nvme-pci: fix race between poll and IRQ completions

If polling completions are racing with the IRQ triggered by a
completion, the IRQ handler will find no work and return IRQ_NONE.
This can trigger complaints about spurious interrupts:

[  560.169153] irq 630: nobody cared (try booting with the "irqpoll" option)
[  560.175988] CPU: 40 PID: 0 Comm: swapper/40 Not tainted 4.17.0-rc2+ #65
[  560.175990] Hardware name: Intel Corporation S2600STB/S2600STB, BIOS SE5C620.86B.00.01.0010.010920180151 01/09/2018
[  560.175991] Call Trace:
[  560.175994]  <IRQ>
[  560.176005]  dump_stack+0x5c/0x7b
[  560.176010]  __report_bad_irq+0x30/0xc0
[  560.176013]  note_interrupt+0x235/0x280
[  560.176020]  handle_irq_event_percpu+0x51/0x70
[  560.176023]  handle_irq_event+0x27/0x50
[  560.176026]  handle_edge_irq+0x6d/0x180
[  560.176031]  handle_irq+0xa5/0x110
[  560.176036]  do_IRQ+0x41/0xc0
[  560.176042]  common_interrupt+0xf/0xf
[  560.176043]  </IRQ>
[  560.176050] RIP: 0010:cpuidle_enter_state+0x9b/0x2b0
[  560.176052] RSP: 0018:ffffa0ed4659fe98 EFLAGS: 00000246 ORIG_RAX: ffffffffffffffdd
[  560.176055] RAX: ffff9527beb20a80 RBX: 000000826caee491 RCX: 000000000000001f
[  560.176056] RDX: 000000826caee491 RSI: 00000000335206ee RDI: 0000000000000000
[  560.176057] RBP: 0000000000000001 R08: 00000000ffffffff R09: 0000000000000008
[  560.176059] R10: ffffa0ed4659fe78 R11: 0000000000000001 R12: ffff9527beb29358
[  560.176060] R13: ffffffffa235d4b8 R14: 0000000000000000 R15: 000000826caed593
[  560.176065]  ? cpuidle_enter_state+0x8b/0x2b0
[  560.176071]  do_idle+0x1f4/0x260
[  560.176075]  cpu_startup_entry+0x6f/0x80
[  560.176080]  start_secondary+0x184/0x1d0
[  560.176085]  secondary_startup_64+0xa5/0xb0
[  560.176088] handlers:
[  560.178387] [<00000000efb612be>] nvme_irq [nvme]
[  560.183019] Disabling IRQ torvalds#630

A previous commit removed ->cqe_seen that was handling this case,
but we need to handle this a bit differently due to completions
now running outside the queue lock. Return IRQ_HANDLED from the
IRQ handler, if the completion ring head was moved since we last
saw it.

Fixes: 5cb525c ("nvme-pci: handle completions outside of the queue lock")
Reported-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Tested-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

drivers/nvme/host/pci.c

@@ -160,6 +160,7 @@ struct nvme_queue {
 	s16 cq_vector;
 	u16 sq_tail;
 	u16 cq_head;
+	u16 last_cq_head;
 	u16 qid;
 	u8 cq_phase;
 	u32 *dbbuf_sq_db;
@@ -999,16 +1000,22 @@ static inline bool nvme_process_cq(struct nvme_queue *nvmeq, u16 *start,
 static irqreturn_t nvme_irq(int irq, void *data)
 {
 	struct nvme_queue *nvmeq = data;
+	irqreturn_t ret = IRQ_NONE;
 	u16 start, end;
 
 	spin_lock(&nvmeq->cq_lock);
+	if (nvmeq->cq_head != nvmeq->last_cq_head)
+		ret = IRQ_HANDLED;
 	nvme_process_cq(nvmeq, &start, &end, -1);
+	nvmeq->last_cq_head = nvmeq->cq_head;
 	spin_unlock(&nvmeq->cq_lock);
 
-	if (start == end)
-		return IRQ_NONE;
-	nvme_complete_cqes(nvmeq, start, end);
-	return IRQ_HANDLED;
+	if (start != end) {
+		nvme_complete_cqes(nvmeq, start, end);
+		return IRQ_HANDLED;
+	}
+
+	return ret;
 }
 
 static irqreturn_t nvme_irq_check(int irq, void *data)