Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 12 Aug 2018 01:35:21 +0800
From:      Li-Wen Hsu <lwhsu@freebsd.org>
To:        asomers@freebsd.org
Cc:        mmacy@freebsd.org, brd@freebsd.org, freebsd-current@freebsd.org,  marklmi@yahoo.com
Subject:   Re: ci.freebsd.org 's FreeBSD-head-{amd64, i386}-test started failing after -r337332 (last good), inp_gcmoptions involved
Message-ID:  <CAKBkRUxM%2BbrPNSbUF5LjLgMo9JhMCQu%2BBzYxcsuP8fOXFrj6aA@mail.gmail.com>
In-Reply-To: <CAOtMX2imOhs-UOpe93js5THPYcf_=o6MfM_CFi8hVs3OmMdmqQ@mail.gmail.com>
References:  <5D25CBEE-E432-42B3-B9E9-AE344785156F@yahoo.com> <CAKBkRUxX3_-JxDz3pCkRuN9RchsYQRb8Ws0DuAKxNa2XgSGG-Q@mail.gmail.com> <CAPrugNpamceDrqRYJVDnRo5m-n%2BjiEj3Wm8KXQ=1TeK%2Bg%2BdbfQ@mail.gmail.com> <CAOtMX2imOhs-UOpe93js5THPYcf_=o6MfM_CFi8hVs3OmMdmqQ@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
With the VMs images on artifact.ci.freebsd.org, I can reproduce this with:

root@:/usr/tests/sys/netinet # kyua debug fibs_test:slaac_on_nondefault_fib6
fib is 1
fib is 2
net.inet6.ip6.forwarding: 0 -> 1
net.inet6.ip6.rfc6204w3: 0 -> 1
/sbin/pfctl
/sbin/ipf
ipf: IP Filter: v5.1.2 (608)
setfib 1 ifconfig epair0a inet6 2001:db8:115e:fc32::2/64 fib 1
setfib 2 ifconfig epair0b inet6 -ifdisabled accept_rtadv fib 2 up
Executing command [ ifconfig epair0b ]
Executing co

Fatal trap 9: general protection fault while in kernel mode
cpuid = 0; apic id = 00
instruction pointer     = 0x20:0xffffffff80ded513
stack pointer           = 0x28:0xfffffe0012158860
frame pointer           = 0x28:0xfffffe00121588a0
code segment            = base 0x0, limit 0xfffff, type 0x1b
                        = DPL 0, pres 1, long 1, def32 0, gran 1
processor eflags        = interrupt enabled, resume, IOPL = 0
current process         = 0 (softirq_0)
[ thread pid 0 tid 100013 ]
Stopped at      inp_gcmoptions+0xe3:    movq    ll+0x33f(%rax),%r9
db> bt
Tracing pid 0 tid 100013 td 0xfffff800031de000
inp_gcmoptions() at inp_gcmoptions+0xe3/frame 0xfffffe00121588a0
epoch_call_task() at epoch_call_task+0x21a/frame 0xfffffe00121588f0
gtaskqueue_run_locked() at gtaskqueue_run_locked+0x139/frame 0xfffffe0012158940
gtaskqueue_thread_loop() at gtaskqueue_thread_loop+0x88/frame 0xfffffe0012158970
fork_exit() at fork_exit+0x84/frame 0xfffffe00121589b0
fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe00121589b0
--- trap 0, rip = 0, rsp = 0, rbp = 0 ---
db>


Li-Wen

On Mon, Aug 6, 2018 at 9:53 AM Alan Somers <asomers@freebsd.org> wrote:
>
> I can't reproduce the failure.  On my VM, with a kernel from Aug-2, the test passes.  But it sure seems to be consistent in Jenkins.
>
> On Sun, Aug 5, 2018 at 6:59 PM, Matthew Macy <mmacy@freebsd.org> wrote:
>>
>> That looks like it is tied to changes I made 3 months ago. I won't be at my desk until the end of the week, but if it's consistent I can take a look.
>>
>> -M
>>
>> On Sun, Aug 5, 2018 at 17:57 Li-Wen Hsu <lwhsu@freebsd.org> wrote:
>>>
>>> On Sun, Aug 5, 2018 at 6:23 PM Mark Millard <marklmi@yahoo.com> wrote:
>>> >
>>> > amd64: #8493 was for -r337342 and #8492 (last good) was for -r337332 .
>>> > more recent builds also failed. -r337342 and laster also failed for
>>> > i386.
>>> >
>>> > All but a sys/gettimeofday.2 change after -r337332 through -r337342
>>> > are from Brad Davis. It is unclear to me how the changes matches up
>>> > with the below example (from the log for amd64). It might not?
>>> >
>>> > For example (i386 is similar):
>>> >
>>> > https://ci.freebsd.org/job/FreeBSD-head-amd64-test/8493/consoleText
>>> >
>>> > sys/netinet/fibs_test:subnet_route_with_multiple_fibs_on_same_subnet  ->
>>> >
>>> > Fatal trap 9: general protection fault while in kernel mode
>>> > cpuid = 0; apic id = 00
>>> > instruction pointer     = 0x20:0xffffffff80ded213
>>> > stack pointer           = 0x28:0xfffffe002648c960
>>> > frame pointer           = 0x28:0xfffffe002648c9a0
>>> > code segment            = base 0x0, limit 0xfffff, type 0x1b
>>> >                         = DPL 0, pres 1, long 1, def32 0, gran 1
>>> > processor eflags        = interrupt enabled, resume, IOPL = 0
>>> > current process         = 0 (softirq_0)
>>> > [ thread pid 0 tid 100013 ]
>>> > Stopped at      inp_gcmoptions+0xe3:    movq    ll+0x33f(%rax),%r9
>>>
>>> I think this is because we are trying to enable more tests:
>>> https://github.com/freebsd/freebsd-ci/pull/25
>>>
>>> I'm looking into that.  If I cannot resolve this quickly I will revert
>>> it temporarily.
>>>
>>> Li-Wen
>>>
>>> --
>>> Li-Wen Hsu <lwhsu@FreeBSD.org>
>>> https://lwhsu.org
>>> _______________________________________________
>>> freebsd-current@freebsd.org mailing list
>>> https://lists.freebsd.org/mailman/listinfo/freebsd-current
>>> To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"
>
>


--
Li-Wen Hsu <lwhsu@FreeBSD.org>
https://lwhsu.org



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAKBkRUxM%2BbrPNSbUF5LjLgMo9JhMCQu%2BBzYxcsuP8fOXFrj6aA>