[Rtai] Oops in testsuite with RTAI 3.7.1 on Geode LX800

[Paolo Mantegazza]

Try: disabling HPET, play with PM/ACPI off, no kernel dubugging/tracing.
Hopying somebody using Geode can say more.

paolo


Marco Pantaleoni wrote:
> Hi,
> 
> I'm using RTAI 3.7.1 with kernel 2.6.29.5 on a Geode LX800. I keep getting
> (multiple) kernel oops whenever I try to run the userspace latency test from
> the testsuite. Every other test in the testsuite seems to run fine, both
> kernel-space and user-space.
> The oops are like the following:
> 
> [  964.722776] I-pipe: Domain RTAI registered.
> [  964.728131] RTAI[hal]: <3.7.1> mounted over IPIPE-NOTHREADS 2.4-03.
> [  964.732106] RTAI[hal]: compiled with gcc version 4.3.2 (Debian 4.3.2-1.1)
> .
> [  964.739271] RTAI[hal]: mounted (IPIPE-NOTHREADS, IMMEDIATE (INTERNAL IRQs
> VECTORED), ISOL_CPUS_MASK: 0).
> [  964.750497] PIPELINE layers:
> [  964.756032] e08eb400 9ac15d93 RTAI 200
> [  964.761468] c03b57a0 0 Linux 100
> [  964.994507] RTAI[malloc]: global heap size = 2097152 bytes, <BSD>.
> [  965.003111] RTAI[sched]: loaded (IMMEDIATE, UP, USER/KERNEL SPACE:
> <without RTAI OWN KTASKs>, <uses LINUX SYSCALLs>, kstacks pool size = 524288
> bytes.
> [  965.014400] RTAI[sched]: hard timer type/freq = 8254-PIT/1193180(Hz);
> default timing: periodic; linear timed lists.
> [  965.025627] RTAI[sched]: Linux timer freq = 250 (Hz), TimeBase freq =
> 498100000 hz.
> [  965.037250] RTAI[sched]: timer setup = 2009 ns, resched latency = 2690
> ns.
> [  965.261722] BUG: Unhandled exception over domain RTAI at 0xc01018ea -
> switching to ROOT
> [  965.263318] Pid: 17257, comm: cut Not tainted 2.6.29.5-geodelx-rtai-tde
> #1
> [  965.263318] Call Trace:
> [  965.263318]  [<c010d0ee>] __ipipe_handle_exception+0x114/0x13f
> [  965.263318]  [<c01018ea>] __switch_to+0x36/0x133
> [  965.263318]  [<c029948b>] error_code+0x6f/0x7c
> [  965.263318]  [<c01018ea>] __switch_to+0x36/0x133
> [  965.263318]  [<c01126ac>] context_switch+0x233/0x289
> [  965.263318]  [<e0909fa5>] rt_schedule+0x549/0x8b4 [rtai_lxrt]
> [  965.263318]  [<e090eb10>] rt_task_wait_period+0x100/0x11d [rtai_lxrt]
> [  965.263318]  [<e090ba2f>] rtai_lxrt_invoke+0x499/0x197a [rtai_lxrt]
> [  965.263318]  [<e090ea10>] rt_task_wait_period+0x0/0x11d [rtai_lxrt]
> [  965.263318]  [<c02981fa>] schedule+0x119/0x14c
> [  965.263318]  [<c02982dd>] schedule_timeout+0x14/0xb1
> [  965.263318]  [<c0111c0d>] activate_task+0x23/0x30
> [  965.263318]  [<c01128de>] try_to_wake_up+0xcc/0xd5
> [  965.263318]  [<c029806b>] wait_for_common+0xe4/0xf5
> [  965.263318]  [<e08e6eb1>] intercept_syscall_prologue+0x28/0xcd [rtai_hal]
> [  965.263318]  [<c0135a83>] __ipipe_dispatch_event+0x28/0x175
> [  965.263318]  [<c012242d>] flush_work+0x53/0x62
> [  965.263318]  [<c0147d25>] do_mlockall+0x6e/0x81
> [  965.263318]  [<c010ce67>] __ipipe_syscall_root+0x3c/0x8f
> [  965.263318]  [<c0102b5f>] system_call+0x2b/0x4c
> [  965.411736] spurious 8259A interrupt: IRQ7.
> [  965.425552] sed[17260]: segfault at b7e68004 ip b7e95551 sp bff58878
> error 7 in libc-2.7.so[b7e68000+138000]
> [  965.440220] BUG: Unhandled exception over domain RTAI at 0xc01018ea -
> switching to ROOT
> [  965.443747] Pid: 17258, comm: rtai-load Not tainted
> 2.6.29.5-geodelx-rtai-tde #1
> [  965.443747] Call Trace:
> [  965.443747]  [<c010d0ee>] __ipipe_handle_exception+0x114/0x13f
> [  965.443747]  [<c01018ea>] __switch_to+0x36/0x133
> ...
> 
> and it repeats many times. From this point on also regular linux commands
> cause similar oops until I rmmod all rtai modules.
> As I said, this doesn't happen with the other tests.
> I've tried also with 2.6.30.5 and the problem persists.
> Do you have any idea why this is happening? May it be related to the 8259A
> spurious interrupt?
> 
> I attach the kernel config, rtai config and kernel log for reference.
> Btw, I've compiled things almost following this guide:
> 
> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Debian_Lenny_Compile_RTAI
> 
> (apart having changed the configure line in debian/rules to fix the mismatch
> from the libraries compilation and the subsequent modules compilation).
> 
> Thanks in advance for your help!
> 
> Ciao,
> Marco
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Rtai mailing list
> Rtai at rtai.org
> https://mail.rtai.org/cgi-bin/mailman/listinfo/rtai