Re: [ANNOUNCE] 2.6.33-rc8-rt1

From: Arnaldo Carvalho de Melo
Date: Tue Feb 23 2010 - 17:04:29 EST

Next message: Jens Axboe: "Re: Testing for dirty buffers on a block device"
Previous message: tip-bot for Suresh Siddha: "[tip:x86/ptrace] x86, ptrace: Remove set_stopped_child_used_math() in [x]fpregs_set"
In reply to: gowrishankar: "Re: [ANNOUNCE] 2.6.33-rc8-rt1"
Next in thread: Fernando Lopez-Lezcano: "Re: [ANNOUNCE] 2.6.31.12-rt21"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Em Mon, Feb 22, 2010 at 12:24:07AM +0100, Thomas Gleixner escreveu:
> We are pleased to announce the next update to our new preempt-rt
> series.
>
> - Forward to 2.6.33-rc8
>
> Finally I managed to get that beast tamed.

mica.ghostprotocols.net login: BUG: unable to handle kernel NULL pointer dereference at (null)
IP: [<(null)>] (null)
PGD 37ea1067 PUD bae52067 PMD 0
Oops: 0010 [#1] PREEMPT SMP
last sysfs file: /sys/class/misc/autofs/dev
CPU 0
Pid: 7341, comm: perf Not tainted 2.6.33-rc8-rt #1 0KN122/PowerEdge 1900
RIP: 0010:[<0000000000000000>] [<(null)>] (null)
RSP: 0018:ffff8800bae01990 EFLAGS: 00010046
RAX: ffffffff8140a060 RBX: ffff88012f0206c0 RCX: 0000000000000000
RDX: 0000000000000001 RSI: ffff88012f0206c0 RDI: ffff8800282d0300
RBP: ffff8800bae019b8 R08: ffff8800282c3e68 R09: ffff8800bac16238
R10: ffffffff81600000 R11: ffff8800bae01b38 R12: 0000000000000001
R13: ffff8800282d0300 R14: ffff8800282d0300 R15: 0000000000000001
FS: 00007f862804b6e0(0000) GS:ffff880028200000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 0000000000000000 CR3: 00000000baf13000 CR4: 00000000000006f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process perf (pid: 7341, threadinfo ffff8800bae00000, task ffff880037f580c0)
Stack:
ffffffff8102caee ffff8800282d0300 ffff88012f0206c0 ffff8800282d0300
<0> ffff8800282d0300 ffff8800bae019d8 ffffffff8102cb29 ffff8800bac16238
<0> 0000000000000003 ffff8800bae01a38 ffffffff8103aeb3 ffff8800bae019f8
Call Trace:
[<ffffffff8102caee>] ? enqueue_task+0x45/0x55
[<ffffffff8102cb29>] activate_task+0x2b/0x33
[<ffffffff8103aeb3>] try_to_wake_up+0x1cc/0x33c
[<ffffffff81330d37>] ? preempt_schedule_irq+0x6a/0x76
[<ffffffff8103b072>] wake_up_process_mutex+0x15/0x17
[<ffffffff8106ac2e>] wakeup_next_waiter+0x85/0x15e
[<ffffffff813319e5>] rt_spin_lock_slowunlock+0x3e/0x5c
[<ffffffff8106a302>] rt_spin_lock_fastunlock+0x35/0x37
[<ffffffff81331f67>] rt_spin_unlock+0x10/0x12
[<ffffffff8105a5a7>] remove_wait_queue+0x49/0x4e
[<ffffffff810f0927>] free_poll_entry+0x19/0x25
[<ffffffff810f095d>] poll_freewait+0x2a/0x6b
[<ffffffff810f0c81>] do_sys_poll+0x2e3/0x3a3
[<ffffffff810f1678>] ? __pollwait+0x0/0xd7
[<ffffffff810f174f>] ? pollwake+0x0/0x5b
[<ffffffff810f174f>] ? pollwake+0x0/0x5b
[<ffffffff810f174f>] ? pollwake+0x0/0x5b
[<ffffffff810f174f>] ? pollwake+0x0/0x5b
[<ffffffff810308c7>] ? rcu_read_unlock+0x9/0xb
[<ffffffff8103091f>] ? cpuacct_charge+0x56/0x5f
[<ffffffff81001d1e>] ? __switch_to+0xc2/0x1ca
[<ffffffff81332239>] ? _raw_spin_unlock+0x23/0x2e
[<ffffffff810351c3>] ? finish_task_switch+0x53/0xb8
[<ffffffff81330caa>] ? __schedule+0x7bb/0x7de
[<ffffffff810308c7>] ? rcu_read_unlock+0x9/0xb
[<ffffffff810308c7>] ? rcu_read_unlock+0x9/0xb
[<ffffffff8103091f>] ? cpuacct_charge+0x56/0x5f
[<ffffffff81001d1e>] ? __switch_to+0xc2/0x1ca
[<ffffffff81031201>] ? __dequeue_entity+0x2c/0x35
[<ffffffff810445bd>] ? timespec_add_safe+0x37/0x64
[<ffffffff810f0888>] ? poll_select_set_timeout+0x5c/0x7f
[<ffffffff810f0d93>] sys_poll+0x52/0xb1
[<ffffffff81002b2b>] system_call_fastpath+0x16/0x1b
Code: Bad RIP value.
RIP [<(null)>] (null)
RSP <ffff8800bae01990>
CR2: 0000000000000000
---[ end trace 55781d66e9d64e79 ]---

Workload (screens were frozen at that time):

make -j128 allyesconfig

[root@mica rt-tests]# perf record -a -f ./cyclictest -a -t -n -p99 -i100 -d50
policy: fifo: loadavg: 0.58 0.19 0.09 1/238 7343

T: 0 ( 5502) P:99 I:100 C:1305514 Min: 2 Act: 4 Avg: 3 Max: 27
T: 1 ( 5503) P:98 I:150 C: 870343 Min: 3 Act: 3 Avg: 4 Max: 25
T: 2 ( 5504) P:97 I:200 C: 652757 Min: 3 Act: 4 Avg: 3 Max: 27
T: 3 ( 5505) P:96 I:250 C: 522206 Min: 3 Act: 5 Avg: 4 Max: 25

[root@mica rt-tests]# perf top
--------------------------------------------------------------------------
PerfTop: 2799 irqs/sec kernel:99.4% [1000Hz cycles], (all, 4 CPUs)
--------------------------------------------------------------------------

samples pcnt function DSO
_______ _____ ___________________________ ________________________________________

9924.00 50.6% need_resched /lib/modules/2.6.33-rc8-rt/build/vmlinux
8276.00 42.2% test_ti_thread_flag /lib/modules/2.6.33-rc8-rt/build/vmlinux
116.00 0.6% _raw_spin_lock_irqsave /lib/modules/2.6.33-rc8-rt/build/vmlinux
102.00 0.5% cpupri_set /lib/modules/2.6.33-rc8-rt/build/vmlinux
90.00 0.5% sha1_block_data_order /lib64/libcrypto.so.0.9.8b
77.00 0.4% __schedule /lib/modules/2.6.33-rc8-rt/build/vmlinux
63.00 0.3% read_tsc /lib/modules/2.6.33-rc8-rt/build/vmlinux
58.00 0.3% get_next_timer_interrupt /lib/modules/2.6.33-rc8-rt/build/vmlinux
57.00 0.3% apic_timer_interrupt /lib/modules/2.6.33-rc8-rt/build/vmlinux
45.00 0.2% native_sched_clock /lib/modules/2.6.33-rc8-rt/build/vmlinux
44.00 0.2% _raw_spin_lock /lib/modules/2.6.33-rc8-rt/build/vmlinux
42.00 0.2% __unlazy_fpu /lib/modules/2.6.33-rc8-rt/build/vmlinux
40.00 0.2% system_call /lib/modules/2.6.33-rc8-rt/build/vmlinux
37.00 0.2% sched_clock_local /lib/modules/2.6.33-rc8-rt/build/vmlinux
30.00 0.2% __switch_to /lib/modules/2.6.33-rc8-rt/build/vmlinux
27.00 0.1% rt_spin_lock_fastlock /lib/modules/2.6.33-rc8-rt/build/vmlinux
26.00 0.1% rt_spin_lock_fastunlock /lib/modules/2.6.33-rc8-rt/build/vmlinux
24.00 0.1% _raw_spin_unlock_irqrestore /lib/modules/2.6.33-rc8-rt/build/vmlinux
22.00 0.1% dequeue_rt_stack /lib/modules/2.6.33-rc8-rt/build/vmlinux
22.00 0.1% timekeeping_get_ns /lib/modules/2.6.33-rc8-rt/build/vmlinux
20.00 0.1% rcu_sched_qs /lib/modules/2.6.33-rc8-rt/build/vmlinux
20.00 0.1% switch_mm /lib/modules/2.6.33-rc8-rt/build/vmlinux
19.00 0.1% mmdrop_delayed /lib/modules/2.6.33-rc8-rt/build/vmlinux
19.00 0.1% sched_clock_cpu /lib/modules/2.6.33-rc8-rt/build/vmlinux
17.00 0.1% __cycles_2_ns /lib/modules/2.6.33-rc8-rt/build/vmlinux
16.00 0.1% tick_nohz_stop_sched_tick /lib/modules/2.6.33-rc8-rt/build/vmlinux
13.00 0.1% try_to_wake_up /lib/modules/2.6.33-rc8-rt/build/vmlinux
13.00 0.1% __math_state_restore /lib/modules/2.6.33-rc8-rt/build/vmlinux
13.00 0.1% __GI_clock_gettime /lib64/librt-2.5.so
11.00 0.1% _raw_spin_lock_irq /lib/modules/2.6.33-rc8-rt/build/vmlinux
10.00 0.1% timerthread /home/acme/git/rt-tests/cyclictest
9.00 0.0% cpumask_clear_cpu /lib/modules/2.6.33-rc8-rt/build/vmlinux
9.00 0.0% __GI_vfprintf /lib64/libc-2.5.so
8.00 0.0% finish_task_switch /lib/modules/2.6.33-rc8-rt/build/vmlinux
8.00 0.0% hrtimer_interrupt /lib/modules/2.6.33-rc8-rt/build/vmlinux
8.00 0.0% ktime_get /lib/modules/2.6.33-rc8-rt/build/vmlinux
8.00 0.0% rb_erase /lib/modules/2.6.33-rc8-rt/build/vmlinux
8.00 0.0% test_ti_thread_flag /lib/modules/2.6.33-rc8-rt/build/vmlinux
7.00 0.0% pick_next_task /lib/modules/2.6.33-rc8-rt/build/vmlinux
7.00 0.0% clock_nanosleep /lib64/librt-2.5.so
6.00 0.0% __remove_hrtimer /lib/modules/2.6.33-rc8-rt/build/vmlinux
6.00 0.0% cpumask_set_cpu /lib/modules/2.6.33-rc8-rt/build/vmlinux
6.00 0.0% __rcu_read_unlock /lib/modules/2.6.33-rc8-rt/build/vmlinux
5.00 0.0% megasas_clear_intr_xscale /lib/modules/2.6.33-rc8-rt/kernel/drivers/scsi/megaraid/megaraid_sas.ko
5.00 0.0% native_apic_mem_write /lib/modules/2.6.33-rc8-rt/build/vmlinux
5.00 0.0% __hrtimer_start_range_ns /lib/modules/2.6.33-rc8-rt/build/vmlinux
5.00 0.0% rb_insert_color /lib/modules/2.6.33-rc8-rt/build/vmlinux
5.00 0.0% update_curr_rt /lib/modules/2.6.33-rc8-rt/build/vmlinux
5.00 0.0% hrtimer_nanosleep /lib/modules/2.6.33-rc8-rt/build/vmlinux

- Arnaldo
--
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/

Next message: Jens Axboe: "Re: Testing for dirty buffers on a block device"
Previous message: tip-bot for Suresh Siddha: "[tip:x86/ptrace] x86, ptrace: Remove set_stopped_child_used_math() in [x]fpregs_set"
In reply to: gowrishankar: "Re: [ANNOUNCE] 2.6.33-rc8-rt1"
Next in thread: Fernando Lopez-Lezcano: "Re: [ANNOUNCE] 2.6.31.12-rt21"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]