[Cake] cake infinite loop(?) with hfsc on one-armed router

[Pete Heist]

Ok, that fixes the compiler warnings, but I get this now. Same as before it’s repeated until reboot, stack sometimes changes but I always see sch_hfsc.c:1427 at the beginning:

root at apu1a:~# [ 5972.967008] ------------[ cut here ]------------
[ 5972.971707] WARNING: CPU: 1 PID: 0 at net/sched/sch_hfsc.c:1427 0xfa02f4ef()
[ 5972.978812] Modules linked in: sch_cake(O) em_meta cls_basic sch_hfsc xt_ACCOUNT(O) ipt_REJECT xt_recent iptable_mangle iptable_nat nf]
[ 5973.032173] CPU: 1 PID: 0 Comm: swapper/1 Tainted: G           O  3.16.7-ckt9-voyage #1
[ 5973.040181] Hardware name: PC Engines APU/APU, BIOS 4.0 09/08/2014
[ 5973.046368]  00000000 00000000 f5c99d94 c13167e9 00000000 f5c99dac c102a7dd fa02f4ef
[ 5973.054323]  f06b6c00 00000000 00000000 f5c99dbc c102a803 00000009 00000000 f5c99ddc
[ 5973.062279]  fa02f4ef f06b6fc8 b48b3e3f 00000015 f06b6c00 00000000 00000040 f5c99df8
[ 5973.070226] Call Trace:
[ 5973.072703]  [<c13167e9>] dump_stack+0x41/0x52
[ 5973.077160]  [<c102a7dd>] warn_slowpath_common+0x5c/0x73
[ 5973.082485]  [<fa02f4ef>] ? 0xfa02f4ee
[ 5973.086249]  [<c102a803>] warn_slowpath_null+0xf/0x13
[ 5973.091308]  [<fa02f4ef>] 0xfa02f4ee
[ 5973.094899]  [<c129edf2>] __qdisc_run+0x81/0xf0
[ 5973.099441]  [<c128b655>] __dev_queue_xmit+0x23d/0x35f
[ 5973.104594]  [<c128b78b>] dev_queue_xmit+0xa/0xc
[ 5973.109224]  [<c12afa93>] ip_finish_output+0x345/0x73d
[ 5973.114372]  [<c12b0e19>] ip_output+0x73/0xaf
[ 5973.118741]  [<c12ad986>] ip_forward_finish+0x66/0x6b
[ 5973.123801]  [<c12adc3b>] ip_forward+0x2b0/0x36d
[ 5973.128429]  [<c12ac467>] ip_rcv_finish+0x267/0x29a
[ 5973.133317]  [<c12aca4c>] ip_rcv+0x2b4/0x338
[ 5973.137601]  [<c12895dd>] __netif_receive_skb_core+0x467/0x4b6
[ 5973.143450]  [<c1289674>] __netif_receive_skb+0x48/0x59
[ 5973.148693]  [<c1289cb9>] netif_receive_skb_internal+0x59/0x85
[ 5973.154533]  [<c1289d6c>] napi_gro_complete+0x87/0x8c
[ 5973.159594]  [<c128a020>] napi_gro_flush+0x3e/0x53
[ 5973.164395]  [<c128a04c>] napi_complete+0x17/0x27
[ 5973.169114]  [<f80841a3>] 0xf80841a2
[ 5973.172709]  [<c128a0b2>] net_rx_action+0x56/0x10e
[ 5973.177517]  [<c102d689>] __do_softirq+0x91/0x175
[ 5973.182233]  [<c102d5f8>] ? __hrtimer_tasklet_trampoline+0x1a/0x1a
[ 5973.188421]  [<c10033c3>] do_softirq_own_stack+0x1d/0x23
[ 5973.193744]  <IRQ>  [<c102d8a9>] irq_exit+0x34/0x75
[ 5973.198679]  [<c1002f30>] do_IRQ+0x92/0xa6
[ 5973.202789]  [<c131a4ec>] common_interrupt+0x2c/0x40
[ 5973.207765]  [<c126add1>] ? cpuidle_enter_state+0x37/0x96
[ 5973.213179]  [<c126aee8>] cpuidle_enter+0xf/0x12
[ 5973.217810]  [<c1051e54>] cpu_startup_entry+0x135/0x1e1
[ 5973.223053]  [<c101d553>] start_secondary+0x1a6/0x1ab
[ 5973.228111] ---[ end trace fe3dc3fad8d2493c ]---


> On Jan 5, 2019, at 1:38 PM, Toke Høiland-Jørgensen <toke at toke.dk> wrote:
> 
> Pete Heist <pete at heistp.net> writes:
> 
>> Quick update to the trace because I had to apply the patch manually
>> and missed one line to remove (qdisc_tree_reduce_backlog...), just so
>> it doesn’t through off the addresses for you, but it still does the
>> same thing:
> 
> Ah, my bad; you need to replace
> 
> qdisc_match_from_root(qdisc_dev(sch), TC_H_MAJ(parentid));
> 
> with
> 
> qdisc_match_from_root(qdisc_dev(sch)->qdisc, TC_H_MAJ(parentid));
> 
> -Toke