Invalid Opcode

Invalid Opcode

[John Ho]

Hi all,
I have encountered an invalid Opcode while running some of my stuff,
and I have no idea where this is coming from, and I would like to ask
for some help.

I have assumed that adding Nopti would help eliminate this invalid
opcode, but it has not helped.

[  100.456173] kernel BUG at kernel/xenomai/synch.c:968!
[  100.456183] invalid opcode: 0000 [#1] PREEMPT SMP NOPTI
[  100.456191] CPU: 0 PID: 2255 Comm: OrbRunner Tainted: G     U
5.4.198-neuron-1.6 #1
[  100.456194] Hardware name: _ _/, BIOS N3FCPDU1 03/22/2023
[  100.456197] I-pipe domain: Linux
[  100.456206] RIP: 0010:xnsynch_requeue_sleeper+0x16/0x1d0
[  100.456210] Code: f6 4c 89 cf 48 89 e5 e8 98 fe ff ff 5d c3 c3 0f
1f 44 00 00 e8 9b c9 a1 00 55 48 8b 97 40 12 00 00 48 89 e5 f6 42 20
01 75 02 <0f> 0b 48 8b 87 18 12 00 00 48 8d 72 28 48 8b 8f 10 12 00 00
4c 8d
[  100.456214] RSP: 0018:ffffb1d8414ebd28 EFLAGS: 00010246
[  100.456218] RAX: 0000000000000000 RBX: ffffb1d840e96240 RCX: 0000000000000002
[  100.456221] RDX: ffffffff84f48ce0 RSI: ffffb1d8414ebd0c RDI: ffffb1d840e96240
[  100.456223] RBP: ffffb1d8414ebd28 R08: ffffb1d840e91a10 R09: ffffb1d840e91a30
[  100.456226] R10: 0000000080000092 R11: 0000000080000092 R12: ffffb1d840e91a08
[  100.456229] R13: ffffb1d840e8fa40 R14: ffffb1d843880c80 R15: 00000000000000c5
[  100.456232] FS:  00007f1b57e4ab00(0000) GS:ffff95360c000000(0000)
knlGS:0000000000000000
[  100.456235] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  100.456238] CR2: 00007f94fc6b6d68 CR3: 00000003de2be006 CR4: 00000000003606b0
[  100.456241] Call Trace:
[  100.456248]  inherit_thread_priority+0x2d/0x40
[  100.456254]  xnsynch_acquire+0x5b7/0x810
[  100.456260]  ? __switch_to_asm+0x40/0x70
[  100.456267]  __cobalt_mutex_acquire_unchecked+0x55/0x80
[  100.456271]  __cobalt_mutex_timedlock_break+0x2cb/0x390
[  100.456277]  ? CoBaLt_mutex_trylock+0x200/0x200
[  100.456281]  CoBaLt_mutex_lock+0x12/0x20
[  100.456286]  handle_root_syscall+0x15b/0x2e0
[  100.456291]  ipipe_syscall_hook+0x38/0x50
[  100.456296]  __ipipe_notify_syscall+0xb4/0x190
[  100.456301]  ipipe_handle_syscall+0x7f/0xb0
[  100.456306]  do_syscall_64+0x2c/0xe0
[  100.456312]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
[  100.456315] RIP: 0033:0x7f1b651cc7c7
[  100.456319] Code: 43 08 75 77 48 8b 15 98 67 21 00 48 03 02 8b 00
25 ff ff ff 5f 39 e8 74 9a be 1d 00 00 10 0f 1f 44 00 00 48 89 df 89
f0 0f 05 <83> f8 fc 89 c2 74 f2 85 c0 75 07 c7 43 04 01 00 00 00 48 83
c4 18
[  100.456322] RSP: 002b:00007f1b57e49f60 EFLAGS: 00000293 ORIG_RAX:
000000001000001d
[  100.456326] RAX: ffffffffffffffda RBX: 00007f1b67690bd8 RCX: 00007f1b651cc7c7
[  100.456329] RDX: 00007f1b653e41b0 RSI: 000000001000001d RDI: 00007f1b67690bd8
[  100.456331] RBP: 00000000000000c5 R08: 0000000000000009 R09: 0000556930a82a90
[  100.456334] R10: 0000000000000000 R11: 0000000000000293 R12: 0000000000000000
[  100.456337] R13: 00007f1b653e420c R14: 00007f1b67690bd8 R15: 00007f1b57e4a240
[  100.456343] Modules linked in: nls_iso8859_1 snd_hda_codec_hdmi
snd_hda_codec_realtek snd_hda_codec_generic ledtrig_audio joydev
input_leds snd_hda_intel snd_intel_nhlt snd_hda_codec snd_hda_core
snd_hwdep snd_pcm hid_generic mei_hdcp snd_seq_midi snd_seq_midi_event
crct10dif_pclmul i915 snd_rawmidi binfmt_misc crc32_pclmul
ghash_clmulni_intel snd_seq aesni_intel crypto_simd cdc_ether cryptd
usbnet drm_kms_helper glue_helper snd_seq_device r8152 snd_timer mii
drm usbhid i2c_algo_bit wmi_bmof serio_raw snd hid fb_sys_fops
syscopyarea soundcore mei_me sysfillrect mei sysimgblt acpi_tad
sch_fq_codel parport_pc ppdev lp parport ip_tables x_tables autofs4
ahci psmouse e1000e igc libahci wmi pinctrl_cannonlake pinctrl_intel
video
[  100.456390] ---[ end trace 7997569d11a0d891 ]---
[  100.456396] RIP: 0010:xnsynch_requeue_sleeper+0x16/0x1d0
[  100.456399] Code: f6 4c 89 cf 48 89 e5 e8 98 fe ff ff 5d c3 c3 0f
1f 44 00 00 e8 9b c9 a1 00 55 48 8b 97 40 12 00 00 48 89 e5 f6 42 20
01 75 02 <0f> 0b 48 8b 87 18 12 00 00 48 8d 72 28 48 8b 8f 10 12 00 00
4c 8d
[  100.456402] RSP: 0018:ffffb1d8414ebd28 EFLAGS: 00010246
[  100.456406] RAX: 0000000000000000 RBX: ffffb1d840e96240 RCX: 0000000000000002
[  100.456408] RDX: ffffffff84f48ce0 RSI: ffffb1d8414ebd0c RDI: ffffb1d840e96240
[  100.456411] RBP: ffffb1d8414ebd28 R08: ffffb1d840e91a10 R09: ffffb1d840e91a30
[  100.456414] R10: 0000000080000092 R11: 0000000080000092 R12: ffffb1d840e91a08
[  100.456417] R13: ffffb1d840e8fa40 R14: ffffb1d843880c80 R15: 00000000000000c5
[  100.456420] FS:  00007f1b57e4ab00(0000) GS:ffff95360c000000(0000)
knlGS:0000000000000000
[  100.456423] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  100.456426] CR2: 00007f94fc6b6d68 CR3: 00000003de2be006 CR4: 00000000003606b0

May I know if there is anything I can do to help resolve this, or can
I get a suggestion on this issue, I have been trying to resolve this
for the past few days. This has been an issue after switching from
xenomai 3.1 to 3.2.1 #eedf4ffe3

yours sincerely,
John

Re: Invalid Opcode

[Jan Kiszka]

On 04.07.23 18:52, John Ho wrote:
> Hi all,
> I have encountered an invalid Opcode while running some of my stuff,
> and I have no idea where this is coming from, and I would like to ask
> for some help.
> 
> I have assumed that adding Nopti would help eliminate this invalid
> opcode, but it has not helped.
> 
> [  100.456173] kernel BUG at kernel/xenomai/synch.c:968!

This is

XENO_BUG_ON(COBALT, !(synch->status & XNSYNCH_PRIO));

in xnsynch_requeue_sleeper. Something very strange happens here.

Can you extract a simple test case that triggers the issue and that you
can share?

Jan

-- 
Siemens AG, Technology
Competence Center Embedded Linux

Re: Invalid Opcode

[Bezdeka, Florian]

On Tue, 2023-07-04 at 18:56 +0200, Jan Kiszka wrote:
> On 04.07.23 18:52, John Ho wrote:
> > Hi all,
> > I have encountered an invalid Opcode while running some of my stuff,
> > and I have no idea where this is coming from, and I would like to ask
> > for some help.
> > 
> > I have assumed that adding Nopti would help eliminate this invalid
> > opcode, but it has not helped.
> > 
> > [  100.456173] kernel BUG at kernel/xenomai/synch.c:968!
> 
> This is
> 
> XENO_BUG_ON(COBALT, !(synch->status & XNSYNCH_PRIO));

Just to be clear: You're not searching for an invalid instruction. The
invalid instruction is injected by the BUG() infrastructure to trigger
the HW exception for the purpose of proper error reporting.

You're searching for the root cause that the error condition mentioned
above is getting true.

> 
> in xnsynch_requeue_sleeper. Something very strange happens here.
> 
> Can you extract a simple test case that triggers the issue and that you
> can share?
> 
> Jan
>