From mboxrd@z Thu Jan  1 00:00:00 1970
Message-ID: <54508FB3.8080600@bmwedler.de>
Date: Wed, 29 Oct 2014 07:56:51 +0100
From: Registrierungen <regist@bmwedler.de>
MIME-Version: 1.0
References: <544F6117.7070101@bmwedler.de> <544F6988.1030005@xenomai.org>
 <544F9DB1.1050504@bmwedler.de> <544FA001.6070305@xenomai.org>
In-Reply-To: <544FA001.6070305@xenomai.org>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Xenomai] Using pipes in 3.0rc
List-Id: Discussions about the Xenomai project <xenomai.xenomai.org>
List-Unsubscribe: <http://www.xenomai.org/mailman/options/xenomai>,
 <mailto:xenomai-request@xenomai.org?subject=unsubscribe>
List-Archive: <http://www.xenomai.org/pipermail/xenomai/>
List-Post: <mailto:xenomai@xenomai.org>
List-Help: <mailto:xenomai-request@xenomai.org?subject=help>
List-Subscribe: <http://www.xenomai.org/mailman/listinfo/xenomai>,
 <mailto:xenomai-request@xenomai.org?subject=subscribe>
To: Philippe Gerum <rpm@xenomai.org>, xenomai@xenomai.org

Addendum:
I am using a x64 Kernel.
Does it matter?


Am 28.10.2014 14:54, schrieb Philippe Gerum:
> On 10/28/2014 02:44 PM, Registrierungen wrote:
>> Thanks for your support.
>>
>> Unfortainly it din'twork for me.
>> I patched and recompiled and installed xenomai and my app,
>> but i have still the same error.
>>
>> Have you got another idea, or need some other informations from my side?
> You may want to trace your app with gdb, configuring with
> --enable-debug=full, then single-stepping in rt_pipe_create() to
> pinpoint the failing call in there.
>
>> Sascha
>>
>> Am 28.10.2014 11:01, schrieb Philippe Gerum:
>>> On 10/28/2014 10:25 AM, Registrierungen wrote:
>>>> Hello,
>>>>
>>>> i try to switch with my working project from Xenomai 2.6 to 3.0rc
>>>> Cobaltusing the native/alchemy skin.
>>>>
>>>> Creating tasks,etc. seems to works fine.
>>>>
>>>> If i try to open a pipe from the realtime taks using rt_pipe_create it
>>>> returns with the error -88.
>>> This is -ENOTSOCK, you can check all error codes there:
>>> /usr/include/asm-generic/errno-base.h
>>> /usr/include/asm-generic/errno.h
>>>
>>> ENOTSOCK is not raised by the alchemy API, so this must be a syscall
>>> taking the wrong route to the glibc in the core lib.
>>> Please try this patch:
>>>
>>> diff --git a/lib/alchemy/pipe.c b/lib/alchemy/pipe.c
>>> index 93da9be..439a00f 100644
>>> --- a/lib/alchemy/pipe.c
>>> +++ b/lib/alchemy/pipe.c
>>> @@ -166,8 +166,8 @@ int rt_pipe_create(RT_PIPE *pipe,
>>>        }
>>>          if (poolsize > 0) {
>>> -        ret = setsockopt(pcb->sock, SOL_XDDP, XDDP_POOLSZ,
>>> -                 &poolsize, sizeof(poolsize));
>>> +        ret = __RT(setsockopt(pcb->sock, SOL_XDDP, XDDP_POOLSZ,
>>> +                      &poolsize, sizeof(poolsize)));
>>>            if (ret)
>>>                goto fail_sockopt;
>>>        }
>>> @@ -181,7 +181,7 @@ int rt_pipe_create(RT_PIPE *pipe,
>>>        memset(&saddr, 0, sizeof(saddr));
>>>        saddr.sipc_family = AF_RTIPC;
>>>        saddr.sipc_port = minor;
>>> -    ret = bind(sock, (struct sockaddr *)&saddr, sizeof(saddr));
>>> +    ret = __RT(bind(sock, (struct sockaddr *)&saddr, sizeof(saddr)));
>>>        if (ret)
>>>            goto fail_sockopt;
>>>   
>>>> Has anything changed handling pipes in 3.0?
>>> All user-visible changes are reported here:
>>> http://localhost/xenomai/migrating-from-xenomai-2-x-to-3-x/
>>>
>>
>