From: Wander Lairson Costa wander@redhat.com

[ Upstream commit d243b34459cea30cfe5f3a9b2feb44e7daff9938 ]

Under PREEMPT_RT, __put_task_struct() indirectly acquires sleeping locks. Therefore, it can't be called from an non-preemptible context.

One practical example is splat inside inactive_task_timer(), which is called in a interrupt context:

CPU: 1 PID: 2848 Comm: life Kdump: loaded Tainted: G W --------- Hardware name: HP ProLiant DL388p Gen8, BIOS P70 07/15/2012 Call Trace: dump_stack_lvl+0x57/0x7d mark_lock_irq.cold+0x33/0xba mark_lock+0x1e7/0x400 mark_usage+0x11d/0x140 __lock_acquire+0x30d/0x930 lock_acquire.part.0+0x9c/0x210 rt_spin_lock+0x27/0xe0 refill_obj_stock+0x3d/0x3a0 kmem_cache_free+0x357/0x560 inactive_task_timer+0x1ad/0x340 __run_hrtimer+0x8a/0x1a0 __hrtimer_run_queues+0x91/0x130 hrtimer_interrupt+0x10f/0x220 __sysvec_apic_timer_interrupt+0x7b/0xd0 sysvec_apic_timer_interrupt+0x4f/0xd0 asm_sysvec_apic_timer_interrupt+0x12/0x20 RIP: 0033:0x7fff196bf6f5

Instead of calling __put_task_struct() directly, we defer it using call_rcu(). A more natural approach would use a workqueue, but since in PREEMPT_RT, we can't allocate dynamic memory from atomic context, the code would become more complex because we would need to put the work_struct instance in the task_struct and initialize it when we allocate a new task_struct.

The issue is reproducible with stress-ng:

while true; do stress-ng --sched deadline --sched-period 1000000000 \ --sched-runtime 800000000 --sched-deadline \ 1000000000 --mmapfork 23 -t 20 done

Reported-by: Hu Chunyu chuhu@redhat.com Suggested-by: Oleg Nesterov oleg@redhat.com Suggested-by: Valentin Schneider vschneid@redhat.com Suggested-by: Peter Zijlstra peterz@infradead.org Signed-off-by: Wander Lairson Costa wander@redhat.com Signed-off-by: Peter Zijlstra (Intel) peterz@infradead.org Link: https://lore.kernel.org/r/20230614122323.37957-2-wander@redhat.com Signed-off-by: Sasha Levin sashal@kernel.org --- include/linux/sched/task.h | 28 +++++++++++++++++++++++++++- kernel/fork.c | 8 ++++++++ 2 files changed, 35 insertions(+), 1 deletion(-)

diff --git a/include/linux/sched/task.h b/include/linux/sched/task.h index 853ab403e77b8..e3c20a4f81f5e 100644 --- a/include/linux/sched/task.h +++ b/include/linux/sched/task.h @@ -113,10 +113,36 @@ static inline struct task_struct *get_task_struct(struct task_struct *t) }

extern void __put_task_struct(struct task_struct *t); +extern void __put_task_struct_rcu_cb(struct rcu_head *rhp);

static inline void put_task_struct(struct task_struct *t) { - if (refcount_dec_and_test(&t->usage)) + if (!refcount_dec_and_test(&t->usage)) + return; + + /* + * under PREEMPT_RT, we can't call put_task_struct + * in atomic context because it will indirectly + * acquire sleeping locks. + * + * call_rcu() will schedule delayed_put_task_struct_rcu() + * to be called in process context. + * + * __put_task_struct() is called when + * refcount_dec_and_test(&t->usage) succeeds. + * + * This means that it can't "conflict" with + * put_task_struct_rcu_user() which abuses ->rcu the same + * way; rcu_users has a reference so task->usage can't be + * zero after rcu_users 1 -> 0 transition. + * + * delayed_free_task() also uses ->rcu, but it is only called + * when it fails to fork a process. Therefore, there is no + * way it can conflict with put_task_struct(). + */ + if (IS_ENABLED(CONFIG_PREEMPT_RT) && !preemptible()) + call_rcu(&t->rcu, __put_task_struct_rcu_cb); + else __put_task_struct(t); }

diff --git a/kernel/fork.c b/kernel/fork.c index 39134effb2bff..1728aa77861cb 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -749,6 +749,14 @@ void __put_task_struct(struct task_struct *tsk) } EXPORT_SYMBOL_GPL(__put_task_struct);

+void __put_task_struct_rcu_cb(struct rcu_head *rhp) +{ + struct task_struct *task = container_of(rhp, struct task_struct, rcu); + + __put_task_struct(task); +} +EXPORT_SYMBOL_GPL(__put_task_struct_rcu_cb); + void __init __weak arch_task_cache_init(void) { }

/*

From: Yicong Yang yangyicong@hisilicon.com

[ Upstream commit 0242737dc4eb9f6e9a5ea594b3f93efa0b12f28d ]

Some HiSilicon SMMU PMCG suffers the erratum 162001900 that the PMU disable control sometimes fail to disable the counters. This will lead to error or inaccurate data since before we enable the counters the counter's still counting for the event used in last perf session.

This patch tries to fix this by hardening the global disable process. Before disable the PMU, writing an invalid event type (0xffff) to focibly stop the counters. Correspondingly restore each events on pmu::pmu_enable().

Signed-off-by: Yicong Yang yangyicong@hisilicon.com Link: https://lore.kernel.org/r/20230814124012.58013-1-yangyicong@huawei.com Signed-off-by: Will Deacon will@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- Documentation/arm64/silicon-errata.rst | 3 ++ drivers/acpi/arm64/iort.c | 5 ++- drivers/perf/arm_smmuv3_pmu.c | 46 +++++++++++++++++++++++++- include/linux/acpi_iort.h | 1 + 4 files changed, 53 insertions(+), 2 deletions(-)

diff --git a/Documentation/arm64/silicon-errata.rst b/Documentation/arm64/silicon-errata.rst index 36a8c01191a07..6b70b6aabcffe 100644 --- a/Documentation/arm64/silicon-errata.rst +++ b/Documentation/arm64/silicon-errata.rst @@ -134,6 +134,9 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | Hisilicon | Hip08 SMMU PMCG | #162001800 | N/A | +----------------+-----------------+-----------------+-----------------------------+ +| Hisilicon | Hip08 SMMU PMCG | #162001900 | N/A | +| | Hip09 SMMU PMCG | | | ++----------------+-----------------+-----------------+-----------------------------+ +----------------+-----------------+-----------------+-----------------------------+ | Qualcomm Tech. | Kryo/Falkor v1 | E1003 | QCOM_FALKOR_ERRATUM_1003 | +----------------+-----------------+-----------------+-----------------------------+ diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c index 553c89b0bdcbb..09eb170f26d27 100644 --- a/drivers/acpi/arm64/iort.c +++ b/drivers/acpi/arm64/iort.c @@ -1393,7 +1393,10 @@ static void __init arm_smmu_v3_pmcg_init_resources(struct resource *res, static struct acpi_platform_list pmcg_plat_info[] __initdata = { /* HiSilicon Hip08 Platform */ {"HISI ", "HIP08 ", 0, ACPI_SIG_IORT, greater_than_or_equal, - "Erratum #162001800", IORT_SMMU_V3_PMCG_HISI_HIP08}, + "Erratum #162001800, Erratum #162001900", IORT_SMMU_V3_PMCG_HISI_HIP08}, + /* HiSilicon Hip09 Platform */ + {"HISI ", "HIP09 ", 0, ACPI_SIG_IORT, greater_than_or_equal, + "Erratum #162001900", IORT_SMMU_V3_PMCG_HISI_HIP09}, { } };

diff --git a/drivers/perf/arm_smmuv3_pmu.c b/drivers/perf/arm_smmuv3_pmu.c index 0b6af77196418..de85e9191947a 100644 --- a/drivers/perf/arm_smmuv3_pmu.c +++ b/drivers/perf/arm_smmuv3_pmu.c @@ -95,6 +95,7 @@ #define SMMU_PMCG_PA_SHIFT 12

#define SMMU_PMCG_EVCNTR_RDONLY BIT(0) +#define SMMU_PMCG_HARDEN_DISABLE BIT(1)

static int cpuhp_state_num;

@@ -138,6 +139,20 @@ static inline void smmu_pmu_enable(struct pmu *pmu) writel(SMMU_PMCG_CR_ENABLE, smmu_pmu->reg_base + SMMU_PMCG_CR); }

+static int smmu_pmu_apply_event_filter(struct smmu_pmu *smmu_pmu, + struct perf_event *event, int idx); + +static inline void smmu_pmu_enable_quirk_hip08_09(struct pmu *pmu) +{ + struct smmu_pmu *smmu_pmu = to_smmu_pmu(pmu); + unsigned int idx; + + for_each_set_bit(idx, smmu_pmu->used_counters, smmu_pmu->num_counters) + smmu_pmu_apply_event_filter(smmu_pmu, smmu_pmu->events[idx], idx); + + smmu_pmu_enable(pmu); +} + static inline void smmu_pmu_disable(struct pmu *pmu) { struct smmu_pmu *smmu_pmu = to_smmu_pmu(pmu); @@ -146,6 +161,22 @@ static inline void smmu_pmu_disable(struct pmu *pmu) writel(0, smmu_pmu->reg_base + SMMU_PMCG_IRQ_CTRL); }

+static inline void smmu_pmu_disable_quirk_hip08_09(struct pmu *pmu) +{ + struct smmu_pmu *smmu_pmu = to_smmu_pmu(pmu); + unsigned int idx; + + /* + * The global disable of PMU sometimes fail to stop the counting. + * Harden this by writing an invalid event type to each used counter + * to forcibly stop counting. + */ + for_each_set_bit(idx, smmu_pmu->used_counters, smmu_pmu->num_counters) + writel(0xffff, smmu_pmu->reg_base + SMMU_PMCG_EVTYPER(idx)); + + smmu_pmu_disable(pmu); +} + static inline void smmu_pmu_counter_set_value(struct smmu_pmu *smmu_pmu, u32 idx, u64 value) { @@ -719,7 +750,10 @@ static void smmu_pmu_get_acpi_options(struct smmu_pmu *smmu_pmu) switch (model) { case IORT_SMMU_V3_PMCG_HISI_HIP08: /* HiSilicon Erratum 162001800 */ - smmu_pmu->options |= SMMU_PMCG_EVCNTR_RDONLY; + smmu_pmu->options |= SMMU_PMCG_EVCNTR_RDONLY | SMMU_PMCG_HARDEN_DISABLE; + break; + case IORT_SMMU_V3_PMCG_HISI_HIP09: + smmu_pmu->options |= SMMU_PMCG_HARDEN_DISABLE; break; }

@@ -808,6 +842,16 @@ static int smmu_pmu_probe(struct platform_device *pdev)

smmu_pmu_get_acpi_options(smmu_pmu);

+ /* + * For platforms suffer this quirk, the PMU disable sometimes fails to + * stop the counters. This will leads to inaccurate or error counting. + * Forcibly disable the counters with these quirk handler. + */ + if (smmu_pmu->options & SMMU_PMCG_HARDEN_DISABLE) { + smmu_pmu->pmu.pmu_enable = smmu_pmu_enable_quirk_hip08_09; + smmu_pmu->pmu.pmu_disable = smmu_pmu_disable_quirk_hip08_09; + } + /* Pick one CPU to be the preferred one to use */ smmu_pmu->on_cpu = raw_smp_processor_id(); WARN_ON(irq_set_affinity_hint(smmu_pmu->irq, diff --git a/include/linux/acpi_iort.h b/include/linux/acpi_iort.h index 8e7e2ec37f1b2..64f700254ca0f 100644 --- a/include/linux/acpi_iort.h +++ b/include/linux/acpi_iort.h @@ -21,6 +21,7 @@ */ #define IORT_SMMU_V3_PMCG_GENERIC 0x00000000 /* Generic SMMUv3 PMCG */ #define IORT_SMMU_V3_PMCG_HISI_HIP08 0x00000001 /* HiSilicon HIP08 PMCG */ +#define IORT_SMMU_V3_PMCG_HISI_HIP09 0x00000002 /* HiSilicon HIP09 PMCG */

int iort_register_domain_token(int trans_id, phys_addr_t base, struct fwnode_handle *fw_node);