[PATCH v2 10/11] mm/memory-failure: increase queued recovery work's priority

[James Morse]

arm64 can take an NMI-like error notification when user-space steps in
some corrupt memory. APEI's GHES code will call memory_failure_queue()
to schedule the recovery work. We then return to user-space, possibly
taking the fault again.

Currently the arch code unconditionally signals user-space from this
path, so we don't get stuck in this loop, but the affected process
never benefits from memory_failure()s recovery work. To fix this we
need to know the recovery work will run before we get back to user-space.

Increase the priority of the recovery work by scheduling it on the
system_highpri_wq, then try to bump the current task off this CPU
so that the recover work starts immediately.

Reported-by: Xie XiuQi <xiexiuqi at huawei.com>
Signed-off-by: James Morse <james.morse at arm.com>
Reviewed-by: Punit Agrawal <punit.agrawal at arm.com>
CC: Xie XiuQi <xiexiuqi at huawei.com>
CC: gengdongjiu <gengdongjiu at huawei.com>
---
It is possible that the task will migrate to another CPU which will take
RAS error again until memory_failure() has completed its work. I don't
think this is any different from a process being scheduled while
memory_failure() is running, which we should tackle separately.

 mm/memory-failure.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 4b80ccee4535..14f44d841e8b 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -55,6 +55,7 @@
 #include <linux/hugetlb.h>
 #include <linux/memory_hotplug.h>
 #include <linux/mm_inline.h>
+#include <linux/preempt.h>
 #include <linux/kfifo.h>
 #include <linux/ratelimit.h>
 #include "internal.h"
@@ -1319,6 +1320,7 @@ static DEFINE_PER_CPU(struct memory_failure_cpu, memory_failure_cpu);
  */
 void memory_failure_queue(unsigned long pfn, int flags)
 {
+	int cpu = smp_processor_id();
 	struct memory_failure_cpu *mf_cpu;
 	unsigned long proc_flags;
 	struct memory_failure_entry entry = {
@@ -1328,11 +1330,14 @@ void memory_failure_queue(unsigned long pfn, int flags)
 
 	mf_cpu = &get_cpu_var(memory_failure_cpu);
 	spin_lock_irqsave(&mf_cpu->lock, proc_flags);
-	if (kfifo_put(&mf_cpu->fifo, entry))
-		schedule_work_on(smp_processor_id(), &mf_cpu->work);
-	else
+	if (kfifo_put(&mf_cpu->fifo, entry)) {
+		queue_work_on(cpu, system_highpri_wq, &mf_cpu->work);
+		set_tsk_need_resched(current);
+		preempt_set_need_resched();
+	} else {
 		pr_err("Memory failure: buffer overflow when queuing memory failure at %#lx\n",
 		       pfn);
+	}
 	spin_unlock_irqrestore(&mf_cpu->lock, proc_flags);
 	put_cpu_var(memory_failure_cpu);
 }
-- 
2.16.2