From 320e77acb32774b6addfe38d562956fad70c5f0d Mon Sep 17 00:00:00 2001 From: Suganath Prabu Date: Fri, 15 Feb 2019 02:40:27 -0500 Subject: [PATCH] scsi: mpt3sas: Irq poll to avoid CPU hard lockups Issue Description: We have seen cpu lock up issue from fields if system has greater (more than 96) logical cpu count. SAS3.0 controller (Invader series) supports at max 96 msix vector and SAS3.5 product (Ventura) supports at max 128 msix vectors. This may be a generic issue (if PCI device supports completion on multiple reply queues). Let me explain it w.r.t to mpt3sas supported h/w just to simplify the problem and possible changes to handle such issues. IT HBA (mpt3sas) supports multiple reply queues in completion path. Driver creates MSI-x vectors for controller as "min of (FW supported Reply queue, Logical CPUs)". If submitter is not interrupted via completion on same CPU, there is a loop in the IO path. This behavior can cause hard/soft CPU lockups, IO timeout, system sluggish etc. Example - one CPU (e.g. CPU A) is busy submitting the IOs and another CPU (e.g. CPU B) is busy with processing the corresponding IO's reply descriptors from reply descriptor queue upon receiving the interrupts from HBA. If the CPU A is continuously pumping the IOs then always CPU B (which is executing the ISR) will see the valid reply descriptors in the reply descriptor queue and it will be continuously processing those reply descriptor in a loop without quitting the ISR handler. Mpt3sas driver will exit ISR handler if it finds unused reply descriptor in the reply descriptor queue. Since CPU A will be continuously sending the IOs, CPU B may always see a valid reply descriptor (posted by HBA Firmware after processing the IO) in the reply descriptor queue. In worst case, driver will not quit from this loop in the ISR handler. Eventually, CPU lockup will be detected by watchdog. Above mentioned behavior is not common if "rq_affinity" set to 2 or affinity_hint is honored by irqbalance as "exact". If rq_affinity is set to 2, submitter will be always interrupted via completion on same CPU. If irqbalance is using "exact" policy, interrupt will be delivered to submitter CPU. If CPU counts to MSI-X vectors (reply descriptor Queues) count ratio is not 1:1, we still have exposure of issue explained above and for that we don't have any solution. Exposure of soft/hard lockup if CPU count is more than MSI-x supported by device. If CPUs count to MSI-x vectors count ratio is not 1:1, (Other way, if CPU counts to MSI-x vector count ratio is something like X:1, where X > 1) then 'exact' irqbalance policy OR rq_affinity = 2 won't help to avoid CPU hard/soft lockups. There won't be any one to one mapping between CPU to MSI-x vector instead one MSI-x interrupt (or reply descriptor queue) is shared with group/set of CPUs and there is a possibility of having a loop in the IO path within that CPU group and may observe lockups. For example: Consider a system having two NUMA nodes and each node having four logical CPUs and also consider that number of MSI-x vectors enabled on the HBA is two, then CPUs count to MSI-x vector count ratio as 4:1. e.g. MSIx vector 0 is affinity to CPU 0, CPU 1, CPU 2 & CPU 3 of NUMA node 0 and MSI-x vector 1 is affinity to CPU 4, CPU 5, CPU 6 & CPU 7 of NUMA node 1. numactl --hardware available: 2 nodes (0-1) node 0 cpus: 0 1 2 3 --> MSI-x 0 node 0 size: 65536 MB node 0 free: 63176 MB node 1 cpus: 4 5 6 7 -->MSI-x 1 node 1 size: 65536 MB node 1 free: 63176 MB Assume that user started an application which uses all the CPUs of NUMA node 0 for issuing the IOs. Only one CPU from affinity list (it can be any cpu since this behavior depends upon irqbalance) CPU0 will receive the interrupts from MSIx vector 0 for all the IOs. Eventually, CPU 0 IO submission percentage will be decreasing and ISR processing percentage will be increasing as it is more busy with processing the interrupts. Gradually IO submission percentage on CPU 0 will be zero and it's ISR processing percentage will be 100 percentage as IO loop has already formed within the NUMA node 0, i.e. CPU 1, CPU 2 & CPU 3 will be continuously busy with submitting the heavy IOs and only CPU 0 is busy in the ISR path as it always find the valid reply descriptor in the reply descriptor queue. Eventually, we will observe the hard lockup here. Chances of occurring of hard/soft lockups are directly proportional to value of X. If value of X is high, then chances of observing CPU lockups is high. Solution: Use IRQ poll interface defined in " irq_poll.c". mpt3sas driver will execute ISR routine in Softirq context and it will always quit the loop based on budget provided in IRQ poll interface. In these scenarios (i.e. where CPUs count to MSI-X vectors count ratio is X:1 (where X > 1)), IRQ poll interface will avoid CPU hard lockups due to voluntary exit from the reply queue processing based on budget. Note - Only one MSI-x vector is busy doing processing. Irqstat output: IRQs / 1 second(s) IRQ# TOTAL NODE0 NODE1 NODE2 NODE3 NAME 44 122871 122871 0 0 0 IR-PCI-MSI-edge mpt3sas0-msix0 45 0 0 0 0 0 IR-PCI-MSI-edge mpt3sas0-msix1 We use this approach only if cpu count is more than FW supported MSI-x vector Signed-off-by: Suganath Prabu Signed-off-by: Martin K. Petersen --- drivers/scsi/mpt3sas/Kconfig | 1 + drivers/scsi/mpt3sas/mpt3sas_base.c | 78 ++++++++++++++++++++++++++++- drivers/scsi/mpt3sas/mpt3sas_base.h | 8 +++ 3 files changed, 85 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/mpt3sas/Kconfig b/drivers/scsi/mpt3sas/Kconfig index b736dbc80485..a072187875df 100644 --- a/drivers/scsi/mpt3sas/Kconfig +++ b/drivers/scsi/mpt3sas/Kconfig @@ -45,6 +45,7 @@ config SCSI_MPT3SAS depends on PCI && SCSI select SCSI_SAS_ATTRS select RAID_ATTRS + select IRQ_POLL ---help--- This driver supports PCI-Express SAS 12Gb/s Host Adapters. diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.c b/drivers/scsi/mpt3sas/mpt3sas_base.c index f85c8eb154ef..9a36ea93cfbd 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_base.c +++ b/drivers/scsi/mpt3sas/mpt3sas_base.c @@ -1504,7 +1504,12 @@ _base_process_reply_queue(struct adapter_reply_queue *reply_q) MPI2_RPHI_MSIX_INDEX_SHIFT), &ioc->chip->ReplyPostHostIndex); } - completed_cmds = 1; + if (!reply_q->irq_poll_scheduled) { + reply_q->irq_poll_scheduled = true; + irq_poll_sched(&reply_q->irqpoll); + } + atomic_dec(&reply_q->busy); + return completed_cmds; } if (request_descript_type == MPI2_RPY_DESCRIPT_FLAGS_UNUSED) goto out; @@ -1570,11 +1575,66 @@ _base_interrupt(int irq, void *bus_id) if (ioc->mask_interrupts) return IRQ_NONE; - + if (reply_q->irq_poll_scheduled) + return IRQ_HANDLED; return ((_base_process_reply_queue(reply_q) > 0) ? IRQ_HANDLED : IRQ_NONE); } +/** + * _base_irqpoll - IRQ poll callback handler + * @irqpoll - irq_poll object + * @budget - irq poll weight + * + * returns number of reply descriptors processed + */ +static int +_base_irqpoll(struct irq_poll *irqpoll, int budget) +{ + struct adapter_reply_queue *reply_q; + int num_entries = 0; + + reply_q = container_of(irqpoll, struct adapter_reply_queue, + irqpoll); + if (reply_q->irq_line_enable) { + disable_irq(reply_q->os_irq); + reply_q->irq_line_enable = false; + } + num_entries = _base_process_reply_queue(reply_q); + if (num_entries < budget) { + irq_poll_complete(irqpoll); + reply_q->irq_poll_scheduled = false; + reply_q->irq_line_enable = true; + enable_irq(reply_q->os_irq); + } + + return num_entries; +} + +/** + * _base_init_irqpolls - initliaze IRQ polls + * @ioc: per adapter object + * + * returns nothing + */ +static void +_base_init_irqpolls(struct MPT3SAS_ADAPTER *ioc) +{ + struct adapter_reply_queue *reply_q, *next; + + if (list_empty(&ioc->reply_queue_list)) + return; + + list_for_each_entry_safe(reply_q, next, &ioc->reply_queue_list, list) { + irq_poll_init(&reply_q->irqpoll, + ioc->hba_queue_depth/4, _base_irqpoll); + reply_q->irq_poll_scheduled = false; + reply_q->irq_line_enable = true; + reply_q->os_irq = pci_irq_vector(ioc->pdev, + reply_q->msix_index); + } +} + /** * _base_is_controller_msix_enabled - is controller support muli-reply queues * @ioc: per adapter object @@ -1613,6 +1673,17 @@ mpt3sas_base_sync_reply_irqs(struct MPT3SAS_ADAPTER *ioc) /* TMs are on msix_index == 0 */ if (reply_q->msix_index == 0) continue; + if (reply_q->irq_poll_scheduled) { + /* Calling irq_poll_disable will wait for any pending + * callbacks to have completed. + */ + irq_poll_disable(&reply_q->irqpoll); + irq_poll_enable(&reply_q->irqpoll); + reply_q->irq_poll_scheduled = false; + reply_q->irq_line_enable = true; + enable_irq(reply_q->os_irq); + continue; + } synchronize_irq(pci_irq_vector(ioc->pdev, reply_q->msix_index)); } } @@ -3032,6 +3103,8 @@ mpt3sas_base_map_resources(struct MPT3SAS_ADAPTER *ioc) if (r) goto out_fail; + if (!ioc->is_driver_loading) + _base_init_irqpolls(ioc); /* Use the Combined reply queue feature only for SAS3 C0 & higher * revision HBAs and also only when reply queue count is greater than 8 */ @@ -6518,6 +6591,7 @@ mpt3sas_base_attach(struct MPT3SAS_ADAPTER *ioc) if (r) goto out_free_resources; + _base_init_irqpolls(ioc); init_waitqueue_head(&ioc->reset_wq); /* allocate memory pd handle bitmask list */ diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.h b/drivers/scsi/mpt3sas/mpt3sas_base.h index 19158cb59101..fb572cdb751e 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_base.h +++ b/drivers/scsi/mpt3sas/mpt3sas_base.h @@ -67,6 +67,7 @@ #include #include #include +#include #include "mpt3sas_debug.h" #include "mpt3sas_trigger_diag.h" @@ -882,6 +883,9 @@ struct _event_ack_list { * @reply_post_free: reply post base virt address * @name: the name registered to request_irq() * @busy: isr is actively processing replies on another cpu + * @os_irq: irq number + * @irqpoll: irq_poll object + * @irq_poll_scheduled: Tells whether irq poll is scheduled or not * @list: this list */ struct adapter_reply_queue { @@ -891,6 +895,10 @@ struct adapter_reply_queue { Mpi2ReplyDescriptorsUnion_t *reply_post_free; char name[MPT_NAME_LENGTH]; atomic_t busy; + u32 os_irq; + struct irq_poll irqpoll; + bool irq_poll_scheduled; + bool irq_line_enable; struct list_head list; };