[Xenomai-help] xenomai installation issue.

[Xenomai-help] xenomai installation issue.

[garryt]

[-- Attachment #1: Type: text/plain, Size: 849 bytes --]


Hello,

I am facing issue with the installation of xenomai on x86_64 system. Obviously i
should have configured something not as required, but i cannot figure what.
This is a widespread optiplex DELL PC.(I hope someone successfully installed
xenomai on this and might help :)

Compilation works fine, then after few minutes system hangs (for example when
using xeno-test..).

I have search around the net, look at the TROUBLESHOOTING guide, modify kernel
compilation options to reflect as close as possible to the hw ( Core2 Duo
processor) and try different configuration (2.6.24 / 2.6.25 ) and IPIPE patches,
but still have issues.

Here is attached an archive with latest traces and configuration options.

By the way I had no problem with previous installation on old UP P4 PC...

Help will be appreciated.

Best regards,

[-- Attachment #2: xeno.tar.gz --]
[-- Type: application/gzip, Size: 23140 bytes --]

Re: [Xenomai-help] xenomai installation issue.

[Gilles Chanteperdrix]

garryt wrote:
> Hello,
> 
> I am facing issue with the installation of xenomai on x86_64 system. Obviously i
> should have configured something not as required, but i cannot figure what.
> This is a widespread optiplex DELL PC.(I hope someone successfully installed
> xenomai on this and might help :)
> 
> Compilation works fine, then after few minutes system hangs (for example when
> using xeno-test..).
> 
> I have search around the net, look at the TROUBLESHOOTING guide, modify kernel
> compilation options to reflect as close as possible to the hw ( Core2 Duo
> processor) and try different configuration (2.6.24 / 2.6.25 ) and IPIPE patches,
> but still have issues.
> 
> Here is attached an archive with latest traces and configuration options.

The most important thing is missing from this archive: the kernel logs.
If your processor has no serial port, please use netconsole (see
Documentation/networking/netconsole.txt in the kernel sources).

-- 
                                                 Gilles.

Re: [Xenomai-help] xenomai installation issue.

[Gilles Chanteperdrix]

garryt wrote:
> Hello,
> 
> I am facing issue with the installation of xenomai on x86_64 system. Obviously i
> should have configured something not as required, but i cannot figure what.
> This is a widespread optiplex DELL PC.(I hope someone successfully installed
> xenomai on this and might help :)
> 
> Compilation works fine, then after few minutes system hangs (for example when
> using xeno-test..).
> 
> I have search around the net, look at the TROUBLESHOOTING guide, modify kernel
> compilation options to reflect as close as possible to the hw ( Core2 Duo
> processor) and try different configuration (2.6.24 / 2.6.25 ) and IPIPE patches,
> but still have issues.
> 
> Here is attached an archive with latest traces and configuration options.

Could you try enabling CONFIG_ACPI (the only dangerous ACPI option is
CONFIG_ACPI_PROCESSOR), could you also try to disable CONFIG_AUDITSYSCALL ?

-- 
                                                 Gilles.

Re: [Xenomai-help] xenomai installation issue.

[garryt]

[-- Attachment #1: Type: text/plain, Size: 1296 bytes --]

Oups,

It seems i only reply to you and not to the list.
For information:
Here is the log (if someone has the same problem).

These new options didn't solve the issue.



Quoting Gilles Chanteperdrix <gilles.chanteperdrix@xenomai.org>:

> garryt wrote:
> > Hello,
> >
> > I am facing issue with the installation of xenomai on x86_64 system.
> Obviously i
> > should have configured something not as required, but i cannot figure what.
> > This is a widespread optiplex DELL PC.(I hope someone successfully
> installed
> > xenomai on this and might help :)
> >
> > Compilation works fine, then after few minutes system hangs (for example
> when
> > using xeno-test..).
> >
> > I have search around the net, look at the TROUBLESHOOTING guide, modify
> kernel
> > compilation options to reflect as close as possible to the hw ( Core2 Duo
> > processor) and try different configuration (2.6.24 / 2.6.25 ) and IPIPE
> patches,
> > but still have issues.
> >
> > Here is attached an archive with latest traces and configuration options.
>
> Could you try enabling CONFIG_ACPI (the only dangerous ACPI option is
> CONFIG_ACPI_PROCESSOR), could you also try to disable CONFIG_AUDITSYSCALL ?
>
> --
>                                                  Gilles.
>



[-- Attachment #2: logs --]
[-- Type: application/octet-stream, Size: 4921 bytes --]

mtrr: type mismatch for d0000000,10000000 old: write-back new: write-combining
Xenomai: POSIX: destroyed thread ffff81007cb80630
Xenomai: POSIX: destroyed thread ffff81007cb80630
Xenomai: POSIX: destroyed thread ffff81007cb80630
Xenomai: POSIX: destroyed thread ffff81007cb80630
Clocksource tsc unstable (delta = 739968683 ns)
BUG: unable to handle kernel NULL pointer dereference at 0000000000000008
IP: [<ffffffff8027b1d6>] __xntimer_init+0x280/0x3a2
PGD 7cc63067 PUD 714b1067 PMD 0 
Oops: 0002 [1] SMP 
CPU 0 
Modules linked in:<3>Xenomai: stuck on nucleus lock ffffffff80530a60
       waiter = kernel/xenomai/nucleus/intr.c:105 (xnintr_clock_handler(), CPU #1)
       owner  = kernel/xenomai/nucleus/timer.c:791 (__xntimer_init(), CPU #0)
 ffff81007e09feb8<3>I-pipe: Detected illicit call from domain 'Xenomai'
        into a service reserved for domain 'Linux' and below.
Pid: 4371, comm: dd Not tainted 2.6.25.19 #5

Call Trace:
 <IRQ>  [<ffffffff8026550f>] ipipe_check_context+0xd5/0xdd
 [<ffffffff8022b0dc>] cpu_clock+0x27/0xa0
 [<ffffffff80261106>] get_timestamp+0x9/0xf
 [<ffffffff8020f3ee>] do_softirq+0x5e/0xac
 [<ffffffff80261130>] touch_softlockup_watchdog+0x24/0x2d
 [<ffffffff8020d54f>] _show_stack+0xc8/0xf9
 [<ffffffff8026cdb4>] xnintr_clock_handler+0xee/0x33e
 [<ffffffff80266981>] __ipipe_dispatch_wired+0x11e/0x160
 [<ffffffff8021b986>] __ipipe_handle_irq+0x95/0x1c6
 [<ffffffff8020c311>] common_interrupt+0x61/0x7d
 <EOI>  [<ffffffff8021bc9c>] __ipipe_syscall_root+0x1c/0x156
 [<ffffffff802ff8bd>] vfs_write+0x121/0x156
 [<ffffffff8045b523>] __ipipe_syscall_root_thunk+0x35/0x6a
 [<ffffffff8020bcf4>] system_call_after_swapgs+0x54/0x94

 ffffffff80b83f60 0000000035561e5e 0000000000000162
 ffffffff80b84508 ffffffff80b51288 0000000000000000 ffffffff8026cdb4
 ffffffff804db2af ffff810000000317 ffffffff80478c10 ffffffff00000000
Call Trace:
 <IRQ>  [<ffffffff8026cdb4>] xnintr_clock_handler+0xee/0x33e
 [<ffffffff80266981>] __ipipe_dispatch_wired+0x11e/0x160
 [<ffffffff8021b986>] __ipipe_handle_irq+0x95/0x1c6
 [<ffffffff8020c311>] common_interrupt+0x61/0x7d
 <EOI>  [<ffffffff8021bc9c>] __ipipe_syscall_root+0x1c/0x156
 [<ffffffff802ff8bd>] vfs_write+0x121/0x156
 [<ffffffff8045b523>] __ipipe_syscall_root_thunk+0x35/0x6a
 [<ffffffff8020bcf4>] system_call_after_swapgs+0x54/0x94

 xeno_timerbench i915 drm ipv6 dm_snapshot dm_mirror dm_mod loop snd_hda_intel snd_pcm_oss snd_pcm snd_mixer_oss snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_timer snd_seq_device sg serio_raw ide_pci_generic snd i2c_i801 pcspkr soundcore snd_page_alloc psmouse i2c_core e1000e ide_core sr_mod intel_agp cdrom dcdbas evdev ext3 jbd mbcache sd_mod usbhid hid ata_piix 3c59x mii ata_generic libata ehci_hcd uhci_hcd scsi_mod [last unloaded: xeno_switchtest]
Pid: 4539, comm: display-4538 Not tainted 2.6.25.19 #5
RIP: 0010:[<ffffffff8027b1d6>]  [<ffffffff8027b1d6>] __xntimer_init+0x280/0x3a2
RSP: 0018:ffff81007147dca8  EFLAGS: 00050002
RAX: ffff81007cb80a08 RBX: ffff81007cb80988 RCX: ffff8100714b1028
RDX: 0000000000000000 RSI: ffffffff80530b10 RDI: 0000000000000000
RBP: ffffffff80530ac0 R08: 00000000351f9578 R09: 0000000000001000
R10: ffff81007cb80e30 R11: ffff81007cb80c30 R12: 0000000000000000
R13: ffffffff80530ac0 R14: 0000000000300000 R15: 0000000000001000
FS:  0000000041dfe960(0063) GS:ffffffff80562000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 0000000000000008 CR3: 00000000714bb000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process display-4538 (pid: 4539, threadinfo ffff81007147c000, task ffff810071436000)
Stack:  ffff81007147dcb8 0000000200000002 0000000000000064 ffff81007147dcc0
 ffff81007147dcc0 ffff81007cb80b60 ffff81007cb80650 ffffffff8027a274
 ffff81007147dde0 0000000000300000 ffff81007cb80650 0000000000300000
Call Trace:
 [<ffffffff8027a274>] ? xnthread_init+0x68/0x2dc
 [<ffffffff80275a25>] ? xnpod_init_thread+0x4b/0x340
 [<ffffffff8028f46e>] ? rt_task_create+0xc4/0x48d
 [<ffffffff8026a429>] ? xnheap_alloc+0x552/0x568
 [<ffffffff8037ef6f>] ? __up_read+0x13/0x8a
 [<ffffffff8029802f>] ? __rt_task_create+0x136/0x1d6
 [<ffffffff8027fed3>] ? losyscall_event+0xb2/0x189
 [<ffffffff80266751>] ? __ipipe_dispatch_event+0xcd/0x1df
 [<ffffffff8021bd17>] ? __ipipe_syscall_root+0x97/0x156
 [<ffffffff8045b523>] ? __ipipe_syscall_root_thunk+0x35/0x6a
 [<ffffffff8020bcf4>] ? system_call_after_swapgs+0x54/0x94


Code: 00 00 89 05 c9 58 2b 00 eb 04 48 83 cf 02 48 8d 75 50 48 8d 83 80 00 00 00 48 8b 4e 08 48 8b 11 48 89 48 08 48 89 93 80 00 00 00 <48> 89 42 08 48 89 01 ff 46 10 ff 45 68 40 f6 c7 02 0f 85 d8 00 
RIP  [<ffffffff8027b1d6>] __xntimer_init+0x280/0x3a2
 RSP <ffff81007147dca8>
CR2: 0000000000000008
---[ end trace 9a45efdc794cc264 ]---

Re: [Xenomai-help] xenomai installation issue.

[Gilles Chanteperdrix]

garryt wrote:
> Oups,
> 
> It seems i only reply to you and not to the list.
> For information:
> Here is the log (if someone has the same problem).
> 
> These new options didn't solve the issue.

Call Trace:
 [<ffffffff8027a274>] ? xnthread_init+0x68/0x2dc
 [<ffffffff80275a25>] ? xnpod_init_thread+0x4b/0x340
 [<ffffffff8028f46e>] ? rt_task_create+0xc4/0x48d
 [<ffffffff8026a429>] ? xnheap_alloc+0x552/0x568
 [<ffffffff8037ef6f>] ? __up_read+0x13/0x8a
 [<ffffffff8029802f>] ? __rt_task_create+0x136/0x1d6
 [<ffffffff8027fed3>] ? losyscall_event+0xb2/0x189
 [<ffffffff80266751>] ? __ipipe_dispatch_event+0xcd/0x1df
 [<ffffffff8021bd17>] ? __ipipe_syscall_root+0x97/0x156
 [<ffffffff8045b523>] ? __ipipe_syscall_root_thunk+0x35/0x6a
 [<ffffffff8020bcf4>] ? system_call_after_swapgs+0x54/0x94


Code: 00 00 89 05 c9 58 2b 00 eb 04 48 83 cf 02 48 8d 75 50 48 8d 83 80
00 00 00 48 8b 4e 08 48 8b 11 48 89 48 08 48 89 93 80 00 00 00 <48> 89
42 08 48 89 01 ff 46 10 ff 45 68 40 f6 c7 02 0f 85 d8 00
RIP  [<ffffffff8027b1d6>] __xntimer_init+0x280/0x3a2
 RSP <ffff81007147dca8>
CR2: 0000000000000008
---[ end trace 9a45efdc794cc264 ]---


This is a known issue which has been fixed in trunk and v2.4.x branch.
It will be in the upcoming v2.4.6 release.


-- 
                                                 Gilles.