real time hang w/ latency tracing

[Daniel Walker]

With 2.6.21-rc5-rt8 / i386 . If I disable wakeup timing , and enable
interrupt off timing with latency tracing I can cause,

BUG: scheduling with irqs disabled: IRQ-16/0x00000000/909
caller is rt_spin_lock_slowlock+0x94/0x190
 [<c010577a>] show_trace_log_lvl+0x3a/0x60
 [<c0105fdd>] show_trace+0x2d/0x30
 [<c01060c7>] dump_stack+0x27/0x30
 [<c03f52db>] schedule+0x12b/0x130
 [<c03f6314>] rt_spin_lock_slowlock+0x94/0x190
 [<c03f6b35>] rt_spin_lock+0x25/0x30
 [<c02f19bf>] serial8250_console_write+0x16f/0x190
 [<c012b3aa>] __call_console_drivers+0x7a/0xa0
 [<c012b427>] _call_console_drivers+0x57/0xa0
 [<c012b58d>] release_console_sem+0xdd/0x1e0
 [<c012bd46>] vprintk+0x226/0x3b0
 [<c012bef0>] printk+0x20/0x30
 [<c0158d29>] print_traces+0x49/0x100
 [<c0105792>] show_trace_log_lvl+0x52/0x60
 [<c0105fdd>] show_trace+0x2d/0x30
 [<c01060c7>] dump_stack+0x27/0x30
 [<c0158481>] check_critical_timing+0x2b1/0x380
 [<c01585f6>] trace_hardirqs_on+0xa6/0xc0
 [<c01046e5>] restore_nocheck+0x12/0x15
 =======================

I do this by running,

echo 1 > /proc/sys/kernel/preempt_thresh

which clearly causes loads of output .. Then while the output is
screaming along I hold down the left arrow key and eventually the
message above starts to appear often .. Then the system hangs.

Also I'm using an NFS root which might help to bring out the error ..

Daniel

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/