
>   |
>   |                STACKTRACE_SUPPORT [=y] && LOCKDEP_SUPPORT [=y
>  |
>   |
>  |
>   |
>  |
>   | Symbol: DEBUG_LOCKDEP [=n]
>   |
>   | Type  : boolean
>  |
>   | Prompt: Lock dependency engine debugging
>   |
>   |   Location:
>  |
>   |     -> Kernel hacking
>  |
>   | (1)   -> Lock Debugging (spinlocks, mutexes, etc...)
>   |
>   |   Defined at lib/Kconfig.debug:957
>   |
>   |   Depends on: DEBUG_KERNEL [=y] && LOCKDEP [=n]
>  |
>   |
>  |
>   |
>  |
>   | Symbol: LOCKDEP_SUPPORT [=y]
>   |
>   | Type  : boolean
>  |
>   |   Defined at arch/arm/Kconfig:169
>  |
>   |
>  |
>
> +-----------------------------------------------------------------(100%)--+
>   |                                < Exit >
>  |
>
> +-------------------------------------------------------------------------+
>
> So in this case since the 'Selected by' says DEBUG_LOCK_ALLOC is off,
> then the first block (up to ||) is false, and PROVE_LOCKING is off,
> so the next block is false, which means all selectors of LOCKDEP are
> false and it is itself false.
>

Tried it here.

kernel hacking --->
              Lock Debugging (spinlocks, mutexes, etc...)  --->
                 [ ] RT Mutex debugging, deadlock
detection                       │ │
  │ │    [ ] Spinlock and rw-lock debugging: basic checks                 │
│
  │ │    [ ] Mutex debugging: basic checks                                │
│
  │ │    [ ] Wait/wound mutex debugging: Slowpath testing                 │
│
  │ │    [ ] Lock debugging: detect incorrect freeing of live locks       │
│
  │ │    [ ] Lock debugging: prove locking correctness                    │
│
  │ │    [ ] Lock usage statistics                                        │
│
  │ │    [ ] Sleep inside atomic section checking                         │
│
  │ │    [ ] Locking API boot-time self-tests                             │
│
  │ │    < > torture tests for locking

All items under Lock Debugging are disabled, yet LOCKDEP and TRACE_IRQFLAGS
remain unchanged.;-(



> --
> Len Sorensen
>

Re: [Xenomai] Xenomai tests

[Lennart Sorensen]

On Thu, Sep 24, 2015 at 04:11:25AM +1200, vibnwis wrote:
> On 24 September 2015 at 03:34, Lennart Sorensen <
> lsorense@csclub.uwaterloo.ca> wrote:
> 
> > On Thu, Sep 24, 2015 at 01:26:31AM +1200, vibnwis wrote:
> > > To do a quick test if these CONFIG_LOCKDEP_SUPPORT
> > > CONFIG_TRACE_IRQFLAGS_SUPPORT ever able to be changed. I simply turn off
> > > everything in make menuconfig. I mean everything and these two flags are
> > > still there unchanged.
> > > FYI, I had the .config opened in a editor. Whenever changes were made, it
> > > would prompt in the editor to reload the contents.
> >
> > Certainly looking at lib/Kconfig.debug, nowhere else can turn on
> > CONFIG_LOCKDEP except stuff in there.  Doing a search in make menuconfig
> > for LOCKDEP should show you that symbol and even explain (at least
> > usually it seems to) why the setting is what it is.
> >
> > Here is what I see on 3.14 where I do have LOCKDEP disabled and
> > ipipe/xenomai in use:
> >
> >  .config - Linux/arm 3.14.51 Kernel Configuration
> >  > Search (LOCKDEP)
> > -----------------------------------------------------------
> >   +---------------------------- Search Results
> > -----------------------------+
> >   | Symbol: LOCKDEP [=n]
> >   |
> >   | Type  : boolean
> >  |
> >   |   Defined at lib/Kconfig.debug:928
> >   |
> >   |   Depends on: DEBUG_KERNEL [=y] && TRACE_IRQFLAGS_SUPPORT [=y] &&
> >  |
> >   |               STACKTRACE_SUPPORT [=y] && LOCKDEP_SUPPORT [=y]
> >  |
> >   |   Selects: STACKTRACE [=y] && FRAME_POINTER [=n] && KALLSYMS [=y] &&
> >   |
> >   |            KALLSYMS_ALL [=y]
> >   |
> >   |   Selected by: DEBUG_LOCK_ALLOC [=n] && DEBUG_KERNEL [=y] &&
> >   |
> >   |                TRACE_IRQFLAGS_SUPPORT [=y] && STACKTRACE_SUPPORT [=y]
> > &&|
> >   |                LOCKDEP_SUPPORT [=y] || PROVE_LOCKING [=n] &&
> >   |
> >   |                DEBUG_KERNEL [=y] && TRACE_IRQFLAGS_SUPPORT [=y] &&
> >   |
> >   |                STACKTRACE_SUPPORT [=y] && LOCKDEP_SUPPORT [=y
> >  |
> >   |
> >  |
> >   |
> >  |
> >   | Symbol: DEBUG_LOCKDEP [=n]
> >   |
> >   | Type  : boolean
> >  |
> >   | Prompt: Lock dependency engine debugging
> >   |
> >   |   Location:
> >  |
> >   |     -> Kernel hacking
> >  |
> >   | (1)   -> Lock Debugging (spinlocks, mutexes, etc...)
> >   |
> >   |   Defined at lib/Kconfig.debug:957
> >   |
> >   |   Depends on: DEBUG_KERNEL [=y] && LOCKDEP [=n]
> >  |
> >   |
> >  |
> >   |
> >  |
> >   | Symbol: LOCKDEP_SUPPORT [=y]
> >   |
> >   | Type  : boolean
> >  |
> >   |   Defined at arch/arm/Kconfig:169
> >  |
> >   |
> >  |
> >
> > +-----------------------------------------------------------------(100%)--+
> >   |                                < Exit >
> >  |
> >
> > +-------------------------------------------------------------------------+
> >
> > So in this case since the 'Selected by' says DEBUG_LOCK_ALLOC is off,
> > then the first block (up to ||) is false, and PROVE_LOCKING is off,
> > so the next block is false, which means all selectors of LOCKDEP are
> > false and it is itself false.
> >
> 
> Tried it here.
> 
> kernel hacking --->
>               Lock Debugging (spinlocks, mutexes, etc...)  --->
>                  [ ] RT Mutex debugging, deadlock
> detection                       │ │
>   │ │    [ ] Spinlock and rw-lock debugging: basic checks                 │
> │
>   │ │    [ ] Mutex debugging: basic checks                                │
> │
>   │ │    [ ] Wait/wound mutex debugging: Slowpath testing                 │
> │
>   │ │    [ ] Lock debugging: detect incorrect freeing of live locks       │
> │
>   │ │    [ ] Lock debugging: prove locking correctness                    │
> │
>   │ │    [ ] Lock usage statistics                                        │
> │
>   │ │    [ ] Sleep inside atomic section checking                         │
> │
>   │ │    [ ] Locking API boot-time self-tests                             │
> │
>   │ │    < > torture tests for locking
> 
> All items under Lock Debugging are disabled, yet LOCKDEP and TRACE_IRQFLAGS
> remain unchanged.;-(

Hit '/' Then type 'LOCKDEP' and see what it says is causing it to be on.
Looking at the menu will not help you.

-- 
Len Sorensen

Re: [Xenomai] Xenomai tests

[Gilles Chanteperdrix]

On Thu, Sep 24, 2015 at 12:52:14AM +1200, vibnwis wrote:
> On 23 September 2015 at 18:55, Gilles Chanteperdrix <
> gilles.chanteperdrix@xenomai.org> wrote:
> 
> > On Wed, Sep 23, 2015 at 06:29:25PM +1200, vibnwis wrote:
> > > On 23 September 2015 at 17:38, Gilles Chanteperdrix <
> > > gilles.chanteperdrix@xenomai.org> wrote:
> > >
> > > > On Wed, Sep 23, 2015 at 12:41:38PM +1200, vibnwis wrote:
> > > > > On 23 September 2015 at 05:44, Gilles Chanteperdrix <
> > > > > gilles.chanteperdrix@xenomai.org> wrote:
> > > > >
> > > > > > On Wed, Sep 23, 2015 at 12:32:11AM +1200, vibnwis wrote:
> > > > > > > By regenerating ./configure using autoreconf -fi, I managed to
> > > > compile
> > > > > > > xenomai-3-3.0-rc7 natively on PandaBoard. All things are put in
> > > > their own
> > > > > > > place by make install. However, when running latency, it produces
> > > > > > >
> > > > > > > > latency
> > > > > > > 0"000.000| BUG in low_init(): [main] mlockall: Cannot allocate
> > memory
> > > > > >
> > > > > > You should either run the latency test as root, or have a look at
> > > > > > this page:
> > > > > >
> > > >
> > http://xenomai.org/2014/06/running-a-xenomai-application-as-a-regular-user/
> > > > > >
> > > > > > By root I get this.
> > > > > root@arm:/usr/xenomai# latency
> > > > >    0"000.000| BUG in low_init(): [main] binding failed: Function not
> > > > > implemented
> > > > >
> > > > > However, the PATH has set the Lib
> > > >
> > > > Running as root was necessary to avoid the mlockall issue. The issue
> > > > you have now is different: the logs you sent show that the last
> > > > kernel you booted is not patched with Xenomai. The kernel is patched
> > > > with the I-pipe patch, but this is not sufficient, it is not
> > > > prepared with prepare-kernel.sh
> > > >
> > >
> > > I had thought of that too. However, the patch namely
> > > ipipe-core-3.18.20-arm-5.patch I used is meant for kernel 3.18.20. And
> > > there is no such patch available in the latest xenomai-3-3.0-rc7. ie
> > under
> > > the kernel/Cobalt/arch/arm/patches/, there is such patch available for
> > > 3.18.20. With the observation, I concluded that
> > > ipipe-core-3.18.20-arm-5.patch would patch all is needed for xenomai to
> > > work. In addition, I checked in its make menuconfig, I did not see Cobalt
> > > feature, but I did see ipipe was integrated. Obviously now I know it was
> > > wrong. However, I still yet found a xenomai which has patches for 3.18.20
> > > kernel. Any comments?
> >
> > You can use Xenomai 3.0-rc7 with the I-pipe patch for 3.18.20.
> > Simply pass the path to the patch to the prepare-kernel.sh --adeos
> > option.
> >
> If I am not wrong, prepare-kernel.sh is for preparing kernel by patching
> the patch files under the ..arm/patches/ipipe-core...3.18.12.patch, and in
> my case I patch kernel 3.18.12 with ipipe-core-3.18.12.patch which I did
> not use --adeos but --ipipe.
> 
> The other issue I pointed out above, Xenomai-3.0.rc7 does not have 3.18.20
> patch. Hope this clarifies things. I am happy to use kernel 3.18.12 anyway.

Ok, answering you is beyond my competences. I have been unable to
give you answers you understand. Therefore, I will stop answering
your questions, and hope someone else on the list will.

-- 
					    Gilles.
https://click-hack.org

Re: [Xenomai] Xenomai tests

[vibnwis]

Do you set/unset the items in "make menuconfig" or edit them manually?

On 24 September 2015 at 04:12, Lennart Sorensen <
lsorense@csclub.uwaterloo.ca> wrote:

> On Thu, Sep 24, 2015 at 04:11:25AM +1200, vibnwis wrote:
> > On 24 September 2015 at 03:34, Lennart Sorensen <
> > lsorense@csclub.uwaterloo.ca> wrote:
> >
> > > On Thu, Sep 24, 2015 at 01:26:31AM +1200, vibnwis wrote:
> > > > To do a quick test if these CONFIG_LOCKDEP_SUPPORT
> > > > CONFIG_TRACE_IRQFLAGS_SUPPORT ever able to be changed. I simply turn
> off
> > > > everything in make menuconfig. I mean everything and these two flags
> are
> > > > still there unchanged.
> > > > FYI, I had the .config opened in a editor. Whenever changes were
> made, it
> > > > would prompt in the editor to reload the contents.
> > >
> > > Certainly looking at lib/Kconfig.debug, nowhere else can turn on
> > > CONFIG_LOCKDEP except stuff in there.  Doing a search in make
> menuconfig
> > > for LOCKDEP should show you that symbol and even explain (at least
> > > usually it seems to) why the setting is what it is.
> > >
> > > Here is what I see on 3.14 where I do have LOCKDEP disabled and
> > > ipipe/xenomai in use:
> > >
> > >  .config - Linux/arm 3.14.51 Kernel Configuration
> > >  > Search (LOCKDEP)
> > > -----------------------------------------------------------
> > >   +---------------------------- Search Results
> > > -----------------------------+
> > >   | Symbol: LOCKDEP [=n]
> > >   |
> > >   | Type  : boolean
> > >  |
> > >   |   Defined at lib/Kconfig.debug:928
> > >   |
> > >   |   Depends on: DEBUG_KERNEL [=y] && TRACE_IRQFLAGS_SUPPORT [=y] &&
> > >  |
> > >   |               STACKTRACE_SUPPORT [=y] && LOCKDEP_SUPPORT [=y]
> > >  |
> > >   |   Selects: STACKTRACE [=y] && FRAME_POINTER [=n] && KALLSYMS [=y]
> &&
> > >   |
> > >   |            KALLSYMS_ALL [=y]
> > >   |
> > >   |   Selected by: DEBUG_LOCK_ALLOC [=n] && DEBUG_KERNEL [=y] &&
> > >   |
> > >   |                TRACE_IRQFLAGS_SUPPORT [=y] && STACKTRACE_SUPPORT
> [=y]
> > > &&|
> > >   |                LOCKDEP_SUPPORT [=y] || PROVE_LOCKING [=n] &&
> > >   |
> > >   |                DEBUG_KERNEL [=y] && TRACE_IRQFLAGS_SUPPORT [=y] &&
> > >   |
> > >   |                STACKTRACE_SUPPORT [=y] && LOCKDEP_SUPPORT [=y
> > >  |
> > >   |
> > >  |
> > >   |
> > >  |
> > >   | Symbol: DEBUG_LOCKDEP [=n]
> > >   |
> > >   | Type  : boolean
> > >  |
> > >   | Prompt: Lock dependency engine debugging
> > >   |
> > >   |   Location:
> > >  |
> > >   |     -> Kernel hacking
> > >  |
> > >   | (1)   -> Lock Debugging (spinlocks, mutexes, etc...)
> > >   |
> > >   |   Defined at lib/Kconfig.debug:957
> > >   |
> > >   |   Depends on: DEBUG_KERNEL [=y] && LOCKDEP [=n]
> > >  |
> > >   |
> > >  |
> > >   |
> > >  |
> > >   | Symbol: LOCKDEP_SUPPORT [=y]
> > >   |
> > >   | Type  : boolean
> > >  |
> > >   |   Defined at arch/arm/Kconfig:169
> > >  |
> > >   |
> > >  |
> > >
> > >
> +-----------------------------------------------------------------(100%)--+
> > >   |                                < Exit >
> > >  |
> > >
> > >
> +-------------------------------------------------------------------------+
> > >
> > > So in this case since the 'Selected by' says DEBUG_LOCK_ALLOC is off,
> > > then the first block (up to ||) is false, and PROVE_LOCKING is off,
> > > so the next block is false, which means all selectors of LOCKDEP are
> > > false and it is itself false.
> > >
> >
> > Tried it here.
> >
> > kernel hacking --->
> >               Lock Debugging (spinlocks, mutexes, etc...)  --->
> >                  [ ] RT Mutex debugging, deadlock
> > detection                       │ │
> >   │ │    [ ] Spinlock and rw-lock debugging: basic checks
>  │
> > │
> >   │ │    [ ] Mutex debugging: basic checks
>   │
> > │
> >   │ │    [ ] Wait/wound mutex debugging: Slowpath testing
>  │
> > │
> >   │ │    [ ] Lock debugging: detect incorrect freeing of live locks
>  │
> > │
> >   │ │    [ ] Lock debugging: prove locking correctness
>   │
> > │
> >   │ │    [ ] Lock usage statistics
>   │
> > │
> >   │ │    [ ] Sleep inside atomic section checking
>  │
> > │
> >   │ │    [ ] Locking API boot-time self-tests
>  │
> > │
> >   │ │    < > torture tests for locking
> >
> > All items under Lock Debugging are disabled, yet LOCKDEP and
> TRACE_IRQFLAGS
> > remain unchanged.;-(
>
> Hit '/' Then type 'LOCKDEP' and see what it says is causing it to be on.
> Looking at the menu will not help you.
>
> --
> Len Sorensen
>

Re: [Xenomai] Xenomai tests

[Lennart Sorensen]

On Fri, Sep 25, 2015 at 11:55:39AM +1200, vibnwis wrote:
> Do you set/unset the items in "make menuconfig" or edit them manually?

In make menuconfig.

But using the saerch and seeing the state and logic it is using helps
to go change the right dependancies.

-- 
Len Sorensen

Re: [Xenomai] Xenomai tests

[vibnwis]

>
>
> In make menuconfig.
>
> But using the saerch and seeing the state and logic it is using helps
> to go change the right dependancies.
>
Using the boolean logic it should be quite easy to do that. However, I
seriously not able to achieve that.
For instance, DEBUG_KERNEL = "kernel debugging" has been enabled but greyed
out. "/" key and searched it, showing it is of boolean and no dependencies
info. I can't get it disabled. Mouse-clicking ( I use xconfig is very
helpful  and useful) as well as space-bar, (use menuconfig) not able to
change the status. Someone please help.

Thank you in advance.

>
> --
> Len Sorensen
>