Re: Error reproduecd again when using IDDP: BUG in low_init(): [main] binding failed: Cannot allocate memory.

to this matter. Look forward to hearing from you. Best regards. Jan Kiszka 于2020年6月16日周二 下午5:39写道： > On 16.06.20 09:51, 孙世龙 via Xenomai wrote: > > Hi, > > As the subject, I encounter this error when I restarted a realtime > > process that uses *IDDP *on x86 embedded syste

Error reproduecd again when using IDDP: BUG in low_init(): [main] binding failed: Cannot allocate memory.

Hi, As the subject, I encounter this error when I restarted a realtime process that uses *IDDP *on x86 embedded system. I am using xenomai3.1.0 and linux4.19.84. There are two points need to be emphasized: 1. After this error occurred, I could restart the program successfully again.

Questions about XDDP.

As per the documentation ( *https://xenomai.org/documentation/xenomai-3/html/xeno3prm/ group__trank.html#details*),i can draw the conclusion that the non real-time side could block upon attempt to open the special device file until the

Could interuption be handled while process context is switching?

Could interuption be handled while process context is switching?

Re: Question about the complie skin.

gt; According to > >> > https://xenomai.org/documentation/xenomai-3/html/man1/xeno-config/index.html > >> > >> > --native and --skin=native are accepted for backward compatibility > purpose. They are stricly equivalent to passing --alchemy --compat. > &

Re: Question about the complie skin.

ot use --native in new xenomai3 programs, use --alchemy > instead. > > > > --compat > >Enable the Xenomai 2.x compatibility mode for the API/skin selected. > This switch affects the Alchemy and POSIX APIs, turning on a set of > source-level compatibility wrappers when

Question about the complie skin.

Hi, I am using xenomai-3.1. I complied source code with xenomai *native skin* for months. I did not notice any complie error indeed and the * final binary programs could run well*. But, i encoutered some problems this time while the source code had called *rt_alarm_create*(still

Question about the xenomai.

Hi, I am using xenomai-3.1. I complied source code with xenomai native skin for months. I did not notice any complie error indeed and the final binary programs could run well. But, i encoutered some problems this time while the source code had called rt_alarm_create(still complie the

Could xenomai real time thread share the rt_mutex\rt_sem with the routine which is used as input argument of rt_alarm_create？

Hi, Could the main thread which has been shadowed(or *xenomai real time thread* whose scheduling policy is *SCHED_FIFO*) *share the rt_mutex\rt_sem* *with the routine* which is used as input argument of *rt_alarm_create*？ Thank you for you attention to this matter. Look forward to hearing from

Could nomal linux threads and xenomai threads run together in the same process?

Hi, As per the documentation (https://research.idi.ntnu.no/multicore/_media/ linux_xenomai_hard_real-time_en_read-only_.pdf *page number:11*), it give me the sense that *nomal linux threads *which are not normal xenomai thread whose scheduling policy is SCHED_OTHER and *xenomai threads* could run

Which way could guarantee more precise delay?'rt_task_sleep' or 'rt_task_wait_period'?

Hi, I need to run a simpe routline periodically.For example, the routline needs to run every 1ms. Which way could guarantee more precise delay?'rt_task_sleep' or 'rt_task_wait_period'? If there is a better solution,please let me know. Thank you for attention to this

Question about rt_task_sleep and rt_alarm_start.

Hi， Assuming that, there is a real time process which has only a single thread.It creates an object triggering an alarm routine at a specified time in the future.Then it triggers the alarm and has been put to sleep for *a long long duration* through calling to the *rt_task_sleep*.

When would a real time process be able to resume running again after it have called rt_task_sleep?

Hi, I wonder that when a *real time process* would be able to resume running again after it have called rt_task_sleep. Assuming that many *real time processes* with higher priority are running in parallel，I think the process(which have called rt_task_sleep) have to wait for available cpu

BUG in low_init(): [main] binding failed: Cannot allocate memory.

Hi， I encounter a error when starts a realtime process on x86 embedded system . The point to be emphasized, all the realtime processes have worked well for more then one day.I have terminated the realtime processes, started them again now and then. When i met the error showed

Does the xenomai regular thread whose schedule policy is SCHED_OTHER has more opportunities to occupy the cpu resources then the normal linux thread which managed by linux?

Hi, I wonder that whether the xenomai regular thread whose schedule policy is *SCHED_OTHER* *has more opportunities to occupy the cpu resources* then the normal linux thread which managed by linux? Even if the linux thread with *SCHED_FIFO * schedule policy? Thank you for your

Re: When the linux process could recieve the message transport through XDDP？

s between the RT/NRT schedulers >involved in that scenario. What do you mean by " employ some non-standard logic"？Could you make me more clearlly? Thank you for your attention. Look forward to hearing from you. steve freyder 于2020年5月12日周二 上午11:43写道： > On 5/11/2020

When the linux process could recieve the message transport through XDDP？

Hi, As i am using XDDP to transport messages bettween the head domain and linux domain(i mean there are two processes, one is rt process, the other is linux process.),I wonder that when the linux process could recieve the message？ *Will ADOES wake up the related linux process at once

Re: Why xddp-stream demo process could not be terminate by ctrl + C.

Hi, Sorry for disturbing you. I found the answer allready. There are apis used to capture the ctrl+c signal indeed. 孙世龙 于2020年5月11日周一 下午3:51写道： > Hi, > I could run the xddp-stream demo process normally indeed. > But when i intend to terminate the process by ctrl+c, nothing

Why xddp-stream demo process could not be terminate by ctrl + C.

Hi, I could run the xddp-stream demo process normally indeed. But when i intend to terminate the process by ctrl+c, nothing has happened. The only way i found useful is to use the shell command(sudo killall -9 xddp-stream) to terminate it. I*'m really confused.* * As per

What are the relations bettween COBALT_SYSCALL(ioctl, handover, args...) and COBALT_IMPL(int, ioctl, args...)?d

Hi, What are the relations bettween* COBALT_SYSCALL(ioctl, handover, args...) *and* COBALT_IMPL(int, ioctl, args...)*? The related source code could be seen in the urls listed below. https://gitlab.denx.de/Xenomai/xenomai/-/blob/v3.0.9/lib/cobalt/rtdm.c （COBALT_SYSCALL( ioctl )）

Could anybody help to reslove the question about _RTIOC_LISTEN?

Hi, As per the source code( https://gitlab.denx.de/Xenomai/xenomai/-/blob/v3.0.9/kernel/drivers/net/stack/ipv4/tcp/tcp.c), I could see that some code have relation to _RTIOC_LISTEN.For details, see below. */*** * rt_tcp_ioctl */static int rt_tcp_ioctl(struct rtdm_fd *fd,

Re: Q: about mutex lock bettween rt thread and nrt thread.

Hi, Thank you for your response. >>If Head domain's interrupts & threads occupied CPU, the system may deadlock there since nothing can kill them. What do you mean? I could not get the idea.When this happen? And how?Does it have relation to XDDP? Thank you for your attention.

Q: about mutex lock bettween rt thread and nrt thread.

Hi, I think mutex lock should not be shared bettween rt thread and nrt thread, because if i use the posix mutex api implemented by linux,it causes the rt thread switch to the secondary mode. As per the documentation(https://xenomai.org/documentation/

Could std::atomic be shared bettween rt thread and nrt thread?

Hi, Could std::atomic be shared bettween rt thread and nrt thread? Is it cause latency problem? Is it cause rt thread to switch to secondary mode? Thank you for your attention. Look forward to hearing from you.

Must I implement the vm_operations while rtdm_mmap_to_user function is to be called?

Hi, Must I implement the vm_operations while rtdm_mmap_to_user function is to be called? Vm_operations is to be executed on the vm_area of the user memory range, as per the documentation,

What' the meaning of snapshot-driven Which is seen in the comments of the code of xenomai-v3.1.

Hi， What' the meaning of snapshot-driven Which is seen in the comments of the code of xenomai-v3.1. For your convenice, I list the related code below. Very few useful information could be aquired through the search engines. I have seen many function names which contain the

[no subject]

Hi， What' the meaning of snapshot-driven Which is seen in the comments of the code of xenomai-v3.1. For your convenice, I list the related code below. Very few useful information could be aquired through the search engines. I have seen many function names which contain the

I wonder what “apc” stands for?

Hi, I wonder what “apc” stands for? There are some fileds named start with "apc" in struct cobalt_pipeline. For instance, "apc_virq"、" apc_map"、" apc_table". For your convenience, I listed the struct mentioned in this email. struct cobalt_pipeline { struct

Could anybody tell me the function of the system heap and the shared heap whose capacity been aquired from /proc/xenmai/heap.

Hi， Could anybody tell me the functions of the system heap and the shared heap whose capacity can be aquired from /proc/xenmai/heap. I mean what the heaps meationed above are used for. Looking forward to hearing from you. Thanks a lot.

Re: linux4.14+xenomai3.1 real-time

Hi， >When we tested on the Intel(r) Celeron(r) processor 1047ue (third >generation) motherboard (using kernel version: 4.1.18, xenomai version: 2.6. >5, patch: ipipe-core-4.1.18-x86-3. ), The real-time test results meet the >requirements.But the 4.1.18 version of the kernel (xenomai version 2.6.5)

Re: Could give me some guideline on how to locate the source code of the xenomai system calls.

c_cobalt_sched_yield] = cobalt_sched_yield,*/ Best regards. Richard Weinberger 于2020年4月16日周四 下午8:48写道： > On Thu, Apr 16, 2020 at 2:45 PM 孙世龙 via Xenomai > wrote: > > > > Could give me some guideline on how to locate the source code of > > the xenomai system calls. &g

Could give me some guideline on how to locate the source code of the xenomai system calls.

Could give me some guideline on how to locate the source code of the xenomai system calls. I confirm that there is corresponding binary file after compiling the source code which include xenomai and linux.You can see the log listed below for detail. What puzzled me is that i could

Re: ERR MSG： called regular read() on /dev/rtdm/fake_dev

> > > > > Look forward to hearing from you. > > Thanks a lot. > > > > Jan Kiszka mailto:jan.kis...@siemens.com>> 于 > > 2020年4月16日周四 下午5:01写道： > > > > On 16.04.20 10:57, 孙世龙 via Xenomai wrote: > > > Hi,

Re: ERR MSG： called regular read() on /dev/rtdm/fake_dev

$(XENO_CONFIG) --skin=native --ldflags) Look forward to hearing from you. Thanks a lot. Jan Kiszka 于2020年4月16日周四 下午5:01写道： > On 16.04.20 10:57, 孙世龙 via Xenomai wrote: > > Hi, > > I got the error message listed below when I called read > operation to > > an opened

ERR MSG： called regular read() on /dev/rtdm/fake_dev

Hi, I got the error message listed below when I called read operation to an opened rtdm file handle in main function. I have implemented the rt read operation in my rtdm driver allready. I think the main thread is allready turned into xenomai thread because I have called

I wonder that wheather the the rtdm interface could be called in user space?

Hi, I wonder that whether the the rtdm interface could be called in user space?I think it's impossible to call rtdm interface in main function.Am i right? Look forward to hearing from you. Thanks a lot.

Must I provide both rt function and nrt function for the ops of rtdm driver ?

Hi, Must I provide both rt function and nrt function for the ops of rtdm driver ? Must I implement the rt read and nrt read operation if i need provide read operation for my rtdm driver?What I have to consider? I have searched all over the source code of xenomai-v3.1.I find

Re: I wonder why it calls pipeline_syscall function firstly when i call the open function to aquire the handle of a rtdm driver?

] ? SyS_write+0x79/0xc0 [103376.536505] [] pipeline_syscall+0x8/0x1b [103376.536540] ---[ end trace 41f9f284a6c92035 ]--- Jan Kiszka 于2020年4月15日周三 下午3:29写道： > On 15.04.20 04:45, 孙世龙 via Xenomai wrote: > > Hi， > > > > I am using xenomai-v3.0.9 with linux3.0.9. > > &g

I wonder why it calls pipeline_syscall function firstly when i call the open function to aquire the handle of a rtdm driver?

Hi， I am using xenomai-v3.0.9 with linux3.0.9. I wonder that why it calls pipeline_syscall function firstly when i call the open function to aquire the handle of a rtdm driver? The related call trace log for kernel is listed below. Some notes. I am quite conscious of the reason which causes

I wonder why it calls pipeline_syscall function firstly when i call the open function to aquire the handle of a rtdm driver?

Hi， I am using xenomai-v3.0.9 with linux3.0.9. I wonder that why it calls pipeline_syscall function firstly when i call the open function to aquire the handle of a rtdm driver? The related call trace log for kernel is listed below. Some notes. I am quite conscious of the reason which causes

What are the differences bettween rtdm_open function and open function used to get the file handle of the rtdm driver?

Hi， The rtdm_open function is described as inter-driver interface. What are the differences between rtdm_open function and open function used to get the file handle of the rtdm driver. I have searched all the source code of xenomai-3.1.I found no usage of this function.

I think there is no way to sync objects which are used in both rt rtdm ops and nrt rtdm ops.Am i right?

Hi, I think there is no way to sync objects which are used in both rt rtdm ops and nrt rtdm ops.Am i right? Best regards.

Re: How could I allocate a huge memory in rtdm driver?

Hi Meng, Thanks a lot for your reply. I will follow your advice not to throw out multi questions at once any more. I have some industry PCs in hand which allready installed xenoami. I use vmware to verify my code. It's more convenient to use word etc on windows while ubuntu

Re: How could I allocate a huge memory in rtdm driver?

contiguous memory indeed. Best regards. Richard Weinberger 于2020年4月13日周一 下午4:17写道： > On Mon, Apr 13, 2020 at 9:21 AM 孙世龙 via Xenomai > wrote: > > > > Hi, > > I am using xenomai 3.1. > > > > How could I allocate a huge memory which may be

How to sync objects which are used in rt ops and nrt ops which are contained in rtdm driver?

Hi, I am using xenomai-3.1. I wonder how to sync objects which are used in rt ops and nrt ops contained in rtdm driver? Best regards.

Re: Why X11 influences the latency of xenomai？

n order to make it visible to the GPU. > > Jan > > > Best regards. > > > > Jan Kiszka mailto:jan.kis...@siemens.com>> 于 > > 2020年4月8日周三 下午5:05写道： > > > > On 08.04.20 10:39, 孙世龙 via Xenomai wrote: > > > Hi, > > >

I think lib/copperplate depends on glibc while using dual kernel configuration.Am i right?

Hi, I am using xenomai-3.1. Two figures are contained in url https://gitlab.denx.de/Xenomai/xenomai/-/wikis/Start_Here. Since the architecture showed in figure 2 named xenomai 3 single kernel configuration, i could draw the conclusion that copperplate interface depends on the glibc.

I think lib/copperplate depends on glic while using dual kernel configuration.Am i right?

Hi, Two figures are contains in url https://gitlab.denx.de/Xenomai/xenomai/-/wikis/Start_Here. Since the architecture showed in Figure 2. Xenomai 3 single kernel configuration, i could draw the conclusion that copperplate interface depends on the glibc. But i don't think i can draw the

Re: Why X11 influences the latency of xenomai?

Hi, I am interested that how to disable all DRM/KMS drivers while the functions of X ares not inflenced? I am surprised that X can cause spikes without any GPU involvement.It's amazing! Thanks a lot.Looking for your reply. Best regards. >>From: Alec Ari >>To: Jan Kiszka ,

Re: Why X11 influences the latency of xenomai？

： > On 08.04.20 10:39, 孙世龙 via Xenomai wrote: > > Hi, > > Why X11 influences the latency of xenomai？I guess it has some > relation > > to GPU.Could you make me more clearly? > > I have solved the latency . I am using QT runing on linux > framebuffer > > ,d

What function does cobalt_thread_probe() provides.

Hi， I am using xenomai-v3.0.9. I wonder the function cobalt_thread_probe() provides. The source code is really simple which is listed below. I could not find any clue in the entrie project. I don't think the function is really to kill some process.

What function does cobalt_thread_probe() provides.

Hi， I am using xenomai-v3.0.9. I wonder the function cobalt_thread_probe() provides.The source code is really simple which is listed below.I could not find any clue in the entrie project. I don't think the function is really to kill some process. Best regards.

Why X11 influences the latency of xenomai？

Hi, Why X11 influences the latency of xenomai？I guess it has some relation to GPU.Could you make me more clearly? I have solved the latency . I am using QT runing on linux framebuffer ,disabled X11 at all.I really hope to why X11 influences the latency of xenomai Thanks a lot.Looking

Re: Is anybody can show me a simple rtdm driver to read/write virtual memory device?

Kiszka via Xenomai > a écrit : > >> On 06.04.20 09:22, 孙世龙 via Xenomai wrote: >> > Is anybody could show me a simple xenomai3 rtdm driver to read/write >> > virtual memory device? >> > A virtual memory device to: >> > write "hello world" to memory devic

Is anybody could tell me the relation between lib/boilerplate folder and lib/copperplate folder?

Is anybody could tell me the relation between lib/boilerplate and lib/copperplate? I am using xenomai-v3.1. Thanks a lot.Looking forward to your replay.

Does the function malloc() uses the real malloc or the wrap one in heapobj_pkg_init_private(）.

The source code url is https://gitlab.denx.de/Xenomai/xenomai/-/blob/v3.1/lib/copperplate/ heapobj-heapmem.c line:107 In the source code listed above, function heapobj_pkg_init_private() calls malloc(). I hope to know whether it calls the wrapper function or the real malloc function which

Re: question about the calculation of overload bytes in function add_extent()

> > Thanks a lot for your replay. > I am sorry for the bad experience in reading my last letter. > I rewrite the letter below. > > The question i met is when i enabled --enable-pshared option in compiling > the xenomai-v3.1. > > In the source code

How to use the rtdm udd?

How to use the rtdm udd? Is there some examples to show how to use it?Is there some material on how to use it? Thanks.Looking forward to your replay.

Is anybody can show me a simple rtdm driver to read/write virtual memory device?

Is anybody could show me a simple xenomai3 rtdm driver to read/write virtual memory device? A virtual memory device to: write "hello world" to memory device in real real time(It's appricate if the write operation could wake up the read task which allready waited to read) read the memory

question about the calculation of overload bytes in function add_extent()

Question about the calculation of overload memory in function add_extent() in heapmem.c line:607(the source is listed below) In the source code, notes that"overhead memory we need for storing our meta data as calculated by HEAPMEM_ARENA_SIZE()".(The source of HEAPMEM_ARENA_SIZE macro is also