Messages in this thread | | | Date | Tue, 10 Jul 2012 23:14:51 -0700 | Subject | Re: [PATCH 05/13] rbtree: performance and correctness test | From | Michel Lespinasse <> |
| |
On Tue, Jul 10, 2012 at 5:27 AM, Michal Nazarewicz <mina86@mina86.com> wrote: > On Tue, 10 Jul 2012 01:35:15 +0200, Michel Lespinasse <walken@google.com> wrote: >> + u32 prev_key = 0; >> + >> + for (rb = rb_first(&root); rb; rb = rb_next(rb)) { >> + struct test_node *node = rb_entry(rb, struct test_node, >> rb); >> + WARN_ON_ONCE(node->key < prev_key); > > What if for some reason we generate node with key equal zero or two keys > with the same value? It may not be the case for current code, but someone > might change it in the future. I think <= is safer here.
No, it's not illegal for two nodes to have the same key; the second one to be inserted will just get placed after the first one. The rbtree library doesn't care either way as it's not even aware of the key values :)
-- Michel "Walken" Lespinasse A program is never fully debugged until the last user dies.
| |