Re: [PATCH] Oprofile: Loongson: Fixup of irq handler

[Wu Zhangjin <wuzhangjin@gmail.com>]

Hi, Ralf

This one is urgent, could you please review it, thanks ;)

Regards,
	Wu Zhangjin

On Fri, 2010-05-07 at 00:59 +0800, Wu Zhangjin wrote:
> The interrupt enable bit of performance counters of Loongson is in the
> control register($24), not in the counter register, so, in
> loongson2_perfcount_handler(), we need to use
> 
> 	enabled = read_c0_perfctrl() & LOONGSON2_PERFCNT_INT_EN;
> 
> instead of
> 
> 	enabled = read_c0_perfcnt() & LOONGSON2_PERFCNT_INT_EN;
> 
> Reported-by: Xu Hengyang <hengyang@mail.ustc.edu.cn>
> Signed-off-by: Wu Zhangjin <wuzhangjin@gmail.com>
> ---
>  arch/mips/oprofile/op_model_loongson2.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/arch/mips/oprofile/op_model_loongson2.c b/arch/mips/oprofile/op_model_loongson2.c
> index 29e2326..fa3bf66 100644
> --- a/arch/mips/oprofile/op_model_loongson2.c
> +++ b/arch/mips/oprofile/op_model_loongson2.c
> @@ -122,7 +122,7 @@ static irqreturn_t loongson2_perfcount_handler(int irq, void *dev_id)
>  	 */
>  
>  	/* Check whether the irq belongs to me */
> -	enabled = read_c0_perfcnt() & LOONGSON2_PERFCNT_INT_EN;
> +	enabled = read_c0_perfctrl() & LOONGSON2_PERFCNT_INT_EN;
>  	if (!enabled)
>  		return IRQ_NONE;
>  	enabled = reg.cnt1_enabled | reg.cnt2_enabled;