This is a note to let you know that I've just added the patch titled xen/events: avoid handling the same event on two cpus at the same time to the 4.4-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary The filename of the patch is: xen-events-avoid-handling-the-same-event-on-two-cpus-at-the-same-time.patch and it can be found in the queue-4.4 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable@xxxxxxxxxxxxxxx> know about it. >From foo@baz Mon Mar 15 01:16:07 PM CET 2021 From: Juergen Gross <jgross@xxxxxxxx> Date: Mon, 15 Mar 2021 10:06:31 +0100 Subject: xen/events: avoid handling the same event on two cpus at the same time From: Juergen Gross <jgross@xxxxxxxx> commit b6622798bc50b625a1e62f82c7190df40c1f5b21 upstream. When changing the cpu affinity of an event it can happen today that (with some unlucky timing) the same event will be handled on the old and the new cpu at the same time. Avoid that by adding an "event active" flag to the per-event data and call the handler only if this flag isn't set. Cc: stable@xxxxxxxxxxxxxxx Reported-by: Julien Grall <julien@xxxxxxx> Signed-off-by: Juergen Gross <jgross@xxxxxxxx> Reviewed-by: Julien Grall <jgrall@xxxxxxxxxx> Link: https://lore.kernel.org/r/20210306161833.4552-4-jgross@xxxxxxxx Signed-off-by: Boris Ostrovsky <boris.ostrovsky@xxxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- drivers/xen/events/events_base.c | 25 +++++++++++++++++-------- drivers/xen/events/events_internal.h | 1 + 2 files changed, 18 insertions(+), 8 deletions(-) --- a/drivers/xen/events/events_base.c +++ b/drivers/xen/events/events_base.c @@ -702,6 +702,12 @@ static void xen_evtchn_close(unsigned in BUG(); } +static void event_handler_exit(struct irq_info *info) +{ + smp_store_release(&info->is_active, 0); + clear_evtchn(info->evtchn); +} + static void pirq_query_unmask(int irq) { struct physdev_irq_status_query irq_status; @@ -732,13 +738,13 @@ static void eoi_pirq(struct irq_data *da likely(!irqd_irq_disabled(data))) { do_mask(info, EVT_MASK_REASON_TEMPORARY); - clear_evtchn(evtchn); + event_handler_exit(info); irq_move_masked_irq(data); do_unmask(info, EVT_MASK_REASON_TEMPORARY); } else - clear_evtchn(evtchn); + event_handler_exit(info); if (pirq_needs_eoi(data->irq)) { rc = HYPERVISOR_physdev_op(PHYSDEVOP_eoi, &eoi); @@ -1573,6 +1579,8 @@ void handle_irq_for_port(evtchn_port_t p } info = info_for_irq(irq); + if (xchg_acquire(&info->is_active, 1)) + return; if (ctrl->defer_eoi) { info->eoi_cpu = smp_processor_id(); @@ -1749,13 +1757,13 @@ static void ack_dynirq(struct irq_data * likely(!irqd_irq_disabled(data))) { do_mask(info, EVT_MASK_REASON_TEMPORARY); - clear_evtchn(evtchn); + event_handler_exit(info); irq_move_masked_irq(data); do_unmask(info, EVT_MASK_REASON_TEMPORARY); } else - clear_evtchn(evtchn); + event_handler_exit(info); } static void mask_ack_dynirq(struct irq_data *data) @@ -1771,7 +1779,7 @@ static void lateeoi_ack_dynirq(struct ir if (VALID_EVTCHN(evtchn)) { do_mask(info, EVT_MASK_REASON_EOI_PENDING); - clear_evtchn(evtchn); + event_handler_exit(info); } } @@ -1782,7 +1790,7 @@ static void lateeoi_mask_ack_dynirq(stru if (VALID_EVTCHN(evtchn)) { do_mask(info, EVT_MASK_REASON_EXPLICIT); - clear_evtchn(evtchn); + event_handler_exit(info); } } @@ -1891,10 +1899,11 @@ static void restore_cpu_ipis(unsigned in /* Clear an irq's pending state, in preparation for polling on it */ void xen_clear_irq_pending(int irq) { - int evtchn = evtchn_from_irq(irq); + struct irq_info *info = info_for_irq(irq); + evtchn_port_t evtchn = info ? info->evtchn : 0; if (VALID_EVTCHN(evtchn)) - clear_evtchn(evtchn); + event_handler_exit(info); } EXPORT_SYMBOL(xen_clear_irq_pending); void xen_set_irq_pending(int irq) --- a/drivers/xen/events/events_internal.h +++ b/drivers/xen/events/events_internal.h @@ -40,6 +40,7 @@ struct irq_info { #define EVT_MASK_REASON_EXPLICIT 0x01 #define EVT_MASK_REASON_TEMPORARY 0x02 #define EVT_MASK_REASON_EOI_PENDING 0x04 + u8 is_active; /* Is event just being handled? */ unsigned irq; unsigned int evtchn; /* event channel */ unsigned short cpu; /* cpu bound */ Patches currently in stable-queue which might be from jgross@xxxxxxxx are queue-4.4/xen-events-don-t-unmask-an-event-channel-when-an-eoi-is-pending.patch queue-4.4/xen-events-reset-affinity-of-2-level-event-when-tearing-it-down.patch queue-4.4/xen-events-avoid-handling-the-same-event-on-two-cpus-at-the-same-time.patch