hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I6WPFT
--------------------------------
Recently, We have a LPI migration issue on the ARM SMP platform.
For example, NIC device generates MSI and sends LPI to CPU0 via ITS, meanwhile irqbalance running on CPU1 set irq affinity of NIC to CPU1, the next interrupt will be sent to CPU2, due to the state of irq is still in progress, kernel does not end up performing irq handler on CPU2, which results in some userland service timeouts, the sequence of events is shown as follows:
NIC CPU0 CPU1
Generate IRQ#1 READ_IAR Lock irq_desc Set IRQD_IN_PROGRESS Unlock irq_desc Lock irq_desc Change LPI Affinity Unlock irq_desc Call irq_handler Generate IRQ#2 READ_IAR Lock irq_desc Check IRQD_IN_PROGRESS Unlock irq_desc Return from interrupt#2 Lock irq_desc Clear IRQD_IN_PROGRESS Unlock irq_desc return from interrupt#1
For this scenario, The IRQ#2 will be lost. This does cause some exceptions.
For further information, see [1].
This patch introduced a new flow handler which combines fasteoi and edge type as a workaround. An additional loop will be executed if the IRQS_PENDING has been setup.
Fixes: cc2d3216f53c ("irqchip: GICv3: ITS command queue") Signed-off-by: Yipeng Zou zouyipeng@huawei.com --- drivers/irqchip/irq-gic-v3.c | 2 +- include/linux/irq.h | 1 + kernel/irq/chip.c | 67 ++++++++++++++++++++++++++++++++++++ 3 files changed, 69 insertions(+), 1 deletion(-)
diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c index 8b2a45fbc9f7..9bf324aeabc7 100644 --- a/drivers/irqchip/irq-gic-v3.c +++ b/drivers/irqchip/irq-gic-v3.c @@ -1475,7 +1475,7 @@ static int gic_irq_domain_map(struct irq_domain *d, unsigned int irq, if (!gic_dist_supports_lpis()) return -EPERM; irq_domain_set_info(d, irq, hw, chip, d->host_data, - handle_fasteoi_irq, NULL, NULL); + handle_fasteoi_edge_irq, NULL, NULL); break;
default: diff --git a/include/linux/irq.h b/include/linux/irq.h index e79297d16785..9c15e550ddcf 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -667,6 +667,7 @@ static inline int irq_set_parent(int irq, int parent_irq) */ extern void handle_level_irq(struct irq_desc *desc); extern void handle_fasteoi_irq(struct irq_desc *desc); +extern void handle_fasteoi_edge_irq(struct irq_desc *desc); extern void handle_percpu_devid_fasteoi_ipi(struct irq_desc *desc); extern void handle_edge_irq(struct irq_desc *desc); extern void handle_edge_eoi_irq(struct irq_desc *desc); diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index bd31fff8de5a..b82dcb49861d 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -680,6 +680,73 @@ static void cond_unmask_eoi_irq(struct irq_desc *desc, struct irq_chip *chip) } }
+/** + * handle_fasteoi_edge_irq - irq handler for transparent controllers + * edge type IRQ. + * @desc: the interrupt description structure for this irq + */ +void handle_fasteoi_edge_irq(struct irq_desc *desc) +{ + struct irq_chip *chip = desc->irq_data.chip; + + raw_spin_lock(&desc->lock); + + if (!irq_may_run(desc)) { + desc->istate |= IRQS_PENDING; + mask_irq(desc); + goto out; + } + + desc->istate &= ~(IRQS_REPLAY | IRQS_WAITING); + + /* + * If its disabled or no action available + * then mask it and get out of here: + */ + if (unlikely(!desc->action || irqd_irq_disabled(&desc->irq_data))) { + desc->istate |= IRQS_PENDING; + mask_irq(desc); + goto out; + } + + kstat_incr_irqs_this_cpu(desc); + + if (desc->istate & IRQS_ONESHOT) + mask_irq(desc); + + do { + if (unlikely(!desc->action)) { + mask_irq(desc); + goto out; + } + + /* + * When another irq arrived while we were handling + * one, we could have masked the irq. + * Reenable it, if it was not disabled in meantime. + */ + if (unlikely(desc->istate & IRQS_PENDING)) { + if (!irqd_irq_disabled(&desc->irq_data) && + irqd_irq_masked(&desc->irq_data)) + unmask_irq(desc); + } + + handle_irq_event(desc); + + } while ((desc->istate & IRQS_PENDING) && + !irqd_irq_disabled(&desc->irq_data)); + + cond_unmask_eoi_irq(desc, chip); + + raw_spin_unlock(&desc->lock); + return; +out: + if (!(chip->flags & IRQCHIP_EOI_IF_HANDLED)) + chip->irq_eoi(&desc->irq_data); + raw_spin_unlock(&desc->lock); +} +EXPORT_SYMBOL_GPL(handle_fasteoi_edge_irq); + /** * handle_fasteoi_irq - irq handler for transparent controllers * @desc: the interrupt description structure for this irq