On Tue, Dec 18, 2018 at 06:51:41PM +0800, He Zhe wrote: > On 2018/12/6 03:14, Sebastian Andrzej Siewior wrote: > > With raw locks you wouldn't have multiple readers at the same time. > > Maybe you wouldn't have recursion but since you can't have multiple > > readers you would add lock contention where was none (because you could > > have two readers at the same time). > > OK. I understand your concern finally. At the commit log said, I wanted to use raw > rwlock but didn't find the DEFINE helper for it. Thinking it would not be expected to > have great performance, I turn to use raw spinlock instead. And yes, this would > introduce worse performance. Looking through the kmemleak code, I can't really find significant reader contention. The longest holder of this lock (read) is the scan thread which is also protected by a scan_mutex, so can't run concurrently with another scanner (triggered through debugfs). The other read_lock(&kmemleak_lock) user is find_and_get_object() called from a few places. However, all such places normally follow a create_object() call (kmemleak_alloc() and friends) which already performs a write_lock(&kmemleak_lock), so it needs to wait for the scan thread to release the kmemleak_lock. It may be worth running some performance/latency tests during kmemleak scanning (echo scan > /sys/kernel/debug/kmemleak) but at a quick look, I don't think we'd see any difference with a raw_spin_lock_t. With a bit more thinking (though I'll be off until the new year), we could probably get rid of the kmemleak_lock entirely in scan_block() and make lookup_object() and the related rbtree code in kmemleak RCU-safe. -- Catalin
