powerpc/64s: Fix lost pending interrupt due to race causing lost update to irq_happened

force_external_irq_replay() can be called in the do_IRQ path with interrupts hard enabled and soft disabled if may_hard_irq_enable() set MSR[EE]=1. It updates local_paca->irq_happened with a load, modify, store sequence. If a maskable interrupt hits during this sequence, it will go to the masked handler to be marked pending in irq_happened. This update will be lost when the interrupt returns and the store instruction executes. This can result in unpredictable latencies, timeouts, lockups, etc.

Fix this by ensuring hard interrupts are disabled before modifying irq_happened.

This could cause any maskable asynchronous interrupt to get lost, but it was noticed on P9 SMP system doing RDMA NVMe target over 100GbE, so very high external interrupt rate and high IPI rate. The hang was bisected down to enabling doorbell interrupts for IPIs. These provided an interrupt type that could run at high rates in the do_IRQ path, stressing the race.

Fixes 1d607bb3bd ("powerpc/irq: Add mechanism to force a replay of interrupts")

Reported-by Carol L. Soto <[email protected]>

Cc Benjamin Herrenschmidt <[email protected]>

Signed-off-by Nicholas Piggin <[email protected]>

---

This has survived stress testing quite well so far, may need a little more testing but I'd like to post it now to get some more comments.

We can optimise the mtmsr a bit more (e.g., skip it if interrupts are already disabled or EE alrady set), but I've got some other patches pending which change things there slightly, so I prefer to have this minimal fix now, then make such changes upstream later.

---
 arch/powerpc/kernel/irq.c | 8 ++++++++
 1 file changed, 8 insertions(+)

arch/powerpc/kernel/irq.c +8 -0

index f88038847790..061aa0f47bb1 100644
@@ -476,6 +476,14 @@ void force_external_irq_replay(void)
 	 */
 	WARN_ON(!arch_irqs_disabled());
 
+	/*
+	 * Interrupts must always be hard disabled before irq_happened is
+	 * modified (to prevent lost update in case of interrupt between
+	 * load and store).
+	 */
+	__hard_irq_disable();
+	local_paca->irq_happened |= PACA_IRQ_HARD_DIS;
+
 	/* Indicate in the PACA that we have an interrupt to replay */
 	local_paca->irq_happened |= PACA_IRQ_EE;
 }

Quoted 19 lines · Nicholas Piggin

force_external_irq_replay() can be called in the do_IRQ path with interrupts hard enabled and soft disabled if may_hard_irq_enable() set MSR[EE]=1. It updates local_paca->irq_happened with a load, modify, store sequence. If a maskable interrupt hits during this sequence, it will go to the masked handler to be marked pending in irq_happened. This update will be lost when the interrupt returns and the store instruction executes. This can result in unpredictable latencies, timeouts, lockups, etc.

Fix this by ensuring hard interrupts are disabled before modifying irq_happened.

This could cause any maskable asynchronous interrupt to get lost, but it was noticed on P9 SMP system doing RDMA NVMe target over 100GbE, so very high external interrupt rate and high IPI rate. The hang was bisected down to enabling doorbell interrupts for IPIs. These provided an interrupt type that could run at high rates in the do_IRQ path, stressing the race.

Quoted 5 lines

Fixes 1d607bb3bd ("powerpc/irq: Add mechanism to force a replay of interrupts")

Reported-by Carol L. Soto <[email protected]>

Cc Benjamin Herrenschmidt <[email protected]>

Signed-off-by Nicholas Piggin <[email protected]>

---

Nice one. We need that back into the distros asap.

Yes Carol is working on this with distros. Backporting should be simple, so hoping to get this upstream soon to help the process. We can take that as a Reviewed-by:? :)

force_external_irq_replay() can be called in the do_IRQ path with interrupts hard enabled and soft disabled if may_hard_irq_enable() set MSR[EE]=1. It updates local_paca->irq_happened with a load, modify, store sequence. If a maskable interrupt hits during this sequence, it will go to the masked handler to be marked pending in irq_happened. This update will be lost when the interrupt returns and the store instruction executes. This can result in unpredictable latencies, timeouts, lockups, etc.

Fix this by ensuring hard interrupts are disabled before modifying irq_happened.

This could cause any maskable asynchronous interrupt to get lost, but it was noticed on P9 SMP system doing RDMA NVMe target over 100GbE, so very high external interrupt rate and high IPI rate. The hang was bisected down to enabling doorbell interrupts for IPIs. These provided an interrupt type that could run at high rates in the do_IRQ path, stressing the race.

Fixes 1d607bb3bd ("powerpc/irq: Add mechanism to force a replay of interrupts")

Reported-by Carol L. Soto <[email protected]>

Cc Benjamin Herrenschmidt <[email protected]>

Signed-off-by Nicholas Piggin <[email protected]>