Re: [patch] Real-Time Preemption, -RT-2.6.9-rc4-mm1-U6
From: Adam Heath
Date: Wed Oct 20 2004 - 11:40:57 EST
On Tue, 19 Oct 2004, Ingo Molnar wrote:
>
> i have released the -U6 Real-Time Preemption patch:
>
> http://redhat.com/~mingo/realtime-preempt/realtime-preempt-2.6.9-rc4-mm1-U6
Got these high-latency values during the night on U6(haven't booted U8 yet).
IRQ 5/431/CPU#0): 612 us critical section violates 100 us threshold.
=> started at timestamp 4167601478: <call_console_drivers+0x76/0x140>
=> ended at timestamp 4167602090: <finish_task_switch+0x43/0xb0>
[<c01327f0>] sub_preempt_count+0x60/0x90
[<c01324de>] check_preempt_timing+0x15e/0x270
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c01327f0>] sub_preempt_count+0x60/0x90
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c02a5717>] __sched_text_start+0x2d7/0x5d0
[<c0113104>] mcount+0x14/0x18
[<c013bdfa>] do_irqd+0x5a/0x80
[<c01309ea>] kthread+0xaa/0xb0
[<c013bda0>] do_irqd+0x0/0x80
[<c0130940>] kthread+0x0/0xb0
[<c0104099>] kernel_thread_helper+0x5/0xc
preempt count: 00000002
. 2-level deep critical section nesting:
.. entry 1: __sched_text_start+0x3b/0x5d0 / (do_irqd+0x5a/0x80)
.. entry 2: print_traces+0x1d/0x80 / (dump_stack+0x23/0x30)
=> dump-end timestamp 4167602447
(IRQ 5/431/CPU#0): 34875 us critical section violates 100 us threshold.
=> started at timestamp 4167608224: <call_console_drivers+0x76/0x140>
=> ended at timestamp 4167643099: <finish_task_switch+0x43/0xb0>
[<c01327f0>] sub_preempt_count+0x60/0x90
[<c01324de>] check_preempt_timing+0x15e/0x270
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c01327f0>] sub_preempt_count+0x60/0x90
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c02a5717>] __sched_text_start+0x2d7/0x5d0
[<c0113104>] mcount+0x14/0x18
[<c013bdfa>] do_irqd+0x5a/0x80
[<c01309ea>] kthread+0xaa/0xb0
[<c013bda0>] do_irqd+0x0/0x80
[<c0130940>] kthread+0x0/0xb0
[<c0104099>] kernel_thread_helper+0x5/0xc
preempt count: 00000002
. 2-level deep critical section nesting:
.. entry 1: __sched_text_start+0x3b/0x5d0 / (do_irqd+0x5a/0x80)
.. entry 2: print_traces+0x1d/0x80 / (dump_stack+0x23/0x30)
=> dump-end timestamp 4167643459
(IRQ 1/18/CPU#0): 30560 us critical section violates 100 us threshold.
=> started at timestamp 4167647182: <call_console_drivers+0x76/0x140>
=> ended at timestamp 4167677742: <finish_task_switch+0x43/0xb0>
[<c01327f0>] sub_preempt_count+0x60/0x90
[<c01324de>] check_preempt_timing+0x15e/0x270
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c01327f0>] sub_preempt_count+0x60/0x90
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c02a5717>] __sched_text_start+0x2d7/0x5d0
[<c0113104>] mcount+0x14/0x18
[<c013bdfa>] do_irqd+0x5a/0x80
[<c01309ea>] kthread+0xaa/0xb0
[<c013bda0>] do_irqd+0x0/0x80
[<c0130940>] kthread+0x0/0xb0
[<c0104099>] kernel_thread_helper+0x5/0xc
preempt count: 00000002
. 2-level deep critical section nesting:
.. entry 1: __sched_text_start+0x3b/0x5d0 / (do_irqd+0x5a/0x80)
.. entry 2: print_traces+0x1d/0x80 / (dump_stack+0x23/0x30)
=> dump-end timestamp 4167678099
(bash/10595/CPU#0): 33546 us critical section violates 100 us threshold.
=> started at timestamp 4167681248: <call_console_drivers+0x76/0x140>
=> ended at timestamp 4167714794: <finish_task_switch+0x43/0xb0>
[<c01327f0>] sub_preempt_count+0x60/0x90
[<c01324de>] check_preempt_timing+0x15e/0x270
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c01327f0>] sub_preempt_count+0x60/0x90
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c0117ca3>] finish_task_switch+0x43/0xb0
[<c02a5717>] __sched_text_start+0x2d7/0x5d0
[<c02a684f>] down_write+0x12f/0x1e0
[<c0113104>] mcount+0x14/0x18
[<c02a684f>] down_write+0x12f/0x1e0
[<c01182bb>] lock_kernel+0x2b/0x40
[<c016f122>] sys_ioctl+0x52/0x230
[<c0106013>] syscall_call+0x7/0xb
preempt count: 00000002
. 2-level deep critical section nesting:
.. entry 1: __sched_text_start+0x3b/0x5d0 / (down_write+0x12f/0x1e0)
.. entry 2: print_traces+0x1d/0x80 / (dump_stack+0x23/0x30)
=> dump-end timestamp 4167715168
-
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/