[bug report] driver/perf: Add PMU driver for the ARM DMC-620 memory controller

From: Dan Carpenter <dan.carpenter@oracle.com>
To: tuanphan@os.amperecomputing.com
Cc: linux-arm-kernel@lists.infradead.org
Subject: [bug report] driver/perf: Add PMU driver for the ARM DMC-620 memory controller
Date: Mon, 7 Dec 2020 20:56:48 +0300	[thread overview]
Message-ID: <X85s4Haz+LBzhKnF@mwanda> (raw)

Hello Tuan Phan,

The patch 53c218da220c: "driver/perf: Add PMU driver for the ARM
DMC-620 memory controller" from Nov 4, 2020, leads to the following
static checker warning:

	drivers/perf/arm_dmc620_pmu.c:612 dmc620_pmu_cpu_teardown()
	error: potential NULL dereference 'irq'.

drivers/perf/arm_dmc620_pmu.c
   604  static int dmc620_pmu_cpu_teardown(unsigned int cpu,
   605                                     struct hlist_node *node)
   606  {
   607          struct dmc620_pmu_irq *irq;
   608          struct dmc620_pmu *dmc620_pmu;
   609          unsigned int target;
   610  
   611          irq = hlist_entry_safe(node, struct dmc620_pmu_irq, node);
   612          if (cpu != irq->cpu)

There isn't any point to using hlist_entry_safe() instead of hlist_entry()
when we're not going to check for NULL returns.

   613                  return 0;
   614  
   615          target = cpumask_any_but(cpu_online_mask, cpu);
   616          if (target >= nr_cpu_ids)
   617                  return 0;
   618  
   619          /* We're only reading, but this isn't the place to be involving RCU */
   620          mutex_lock(&dmc620_pmu_irqs_lock);
   621          list_for_each_entry(dmc620_pmu, &irq->pmus_node, pmus_node)
   622                  perf_pmu_migrate_context(&dmc620_pmu->pmu, irq->cpu, target);
   623          mutex_unlock(&dmc620_pmu_irqs_lock);

regards,
dan carpenter

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel