Re: sched: problem with paging request on linux 3.16

Hello Paolo,

Using the calibration data from the running system doesn't help.
I have tried RTAI from CVS and the new released RTAI 5.0 without success.
Even when I use the same kernel settings you have posted a few days ago, the kernel ops with the second hardware still exists.

The hal patch I use is identically to the one that was in VULCANO CVS.
But the patch was removed. Do you know why?

Unfortunately I can't use a newer kernel: The Igh Ethercat Master has only patched diver for network interfaces up to kernel 3.16.

Josef


Von: Paolo Mantegazza [mailto:[email protected]]
Gesendet: Mittwoch, 3. Mai 2017 14:57
An: PAUL-SOFTWARE; [email protected]
Betreff: Re: [Rtai] sched: problem with paging request on linux 3.16

Have you configured your kernel for a UP setup?
The latency calibration failed. you can try copying the file RTAIINSTALLDIR/calibration/latencies that succeeded in the new installation and see what happens.
Moreover, I do not think that the patch you use is distributed with the test version you are using. Try with one in test2 tarball.
All in all we are about to release the final 5.0, which is what is in VULCANO CVS. TRy it too.
Paolo.


On 05/03/2017 01:53 PM, PAUL-SOFTWARE wrote:
Hello,

I installed rtai-5.0-test2 on Debian Jessie with 32Bit kernel 3.16.7, using the hal-patch from magma (hal-linux-3.16.7-x86-5x.patch<https://github.com/ShabbyX/RTAI/blob/master/base/arch/x86/patches/hal-linux-3.16.7-x86-5x.patch>).

Everything works fine on a board with Intel i3-4130 CPU.
On a system with Intel i3-6110 CPU I get an error when I insert rtai_sched.ko.

This is the kernel log when I start testsuite/kernel/latency:

I-pipe: head domain RTAI registered.
RTAI[hal]: mounted. ISOL_CPUS_MASK: 0.
SYSINFO - # CPUs: 1, TIMER NAME: 'lapic', TIMER IRQ: 2305, TIMER FREQ: 99758000, CLOCK NAME: 'tsc', CLOCK FREQ: 3192285000, CPU FREQ: 3192285000, LINUX TIMER IRQ: 2305.
RTAI[malloc]: global heap size = 2097152 bytes, <BSD>.
, kstacks pool size = 524288 bytes.
RTAI[sched]: hard timer type/freq = lapic/99758000(Hz); timing: oneshot; linear timed lists.
RTAI[sched]: Linux timer freq = 250 (Hz), TimeBase freq = 3192285000 hz.
RTAI[sched]: timer setup = 64 ns, resched latency = 0 ns.
RTAI[usi]: enabled.
BUG: unable to handle kernel paging request at bfebea1c
IP: [<f86e2845>] rtai_lxrt_invoke+0xae5/0x13c0 [rtai_sched]
*pde = 108de067 *pte = 10c15067
Oops: 0001 [#1] PREEMPT
Modules linked in: rtai_sched(O+) rtai_hal(O) i915 drm_kms_helper drm i2c_algo_bit cfg80211 rfkill nfsd auth_rpcgss oid_registry nfs_acl nfs lockd fscache sunrpc snd_hda_codec_hdmi coretemp snd_hda_codec_realtek snd_hda_codec_generic kvm snd_hda_intel snd_hda_controller snd_hda_codec crc32_pclmul snd_hwdep aesni_intel snd_pcsp aes_i586 xts snd_pcm snd_timer lrw gf128mul snd soundcore ablk_helper cryptd mxm_wmi battery evdev serio_raw tpm_tis tpm shpchp wmi video button fuse parport_pc ppdev lp parport autofs4 ext4 crc16 mbcache jbd2 hid_generic usbhid sg sd_mod crc_t10dif crct10dif_generic crct10dif_common crc32c_intel ahci libahci xhci_hcd r8169 8139cp mii libata scsi_mod usbcore usb_common fan thermal_sys i2c_hid hid i2c_core
CPU: 0 PID: 1346 Comm: calibrate Tainted: G           O  3.16-rtai #8
Hardware name: System manufacturer System Product Name/H110M-C, BIOS 3402 04/06/2017
task: d08dc010 ti: d08f8000 task.ti: d08f8000
EIP: 0060:[<f86e2845>] EFLAGS: 00010206 CPU: 0
EIP is at rtai_lxrt_invoke+0xae5/0x13c0 [rtai_sched]
EAX: 0040780c EBX: 00000000 ECX: 00000407 EDX: bfebea1c
ESI: bfebea1c EDI: 0040780c EBP: d08f9f84 ESP: d08f9f34
DS: 007b ES: 007b FS: 0000 GS: 00e0 SS: 0068
CR0: 8005003b CR2: bfebea1c CR3: 108dd000 CR4: 00340790
I-pipe domain Linux
Stack:
d08f9f58 c10863ee c15e4180 c1087745 d08f9f98 c15e4240 d08f9f98 00000001
fffffffe d08f9f64 c108789c 00000000 d08f9fac c1145b90 d08f9f98 ed94bc40
e19b3ac0 d08f9fb4 bfebe978 b74f6848 d08f9fa4 f86e09dd 00000000 bfebff48
Call Trace:
[<c10863ee>] ? rcu_is_watching+0x1e/0x40
[<c1087745>] ? __call_rcu.constprop.60+0xa5/0x1e0
[<c108789c>] ? call_rcu+0x1c/0x20
[<c1145b90>] ? SyS_access+0x190/0x1f0
[<f86e09dd>] ? lxrt_intercept_syscall+0x8d/0x190 [rtai_sched]
[<c10bdcd4>] ? __ipipe_notify_syscall+0x14/0x30
[<c1468115>] ? pipeline_sysenter+0x7/0x1e
Code: 10 8b 56 0c 89 54 24 0c 8b 56 08 89 54 24 08 8b 56 04 89 54 24 04 8b 16 89 14 24 ff 50 04 89 44 24 2c 89 54 24 30 e9 37 f6 ff ff <8b> 1a 8d 81 18 fc ff ff 83 f8 20 89 da 0f 87 97 05 00 00 ff 24
EIP: [<f86e2845>] rtai_lxrt_invoke+0xae5/0x13c0 [rtai_sched] SS:ESP 0068:d08f9f34
CR2: 00000000bfebea1c
---[ end trace 426b8ee3f3f6e396 ]---
USERMODE CHECK: ERROR.
USERMODE CHECK PROVIDED (ns): KernelLatency 0, UserLatency 0.
FINAL CALIBRATION SUMMARY (ns): KernelLatency 1, UserLatency 1.


Has anybody an idea what causes this ops and how I can avoid it?

Best regards,
Josef





_______________________________________________

Rtai mailing list

[email protected]<mailto:[email protected]>

https://mail.rtai.org/cgi-bin/mailman/listinfo/rtai

_______________________________________________
Rtai mailing list
[email protected]
https://mail.rtai.org/cgi-bin/mailman/listinfo/rtai