[PATCH] cpu/hotplug: dont offline the last non-isolated CPU

[yang.yang29]

From: Ran Xiaokai <ran.xiaokai@xxxxxxxxxx>

If system has some isolate cpus with "isolcpus=" parameter,
and user try to offline all the non-isolated CPUs, kernel will
first report a warning and then panic when taking the last
non-isolated CPU offline.

Kernel should not panic when user managing CPUs with this
meaningless behavior, so we add a check for this, which
prevent user taking the last non-isolated CPU offline.

[   19.415123] WARNING: CPU: 4 PID: 80 at kernel/sched/topology.c:2366 build_sched_domains+0x120c/0x1408
[   19.430451] Call trace:
[   19.430792]  build_sched_domains+0x120c/0x1408
[   19.431334]  partition_sched_domains_locked+0x234/0x880
[   19.431878]  rebuild_sched_domains_locked+0x37c/0x798
[   19.432436]  rebuild_sched_domains+0x30/0x58
[   19.432902]  cpuset_hotplug_workfn+0x2a8/0x930
[   19.433383]  process_scheduled_works+0x178/0x3e0
[   19.433878]  worker_thread+0x174/0x2f0
[   19.435204] ---[ end trace 0000000000000000 ]---
[   19.438650] Unable to handle kernel paging request at virtual address fffe80027ab37080
[   19.456414]  partition_sched_domains_locked+0x318/0x880
[   19.456899]  rebuild_sched_domains_locked+0x37c/0x798
[   19.457361]  rebuild_sched_domains+0x30/0x58
[   19.457761]  cpuset_hotplug_workfn+0x2a8/0x930
[   19.458175]  process_scheduled_works+0x178/0x3e0
[   19.458599]  worker_thread+0x174/0x2f0
[   19.458948]  kthread+0x10c/0x128
[   19.459268]  ret_from_fork+0x10/0x20
[   19.459728] Code: 1a850042 b9441883 f862dae0 8b000021 (f945003b)

Signed-off-by: Ran Xiaokai <ran.xiaokai@xxxxxxxxxx>
Cc: Xuexin Jiang <jiang.xuexin@xxxxxxxxxx>
Cc: Yang Yang <yang.yang29@xxxxxxxxxx>
---
 kernel/cpu.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/kernel/cpu.c b/kernel/cpu.c
index 6de7c6bb74ee..09ecc19d2999 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -1502,6 +1502,7 @@ static long __cpu_down_maps_locked(void *arg)
 static int cpu_down_maps_locked(unsigned int cpu, enum cpuhp_state target)
 {
 	struct cpu_down_work work = { .cpu = cpu, .target = target, };
+	struct cpumask tmp_mask;

 	/*
 	 * If the platform does not support hotplug, report it explicitly to
@@ -1512,11 +1513,16 @@ static int cpu_down_maps_locked(unsigned int cpu, enum cpuhp_state target)
 	if (cpu_hotplug_disabled)
 		return -EBUSY;

+	/*
+	 * Ensure the last non-isolated CPU is not offlined.
+	 */
+	cpumask_and(&tmp_mask, cpu_online_mask, housekeeping_cpumask(HK_TYPE_DOMAIN));
+
 	/*
 	 * Ensure that the control task does not run on the to be offlined
 	 * CPU to prevent a deadlock against cfs_b->period_timer.
 	 */
-	cpu = cpumask_any_but(cpu_online_mask, cpu);
+	cpu = cpumask_any_but(&tmp_mask, cpu);
 	if (cpu >= nr_cpu_ids)
 		return -EBUSY;
 	return work_on_cpu(cpu, __cpu_down_maps_locked, &work);
-- 
2.15.2