@@ -36,7 +36,7 @@ struct svc_pool {
struct list_head sp_sockets; /* pending sockets */
unsigned int sp_nrthreads; /* # of threads in pool */
struct list_head sp_all_threads; /* all server threads */
- struct list_head sp_idle_threads; /* idle server threads */
+ struct llist_head sp_idle_threads; /* idle server threads */
/* statistics on pool operation */
struct percpu_counter sp_messages_arrived;
@@ -199,7 +199,7 @@ extern u32 svc_max_payload(const struct svc_rqst *rqstp);
*/
struct svc_rqst {
struct list_head rq_all; /* all threads list */
- struct list_head rq_idle; /* On the idle list */
+ struct llist_node rq_idle; /* On the idle list */
struct rcu_head rq_rcu_head; /* for RCU deferred kfree */
struct svc_xprt * rq_xprt; /* transport ptr */
@@ -286,22 +286,24 @@ enum {
* svc_thread_set_busy - mark a thread as busy
* @rqstp: the thread which is now busy
*
- * If rq_idle is "empty", the thread must be busy.
+ * By convention a thread is busy if rq_idle.next points to rq_idle.
+ * This ensures it is not on the idle list.
*/
static inline void svc_thread_set_busy(struct svc_rqst *rqstp)
{
- INIT_LIST_HEAD(&rqstp->rq_idle);
+ smp_store_release(&rqstp->rq_idle.next, &rqstp->rq_idle);
}
/**
* svc_thread_busy - check if a thread as busy
* @rqstp: the thread which might be busy
*
- * If rq_idle is "empty", the thread must be busy.
+ * By convention a thread is busy if rq_idle.next points to rq_idle.
+ * This ensures it is not on the idle list.
*/
static inline bool svc_thread_busy(struct svc_rqst *rqstp)
{
- return list_empty(&rqstp->rq_idle);
+ return smp_load_acquire(&rqstp->rq_idle.next) == &rqstp->rq_idle;
}
#define SVC_NET(rqst) (rqst->rq_xprt ? rqst->rq_xprt->xpt_net : rqst->rq_bc_net)
@@ -702,15 +702,16 @@ void svc_pool_wake_idle_thread(struct svc_serv *serv,
struct svc_pool *pool)
{
struct svc_rqst *rqstp;
+ struct llist_node *ln;
rcu_read_lock();
spin_lock_bh(&pool->sp_lock);
- rqstp = list_first_entry_or_null(&pool->sp_idle_threads,
- struct svc_rqst, rq_idle);
- if (rqstp)
- list_del_init(&rqstp->rq_idle);
+ ln = llist_del_first(&pool->sp_idle_threads);
spin_unlock_bh(&pool->sp_lock);
- if (rqstp) {
+ if (ln) {
+ rqstp = llist_entry(ln, struct svc_rqst, rq_idle);
+ svc_thread_set_busy(rqstp);
+
WRITE_ONCE(rqstp->rq_qtime, ktime_get());
wake_up_process(rqstp->rq_task);
rcu_read_unlock();
@@ -704,7 +704,7 @@ rqst_should_sleep(struct svc_rqst *rqstp)
struct svc_pool *pool = rqstp->rq_pool;
/* did someone call svc_wake_up? */
- if (test_and_clear_bit(SP_TASK_PENDING, &pool->sp_flags))
+ if (test_bit(SP_TASK_PENDING, &pool->sp_flags))
return false;
/* was a socket queued? */
@@ -715,10 +715,6 @@ rqst_should_sleep(struct svc_rqst *rqstp)
if (svc_thread_should_stop(rqstp))
return false;
- /* are we freezing? */
- if (freezing(current))
- return false;
-
#if defined(CONFIG_SUNRPC_BACKCHANNEL)
if (svc_is_backchannel(rqstp)) {
if (!list_empty(&rqstp->rq_server->sv_cb_list))
@@ -733,27 +729,40 @@ static void svc_wait_for_work(struct svc_rqst *rqstp)
{
struct svc_pool *pool = rqstp->rq_pool;
- set_current_state(TASK_IDLE);
- spin_lock_bh(&pool->sp_lock);
- list_add(&rqstp->rq_idle, &pool->sp_idle_threads);
- spin_unlock_bh(&pool->sp_lock);
+ if (rqst_should_sleep(rqstp)) {
+ set_current_state(TASK_IDLE);
+ llist_add(&rqstp->rq_idle, &pool->sp_idle_threads);
+
+ if (unlikely(!rqst_should_sleep(rqstp)))
+ /* maybe there were no idle threads when some work
+ * became ready and so nothing was woken. We've just
+ * become idle so someone can to the work - maybe us.
+ * But we cannot reliably remove ourselves from the
+ * idle list - we can only remove the first task which
+ * might be us, and might not.
+ * So remove and wake it, then schedule(). If it was
+ * us, we won't sleep. If it is some other thread, they
+ * will do the work.
+ */
+ svc_pool_wake_idle_thread(rqstp->rq_server, pool);
- if (likely(rqst_should_sleep(rqstp)))
- schedule();
- else
+ /* We mustn't continue while on the idle list, and we
+ * cannot remove outselves reliably. The only "work"
+ * we can do while on the idle list is to freeze.
+ * So loop until someone removes us
+ */
+ while (!svc_thread_busy(rqstp)) {
+ try_to_freeze();
+ schedule();
+ set_current_state(TASK_IDLE);
+ }
__set_current_state(TASK_RUNNING);
-
- /* We *must* be removed from the list before we can continue.
- * If we were woken, this is already done
- */
- if (!svc_thread_busy(rqstp)) {
- spin_lock_bh(&pool->sp_lock);
- list_del_init(&rqstp->rq_idle);
- spin_unlock_bh(&pool->sp_lock);
}
try_to_freeze();
+ clear_bit(SP_TASK_PENDING, &pool->sp_flags);
+
rqstp->rq_xprt = svc_xprt_dequeue(pool);
if (rqstp->rq_xprt) {
trace_svc_pool_awoken(rqstp);
@@ -785,7 +794,7 @@ static void svc_wait_for_work(struct svc_rqst *rqstp)
/* Normally we will wait up to 5 seconds for any required
* cache information to be provided.
*/
- if (!list_empty(&pool->sp_idle_threads))
+ if (pool->sp_idle_threads.first != NULL)
rqstp->rq_chandle.thread_wait = 5*HZ;
else
rqstp->rq_chandle.thread_wait = 1*HZ;
With an llist we don't need to take a lock to add a thread to the list, though we still need a lock to remove it. That will go in the next patch. Unlike double-linked lists, a thread cannot remove itself from the list. Only the first thread can be removed, and that can change asynchronously. So some care is needed. Firstly, we don't add ourselves to the list of there is more work to do - we just go ahead and do it. So we remain in the busy state. That already applied to xprts being ready, it now applies to all possible triggers. We DON'T keep the likely annotation on this test for "more work to do". It isn't particularly likely that there is no work to do, and there is definitely no need to optimise for that case. Secondly if we do find something needs to be done after adding ourselves to the list, we simply wake up the first thread on the list. If that was us, we have been removed and can continue. If it was some other thread, they will do the work that needs to be done. We can safely sleep until woken. With this a thread could call rqst_should_sleep() without actually proceeding to do work if told not to sleep. So that function mustn't clear SP_TASK_PENDING. The flag must stay set until some thread commits to not sleeping. We also remove the test on freezing() from rqst_should_sleep(). Instead we always try_to_freeze() before scheduling, which is needed as we now schedule() in a loop waiting to be removed from the idle queue. Signed-off-by: NeilBrown <neilb@suse.de> --- include/linux/sunrpc/svc.h | 14 +++++++----- net/sunrpc/svc.c | 11 +++++---- net/sunrpc/svc_xprt.c | 51 ++++++++++++++++++++++++++------------------ 3 files changed, 44 insertions(+), 32 deletions(-)