[tip:irq/urgent] irqchip/gic-v3-its: Don't bind LPI to unavailable NUMA node

From: tip-bot for Yang Yingliang
Date: Fri Jun 22 2018 - 08:26:54 EST


Commit-ID: c1797b11a09c8323c92b074fd48b89a936c991d0
Gitweb: https://git.kernel.org/tip/c1797b11a09c8323c92b074fd48b89a936c991d0
Author: Yang Yingliang <yangyingliang@xxxxxxxxxx>
AuthorDate: Fri, 22 Jun 2018 10:52:51 +0100
Committer: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
CommitDate: Fri, 22 Jun 2018 14:22:01 +0200

irqchip/gic-v3-its: Don't bind LPI to unavailable NUMA node

On a NUMA system, if an ITS is local to an offline node, the ITS driver may
pick an offline CPU to bind the LPI. In this case, pick an online CPU (and
the first one will do).

But on some systems, binding an LPI to non-local node CPU may cause
deadlock (see Cavium erratum 23144). In this case, just fail the activate
and return an error code.

Signed-off-by: Yang Yingliang <yangyingliang@xxxxxxxxxx>
Signed-off-by: Marc Zyngier <marc.zyngier@xxxxxxx>
Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Jason Cooper <jason@xxxxxxxxxxxxxx>
Cc: Alexandre Belloni <alexandre.belloni@xxxxxxxxxxx>
Cc: Sumit Garg <sumit.garg@xxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx
Link: https://lkml.kernel.org/r/20180622095254.5906-5-marc.zyngier@xxxxxxx

---
drivers/irqchip/irq-gic-v3-its.c | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c
index 5377d7e2afba..cae53937feeb 100644
--- a/drivers/irqchip/irq-gic-v3-its.c
+++ b/drivers/irqchip/irq-gic-v3-its.c
@@ -2310,7 +2310,14 @@ static int its_irq_domain_activate(struct irq_domain *domain,
cpu_mask = cpumask_of_node(its_dev->its->numa_node);

/* Bind the LPI to the first possible CPU */
- cpu = cpumask_first(cpu_mask);
+ cpu = cpumask_first_and(cpu_mask, cpu_online_mask);
+ if (cpu >= nr_cpu_ids) {
+ if (its_dev->its->flags & ITS_FLAGS_WORKAROUND_CAVIUM_23144)
+ return -EINVAL;
+
+ cpu = cpumask_first(cpu_online_mask);
+ }
+
its_dev->event_map.col_map[event] = cpu;
irq_data_update_effective_affinity(d, cpumask_of(cpu));