28 Jul
2025
28 Jul
'25
3:29 a.m.
On Mon, Jul 28, 2025 at 10:06:42AM +0800, liqiong wrote:
In this case it's an object pointer, not a freelist pointer. Or am I misunderstanding something?
Actually, in alloc_debug_processing() the pointer came from slab->freelist, so I think saying either "invalid freelist pointer" or "invalid object pointer" make sense...
free_consistency_checks() has 'slab_err(s, slab, "Invalid object pointer 0x%p", object);' Maybe it is better, alloc_consisency_checks() has the same message.
No. Think about it.