Re: Subject: [tip:irq/genirq] genirq: assert that irq handlers are indeed running in hardirq context

[David Brownell <david-b@pacbell.net>]

> Commit-ID:  044d408409cc4e1bc75c886e27ca85c270db104c
> Gitweb:    
> http://git.kernel.org/tip/044d408409cc4e1bc75c886e27ca85c270db104c Author: 
>    Peter Zijlstra <a.p.zijlstra@chello.nl>
> AuthorDate: Mon, 2 Mar 2009 16:13:32 +0100
> Commit:     Ingo Molnar <mingo@elte.hu>
> CommitDate: Tue, 3 Mar 2009 00:05:45 +0100
>
> genirq: assert that irq handlers are indeed running in hardirq context
>
> Make sure the genirq layer handlers are indeed running handlers
> in hardirq context. That is the genirq expectation and doing
> anything else is broken.

Actually, it's more like Peter Zijlstra's expectation.
Genirq works perfectly well without this.

And ... this causes regressions, so I suggest it get
reverted ASAP.

http://marc.info/?l=linux-kernel&m=124082644212584&w=2


>
> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> LKML-Reference: <1236006812.5330.632.camel@laptop>
> Signed-off-by: Ingo Molnar <mingo@elte.hu>
>
>
> ---
>  kernel/irq/handle.c |    2 ++
>  1 files changed, 2 insertions(+), 0 deletions(-)
>
> diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
> index 3aba8d1..a2ee682 100644
> --- a/kernel/irq/handle.c
> +++ b/kernel/irq/handle.c
> @@ -328,6 +328,8 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct
> irqaction *action) irqreturn_t ret, retval = IRQ_NONE;
>  	unsigned int status = 0;
>
> +	WARN_ONCE(!in_irq(), "BUG: IRQ handler called from non-hardirq context!");

Notice also:  clearly untested.  No newline on the message.

> + 
>  	if (!(action->flags & IRQF_DISABLED))
>  		local_irq_enable_in_hardirq();