diff options
Diffstat (limited to 'kernel/bpf/memalloc.c')
| -rw-r--r-- | kernel/bpf/memalloc.c | 91 |
1 files changed, 57 insertions, 34 deletions
diff --git a/kernel/bpf/memalloc.c b/kernel/bpf/memalloc.c index bd45dda9dc35..e9662db7198f 100644 --- a/kernel/bpf/memalloc.c +++ b/kernel/bpf/memalloc.c @@ -102,6 +102,8 @@ struct bpf_mem_cache { int percpu_size; bool draining; struct bpf_mem_cache *tgt; + void (*dtor)(void *obj, void *ctx); + void *dtor_ctx; /* list of objects to be freed after RCU GP */ struct llist_head free_by_rcu; @@ -260,12 +262,14 @@ static void free_one(void *obj, bool percpu) kfree(obj); } -static int free_all(struct llist_node *llnode, bool percpu) +static int free_all(struct bpf_mem_cache *c, struct llist_node *llnode, bool percpu) { struct llist_node *pos, *t; int cnt = 0; llist_for_each_safe(pos, t, llnode) { + if (c->dtor) + c->dtor((void *)pos + LLIST_NODE_SZ, c->dtor_ctx); free_one(pos, percpu); cnt++; } @@ -276,21 +280,10 @@ static void __free_rcu(struct rcu_head *head) { struct bpf_mem_cache *c = container_of(head, struct bpf_mem_cache, rcu_ttrace); - free_all(llist_del_all(&c->waiting_for_gp_ttrace), !!c->percpu_size); + free_all(c, llist_del_all(&c->waiting_for_gp_ttrace), !!c->percpu_size); atomic_set(&c->call_rcu_ttrace_in_progress, 0); } -static void __free_rcu_tasks_trace(struct rcu_head *head) -{ - /* If RCU Tasks Trace grace period implies RCU grace period, - * there is no need to invoke call_rcu(). - */ - if (rcu_trace_implies_rcu_gp()) - __free_rcu(head); - else - call_rcu(head, __free_rcu); -} - static void enque_to_free(struct bpf_mem_cache *c, void *obj) { struct llist_node *llnode = obj; @@ -308,7 +301,7 @@ static void do_call_rcu_ttrace(struct bpf_mem_cache *c) if (atomic_xchg(&c->call_rcu_ttrace_in_progress, 1)) { if (unlikely(READ_ONCE(c->draining))) { llnode = llist_del_all(&c->free_by_rcu_ttrace); - free_all(llnode, !!c->percpu_size); + free_all(c, llnode, !!c->percpu_size); } return; } @@ -322,12 +315,12 @@ static void do_call_rcu_ttrace(struct bpf_mem_cache *c) return; } - /* Use call_rcu_tasks_trace() to wait for sleepable progs to finish. - * If RCU Tasks Trace grace period implies RCU grace period, free - * these elements directly, else use call_rcu() to wait for normal - * progs to finish and finally do free_one() on each element. + /* + * Use call_rcu_tasks_trace() to wait for sleepable progs to finish. + * RCU Tasks Trace grace period implies RCU grace period, so pass + * __free_rcu directly as the callback. */ - call_rcu_tasks_trace(&c->rcu_ttrace, __free_rcu_tasks_trace); + call_rcu_tasks_trace(&c->rcu_ttrace, __free_rcu); } static void free_bulk(struct bpf_mem_cache *c) @@ -417,7 +410,7 @@ static void check_free_by_rcu(struct bpf_mem_cache *c) dec_active(c, &flags); if (unlikely(READ_ONCE(c->draining))) { - free_all(llist_del_all(&c->waiting_for_gp), !!c->percpu_size); + free_all(c, llist_del_all(&c->waiting_for_gp), !!c->percpu_size); atomic_set(&c->call_rcu_in_progress, 0); } else { call_rcu_hurry(&c->rcu, __free_by_rcu); @@ -635,13 +628,13 @@ static void drain_mem_cache(struct bpf_mem_cache *c) * Except for waiting_for_gp_ttrace list, there are no concurrent operations * on these lists, so it is safe to use __llist_del_all(). */ - free_all(llist_del_all(&c->free_by_rcu_ttrace), percpu); - free_all(llist_del_all(&c->waiting_for_gp_ttrace), percpu); - free_all(__llist_del_all(&c->free_llist), percpu); - free_all(__llist_del_all(&c->free_llist_extra), percpu); - free_all(__llist_del_all(&c->free_by_rcu), percpu); - free_all(__llist_del_all(&c->free_llist_extra_rcu), percpu); - free_all(llist_del_all(&c->waiting_for_gp), percpu); + free_all(c, llist_del_all(&c->free_by_rcu_ttrace), percpu); + free_all(c, llist_del_all(&c->waiting_for_gp_ttrace), percpu); + free_all(c, __llist_del_all(&c->free_llist), percpu); + free_all(c, __llist_del_all(&c->free_llist_extra), percpu); + free_all(c, __llist_del_all(&c->free_by_rcu), percpu); + free_all(c, __llist_del_all(&c->free_llist_extra_rcu), percpu); + free_all(c, llist_del_all(&c->waiting_for_gp), percpu); } static void check_mem_cache(struct bpf_mem_cache *c) @@ -680,6 +673,9 @@ static void check_leaked_objs(struct bpf_mem_alloc *ma) static void free_mem_alloc_no_barrier(struct bpf_mem_alloc *ma) { + /* We can free dtor ctx only once all callbacks are done using it. */ + if (ma->dtor_ctx_free) + ma->dtor_ctx_free(ma->dtor_ctx); check_leaked_objs(ma); free_percpu(ma->cache); free_percpu(ma->caches); @@ -689,20 +685,18 @@ static void free_mem_alloc_no_barrier(struct bpf_mem_alloc *ma) static void free_mem_alloc(struct bpf_mem_alloc *ma) { - /* waiting_for_gp[_ttrace] lists were drained, but RCU callbacks + /* + * waiting_for_gp[_ttrace] lists were drained, but RCU callbacks * might still execute. Wait for them. * * rcu_barrier_tasks_trace() doesn't imply synchronize_rcu_tasks_trace(), * but rcu_barrier_tasks_trace() and rcu_barrier() below are only used - * to wait for the pending __free_rcu_tasks_trace() and __free_rcu(), - * so if call_rcu(head, __free_rcu) is skipped due to - * rcu_trace_implies_rcu_gp(), it will be OK to skip rcu_barrier() by - * using rcu_trace_implies_rcu_gp() as well. + * to wait for the pending __free_by_rcu(), and __free_rcu(). RCU Tasks + * Trace grace period implies RCU grace period, so all __free_rcu don't + * need extra call_rcu() (and thus extra rcu_barrier() here). */ rcu_barrier(); /* wait for __free_by_rcu */ rcu_barrier_tasks_trace(); /* wait for __free_rcu */ - if (!rcu_trace_implies_rcu_gp()) - rcu_barrier(); free_mem_alloc_no_barrier(ma); } @@ -1014,3 +1008,32 @@ int bpf_mem_alloc_check_size(bool percpu, size_t size) return 0; } + +void bpf_mem_alloc_set_dtor(struct bpf_mem_alloc *ma, void (*dtor)(void *obj, void *ctx), + void (*dtor_ctx_free)(void *ctx), void *ctx) +{ + struct bpf_mem_caches *cc; + struct bpf_mem_cache *c; + int cpu, i; + + ma->dtor_ctx_free = dtor_ctx_free; + ma->dtor_ctx = ctx; + + if (ma->cache) { + for_each_possible_cpu(cpu) { + c = per_cpu_ptr(ma->cache, cpu); + c->dtor = dtor; + c->dtor_ctx = ctx; + } + } + if (ma->caches) { + for_each_possible_cpu(cpu) { + cc = per_cpu_ptr(ma->caches, cpu); + for (i = 0; i < NUM_CACHES; i++) { + c = &cc->cache[i]; + c->dtor = dtor; + c->dtor_ctx = ctx; + } + } + } +} |