smp_call_function_any() doesn't allow nested calls, since it runs in interrupt context, and nesting would trigger a warning due to interrupt disabled. In mon_event_read(), when all CPUs of the target domain are in TICK-NOHZ mode, no house-keeping cpu can invoke smp_call_on_cpu(). When the IPI fallback is used on machines where MPAM needs to make an access on multiple CPUs, the counter read will always fail. Therefore, must use smp_call_on_cpu() to invoke mon_event_read() in a thread context to ensure the success of MPAM counter read. Fixes: 158e5bec7d72 ("x86/resctrl: Queue mon_event_read() instead of sending an IPI") Signed-off-by: Zeng Heng <zengheng4@huawei.com> --- fs/resctrl/ctrlmondata.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/resctrl/ctrlmondata.c b/fs/resctrl/ctrlmondata.c index 3f0380a2386a..bf87eed826ff 100644 --- a/fs/resctrl/ctrlmondata.c +++ b/fs/resctrl/ctrlmondata.c @@ -507,7 +507,7 @@ void mon_event_read(struct rmid_read *rr, struct rdt_resource *r, * MPAM's resctrl_arch_rmid_read() is unable to read the * counters on some platforms if its called in irq context. */ - if (tick_nohz_full_cpu(cpu)) + if (tick_nohz_full_cpu(cpu) && !IS_ENABLED(CONFIG_ARM64_MPAM)) smp_call_function_any(&d->cpu_mask, mon_event_count, rr, 1); else smp_call_on_cpu(cpu, smp_mon_event_count, rr, false); -- 2.25.1