Re: [BUG][linux-next][ppc] kernel panic - not syncing: stack-protector: Kernel stack is corrupted in: __schedule+0x978/0xa80

[Christophe LEROY <christophe.leroy@c-s.fr>]

Le 12/10/2018 à 11:02, Aneesh Kumar K.V a écrit :
> On 10/12/18 1:38 PM, Christophe LEROY wrote:
>>
>>
>> Le 12/10/2018 à 09:48, Abdul Haleem a écrit :
>>> On Fri, 2018-10-12 at 13:15 +0530, Abdul Haleem wrote:
>>>> Greeting's
>>>>
>>>> Today's linux-next fails to boot on powerpc bare-metal with this error
>>>>
>>>> POWER8 performance monitor hardware support registered
>>>> rcu: Hierarchical SRCU implementation.
>>>> smp: Bringing up secondary CPUs ...
>>>> Kernel panic - not syncing: stack-protector: Kernel stack is 
>>>> corrupted in: __schedule+0x978/0xa80
>>>> CPU: 1 PID: 0 Comm: swapper/1 Not tainted 
>>>> 4.19.0-rc7-next-20181010-autotest-autotest #1
>>>> Call Trace:
>>>> [c000001fed5b3bf0] [c000000000a0ef3c] dump_stack+0xb0/0xf4 (unreliable)
>>>> [c000001fed5b3c30] [c0000000000f9d68] panic+0x140/0x308
>>>> [c000001fed5b3cc0] [c0000000000f9844] __stack_chk_fail+0x24/0x30
>>>> [c000001fed5b3d20] [c000000000a2c3a8] __schedule+0x978/0xa80
>>>> [c000001fed5b3e00] [c000000000a2c9b4] schedule_idle+0x34/0x60
>>>> [c000001fed5b3e30] [c00000000013d344] do_idle+0x224/0x3d0
>>>> [c000001fed5b3ec0] [c00000000013d6e0] cpu_startup_entry+0x30/0x50
>>>> [c000001fed5b3ef0] [c000000000047f34] start_secondary+0x4d4/0x520
>>>> [c000001fed5b3f90] [c00000000000b370] start_secondary_prolog+0x10/0x14
>>>> Rebooting in 10 seconds..
>>>>
>>>> Machine: Power 8 bare-metal
>>>> kernel version: 4.19.0-rc7-next-20181010
>>>> gcc version: 4.8.5 20150623
>>>> config attach
>>>
>>> Attaching the kernel config file
>>>
>>
>> # Linux/powerpc 4.11.0-rc4 Kernel Configuration
>>
>> This is not the correct config file. Can you send the .config ?
>>
>> Christophe
> 
> modified   kernel/sched/idle.c
> @@ -352,7 +352,6 @@ void cpu_startup_entry(enum cpuhp_state state)
>        * make this generic (ARM and SH have never invoked the canary
>        * init for the non boot CPUs!). Will be fixed in 3.11
>        */
> -#ifdef CONFIG_X86
>       /*
>        * If we're the non-boot CPU, nothing set the stack canary up
>        * for us. The boot CPU already has it initialized but no harm
> @@ -361,7 +360,6 @@ void cpu_startup_entry(enum cpuhp_state state)
>        * canaries already on the stack wont ever trigger).
>        */
>       boot_init_stack_canary();
> -#endif
>       arch_cpu_idle_prepare();
>       cpuhp_online_idle(state);
>       while (1)
> 
> 
> This fixes it for me. But we may have to look at the other arch details 
> mentioned there.

I think calling again boot_init_stack_canary() will be an issue for all 
arches using global canary.

As far as I can see, PPC and X86 are the only arches using non global.

Christophe