G5 with tg3 ethernet stops when rsyncing large amounts of data

[Johannes Berg]

On Tue, 2008-09-23 at 17:05 +0200, Johannes Berg wrote:
> On Tue, 2008-09-23 at 16:45 +0200, Johannes Berg wrote:
> > I've been trying to push a few GB of data from my powerbook to my quad
> > G5 with rsync, but after transferring for a bit the machine stops
> > entirely. It doesn't even reply to ping, so this must be happening in an
> > interrupt.

> > The only information I have so far is a picture I took of the console
> > when this happened again:
> > http://johannes.sipsolutions.net/files/dsc_2275.jpg

> This time I got the trace below (system didn't hang)...
> 
> I'd think faulty memory if it weren't for the fact that I never see
> problems when I do random things that don't involve networking.

It seems turning off ftrace and irqtrace helps, at least I haven't been
able to reproduce the problem any more and have transferred more data
than ever before...

johannes


> [  493.257620] ------------[ cut here ]------------
> [  493.270431] kernel BUG at kernel/seccomp.c:50!
> [  493.283978] Oops: Exception in kernel mode, sig: 5 [#1]
> [  493.283984] PREEMPT SMP NR_CPUS=4 PowerMac
> [  493.283997] Modules linked in: binfmt_misc ipv6 dm_crypt dm_snapshot dm_mirror dm_log dm_mod fuse snd_aoa_codec_onyx snd_aoa_fabric_layout snd_aoa snd_aoa_i2sbus snd_aoa_soundbus snd_pcm_oss snd_mixer_oss firewire_ohci firewire_core snd_pcm crc_itu_t snd_page_alloc snd_timer snd rfkill ohci1394 mac80211 ehci_hcd ohci_hcd soundcore ieee1394 usbcore cfg80211
> [  493.284091] NIP: c0000000000bf98c LR: c0000000000bf93c CTR: 0000000000000000
> [  493.284099] REGS: c0000001b0353a90 TRAP: 0700   Not tainted  (2.6.27-rc6-wl-01382-g0bea1f7-dirty)
> [  493.284104] MSR: 9000000000029032 <EE,ME,IR,DR>  CR: 24224828  XER: 00000000
> [  493.284122] TASK = c0000001b0256d00[4894] 'rsync' THREAD: c0000001b0350000 CPU: 1
> [  493.284129] GPR00: 0000000000000000 c0000001b0353d10 c0000000008b23e0 0000000000000001 
> [  493.284140] GPR04: 0000000000000000 c0000001b0256d00 c0000002006e9a30 c0000001b0256d00 
> [  493.284151] GPR08: c0000001b0350000 c0000001b0256d00 0000000001b90000 c0000000007e3ad8 
> [  493.284162] GPR12: 0000000024224824 c000000000943580 0000000010074390 0000000000000004 
> [  493.284173] GPR16: 00000000ff8753f8 0000000010080c18 0000000010073d5c 00000000ff875050 
> [  493.284184] GPR20: 0000000000000000 00000000ff875048 0000000010073d5c 0000000000000020 
> [  493.284195] GPR24: 00000001f7fe3238 0000000010080c18 0000000000000005 00000000ff875050 
> [  493.284205] GPR28: 0000000000000005 000000000000008e c000000000847ab8 c0000001b0353d10 
> [  493.284227] NIP [c0000000000bf98c] .__secure_computing+0x78/0xa8
> [  493.284233] LR [c0000000000bf93c] .__secure_computing+0x28/0xa8
> [  493.284238] Call Trace:
> [  493.284245] [c0000001b0353d10] [c0000000000bf93c] .__secure_computing+0x28/0xa8 (unreliable)
> [  493.284257] [c0000001b0353da0] [c000000000008b7c] .do_syscall_trace_enter+0x40/0x12c
> [  493.284266] [c0000001b0353e30] [c000000000007794] syscall_dotrace+0xc/0x38
> [  493.284272] Instruction dump:
> [  493.284280] 7809e7e3 4182000c 7d695b78 48000008 392b0014 80090000 7f80e800 419e0028 
> [  493.284298] 84090004 2f800000 409effec 4800000c <0fe00000> 48000000 38600009 4bf9f409 
> [  493.285036] BUG: using smp_processor_id() in preemptible [00000000] code: rsync/4894
> [  493.285049] caller is .die+0x198/0x1e4
> [  493.285053] Call Trace:
> [  493.285061] [c0000001b0353500] [c0000000000108ec] .show_stack+0x84/0x190 (unreliable)
> [  493.285074] [c0000001b03535b0] [c000000000463c60] .dump_stack+0x28/0x3c
> [  493.285085] [c0000001b0353630] [c0000000002ac658] .debug_smp_processor_id+0xe8/0x124
> [  493.285094] [c0000001b03536d0] [c000000000023bc8] .die+0x198/0x1e4
> [  493.285103] [c0000001b0353780] [c000000000023f5c] ._exception+0x98/0x1fc
> [  493.285112] [c0000001b0353940] [c000000000024808] .program_check_exception+0x610/0x65c
> [  493.285122] [c0000001b0353a20] [c000000000004a84] program_check_common+0x104/0x180
> [  493.285135] --- Exception: 700 at .__secure_computing+0x78/0xa8
> [  493.285138]     LR = .__secure_computing+0x28/0xa8
> [  493.285146] [c0000001b0353da0] [c000000000008b7c] .do_syscall_trace_enter+0x40/0x12c
> [  493.285155] [c0000001b0353e30] [c000000000007794] syscall_dotrace+0xc/0x38
> [  493.285165] ---[ end trace 488bc840169e13ad ]---
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part
URL: <http://lists.ozlabs.org/pipermail/linuxppc-dev/attachments/20080923/2c96e93d/attachment.pgp>