On Tue, Aug 26, 2014 at 11:03:17AM +0100, Marc Zyngier wrote:
> In order to limit code duplication, convert the architecture specific
> handle_IRQ to use the generic __handle_domain_irq function.
>
> Signed-off-by: Marc Zyngier <marc.zyng...@arm.com>
> ---
> arch/arm64/Kconfig | 1 +
> arch/arm64/kernel/irq.c | 18 +-----------------
> 2 files changed, 2 insertions(+), 17 deletions(-)
>
> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> index fd4e81a..1f16ed9 100644
> --- a/arch/arm64/Kconfig
> +++ b/arch/arm64/Kconfig
> @@ -30,6 +30,7 @@ config ARM64
> select GENERIC_STRNCPY_FROM_USER
> select GENERIC_STRNLEN_USER
> select GENERIC_TIME_VSYSCALL
> + select HANDLE_DOMAIN_IRQ
> select HARDIRQS_SW_RESEND
> select HAVE_ARCH_AUDITSYSCALL
> select HAVE_ARCH_JUMP_LABEL
> diff --git a/arch/arm64/kernel/irq.c b/arch/arm64/kernel/irq.c
> index 0f08dfd..2c0e2a7 100644
> --- a/arch/arm64/kernel/irq.c
> +++ b/arch/arm64/kernel/irq.c
> @@ -48,23 +48,7 @@ int arch_show_interrupts(struct seq_file *p, int prec)
> */
> void handle_IRQ(unsigned int irq, struct pt_regs *regs)
> {
> - struct pt_regs *old_regs = set_irq_regs(regs);
> -
> - irq_enter();
> -
> - /*
> - * Some hardware gives randomly wrong interrupts. Rather
> - * than crashing, do something sensible.
> - */
> - if (unlikely(irq >= nr_irqs)) {
> - pr_warn_ratelimited("Bad IRQ%u\n", irq);
> - ack_bad_irq(irq);
> - } else {
> - generic_handle_irq(irq);
> - }
> -
> - irq_exit();
> - set_irq_regs(old_regs);
> + __handle_domain_irq(NULL, irq, false, regs);
> }
The only thing that's missing is a pr_warn_ratelimited(). Do we still
need it? We could add it to ack_bad_irq() though.
Either way:
Acked-by: Catalin Marinas <catalin.mari...@arm.com>
_______________________________________________
Linux mailing list
Linux@lists.openrisc.net
http://lists.openrisc.net/listinfo/linux
