diff mbox series

[v2,2/2] arm64: Implement panic_smp_self_stop()

Message ID 20190617203519.328-2-aaro.koskinen@iki.fi (mailing list archive)
State New, archived
Headers show
Series [v2,1/2] arm64: Improve parking of stopped CPUs | expand

Commit Message

Aaro Koskinen June 17, 2019, 8:35 p.m. UTC
From: Aaro Koskinen <aaro.koskinen@nokia.com>

Currently arm64 uses the default implementation of panic_smp_self_stop()
where the CPU runs in a cpu_relax() loop unable to receive IPIs anymore.
As a result, when two CPUs panic() simultaneously we get "SMP: failed to
stop secondary CPUs" warnings and extra delays before a reset, because
smp_send_stop() still tries to stop the other paniced CPU.

Provide an implementation of panic_smp_self_stop() that is identical to
the IPI CPU stop handler, so that the online status of stopped CPUs gets
properly updated.

Signed-off-by: Aaro Koskinen <aaro.koskinen@nokia.com>
---

	v2: Update the commit log
	    Rename ipi_cpu_stop() to local_cpu_stop() and make it void func
	    Add a comment for panic_smp_self_stop()

	v1: https://patchwork.kernel.org/patch/10988103/

 arch/arm64/kernel/smp.c | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)

Comments

Will Deacon June 18, 2019, 12:51 p.m. UTC | #1
On Mon, Jun 17, 2019 at 11:35:19PM +0300, Aaro Koskinen wrote:
> From: Aaro Koskinen <aaro.koskinen@nokia.com>
> 
> Currently arm64 uses the default implementation of panic_smp_self_stop()
> where the CPU runs in a cpu_relax() loop unable to receive IPIs anymore.
> As a result, when two CPUs panic() simultaneously we get "SMP: failed to
> stop secondary CPUs" warnings and extra delays before a reset, because
> smp_send_stop() still tries to stop the other paniced CPU.
> 
> Provide an implementation of panic_smp_self_stop() that is identical to
> the IPI CPU stop handler, so that the online status of stopped CPUs gets
> properly updated.
> 
> Signed-off-by: Aaro Koskinen <aaro.koskinen@nokia.com>
> ---
> 
> 	v2: Update the commit log
> 	    Rename ipi_cpu_stop() to local_cpu_stop() and make it void func
> 	    Add a comment for panic_smp_self_stop()
> 
> 	v1: https://patchwork.kernel.org/patch/10988103/
> 
>  arch/arm64/kernel/smp.c | 19 +++++++++++++------
>  1 file changed, 13 insertions(+), 6 deletions(-)

Acked-by: Will Deacon <will.deacon@arm.com>

Will
diff mbox series

Patch

diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c
index 1a1b96a50245..1ac4aa34ffb6 100644
--- a/arch/arm64/kernel/smp.c
+++ b/arch/arm64/kernel/smp.c
@@ -845,18 +845,25 @@  void arch_irq_work_raise(void)
 }
 #endif
 
-/*
- * ipi_cpu_stop - handle IPI from smp_send_stop()
- */
-static void ipi_cpu_stop(unsigned int cpu)
+static void local_cpu_stop(void)
 {
-	set_cpu_online(cpu, false);
+	set_cpu_online(smp_processor_id(), false);
 
 	local_daif_mask();
 	sdei_mask_local_cpu();
 	cpu_park_loop();
 }
 
+/*
+ * We need to implement panic_smp_self_stop() for parallel panic() calls, so
+ * that cpu_online_mask gets correctly updated and smp_send_stop() can skip
+ * CPUs that have already stopped themselves.
+ */
+void panic_smp_self_stop(void)
+{
+	local_cpu_stop();
+}
+
 #ifdef CONFIG_KEXEC_CORE
 static atomic_t waiting_for_crash_ipi = ATOMIC_INIT(0);
 #endif
@@ -907,7 +914,7 @@  void handle_IPI(int ipinr, struct pt_regs *regs)
 
 	case IPI_CPU_STOP:
 		irq_enter();
-		ipi_cpu_stop(cpu);
+		local_cpu_stop();
 		irq_exit();
 		break;