[Xenomai-help] IDDP send-receive example emits SIGXCPU signal

[Xenomai-help] IDDP send-receive example emits SIGXCPU signal

[Guvenc Gulce]

Hi
My environment looks like the following:
Xenomai 2.5.6 
Used skin: Posix

Kernel 2.6.35.3
Target Environment: ARM


I have compiled the example iddp-sendrecv which can be found under /examples/rtdm/profiles/ipc 
(http://www.xenomai.org/documentation/xenomai-head/html/api/iddp-sendrecv_8c-example.html )

with the existing makefile under the same directory. The binary seems to be linking ok with the 

xenomai libraries as you can also see here:


arm-linux-readelf -d iddp-sendrecv

Dynamic section at offset 0x1ed8 contains 32 entries:
  Tag        Type                         Name/Value
 0x00000001 (NEEDED)                     Shared library: [libpthread_rt.so.1]
 0x00000001 (NEEDED)                     Shared library: [libxenomai.so.0]
 0x00000001 (NEEDED)                     Shared library: [libpthread.so.0]
 0x00000001 (NEEDED)                     Shared library: [librt.so.1]
 0x00000001 (NEEDED)                     Shared library: [librtdm.so.1]
 0x00000001 (NEEDED)                     Shared library: [librtdk.so.0]
 0x00000001 (NEEDED)                     Shared library: [libc.so.6]



In order to test the real-timeliness of this piece of binary, I have added 

one line of code to the beginning of the each thread (client and server)

        pthread_set_mode_np(0, PTHREAD_WARNSW);


so that any switch from Primary to Secondary will be reported with a 

SIGXCPU. My expectation was that I wouldnt see any switch to secondary 

mode as IDDP is supposed to work between real time threads without hurting 

their real-timeliness.

But I do get the SIGXCPU after adding the PTHREAD_WARNSW and the 

gdb output says

Program received signal SIGXCPU, CPU time limit exceeded.
[Switching to Thread 0x2bc83470 (LWP 3401)]
0x2aad9894 in __wrap_socket () from /usr/xenomai/lib/libpthread_rt.so.1


so it looks like the __wrap_socket call in the posix skin hurts the realtimeliness 

of the threads when using IDDP. Do you have any idea why this may be 

happening ?

If we look deeper in the posix skin where __wrap_socket is implemented

        ret = XENOMAI_SKINCALL3(__rtdm_muxid,
                                __rtdm_socket,
                                protocol_family, socket_type, protocol);
        if (ret >= 0)
                ret += __rtdm_fd_start;
        else if (ret == -EAFNOSUPPORT || ret == -ENOSYS) {
                ret = __real_socket(protocol_family, socket_type, protocol);

                if (ret >= __rtdm_fd_start) {
                        __real_close(ret);
                        errno = -EMFILE;
                        ret = -1;
                }


looks like XENOMAI_SKINCALL3 macro is failing for some reason but I can't 

explain this behavior as I am using the example code linked with the correct 

xenomai libraries. Any help would be highly appreciated.

Thanks in advance

Guvenc

Re: [Xenomai-help] IDDP send-receive example emits SIGXCPU signal

[Gilles Chanteperdrix]

On 11/11/2011 08:04 AM, Guvenc Gulce wrote:
> looks like XENOMAI_SKINCALL3 macro is failing for some reason but I can't 
> 
> explain this behavior as I am using the example code linked with the correct 
> 
> xenomai libraries. Any help would be highly appreciated.

Do you have support for iddp sockets enabled in kernel configuration? If
yes, do you have the same issue with xenomai 2.6.0?

-- 
                                                                Gilles.

Re: [Xenomai-help] IDDP send-receive example emits SIGXCPU signal

[Guvenc Gulce]

Yes.. I have enabled the support for iddp sockets in the kernel configuration. It looks like that 

it works with xenomai 2.6.0 when I enable the 

pthread_set_mode_np(0, PTHREAD_WARNSW);
after the socket() and bind() calls. (tested again with the example code from xenomai)

My question is:

Is it designed that way that the calls socket() and bind() for IDDP cause a switch to 

secondary mode and the calls recvfrom() and sendto() do not ? In other words 

does the real-timeliness guarantee exist only for recvfrom() and sendto() ?


Thanks & Regards

Guvenc




----- Original Message -----
From: Gilles Chanteperdrix <gilles.chanteperdrix@domain.hid>
To: Guvenc Gulce <gulceg@domain.hid>
Cc: "xenomai@xenomai.org" <xenomai@xenomai.org>
Sent: Friday, November 11, 2011 12:43 PM
Subject: Re: [Xenomai-help] IDDP send-receive example emits SIGXCPU signal

On 11/11/2011 08:04 AM, Guvenc Gulce wrote:
> looks like XENOMAI_SKINCALL3 macro is failing for some reason but I can't 
> 
> explain this behavior as I am using the example code linked with the correct 
> 
> xenomai libraries. Any help would be highly appreciated.

Do you have support for iddp sockets enabled in kernel configuration? If
yes, do you have the same issue with xenomai 2.6.0?

-- 
                                                                Gilles.

Re: [Xenomai-help] IDDP send-receive example emits SIGXCPU signal

[Philippe Gerum]

On 11/23/2011 09:31 AM, Guvenc Gulce wrote:
> Yes.. I have enabled the support for iddp sockets in the kernel configuration. It looks like that
>
> it works with xenomai 2.6.0 when I enable the
>
> pthread_set_mode_np(0, PTHREAD_WARNSW);
> after the socket() and bind() calls. (tested again with the example code from xenomai)
>
> My question is:
>
> Is it designed that way that the calls socket() and bind() for IDDP cause a switch to
>
> secondary mode and the calls recvfrom() and sendto() do not ? In other words
>
> does the real-timeliness guarantee exist only for recvfrom() and sendto() ?

No. socket() or bind() have no reason to switch to secondary mode, 
unless socket() fails for the AF_RTIPC domain, in which case the call is 
routed to the regular kernel to handle the address family. You not only 
need to enable IDDP in the kernel, but you also have to wrap POSIX calls 
to libpthread_rt services.

If socket(AF_RTIPC, ...) failed reaching Xenomai/RTDM, then it will also 
fail with the regular kernel, and this call will return an error.

>
>
> Thanks&  Regards
>
> Guvenc
>
>
>
>
> ----- Original Message -----
> From: Gilles Chanteperdrix<gilles.chanteperdrix@xenomai.org>
> To: Guvenc Gulce<gulceg@domain.hid>
> Cc: "xenomai@xenomai.org"<xenomai@xenomai.org>
> Sent: Friday, November 11, 2011 12:43 PM
> Subject: Re: [Xenomai-help] IDDP send-receive example emits SIGXCPU signal
>
> On 11/11/2011 08:04 AM, Guvenc Gulce wrote:
>> looks like XENOMAI_SKINCALL3 macro is failing for some reason but I can't
>>
>> explain this behavior as I am using the example code linked with the correct
>>
>> xenomai libraries. Any help would be highly appreciated.
>
> Do you have support for iddp sockets enabled in kernel configuration? If
> yes, do you have the same issue with xenomai 2.6.0?
>


-- 
Philippe.