[Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Hadrien Caron]

[-- Attachment #1: Type: text/plain, Size: 440 bytes --]

Hello to all !

I have a small problem using a4l_sync_write : I am using it inside a
program that should write on a digital port.
The function a4l_sync_write returns a value of -38.
But according to the documentation, there is only 3 possible error
possibles :
EINVAL = 22
EFAULT = 14
ENOMEM = 12

So what is error 38 ? Is it missing from the documentation ? Or am I doing
something wrong ?

Thank you very much !

Sincerely

Hadrien Caron

[-- Attachment #2: Type: text/html, Size: 499 bytes --]

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Wolfgang Grandegger]

On 11/21/2011 11:44 AM, Hadrien Caron wrote:
> Hello to all !
> 
> I have a small problem using a4l_sync_write : I am using it inside a
> program that should write on a digital port.
> The function a4l_sync_write returns a value of -38.
> But according to the documentation, there is only 3 possible error
> possibles :
> EINVAL = 22
> EFAULT = 14
> ENOMEM = 12
> 
> So what is error 38 ? Is it missing from the documentation ? Or am I doing
> something wrong ?

With "http://www.xenomai.org/index.php/Resolving_Error_Codes" I get:

$ errno 38
#define	ENOSYS		38	/* Function not implemented */

Wolfgang.

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Hadrien Caron]

[-- Attachment #1: Type: text/plain, Size: 881 bytes --]

Thanks very much Wolfang !
Cool script, I didn't know about it :)
I'll try to correct my code with that information !

Thanks again,

Hadrien

On Mon, Nov 21, 2011 at 12:03 PM, Wolfgang Grandegger <wg@domain.hid>wrote:

> On 11/21/2011 11:44 AM, Hadrien Caron wrote:
> > Hello to all !
> >
> > I have a small problem using a4l_sync_write : I am using it inside a
> > program that should write on a digital port.
> > The function a4l_sync_write returns a value of -38.
> > But according to the documentation, there is only 3 possible error
> > possibles :
> > EINVAL = 22
> > EFAULT = 14
> > ENOMEM = 12
> >
> > So what is error 38 ? Is it missing from the documentation ? Or am I
> doing
> > something wrong ?
>
> With "http://www.xenomai.org/index.php/Resolving_Error_Codes" I get:
>
> $ errno 38
> #define ENOSYS          38      /* Function not implemented */
>
> Wolfgang.
>
>

[-- Attachment #2: Type: text/html, Size: 1425 bytes --]

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Gilles Chanteperdrix]

On 11/21/2011 12:10 PM, Hadrien Caron wrote:
> Thanks very much Wolfang !
> Cool script, I didn't know about it :)
> I'll try to correct my code with that information !
> 
> Thanks again,

ENOSYS usually means that you have not enabled the relevant support in
kernel configuration.


-- 
					    Gilles.

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Alexis Berlemont]

Hi,

On Mon, Nov 21, 2011 at 3:20 PM, Gilles Chanteperdrix
<gilles.chanteperdrix@xenomai.org> wrote:
> On 11/21/2011 12:10 PM, Hadrien Caron wrote:
>> Thanks very much Wolfang !
>> Cool script, I didn't know about it :)
>> I'll try to correct my code with that information !
>>
>> Thanks again,
>
> ENOSYS usually means that you have not enabled the relevant support in
> kernel configuration.
>

Which driver are you using? I think analogy returns ENOSYS if the
driver does not implement the related instruction.

Regards,

Alexis.

>
> --
>                                            Gilles.
>
> _______________________________________________
> Xenomai-help mailing list
> Xenomai-help@domain.hid
> https://mail.gna.org/listinfo/xenomai-help
>

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Anders Blomdell]

On 11/22/2011 01:06 PM, Alexis Berlemont wrote:
> Hi,
>
> On Mon, Nov 21, 2011 at 3:20 PM, Gilles Chanteperdrix
> <gilles.chanteperdrix@xenomai.org>  wrote:
>> On 11/21/2011 12:10 PM, Hadrien Caron wrote:
>>> Thanks very much Wolfang !
>>> Cool script, I didn't know about it :)
>>> I'll try to correct my code with that information !
>>>
>>> Thanks again,
>>
>> ENOSYS usually means that you have not enabled the relevant support in
>> kernel configuration.
>>
>
> Which driver are you using? I think analogy returns ENOSYS if the
> driver does not implement the related instruction.
It can also happen with xenomai compiled with gcc-4.6.x, but without 
-fno-omit-frame-pointer :-(

Regards

Anders

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Hadrien Caron]

[-- Attachment #1: Type: text/plain, Size: 1272 bytes --]

Alexis : yes I am using the Analogy drivers . . . Would that mean that
analog output is implemented but not digital output for National Instrument
acquisition cards ? Has someone tried this already ? Seems strange that
analog output would be implemented but not digital.

Anders : Ok ! I will look into rebulding my xenomai installation then !

Many thanks to both of you for your ideas !

Hadrien

On Tue, Nov 22, 2011 at 1:33 PM, Anders Blomdell <
anders.blomdell@domain.hid> wrote:

> On 11/22/2011 01:06 PM, Alexis Berlemont wrote:
>
>> Hi,
>>
>> On Mon, Nov 21, 2011 at 3:20 PM, Gilles Chanteperdrix
>> <gilles.chanteperdrix@xenomai.org <gilles.chanteperdrix@xenomai.org>>
>>  wrote:
>>
>>> On 11/21/2011 12:10 PM, Hadrien Caron wrote:
>>>
>>>> Thanks very much Wolfang !
>>>> Cool script, I didn't know about it :)
>>>> I'll try to correct my code with that information !
>>>>
>>>> Thanks again,
>>>>
>>>
>>> ENOSYS usually means that you have not enabled the relevant support in
>>> kernel configuration.
>>>
>>>
>> Which driver are you using? I think analogy returns ENOSYS if the
>> driver does not implement the related instruction.
>>
> It can also happen with xenomai compiled with gcc-4.6.x, but without
> -fno-omit-frame-pointer :-(
>
> Regards
>
> Anders
>

[-- Attachment #2: Type: text/html, Size: 2133 bytes --]

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Alexis Berlemont]

On Tue, Nov 22, 2011 at 2:43 PM, Hadrien Caron <hadrien.caron@domain.hid> wrote:
> Alexis : yes I am using the Analogy drivers . . . Would that mean that
> analog output is implemented but not digital output for National Instrument
> acquisition cards ? Has someone tried this already ? Seems strange that
> analog output would be implemented but not digital.
>
> Anders : Ok ! I will look into rebulding my xenomai installation then !

According to the thread:
[Xenomai-core] Problems with gcc 4.6.0 (rt_task_shadow fails with ENOSYS)
https://mail.gna.org/public/xenomai-core/2011-07/msg00010.html
The problem related by Anders Blomdell is supposed to be fixed with
the 2.6 release, isn't it?

Could you tell us which version of Xenomai and which driver
(ni_pcimio, ...) you are using?

Regards,

Alexis.
>
> Many thanks to both of you for your ideas !
>
> Hadrien
>
> On Tue, Nov 22, 2011 at 1:33 PM, Anders Blomdell
> <anders.blomdell@domain.hid> wrote:
>>
>> On 11/22/2011 01:06 PM, Alexis Berlemont wrote:
>>>
>>> Hi,
>>>
>>> On Mon, Nov 21, 2011 at 3:20 PM, Gilles Chanteperdrix
>>> <gilles.chanteperdrix@xenomai.org>  wrote:
>>>>
>>>> On 11/21/2011 12:10 PM, Hadrien Caron wrote:
>>>>>
>>>>> Thanks very much Wolfang !
>>>>> Cool script, I didn't know about it :)
>>>>> I'll try to correct my code with that information !
>>>>>
>>>>> Thanks again,
>>>>
>>>> ENOSYS usually means that you have not enabled the relevant support in
>>>> kernel configuration.
>>>>
>>>
>>> Which driver are you using? I think analogy returns ENOSYS if the
>>> driver does not implement the related instruction.
>>
>> It can also happen with xenomai compiled with gcc-4.6.x, but without
>> -fno-omit-frame-pointer :-(
>>
>> Regards
>>
>> Anders
>
>

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Anders Blomdell]

On 11/22/2011 11:33 PM, Alexis Berlemont wrote:
> On Tue, Nov 22, 2011 at 2:43 PM, Hadrien Caron<hadrien.caron@domain.hid>  wrote:
>> Alexis : yes I am using the Analogy drivers . . . Would that mean that
>> analog output is implemented but not digital output for National Instrument
>> acquisition cards ? Has someone tried this already ? Seems strange that
>> analog output would be implemented but not digital.
>>
>> Anders : Ok ! I will look into rebulding my xenomai installation then !
>
> According to the thread:
> [Xenomai-core] Problems with gcc 4.6.0 (rt_task_shadow fails with ENOSYS)
> https://mail.gna.org/public/xenomai-core/2011-07/msg00010.html
> The problem related by Anders Blomdell is supposed to be fixed with
> the 2.6 release, isn't it?
Well, I recently had problems with analogy (2.6 release, 4.6 compiler), 
but my problems were with analog sync input. Had too much to do at that 
time to investigate further, so I recompiled my old 2.5.x xenomai with 
-fno-omit-frame-pointer just to get that machine shipped shipped away to 
the physics department to do some quantum dot experiments.
>
> Could you tell us which version of Xenomai and which driver
> (ni_pcimio, ...) you are using?
>
> Regards,
>
> Alexis.
>>
>> Many thanks to both of you for your ideas !
>>
>> Hadrien
>>
>> On Tue, Nov 22, 2011 at 1:33 PM, Anders Blomdell
>> <anders.blomdell@domain.hid>  wrote:
>>>
>>> On 11/22/2011 01:06 PM, Alexis Berlemont wrote:
>>>>
>>>> Hi,
>>>>
>>>> On Mon, Nov 21, 2011 at 3:20 PM, Gilles Chanteperdrix
>>>> <gilles.chanteperdrix@xenomai.org>    wrote:
>>>>>
>>>>> On 11/21/2011 12:10 PM, Hadrien Caron wrote:
>>>>>>
>>>>>> Thanks very much Wolfang !
>>>>>> Cool script, I didn't know about it :)
>>>>>> I'll try to correct my code with that information !
>>>>>>
>>>>>> Thanks again,
>>>>>
>>>>> ENOSYS usually means that you have not enabled the relevant support in
>>>>> kernel configuration.
>>>>>
>>>>
>>>> Which driver are you using? I think analogy returns ENOSYS if the
>>>> driver does not implement the related instruction.
>>>
>>> It can also happen with xenomai compiled with gcc-4.6.x, but without
>>> -fno-omit-frame-pointer :-(

>>>
>>> Regards
>>>
>>> Anders
>>
>>

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Gilles Chanteperdrix]

On 11/22/2011 11:33 PM, Alexis Berlemont wrote:
> On Tue, Nov 22, 2011 at 2:43 PM, Hadrien Caron <hadrien.caron@domain.hid> wrote:
>> Alexis : yes I am using the Analogy drivers . . . Would that mean that
>> analog output is implemented but not digital output for National Instrument
>> acquisition cards ? Has someone tried this already ? Seems strange that
>> analog output would be implemented but not digital.
>>
>> Anders : Ok ! I will look into rebulding my xenomai installation then !
> 
> According to the thread:
> [Xenomai-core] Problems with gcc 4.6.0 (rt_task_shadow fails with ENOSYS)
> https://mail.gna.org/public/xenomai-core/2011-07/msg00010.html
> The problem related by Anders Blomdell is supposed to be fixed with
> the 2.6 release, isn't it?
> 
> Could you tell us which version of Xenomai and which driver
> (ni_pcimio, ...) you are using?

Hi,

the problem is not fixed, we can not compile xenomai without frame pointers.

-- 
                                                                Gilles.

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Hadrien Caron]

[-- Attachment #1: Type: text/plain, Size: 1302 bytes --]

I use Xenomai 2.5.6 with drivers nipcimio. Should I try it with the latest
2.6 version ?

Thanks again for all your ideas !

Hadrien

On Wed, Nov 23, 2011 at 9:44 AM, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> On 11/22/2011 11:33 PM, Alexis Berlemont wrote:
> > On Tue, Nov 22, 2011 at 2:43 PM, Hadrien Caron <hadrien.caron@domain.hid>
> wrote:
> >> Alexis : yes I am using the Analogy drivers . . . Would that mean that
> >> analog output is implemented but not digital output for National
> Instrument
> >> acquisition cards ? Has someone tried this already ? Seems strange that
> >> analog output would be implemented but not digital.
> >>
> >> Anders : Ok ! I will look into rebulding my xenomai installation then !
> >
> > According to the thread:
> > [Xenomai-core] Problems with gcc 4.6.0 (rt_task_shadow fails with ENOSYS)
> > https://mail.gna.org/public/xenomai-core/2011-07/msg00010.html
> > The problem related by Anders Blomdell is supposed to be fixed with
> > the 2.6 release, isn't it?
> >
> > Could you tell us which version of Xenomai and which driver
> > (ni_pcimio, ...) you are using?
>
> Hi,
>
> the problem is not fixed, we can not compile xenomai without frame
> pointers.
>
> --
>                                                                Gilles.
>

[-- Attachment #2: Type: text/html, Size: 1946 bytes --]

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Hadrien Caron]

[-- Attachment #1: Type: text/plain, Size: 3920 bytes --]

Hello again to all,

I have been trying to look at the problem and finding it's source following
the various advice you gave me.
My problem is still trying to use the DIO channels of my NI-DAQ card

I updated my xenomai configuration :

xenomai 2.6.0
ubuntu 10.10 kernel 2.6.38-8

with a NI-Daq card : PCI-6221

--------------------
$ lspci | grep National
06:05.0 Unassigned class [ff00]: National Instruments Device 71bc
--------------------
71bc is the device_id of pci-6221_37pin according to pcimio.c . . . So to
me this is a good sign and that means that the device is well recognised.
But in the header of pcimio.c I see :
  "- Digital I/O may not work on 673x.
 * - Information (number of channels, bits, etc.) for some devices may
 *   be incorrect.  Please check this and submit a bug if there are
 *   problems for your device"

So I am starting to be afraid that there might be a bug and that the
information for my device is not accurate (even though the issue reported
is not about my card).

Furthermore, I tried using the cmd_bits program to use the DIO channels.

First, here is the content of my 00-analogy_ni_pcimio file :
jb@domain.hid$ cat 00-analogy_ni_pcimio
--  Subdevices --

| idx | type
|  00 | Analog input subdevice
|  01 | Analog output subdevice
|  02 | Digital input/output subdevice
|  03 | Unused subdevice
|  04 | Unused subdevice
|  05 | Calibration subdevice
|  06 | Memory subdevice
|  07 | Digital input/output subdevice
|  08 | Unused subdevice
|  09 | Serial subdevice
|  10 | Unused subdevice
|  11 | Counter subdevice
|  12 | Counter subdevice
|  13 | Counter subdevice

jb@domain.hid$ sudo ./cmd_bits -d analogy0 -s 1
cmd_bits: selected subdevice is not digital

# Which seems normal : idx 1 is for analog Output. But if I use "-s 2" or
"-s 7" I get :

jb@domain.hid$ sudo ./cmd_bits -d analogy0 -s 7
cmd_bits: a4l_snd_command failed (err=-5)

thant I used the errno.sh tool :
./errno.sh 5
#define    EIO         5    /* I/O error */

and in the Xenomai API documentation I read about a4l_snd_command :
-EIO is returned if the selected subdevice cannot handle command

So I am not sure what to make up of all this. I feel like there is a
problem with the implementation
of analogy_pcmio for my NI-DAQ card. If this is the case, I would be happy
(to at least try) to fix it myself (with a little help to get myself
started maybe ^^), or do you think the problem is elsewhere ?

Thank you very much for reading this :)

Sincerely,

Hadrien

On Wed, Nov 23, 2011 at 10:53 AM, Hadrien Caron <hadrien.caron@domain.hid>wrote:

> I use Xenomai 2.5.6 with drivers nipcimio. Should I try it with the latest
> 2.6 version ?
>
> Thanks again for all your ideas !
>
> Hadrien
>
>
> On Wed, Nov 23, 2011 at 9:44 AM, Gilles Chanteperdrix <
> gilles.chanteperdrix@xenomai.org> wrote:
>
>> On 11/22/2011 11:33 PM, Alexis Berlemont wrote:
>> > On Tue, Nov 22, 2011 at 2:43 PM, Hadrien Caron <hadrien.caron@domain.hid>
>> wrote:
>> >> Alexis : yes I am using the Analogy drivers . . . Would that mean that
>> >> analog output is implemented but not digital output for National
>> Instrument
>> >> acquisition cards ? Has someone tried this already ? Seems strange that
>> >> analog output would be implemented but not digital.
>> >>
>> >> Anders : Ok ! I will look into rebulding my xenomai installation then !
>> >
>> > According to the thread:
>> > [Xenomai-core] Problems with gcc 4.6.0 (rt_task_shadow fails with
>> ENOSYS)
>> > https://mail.gna.org/public/xenomai-core/2011-07/msg00010.html
>> > The problem related by Anders Blomdell is supposed to be fixed with
>> > the 2.6 release, isn't it?
>> >
>> > Could you tell us which version of Xenomai and which driver
>> > (ni_pcimio, ...) you are using?
>>
>> Hi,
>>
>> the problem is not fixed, we can not compile xenomai without frame
>> pointers.
>>
>> --
>>                                                                Gilles.
>>
>
>

[-- Attachment #2: Type: text/html, Size: 5168 bytes --]

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Hadrien Caron]

[-- Attachment #1: Type: text/plain, Size: 4334 bytes --]

Hello again !
I am kind of stuck right now, anybody has an idea regarding my post ?
Anyone able to help ?

Thanks in advance,

Hadrien


On Thu, Dec 8, 2011 at 3:36 PM, Hadrien Caron <hadrien.caron@domain.hid>wrote:

> Hello again to all,
>
> I have been trying to look at the problem and finding it's source
> following the various advice you gave me.
> My problem is still trying to use the DIO channels of my NI-DAQ card
>
> I updated my xenomai configuration :
>
> xenomai 2.6.0
> ubuntu 10.10 kernel 2.6.38-8
>
> with a NI-Daq card : PCI-6221
>
> --------------------
> $ lspci | grep National
> 06:05.0 Unassigned class [ff00]: National Instruments Device 71bc
> --------------------
> 71bc is the device_id of pci-6221_37pin according to pcimio.c . . . So to
> me this is a good sign and that means that the device is well recognised.
> But in the header of pcimio.c I see :
>   "- Digital I/O may not work on 673x.
>  * - Information (number of channels, bits, etc.) for some devices may
>  *   be incorrect.  Please check this and submit a bug if there are
>  *   problems for your device"
>
> So I am starting to be afraid that there might be a bug and that the
> information for my device is not accurate (even though the issue reported
> is not about my card).
>
> Furthermore, I tried using the cmd_bits program to use the DIO channels.
>
> First, here is the content of my 00-analogy_ni_pcimio file :
> jb@domain.hid$ cat 00-analogy_ni_pcimio
> --  Subdevices --
>
> | idx | type
> |  00 | Analog input subdevice
> |  01 | Analog output subdevice
> |  02 | Digital input/output subdevice
> |  03 | Unused subdevice
> |  04 | Unused subdevice
> |  05 | Calibration subdevice
> |  06 | Memory subdevice
> |  07 | Digital input/output subdevice
> |  08 | Unused subdevice
> |  09 | Serial subdevice
> |  10 | Unused subdevice
> |  11 | Counter subdevice
> |  12 | Counter subdevice
> |  13 | Counter subdevice
>
> jb@domain.hid$ sudo ./cmd_bits -d analogy0 -s 1
> cmd_bits: selected subdevice is not digital
>
> # Which seems normal : idx 1 is for analog Output. But if I use "-s 2" or
> "-s 7" I get :
>
> jb@domain.hid$ sudo ./cmd_bits -d analogy0 -s 7
> cmd_bits: a4l_snd_command failed (err=-5)
>
> thant I used the errno.sh tool :
> ./errno.sh 5
> #define    EIO         5    /* I/O error */
>
> and in the Xenomai API documentation I read about a4l_snd_command :
> -EIO is returned if the selected subdevice cannot handle command
>
> So I am not sure what to make up of all this. I feel like there is a
> problem with the implementation
> of analogy_pcmio for my NI-DAQ card. If this is the case, I would be happy
> (to at least try) to fix it myself (with a little help to get myself
> started maybe ^^), or do you think the problem is elsewhere ?
>
> Thank you very much for reading this :)
>
> Sincerely,
>
> Hadrien
>
>
> On Wed, Nov 23, 2011 at 10:53 AM, Hadrien Caron <hadrien.caron@domain.hid>wrote:
>
>> I use Xenomai 2.5.6 with drivers nipcimio. Should I try it with the
>> latest 2.6 version ?
>>
>> Thanks again for all your ideas !
>>
>> Hadrien
>>
>>
>> On Wed, Nov 23, 2011 at 9:44 AM, Gilles Chanteperdrix <
>> gilles.chanteperdrix@xenomai.org> wrote:
>>
>>> On 11/22/2011 11:33 PM, Alexis Berlemont wrote:
>>> > On Tue, Nov 22, 2011 at 2:43 PM, Hadrien Caron <
>>> hadrien.caron@domain.hid> wrote:
>>> >> Alexis : yes I am using the Analogy drivers . . . Would that mean that
>>> >> analog output is implemented but not digital output for National
>>> Instrument
>>> >> acquisition cards ? Has someone tried this already ? Seems strange
>>> that
>>> >> analog output would be implemented but not digital.
>>> >>
>>> >> Anders : Ok ! I will look into rebulding my xenomai installation then
>>> !
>>> >
>>> > According to the thread:
>>> > [Xenomai-core] Problems with gcc 4.6.0 (rt_task_shadow fails with
>>> ENOSYS)
>>> > https://mail.gna.org/public/xenomai-core/2011-07/msg00010.html
>>> > The problem related by Anders Blomdell is supposed to be fixed with
>>> > the 2.6 release, isn't it?
>>> >
>>> > Could you tell us which version of Xenomai and which driver
>>> > (ni_pcimio, ...) you are using?
>>>
>>> Hi,
>>>
>>> the problem is not fixed, we can not compile xenomai without frame
>>> pointers.
>>>
>>> --
>>>                                                                Gilles.
>>>
>>
>>
>

[-- Attachment #2: Type: text/html, Size: 5744 bytes --]

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Gilles Chanteperdrix]

On 12/16/2011 01:39 PM, Hadrien Caron wrote:
> Hello again !
> I am kind of stuck right now, anybody has an idea regarding my post ?
> Anyone able to help ?
> 
> Thanks in advance,

Have you tried compiling with -fno-omit-frame-pointer (like Anders told
you)?

-- 
					    Gilles.

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Anders Blomdell]

On 12/16/2011 02:20 PM, Gilles Chanteperdrix wrote:
> On 12/16/2011 01:39 PM, Hadrien Caron wrote:
>> Hello again !
>> I am kind of stuck right now, anybody has an idea regarding my post ?
>> Anyone able to help ?
>>
>> Thanks in advance,
>
> Have you tried compiling with -fno-omit-frame-pointer (like Anders told
> you)?
How about adding this to be automatically done?

/Anders

-- 
Anders Blomdell                  Email: anders.blomdell@domain.hid
Department of Automatic Control
Lund University                  Phone:    +46 46 222 4625
P.O. Box 118                     Fax:      +46 46 138118
SE-221 00 Lund, Sweden

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Gilles Chanteperdrix]

On 12/16/2011 03:43 PM, Anders Blomdell wrote:
> On 12/16/2011 02:20 PM, Gilles Chanteperdrix wrote:
>> On 12/16/2011 01:39 PM, Hadrien Caron wrote:
>>> Hello again !
>>> I am kind of stuck right now, anybody has an idea regarding my post ?
>>> Anyone able to help ?
>>>
>>> Thanks in advance,
>>
>> Have you tried compiling with -fno-omit-frame-pointer (like Anders told
>> you)?
> How about adding this to be automatically done?

It is on the todo list for 2.6.1.

-- 
					    Gilles.

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Hadrien Caron]

[-- Attachment #1: Type: text/plain, Size: 1206 bytes --]

Hello !

Sorry for the delay, I was out of town for the week.
Hope you all had a nice christmas.
Thanks for the input. In fact I am sorry but I didn't find what it means to
compile with -fno-omit-frame-pointer

I found a SCHED_OMIT_FRAME_POINTER option in the kernel but it was already
enabled so I doubt you were talking about that.

Is it an option to pass to the compiler ? I found -fomit-frame-pointer in
the gcc manual . . . Should I modify xenomai's Makefile to add this option
to the compilation ?

Thanks again ! And happy new year ;)

Hadrien

On Fri, Dec 16, 2011 at 3:50 PM, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> On 12/16/2011 03:43 PM, Anders Blomdell wrote:
> > On 12/16/2011 02:20 PM, Gilles Chanteperdrix wrote:
> >> On 12/16/2011 01:39 PM, Hadrien Caron wrote:
> >>> Hello again !
> >>> I am kind of stuck right now, anybody has an idea regarding my post ?
> >>> Anyone able to help ?
> >>>
> >>> Thanks in advance,
> >>
> >> Have you tried compiling with -fno-omit-frame-pointer (like Anders told
> >> you)?
> > How about adding this to be automatically done?
>
> It is on the todo list for 2.6.1.
>
> --
>                                            Gilles.
>

[-- Attachment #2: Type: text/html, Size: 1695 bytes --]

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Gilles Chanteperdrix]

On 12/29/2011 06:22 PM, Hadrien Caron wrote:
> Hello !
> 
> Sorry for the delay, I was out of town for the week.
> Hope you all had a nice christmas.
> Thanks for the input. In fact I am sorry but I didn't find what it means to
> compile with -fno-omit-frame-pointer
> 
> I found a SCHED_OMIT_FRAME_POINTER option in the kernel but it was already
> enabled so I doubt you were talking about that.
> 
> Is it an option to pass to the compiler ? I found -fomit-frame-pointer in
> the gcc manual . . . Should I modify xenomai's Makefile to add this option
> to the compilation ?
> 
> Thanks again ! And happy new year ;)

It is an option to pass to gcc (every time there is -ffoo there is also
the -fno-foo). You can pass additional flags to gcc using configure
CFLAGS variable.

Using for instance:
/path/to/xenomai/configure CFLAGS="-fno-omit-frame-pointer -g -O2"

-- 
                                                                Gilles.

Re: [Xenomai-help] [xenomai-help] Error return with a4l_sync_write

[Hadrien Caron]

[-- Attachment #1: Type: text/plain, Size: 1346 bytes --]

Thanks, I compiled using the configure as you adviced, it didn't make any
change though :(

For now I need to focus on other developments, I will come back to this
later. Thanks you
very much for the help though !

Happy new year to everybody ;)

Hadrien

On Thu, Dec 29, 2011 at 6:56 PM, Gilles Chanteperdrix <
gilles.chanteperdrix@xenomai.org> wrote:

> On 12/29/2011 06:22 PM, Hadrien Caron wrote:
> > Hello !
> >
> > Sorry for the delay, I was out of town for the week.
> > Hope you all had a nice christmas.
> > Thanks for the input. In fact I am sorry but I didn't find what it means
> to
> > compile with -fno-omit-frame-pointer
> >
> > I found a SCHED_OMIT_FRAME_POINTER option in the kernel but it was
> already
> > enabled so I doubt you were talking about that.
> >
> > Is it an option to pass to the compiler ? I found -fomit-frame-pointer in
> > the gcc manual . . . Should I modify xenomai's Makefile to add this
> option
> > to the compilation ?
> >
> > Thanks again ! And happy new year ;)
>
> It is an option to pass to gcc (every time there is -ffoo there is also
> the -fno-foo). You can pass additional flags to gcc using configure
> CFLAGS variable.
>
> Using for instance:
> /path/to/xenomai/configure CFLAGS="-fno-omit-frame-pointer -g -O2"
>
> --
>                                                                Gilles.
>

[-- Attachment #2: Type: text/html, Size: 1832 bytes --]