On Wed, Apr 20, 2016 at 10:23:31AM +0900, Masahiro Yamada wrote: > Since of_get_cpu_node() increments refcount, the node should be put.
Urgh, I really hate the dt refcounting stuff. > Signed-off-by: Masahiro Yamada <[email protected]> > --- > > arch/arm64/kernel/smp_spin_table.c | 11 ++++++----- > 1 file changed, 6 insertions(+), 5 deletions(-) > > diff --git a/arch/arm64/kernel/smp_spin_table.c > b/arch/arm64/kernel/smp_spin_table.c > index aef3605..18a71bc 100644 > --- a/arch/arm64/kernel/smp_spin_table.c > +++ b/arch/arm64/kernel/smp_spin_table.c > @@ -52,6 +52,7 @@ static void write_pen_release(u64 val) > static int smp_spin_table_cpu_init(unsigned int cpu) > { > struct device_node *dn; > + int ret; > > dn = of_get_cpu_node(cpu, NULL); > if (!dn) > @@ -60,15 +61,15 @@ static int smp_spin_table_cpu_init(unsigned int cpu) > /* > * Determine the address from which the CPU is polling. > */ > - if (of_property_read_u64(dn, "cpu-release-addr", > - &cpu_release_addr[cpu])) { > + ret = of_property_read_u64(dn, "cpu-release-addr", > + &cpu_release_addr[cpu]); > + if (ret) > pr_err("CPU %d: missing or invalid cpu-release-addr property\n", > cpu); > > - return -1; > - } > + of_node_put(dn); > > - return 0; > + return ret; Looks ok to me. The slight change in return code should be harmless for ->cpu_init. Acked-by: Will Deacon <[email protected]> Will
