Re: [PATCH] sched/fair: fix select_task_rq_fair return -1

[Vincent Guittot]

On 4 December 2014 at 11:23, Liu, Chuansheng <chuansheng.liu@xxxxxxxxx> wrote:
>
>
>> -----Original Message-----
>> From: Vincent Guittot [mailto:vincent.guittot@xxxxxxxxxx]
>> Sent: Thursday, December 04, 2014 6:08 PM
>> To: Hillf Danton
>> Cc: Zhang, Jun; Ingo Molnar; Peter Zijlstra; linux-kernel; Liu, Chuansheng; Liu,
>> Changcheng
>> Subject: Re: [PATCH] sched/fair: fix select_task_rq_fair return -1
>>
>> On 4 December 2014 at 10:05, Hillf Danton <hillf.zj@xxxxxxxxxxxxxxx> wrote:
>> >>
>> >> From: zhang jun <jun.zhang@xxxxxxxxx>
>> >>
>> >> when cpu == -1 and sd->child == NULL, select_task_rq_fair return -1, system
>> panic.
>> >>
>> >> [ 0.738326] BUG: unable to handle kernel paging request at
>> ffff8800997ea928
>> >> [ 0.746138] IP: [<ffffffff810b15d3>] wake_up_new_task+0x43/0x1b0
>> >> [ 0.752886] PGD 25df067 PUD 0
>> >> [ 0.756321] Oops: 0000 1 PREEMPT SMP
>> >> [ 0.760743] Modules linked in:
>> >> [ 0.764179] CPU: 0 PID: 6 Comm: kworker/u8:0 Not tainted
>> 3.14.19-quilt-b27ac761 #2
>> >> [ 0.772651] Hardware name: Intel Corporation CHERRYVIEW B1
>> PLATFORM/Cherry Trail CR, BIOS CHTTRVP1.X64.0003.R08.1411110453
>> >> 11/11/2014
>> >> [ 0.786084] Workqueue: khelper __call_usermodehelper
>> >> [ 0.791649] task: ffff88007955a150 ti: ffff88007955c000 task.ti:
>> ffff88007955c000
>> >> [ 0.800021] RIP: 0010:[<ffffffff810b15d3>] [<ffffffff810b15d3>]
>> wake_up_new_task+0x43/0x1b0
>> >> [ 0.809478] RSP: 0000:ffff88007955dd58 EFLAGS: 00010092
>> >> [ 0.815422] RAX: 00000000ffffffff RBX: 0000000000000001 RCX:
>> 0000000000000020
>> >> [ 0.823404] RDX: 00000000ffffffff RSI: 0000000000000020 RDI:
>> 0000000000000020
>> >> [ 0.831386] RBP: ffff88007955dd80 R08: ffff880079604b58 R09:
>> 00000000ffffffff
>> >> [ 0.839368] R10: 0000000000000004 R11: eae0000000000000 R12:
>> ffff8800797ea650
>> >> [ 0.847350] R13: 0000000000004000 R14: ffff8800797ead52 R15:
>> 0000000000000206
>> >> [ 0.855335] FS: 0000000000000000(0000) GS:ffff88007aa00000(0000)
>> knlGS:0000000000000000
>> >> [ 0.864387] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
>> >> [ 0.870817] CR2: ffff8800997ea928 CR3: 000000000220b000 CR4:
>> 00000000001007f0
>> >> [ 0.878796] Stack:
>> >> [ 0.881046] 0000000000000001 ffff8800797ea650 0000000000004000
>> 0000000000000000
>> >> [ 0.889363] 000000000000003c ffff88007955ddf0 ffffffff8107ddfd
>> ffffffff810b6a95
>> >> [ 0.897680] 0000000000000000 ffff8800796beb00 ffff880000000000
>> ffffffff81000000
>> >> [ 0.905998] Call Trace:
>> >> [ 0.908752] [<ffffffff8107ddfd>] do_fork+0x12d/0x3b0
>> >> [ 0.914416] [<ffffffff810b6a95>] ? set_next_entity+0x95/0xb0
>> >> [ 0.920856] [<ffffffff8107e0a6>] kernel_thread+0x26/0x30
>> >> [ 0.926903] [<ffffffff8109703e>] __call_usermodehelper+0x2e/0x90
>> >> [ 0.933730] [<ffffffff8109ad31>] process_one_work+0x171/0x490
>> >> [ 0.940264] [<ffffffff8109ba4b>] worker_thread+0x11b/0x3a0
>> >> [ 0.946508] [<ffffffff8109b930>] ? manage_workers.isra.27+0x2b0/0x2b0
>> >> [ 0.953821] [<ffffffff810a1802>] kthread+0xd2/0xf0
>> >> [ 0.959289] [<ffffffff810a1730>] ? kthread_create_on_node+0x170/0x170
>> >> [ 0.966602] [<ffffffff81af81ac>] ret_from_fork+0x7c/0xb0
>> >> [ 0.972652] [<ffffffff810a1730>] ? kthread_create_on_node+0x170/0x170
>> >> [ 0.979956] Code: 49 89 fc 4c 89 f7 53 e8 bc 5c a4 00 49 8b 54 24 08 31 c9 49
>> 89 c7 49 8b 44 24 60 4c 89 e7 8b 72 18 ba 08 00 00 00 ff 50 40 89
>> >> c2 <49> 0f a3 94 24 e0 02 00 00 19 c9 85 c9 0f 84 34 01 00 00 48 8b
>> >> [ 1.001809] RIP [<ffffffff810b15d3>] wake_up_new_task+0x43/0x1b0
>> >> [ 1.008641] RSP <ffff88007955dd58>
>> >> [ 1.012544] CR2: ffff8800997ea928
>> >> [ 1.016279] --[ end trace 9737aaa337a5ca10 ]--
>> >>
>> >> Signed-off-by: zhang jun <jun.zhang@xxxxxxxxx>
>> >> Signed-off-by: Chuansheng Liu <chuansheng.liu@xxxxxxxxx>
>> >> Signed-off-by: Changcheng Liu <changcheng.liu@xxxxxxxxx>
>> >> ---
>> >>  kernel/sched/fair.c |    2 ++
>> >>  1 file changed, 2 insertions(+)
>> >>
>> >> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
>> >> index 34baa60..123153f 100644
>> >> --- a/kernel/sched/fair.c
>> >> +++ b/kernel/sched/fair.c
>> >> @@ -4587,6 +4587,8 @@ select_task_rq_fair(struct task_struct *p, int
>> prev_cpu, int sd_flag, int wake_f
>> >>               if (new_cpu == -1 || new_cpu == cpu) {
>> >>                       /* Now try balancing at a lower domain level of
>> cpu */
>> >>                       sd = sd->child;
>> >> +                     if ((!sd) && (new_cpu == -1))
>> >> +                             new_cpu = smp_processor_id();
>> >>                       continue;
>> >>               }
>> >>
>> > In 3.18-rc7 is -1 still selected?
>>
>> find_idlest_cpu doesn't return -1 anymore but always a valid cpu. The
>> local cpu will be used if no better cpu has been found
>
> So I guess we can make one similar patch based on 3.14.x branch?
> Latest:
> find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
> return shallowest_idle_cpu != -1 ? shallowest_idle_cpu : least_loaded_cpu;
>
> 3.14.X:
> find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
> return idlest;

The change below  will give a similar behavior than 3.18 for 3.14 and
we still match the condition if (new_cpu == -1 || new_cpu == cpu) in
order to go in the child level

--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -4151,7 +4151,7 @@ static int
 find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
 {
        unsigned long load, min_load = ULONG_MAX;
-       int idlest = -1;
+       int idlest = this_cpu;
        int i;

        /* Traverse only the allowed CPUs */

>
--
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/