On Sat, 2020-02-08 at 15:09 +0100, Christophe JAILLET wrote: > In some error handling path, we should call "of_node_put(np_par)" or > some resource may be leaking in case of error. > > Fixes: 8159df72d43e ("83xx: add support for the kmeter1 board.") > Signed-off-by: Christophe JAILLET <christophe.jaillet@xxxxxxxxxx> > --- > arch/powerpc/platforms/83xx/km83xx.c | 5 ++++- > 1 file changed, 4 insertions(+), 1 deletion(-) > Both patches: Acked-by: Scott Wood <oss@xxxxxxxxxxxx> > diff --git a/arch/powerpc/platforms/83xx/km83xx.c > b/arch/powerpc/platforms/83xx/km83xx.c > index 306be75faec7..bcdc2c203ec9 100644 > --- a/arch/powerpc/platforms/83xx/km83xx.c > +++ b/arch/powerpc/platforms/83xx/km83xx.c > @@ -60,10 +60,12 @@ static void quirk_mpc8360e_qe_enet10(void) > ret = of_address_to_resource(np_par, 0, &res); > if (ret) { > pr_warn("%s couldn't map par_io registers\n", __func__); > - return; > + goto out; > } > > base = ioremap(res.start, resource_size(&res)); > + if (!base) > + goto out; > > /* > * set output delay adjustments to default values according > @@ -111,6 +113,7 @@ static void quirk_mpc8360e_qe_enet10(void) > setbits32((base + 0xac), 0x0000c000); > } > iounmap(base); > +out: > of_node_put(np_par); > } >