diff mbox series

[v9,24/26] arm64: Skip preemption when exiting an NMI

Message ID 1548084825-8803-25-git-send-email-julien.thierry@arm.com (mailing list archive)
State New, archived
Headers show
Series arm64: provide pseudo NMI with GICv3 | expand

Commit Message

Julien Thierry Jan. 21, 2019, 3:33 p.m. UTC
Handling of an NMI should not set any TIF flags. For NMIs received from
EL0 the current exit path is safe to use.

However, an NMI received at EL1 could have interrupted some task context
that has set the TIF_NEED_RESCHED flag. Preempting a task should not
happen as a result of an NMI.

Skip preemption after handling an NMI from EL1.

Signed-off-by: Julien Thierry <julien.thierry@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Marc Zyngier <marc.zyngier@arm.com>
---
 arch/arm64/kernel/entry.S | 8 ++++++++
 1 file changed, 8 insertions(+)

Comments

Marc Zyngier Jan. 28, 2019, 12:34 p.m. UTC | #1
On Mon, 21 Jan 2019 15:33:43 +0000,
Julien Thierry <julien.thierry@arm.com> wrote:
> 
> Handling of an NMI should not set any TIF flags. For NMIs received from
> EL0 the current exit path is safe to use.
> 
> However, an NMI received at EL1 could have interrupted some task context
> that has set the TIF_NEED_RESCHED flag. Preempting a task should not
> happen as a result of an NMI.
> 
> Skip preemption after handling an NMI from EL1.
> 
> Signed-off-by: Julien Thierry <julien.thierry@arm.com>
> Cc: Catalin Marinas <catalin.marinas@arm.com>
> Cc: Will Deacon <will.deacon@arm.com>
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> ---
>  arch/arm64/kernel/entry.S | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S
> index 35a47f6..a0b0a22 100644
> --- a/arch/arm64/kernel/entry.S
> +++ b/arch/arm64/kernel/entry.S
> @@ -624,6 +624,14 @@ el1_irq:
>  
>  #ifdef CONFIG_PREEMPT
>  	ldr	x24, [tsk, #TSK_TI_PREEMPT]	// get preempt count
> +alternative_if ARM64_HAS_IRQ_PRIO_MASKING
> +	/*
> +	 * DA_F were cleared at start of handling. If anything is set in DAIF,
> +	 * we come back from an NMI, so skip preemption
> +	 */
> +	mrs	x0, daif
> +	orr	x24, x24, x0
> +alternative_else_nop_endif
>  	cbnz	x24, 1f				// preempt count != 0
>  	bl	el1_preempt
>  1:

I find this a bit ugly, as what we have in x24 is not the preempt
count anymore. Maybe amend the comment above?

The code being nonetheless correct:

Acked-by: Marc Zyngier <marc.zyngier@arm.com>

	M.
diff mbox series

Patch

diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S
index 35a47f6..a0b0a22 100644
--- a/arch/arm64/kernel/entry.S
+++ b/arch/arm64/kernel/entry.S
@@ -624,6 +624,14 @@  el1_irq:
 
 #ifdef CONFIG_PREEMPT
 	ldr	x24, [tsk, #TSK_TI_PREEMPT]	// get preempt count
+alternative_if ARM64_HAS_IRQ_PRIO_MASKING
+	/*
+	 * DA_F were cleared at start of handling. If anything is set in DAIF,
+	 * we come back from an NMI, so skip preemption
+	 */
+	mrs	x0, daif
+	orr	x24, x24, x0
+alternative_else_nop_endif
 	cbnz	x24, 1f				// preempt count != 0
 	bl	el1_preempt
 1: