On Sun, 4 Mar 2012 13:43:32 -0800 (PST) David Rientjes <rientjes@xxxxxxxxxx> wrote: > BUG() is a no-op when CONFIG_BUG is disabled, so slab_node() needs a > dummy return value to avoid reaching the end of a non-void function. > > Signed-off-by: David Rientjes <rientjes@xxxxxxxxxx> > --- > mm/mempolicy.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/mm/mempolicy.c b/mm/mempolicy.c > --- a/mm/mempolicy.c > +++ b/mm/mempolicy.c > @@ -1611,6 +1611,7 @@ unsigned slab_node(struct mempolicy *policy) > > default: > BUG(); > + return numa_node_id(); > } > } Wait. If the above code generated a warning then surely we get a *lot* of warnings! I'd expect that a lot of code assumes that BUG() never returns? Can we fix this within the BUG() definition? I can't think of a way, unless gcc gives us a way of accessing the return type of the current function, and I don't think it does that. Also, does CONIG_BUG=n even make sense? If we got here and we know that the kernel has malfunctioned, what point is there in pretending otherwise? Odd. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>