[v6][PATCH 5/5] powerpc/book3e/kgdb: Fix a single stgep case of lazy IRQ

[Scott Wood]

On Wed, 2013-10-23 at 17:31 +0800, Tiejun Chen wrote:
> In lazy EE magic, we may have a lazy interrupt occured while
> entering kgdb, but we really don't want to replay that interrupt
> for kgdb, so we have to clear the PACA_IRQ_HARD_DIS force to
> make sure we can exit directly from this debug exception.
> 
> Signed-off-by: Tiejun Chen <tiejun.chen at windriver.com>

s/stgep/step/ in subject

> ---
>  arch/powerpc/kernel/kgdb.c |    8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/arch/powerpc/kernel/kgdb.c b/arch/powerpc/kernel/kgdb.c
> index 447c14b..9872f58 100644
> --- a/arch/powerpc/kernel/kgdb.c
> +++ b/arch/powerpc/kernel/kgdb.c
> @@ -185,6 +185,14 @@ static int kgdb_singlestep(struct pt_regs *regs)
>  		/* Restore current_thread_info lastly. */
>  		memcpy(exception_thread_info, backup_current_thread_info, sizeof *thread_info);
>  
> +#ifdef CONFIG_PPC64
> +	/*
> +	 * Clear the PACA_IRQ_HARD_DIS from the pending mask
> +	 * since we are about to exit this directly from debug
> +	 * exception without any replay interrupt in lazy EE case.
> +	 */
> +	local_paca->irq_happened &= ~PACA_IRQ_HARD_DIS;
> +#endif
>  	return 1;
>  }
>  

What happens to those interrupts you discarded once we get back to a
state when they can be safely replayed?  I don't think just dropping
them is the answer.

I'm not sure what the actual problem is.  I can understand not wanting
kgdb to cause interrupts to appear to run when the interrupted context
has external interrupts disabled, but the replay code in entry_64.S
doesn't run if interrupts are soft-disabled in the context to be
returned to.  What harm does it cause to run the interrupts if we're
returning to an EE=1 context?

Does KGDB enable interrupts in its handler?  If not, how do we even get
into the situation where there are interrupts pending when the
interrupted context has EE soft-enabled (i.e. we went directly from a
context where the interrupt handler should have run, to a hard-disabled
context)?

-Scott