Benjamin Herrenschmidt <b...@kernel.crashing.org> writes:
...
> diff --git a/arch/powerpc/platforms/powernv/opal-irqchip.c
> b/arch/powerpc/platforms/powernv/opal-irqchip.c
> index 9d1b8c0aaf93..46785eaf625d 100644
> --- a/arch/powerpc/platforms/powernv/opal-irqchip.c
> +++ b/arch/powerpc/platforms/powernv/opal-irqchip.c
> @@ -174,24 +175,21 @@ void opal_event_shutdown(void)
>
> /* First free interrupts, which will also mask them */
> for (i = 0; i < opal_irq_count; i++) {
> - if (!opal_irqs[i])
> + if (!opal_irqs || !opal_irqs[i].start)
> continue;
>
> if (in_interrupt())
> - disable_irq_nosync(opal_irqs[i]);
> + disable_irq_nosync(opal_irqs[i].start);
> else
> - free_irq(opal_irqs[i], NULL);
> -
> - opal_irqs[i] = 0;
This ^^^^^^^^^^^^^^
> + free_irq(opal_irqs[i].start, NULL);
> }
causes:
------------[ cut here ]------------
Trying to free already-free IRQ 22
WARNING: CPU: 0 PID: 1295 at ../kernel/irq/manage.c:1583 __free_irq+0xe0/0x420
Modules linked in:
CPU: 0 PID: 1295 Comm: init Tainted: G W
4.18.0-rc3-gcc-7.3.1-00187-g46a3659b3791-dirty #80
NIP: c00000000017ca90 LR: c00000000017ca8c CTR: c000000000771a90
REGS: c00000007552b810 TRAP: 0700 Tainted: G W
(4.18.0-rc3-gcc-7.3.1-00187-g46a3659b3791-dirty)
MSR: 9000000000029033 <SF,HV,EE,ME,IR,DR,RI,LE> CR: 28000222 XER: 20000000
CFAR: c000000000106f10 IRQMASK: 1
GPR00: c00000000017ca8c c00000007552ba90 c000000000fd1600 0000000000000022
GPR04: 0000000000000001 000000000000015e 9000000000009033 0000000000000000
GPR08: 000000007ef50000 c000000000ef0bf0 c000000000ef0bf0 9000000000001003
GPR12: 0000000000000000 c000000001160000 0000000000000000 0000000000000000
GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
GPR20: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
GPR24: 00000000100c0828 0000000000000000 0000000000000016 c000000077118094
GPR28: c000000077118148 0000000000000000 c000000077118000 0000000000000000
NIP [c00000000017ca90] __free_irq+0xe0/0x420
LR [c00000000017ca8c] __free_irq+0xdc/0x420
Call Trace:
[c00000007552ba90] [c00000000017ca8c] __free_irq+0xdc/0x420 (unreliable)
[c00000007552bb30] [c00000000017ced8] free_irq+0x78/0xe0
[c00000007552bb60] [c0000000000a7c48] opal_event_shutdown+0x158/0x160
[c00000007552bbf0] [c00000000009d150] pnv_prepare_going_down+0x20/0x80
[c00000007552bc10] [c00000000009d1d4] pnv_power_off+0x24/0x70
[c00000007552bc40] [c00000000009d240] pnv_restart+0x0/0x70
[c00000007552bc60] [c00000000002c2a0] machine_halt+0x60/0x70
[c00000007552bc80] [c000000000139f74] kernel_halt+0x84/0xa0
[c00000007552bce0] [c00000000013a36c] sys_reboot+0x28c/0x2c0
[c00000007552be30] [c00000000000b9e4] system_call+0x5c/0x70
Instruction dump:
e93f0008 7fa9e840 419e0098 7feafb78 ebea0018 2fbf0000 409effe8 3c62ffd0
7f44d378 386326c8 4bf8a421 60000000 <0fe00000> 7f24cb78 7f63db78 48952bdd
---[ end trace 926f8007a9952304 ]---
Because we're calling opal_event_shutdown() twice, via opal_shutdown()
and then pnv_prepare_going_down().
Did you drop that line on purpose, or was it just collateral damage?
The obvious fix does work:
+ opal_irqs[i].start = 0;
cheers
