On Tue, 2019-07-16 at 22:07 +0200, Michal Hocko wrote: > On Tue 16-07-19 15:21:17, Qian Cai wrote: > [...] > > Thanks to this commit, there are allocation with __GFP_DIRECT_RECLAIM that > > succeeded would keep trying with __GFP_NOFAIL for kmemleak tracking object > > allocations. > > Well, not really. Because low order allocations with > __GFP_DIRECT_RECLAIM basically never fail (they keep retrying) even > without GFP_NOFAIL because that flag is actually to guarantee no > failure. And for high order allocations the nofail mode is actively > harmful. It completely changes the behavior of a system. A light costly > order workload could put the system on knees and completely change the > behavior. I am not really convinced this is a good behavior of a > debugging feature TBH. While I agree your general observation about GFP_NOFAIL, I am afraid the discussion here is about "struct kmemleak_object" slab cache from a single call site create_object(). > > > Otherwise, one kmemleak object allocation failure would kill the > > whole kmemleak. > > Which is not great but quite likely a better than an unpredictable MM > behavior caused by NOFAIL storms. Really, this NOFAIL patch is a > completely broken behavior. There shouldn't be much discussion about > reverting it. I would even argue it shouldn't have been merged in the > first place. It doesn't have any acks nor reviewed-bys while it abuses > __GFP_NOFAIL which is generally discouraged to be used. Again, it seems you are talking about GFP_NOFAIL in general. I don't really see much unpredictable MM behavior which would disrupt the testing or generate false-positive bug reports when "struct kmemleak_object" allocations with GFP_NOFAIL apart from some warnings. All I see is that kmemleak stay alive help find real memory leaks.
