On Fri, 2016-05-06 at 13:00 +1000, Suraj Jitindar Singh wrote:
On 05/05/16 16:50, Michael Ellerman wrote:
quoted
On Tue, 2016-05-03 at 15:32 -0700, Tyrel Datwyler wrote:
quoted
On 04/27/2016 10:34 PM, Suraj Jitindar Singh wrote:
quoted
diff --git a/arch/powerpc/platforms/pseries/mobility.c b/arch/powerpc/platforms/pseries/mobility.c
index ceb18d3..a560a98 100644
--- a/arch/powerpc/platforms/pseries/mobility.c
+++ b/arch/powerpc/platforms/pseries/mobility.c
@@ -191,8 +191,8 @@ static int update_dt_node(__be32 phandle, s32 scope)
break;
case 0x80000000:
- prop = of_find_property(dn, prop_name, NULL);
- of_remove_property(dn, prop);
+ of_remove_property(dn, of_find_property(dn,
+ prop_name, NULL));
prop = NULL;
break;
You haven't removed a NULL check here, as suggested by the changelog,
but instead made a cosmetic change to the code that still leaves behind
a now unnecessary "prop = NULL;" to bit rot.
quoted
Yeah I think you're right. Though it's not very clear how prop is used in that
function.
I didn't delete the prop = NULL; initially as I didn't fully understand
how it was used in the rest of the function and the effect of deleting
it.
Yeah, it's pretty convoluted. I don't think you can actually prove it's safe to
remove the prop = NULL for arbitrary inputs.
cheers