Hard lockup NVRM: os_schedule: Attempted to yield the CPU while in atomic or interrup

[Xevious]

I have this problem with both the gtx 460's I have in my workstation at work along with the gtx 470 I have at home. I am pretty sure I had the problem with a gtx 260 as well.

For me it is reproducible by changing which X session/virtual terminal I am using with control + alt + f7/f8.

It seems like it is far more common on my workstation at work which has two gtx 460s and is outputing on all four DVI ports (no SLI). I have seen many threads about this but nvidia seems to be clueless on the issue still?

I never have this problem except when switching X sesions (it never does it any other time). I would say it does it 1 out of 20 (sometimes more, sometimes less) times I switch X from one session to another. Disabling the video cards from going to low powermode helps a little bit (I think) but I have not verified this 100%.


Here is my bug report (also used the upload applet):

http://box.houkouonchi.jp/nvidia-bug-report-sigoto.txt


I generally get:

NVRM: os_pci_init_handle: invalid context!


or:

NVRM: os_schedule: Attempted to yield the CPU while in atomic or interrupt context

repeated multiple times. When it happens my machine usually becomes totally unresponsive. Sometimes I can still ssh and kill X (like today) and other times my machine runs so slowly that it only periodically responds to pings (let alone anything else) and i can't even ssh in and have to do a hard reset.

When I was able to get back on the system I couldn't start X back up even when unloading/reloading the nvidia module. X would just use 100% CPU and half initialize (make my monitors go blank). I have also seen this on my home behavior. In this case dmesg showed the following:

NVRM: os_schedule: Attempted to yield the CPU while in atomic or interrupt context
NVRM: os_schedule: Attempted to yield the CPU while in atomic or interrupt context
NVRM: os_schedule: Attempted to yield the CPU while in atomic or interrupt context
nvidia 0000:84:00.0: PCI INT A disabled
nvidia 0000:86:00.0: PCI INT A disabled
nvidia 0000:84:00.0: PCI INT A -> GSI 50 (level, low) -> IRQ 50
nvidia 0000:84:00.0: setting latency timer to 64
vgaarb: device changed decodes: PCI:0000:84:00.0,olddecodes=none,decodes=none:owns=none
nvidia 0000:86:00.0: PCI INT A -> GSI 56 (level, low) -> IRQ 56
nvidia 0000:86:00.0: setting latency timer to 64
vgaarb: device changed decodes: PCI:0000:86:00.0,olddecodes=none,decodes=none:owns=none
NVRM: loading NVIDIA UNIX x86_64 Kernel Module  270.41.06  Mon Apr 18 14:53:56 PDT 2011
NVRM: failed to enable MSI,
using PCI-E virtual-wire interrupts!
ioremap error for 0xbf790000-0xbf791000, requested 0x10, got 0x0
NVRM: failed to enable MSI,
using PCI-E virtual-wire interrupts!
BUG: unable to handle kernel NULL pointer dereference at           (null)
IP: [<ffffffffa18c5887>] _nv015958rm+0x6e1/0x9b2 [nvidia]
PGD 623e82067 PUD 623f3d067 PMD 0
Oops: 0002 [#1] PREEMPT SMP
last sysfs file: /sys/devices/pci0000:00/0000:00:01.0/0000:01:00.1/net/eth1/carrier
CPU 2
Modules linked in: nvidia(P) snd_seq snd_pcm_oss snd_mixer_oss snd_emu10k1 snd_rawmidi snd_ac97_codec ac97_bus snd_seq_device snd_util_mem snd_hwdep ipv6 dm_mod snd_hda_codec_hdmi snd_hda_intel snd_hda_codec snd_ctxfi snd_pcm snd_timer snd i2c_i801 snd_page_alloc [last unloaded: nvidia]

Pid: 12101, comm: Xorg Tainted: P            2.6.38-web100 #2 Supermicro X8DTH-i/6/iF/6F/X8DTH
RIP: 0010:[<ffffffffa18c5887>]  [<ffffffffa18c5887>] _nv015958rm+0x6e1/0x9b2 [nvidia]
RSP: 0018:ffff8806252a5ad8  EFLAGS: 00010256
RAX: 0000000000000000 RBX: 0000000000008000 RCX: 0000000000000001
RDX: 0000000000000000 RSI: 0000000000011401 RDI: ffff8805f9524828
RBP: ffff8805fd76b3a0 R08: 0000000000000020 R09: 0000000000000000
R10: ffff8805fc853000 R11: ffffffff81316762 R12: ffff8805fb77c000
R13: 0000000000000001 R14: 0000000000000000 R15: 0000000000008000
FS:  00007fa68dd786f0(0000) GS:ffff8800bee40000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 0000000000000000 CR3: 00000005fb160000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process Xorg (pid: 12101, threadinfo ffff8806252a4000, task ffff88061faa18d0)
Stack:
 ffff8805fb77c000 ffff8805fd76b3f8 ffff8805fae73000 0000000000000000
 ffff8805f9d5a740 ffffffffa18c5ba6 0000000000000000 ffff8805fae73000
 ffff8805fb77c000 ffffffffa18ce542 ffff8805fb77c000 ffff8805fae73000
Call Trace:
 [<ffffffffa18c5ba6>] ? _nv015970rm+0x4e/0x78 [nvidia]
 [<ffffffffa18ce542>] ? _nv015971rm+0x12/0x38 [nvidia]
 [<ffffffffa18868ed>] ? _nv016320rm+0x49/0x6a [nvidia]
 [<ffffffffa1869f4b>] ? _nv014862rm+0xb8/0x55b [nvidia]
 [<ffffffffa1868cf2>] ? _nv015122rm+0xe3/0x45a [nvidia]
 [<ffffffffa153e88e>] ? _nv015300rm+0xd/0x12 [nvidia]
 [<ffffffffa1a43c09>] ? _nv002400rm+0x1e7/0x28a [nvidia]
 [<ffffffffa1a44a0f>] ? _nv002394rm+0x4a7/0x685 [nvidia]
 [<ffffffffa1a4a51d>] ? rm_init_adapter+0x9d/0x111 [nvidia]
 [<ffffffffa1a6791d>] ? nv_kern_open+0x49b/0x5ce [nvidia]
 [<ffffffff810c3455>] ? chrdev_open+0x1d9/0x1f8
 [<ffffffff810c327c>] ? chrdev_open+0x0/0x1f8
 [<ffffffff810bf34f>] ? __dentry_open+0x1b3/0x2af
 [<ffffffff810ca46d>] ? finish_open+0x91/0x141
 [<ffffffff810cc243>] ? do_filp_open+0x178/0x5ad
 [<ffffffff8102f5c9>] ? get_parent_ip+0x9/0x1b
 [<ffffffff8102f5c9>] ? get_parent_ip+0x9/0x1b
 [<ffffffff8159daee>] ? _raw_spin_unlock+0x10/0x2e
 [<ffffffff810d5813>] ? alloc_fd+0x112/0x123
 [<ffffffff810bf0e5>] ? do_sys_open+0x51/0xdd
 [<ffffffff81001f7b>] ? system_call_fastpath+0x16/0x1b
Code: b3 4c 46 50 00 ba 0f 00 00 00 4c 89 e7 41 ff 94 24 28 11 00 00 48 8b 55 38 48 8b 82 10 03 00 00 44 89 f2 4a 8b 04 e8 48 c1 e2 04 <c7> 04 02 00 00 00 00 48 8b 4d 38 48 8b 81 10 03 00 00 4a 8b 04
RIP  [<ffffffffa18c5887>] _nv015958rm+0x6e1/0x9b2 [nvidia]
 RSP <ffff8806252a5ad8>
CR2: 0000000000000000
---[ end trace d734671180180aa1 ]---

So nvidia doing anything about this? it is extremely annoying.

[elsifaka]

I also have this issue on my GTX460M. I don't know which driver version should work.

And if this error is related to xorg/kernel version in use.

I run an updated archlinux as of 12/02/2012.

[Licaon]

did you get a chance to test 295.17 ?