| Date | Thu, 10 Jun 2021 01:29:15 +0300 | From | Cyrill Gorcunov <> | Subject | Re: [RFC v2 29/34] mm: slub: Move flush_cpu_slab() invocations __free_slab() invocations out of IRQ context |
| |
On Wed, Jun 09, 2021 at 01:38:58PM +0200, Vlastimil Babka wrote: > +static DEFINE_MUTEX(flush_lock); > +static DEFINE_PER_CPU(struct slub_flush_work, slub_flush); > + > static void flush_all(struct kmem_cache *s) > { > - on_each_cpu_cond(has_cpu_slab, flush_cpu_slab, s, 1); > + struct slub_flush_work *sfw; > + unsigned int cpu; > + > + cpus_read_lock(); > + mutex_lock(&flush_lock); > +
Hi, Vlastimil! Could you please point why do you lock cpus first and mutex only after? Why not mutex_lock + cpus_read_lock instead?
|