When seq_nr wraps around, the next reorder job with seq 0 is hashed to
the first CPU in padata_do_serial(). Correspondingly, need reset pd->cpu
to the first one when pd->processed wraps around. Otherwise, if the
number of used CPUs is not a power of 2, padata_find_next() will be
checking a wrong list, hence deadlock.
Fixes: 6fc4dbcf0276 ("padata: Replace delayed timer with immediate workqueue in padata_reorder")
Signed-off-by: Xiao Liang <shaw.leon@xxxxxxxxx>
---
kernel/padata.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/kernel/padata.c b/kernel/padata.c
index f85f8bd788d0..833740d75483 100644
--- a/kernel/padata.c
+++ b/kernel/padata.c
@@ -291,8 +291,12 @@ static void padata_reorder(struct padata_priv *padata)
struct padata_serial_queue *squeue;
int cb_cpu;
- cpu = cpumask_next_wrap(cpu, pd->cpumask.pcpu);
processed++;
+ /* When sequence wraps around, reset to the first CPU. */
+ if (unlikely(processed == 0))
+ cpu = cpumask_first(pd->cpumask.pcpu);
+ else
+ cpu = cpumask_next_wrap(cpu, pd->cpumask.pcpu);
cb_cpu = padata->cb_cpu;
squeue = per_cpu_ptr(pd->squeue, cb_cpu);
--
2.50.1
