Around Fri 26 Jun 2015 12:47:18 -0700 or thereabout, tip-bot for Thomas Gleixner wrote:
> avr32/at32ap: Fix race in installing chained IRQ handler
>
<snipp>
> Reported-by: Russell King <[email protected]>
> Signed-off-by: Thomas Gleixner <[email protected]>
> Cc: Julia Lawall <[email protected]>
> Cc: Haavard Skinnemoen <[email protected]>
> Cc: Hans-Christian Egtvedt <[email protected]>
> ---
> arch/avr32/mach-at32ap/extint.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/arch/avr32/mach-at32ap/extint.c b/arch/avr32/mach-at32ap/extint.c
> index cfb298d..2d48b6a 100644
> --- a/arch/avr32/mach-at32ap/extint.c
> +++ b/arch/avr32/mach-at32ap/extint.c
> @@ -231,8 +231,7 @@ static int __init eic_probe(struct platform_device *pdev)
> irq_set_chip_data(eic->first_irq + i, eic);
> }
>
> - irq_set_chained_handler(int_irq, demux_eic_irq);
> - irq_set_handler_data(int_irq, eic);
> + irq_set_chained_handler_and_data(int_irq, demux_eic_irq, eic);
>
> if (pdev->id == 0) {
> nmi_eic = eic;
Good fix, wil you pass it through your git tree, or would you prefer I add it
to the AVR32 tree? I have nothing in the pipe, hence feel free to add it to a
series if you are preparing fixes.
Acked-by: Hans-Christian Egtvedt <[email protected]>
--
mvh
Hans-Christian Egtvedt