Red Hat Bugzilla – Bug 253138
Kernel Opps when running any Cell Application on >=220.127.116.11-41.fc7
Last modified: 2008-02-15 21:35:50 EST
Description of problem:
On kernel 18.104.22.168-41.fc7 when ever a Cell application is run it locks up and a
kernel oops is produced. The system still works and I am able to just ssh into
it(I'm running it headless) but I still can't kill the Cell applications that
have frozen. Cell programs run fine under 2.6.21-1.3228.fc7
Version-Release number of selected component (if applicable):
Cell SDK: 2.1
Run any Cell application
kernel BUG at arch/powerpc/platforms/ps3/spu.c:500!
Oops: Exception in kernel mode, sig: 5 [#1]
SMP NR_CPUS=128 NUMA PS3
Modules linked in: autofs4 hidp rfcomm l2cap sunrpc nf_conntrack_netbios_ns nf_c
onntrack_ipv4 xt_state nf_conntrack nfnetlink ipt_REJECT iptable_filter ip_table
s xt_tcpudp ip6t_REJECT ip6table_filter ip6_tables x_tables spufs dm_mirror dm_m
ultipath dm_mod sd_mod ipv6 snd_ps3 snd_seq_dummy snd_seq_oss snd_seq_midi_event
snd_seq sg snd_seq_device sr_mod snd_pcm_oss cdrom snd_mixer_oss usb_storage hc
i_usb bluetooth snd_pcm snd_page_alloc snd_timer snd ps3rom soundcore scsi_mod p
s3_gelic ps3flash ps3disk ps3stor_lib ext3 jbd mbcache ehci_hcd ohci_hcd uhci_hc
NIP: c00000000006932c LR: d00000000039969c CTR: c000000000069304
REGS: c0006c0065413860 TRAP: 0700 Not tainted (22.214.171.124-41.fc7)
MSR: 8000000000028032 <EE,IR,DR> CR: 22002422 XER: 20000000
TASK = c0000000045ed180 'dma' THREAD: c0006c0065410000 CPU: 1
GPR00: 0000000000000001 c0006c0065413ae0 c000000000730938 c000000007bcf600
GPR04: 000000000000001b c0006c0064c45908 c0006c0064c45908 000000000165d000
GPR08: 0000000000000000 c000000007be32a0 c0000000005ee080 000000000000001f
GPR12: d00000000039bbf8 c000000000617100 000000000000000e 0000000000000000
GPR16: 0000000000000063 0000000001823048 0000000000000000 00000000f7f8bca8
GPR20: 000000000000000a 0000000001823048 0000000001823078 0000000001823078
GPR24: 0000000000000003 d00000000039cc70 00000000f7f8bca8 c0006c0065413d34
GPR28: c0006c0065413d30 c00000000077ad40 d0000000003b5788 c000000007bcf600
NIP [c00000000006932c] .mfc_sr1_set+0x28/0x58
LR [d00000000039969c] .spu_hw_master_stop+0x80/0xd0 [spufs]
[c0006c0065413ae0] [c000000000437d78] ._spin_unlock_irqrestore+0x44/0x58 (unreli
[c0006c0065413b60] [d00000000039969c] .spu_hw_master_stop+0x80/0xd0 [spufs]
[c0006c0065413bf0] [d00000000039a2a0] .spufs_run_spu+0x2d0/0x3bc [spufs]
[c0006c0065413cc0] [d000000000397050] .do_spu_run+0xbc/0x198 [spufs]
[c0006c0065413d80] [c00000000005eae8] .sys_spu_run+0xb0/0x108
[c0006c0065413e30] [c0000000000086c8] syscall_exit+0x0/0x40
e8690040 4e800020 7c0802a6 f8010010 f821ff81 e9230118 e8090040 7c002278
7800efa6 78001fa4 3160ffff 7c0b0110 <0b000000> f8890040 38800000 e9230118
Can you try a kernel from Koji?
Was Koji's kernel pushed out? I did an update tonight my kernel was updated to
126.96.36.199-65.fc7 which is the same as Koji has. In any case 188.8.131.52-65.fc7 has
the same problem.
I'm reviewing this bug as part of the kernel bug triage project, an attempt to
isolate current bugs in the fedora kernel.
I am CC'ing myself to this bug and will try and assist you in resolving it if I can.
There hasn't been much activity on this bug for a while. Could you tell me if
you are still having problems with the latest kernel?
If the problem no longer exists then please close this bug or I'll do so in a
few days if there is no additional information lodged.
It still happens on all recent kernels. I havn't had time to really debug this
myself but I will try to help as much as I can.
Thank you for the update Lee. If it is still happening on a kernel from rawhide,
it is probably a candidate for taking upstream. I am happy to file a bug at the
kernel.org bugzilla for you on your behalf if you wish.
Could I trouble you for an answer to the above? Again, I am happy to file at the
kernel.org bugzilla on your behalf with the information you have provided if you
are willing to be cc'd into that bug.
Closing as per previous comment. Please re-open if this is still an issue for you.