io_uring/napi: add static napi tracking strategy

Add the static napi tracking strategy. That allows the user to manually
manage the napi ids list for busy polling, and eliminate the overhead of
dynamically updating the list from the fast path.

Signed-off-by: Olivier Langlois <olivier@trillion01.com>
Link: https://lore.kernel.org/r/96943de14968c35a5c599352259ad98f3c0770ba.1728828877.git.olivier@trillion01.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

3 files changed, 129 insertions, 24 deletions

diff --git a/io_uring/fdinfo.c b/io_uring/fdinfo.c
index efbec34ccb18..b214e5a407b5 100644
--- a/io_uring/fdinfo.c
+++ b/io_uring/fdinfo.c
@@ -46,6 +46,46 @@ static __cold int io_uring_show_cred(struct seq_file *m, unsigned int id,
 	return 0;
 }
 
+#ifdef CONFIG_NET_RX_BUSY_POLL
+static __cold void common_tracking_show_fdinfo(struct io_ring_ctx *ctx,
+					       struct seq_file *m,
+					       const char *tracking_strategy)
+{
+	seq_puts(m, "NAPI:\tenabled\n");
+	seq_printf(m, "napi tracking:\t%s\n", tracking_strategy);
+	seq_printf(m, "napi_busy_poll_dt:\t%llu\n", ctx->napi_busy_poll_dt);
+	if (ctx->napi_prefer_busy_poll)
+		seq_puts(m, "napi_prefer_busy_poll:\ttrue\n");
+	else
+		seq_puts(m, "napi_prefer_busy_poll:\tfalse\n");
+}
+
+static __cold void napi_show_fdinfo(struct io_ring_ctx *ctx,
+				    struct seq_file *m)
+{
+	unsigned int mode = READ_ONCE(ctx->napi_track_mode);
+
+	switch (mode) {
+	case IO_URING_NAPI_TRACKING_INACTIVE:
+		seq_puts(m, "NAPI:\tdisabled\n");
+		break;
+	case IO_URING_NAPI_TRACKING_DYNAMIC:
+		common_tracking_show_fdinfo(ctx, m, "dynamic");
+		break;
+	case IO_URING_NAPI_TRACKING_STATIC:
+		common_tracking_show_fdinfo(ctx, m, "static");
+		break;
+	default:
+		seq_printf(m, "NAPI:\tunknown mode (%u)\n", mode);
+	}
+}
+#else
+static inline void napi_show_fdinfo(struct io_ring_ctx *ctx,
+				    struct seq_file *m)
+{
+}
+#endif
+
 /*
  * Caller holds a reference to the file already, we don't need to do
  * anything else to get an extra reference.
@@ -219,18 +259,6 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file)
 
 	}
 	spin_unlock(&ctx->completion_lock);
-
-#ifdef CONFIG_NET_RX_BUSY_POLL
-	if (ctx->napi_enabled) {
-		seq_puts(m, "NAPI:\tenabled\n");
-		seq_printf(m, "napi_busy_poll_dt:\t%llu\n", ctx->napi_busy_poll_dt);
-		if (ctx->napi_prefer_busy_poll)
-			seq_puts(m, "napi_prefer_busy_poll:\ttrue\n");
-		else
-			seq_puts(m, "napi_prefer_busy_poll:\tfalse\n");
-	} else {
-		seq_puts(m, "NAPI:\tdisabled\n");
-	}
-#endif
+	napi_show_fdinfo(ctx, m);
 }
 #endif
diff --git a/io_uring/napi.c b/io_uring/napi.c
index 1de1543d8034..b1ade3fda30f 100644
--- a/io_uring/napi.c
+++ b/io_uring/napi.c
@@ -81,6 +81,27 @@ int __io_napi_add_id(struct io_ring_ctx *ctx, unsigned int napi_id)
 	return 0;
 }
 
+static int __io_napi_del_id(struct io_ring_ctx *ctx, unsigned int napi_id)
+{
+	struct hlist_head *hash_list;
+	struct io_napi_entry *e;
+
+	/* Non-NAPI IDs can be rejected. */
+	if (napi_id < MIN_NAPI_ID)
+		return -EINVAL;
+
+	hash_list = &ctx->napi_ht[hash_min(napi_id, HASH_BITS(ctx->napi_ht))];
+	guard(spinlock)(&ctx->napi_lock);
+	e = io_napi_hash_find(hash_list, napi_id);
+	if (!e)
+		return -ENOENT;
+
+	list_del_rcu(&e->list);
+	hash_del_rcu(&e->node);
+	kfree_rcu(e, rcu);
+	return 0;
+}
+
 static void __io_napi_remove_stale(struct io_ring_ctx *ctx)
 {
 	struct io_napi_entry *e;
@@ -136,9 +157,25 @@ static bool io_napi_busy_loop_should_end(void *data,
 	return false;
 }
 
-static bool __io_napi_do_busy_loop(struct io_ring_ctx *ctx,
-				   bool (*loop_end)(void *, unsigned long),
-				   void *loop_end_arg)
+/*
+ * never report stale entries
+ */
+static bool static_tracking_do_busy_loop(struct io_ring_ctx *ctx,
+					 bool (*loop_end)(void *, unsigned long),
+					 void *loop_end_arg)
+{
+	struct io_napi_entry *e;
+
+	list_for_each_entry_rcu(e, &ctx->napi_list, list)
+		napi_busy_loop_rcu(e->napi_id, loop_end, loop_end_arg,
+				   ctx->napi_prefer_busy_poll, BUSY_POLL_BUDGET);
+	return false;
+}
+
+static bool
+dynamic_tracking_do_busy_loop(struct io_ring_ctx *ctx,
+			      bool (*loop_end)(void *, unsigned long),
+			      void *loop_end_arg)
 {
 	struct io_napi_entry *e;
 	bool is_stale = false;
@@ -154,6 +191,16 @@ static bool __io_napi_do_busy_loop(struct io_ring_ctx *ctx,
 	return is_stale;
 }
 
+static inline bool
+__io_napi_do_busy_loop(struct io_ring_ctx *ctx,
+		       bool (*loop_end)(void *, unsigned long),
+		       void *loop_end_arg)
+{
+	if (READ_ONCE(ctx->napi_track_mode) == IO_URING_NAPI_TRACKING_STATIC)
+		return static_tracking_do_busy_loop(ctx, loop_end, loop_end_arg);
+	return dynamic_tracking_do_busy_loop(ctx, loop_end, loop_end_arg);
+}
+
 static void io_napi_blocking_busy_loop(struct io_ring_ctx *ctx,
 				       struct io_wait_queue *iowq)
 {
@@ -195,6 +242,7 @@ void io_napi_init(struct io_ring_ctx *ctx)
 	spin_lock_init(&ctx->napi_lock);
 	ctx->napi_prefer_busy_poll = false;
 	ctx->napi_busy_poll_dt = ns_to_ktime(sys_dt);
+	ctx->napi_track_mode = IO_URING_NAPI_TRACKING_INACTIVE;
 }
 
 /*
@@ -215,6 +263,24 @@ void io_napi_free(struct io_ring_ctx *ctx)
 	INIT_LIST_HEAD_RCU(&ctx->napi_list);
 }
 
+static int io_napi_register_napi(struct io_ring_ctx *ctx,
+				 struct io_uring_napi *napi)
+{
+	switch (napi->op_param) {
+	case IO_URING_NAPI_TRACKING_DYNAMIC:
+	case IO_URING_NAPI_TRACKING_STATIC:
+		break;
+	default:
+		return -EINVAL;
+	}
+	/* clean the napi list for new settings */
+	io_napi_free(ctx);
+	WRITE_ONCE(ctx->napi_track_mode, napi->op_param);
+	WRITE_ONCE(ctx->napi_busy_poll_dt, napi->busy_poll_to * NSEC_PER_USEC);
+	WRITE_ONCE(ctx->napi_prefer_busy_poll, !!napi->prefer_busy_poll);
+	return 0;
+}
+
 /*
  * io_napi_register() - Register napi with io-uring
  * @ctx: pointer to io-uring context structure
@@ -226,7 +292,8 @@ int io_register_napi(struct io_ring_ctx *ctx, void __user *arg)
 {
 	const struct io_uring_napi curr = {
 		.busy_poll_to 	  = ktime_to_us(ctx->napi_busy_poll_dt),
-		.prefer_busy_poll = ctx->napi_prefer_busy_poll
+		.prefer_busy_poll = ctx->napi_prefer_busy_poll,
+		.op_param	  = ctx->napi_track_mode
 	};
 	struct io_uring_napi napi;
 
@@ -234,16 +301,26 @@ int io_register_napi(struct io_ring_ctx *ctx, void __user *arg)
 		return -EINVAL;
 	if (copy_from_user(&napi, arg, sizeof(napi)))
 		return -EFAULT;
-	if (napi.pad[0] || napi.pad[1] || napi.pad[2] || napi.resv)
+	if (napi.pad[0] || napi.pad[1] || napi.resv)
 		return -EINVAL;
 
 	if (copy_to_user(arg, &curr, sizeof(curr)))
 		return -EFAULT;
 
-	WRITE_ONCE(ctx->napi_busy_poll_dt, napi.busy_poll_to * NSEC_PER_USEC);
-	WRITE_ONCE(ctx->napi_prefer_busy_poll, !!napi.prefer_busy_poll);
-	WRITE_ONCE(ctx->napi_enabled, true);
-	return 0;
+	switch (napi.opcode) {
+	case IO_URING_NAPI_REGISTER_OP:
+		return io_napi_register_napi(ctx, &napi);
+	case IO_URING_NAPI_STATIC_ADD_ID:
+		if (curr.op_param != IO_URING_NAPI_TRACKING_STATIC)
+			return -EINVAL;
+		return __io_napi_add_id(ctx, napi.op_param);
+	case IO_URING_NAPI_STATIC_DEL_ID:
+		if (curr.op_param != IO_URING_NAPI_TRACKING_STATIC)
+			return -EINVAL;
+		return __io_napi_del_id(ctx, napi.op_param);
+	default:
+		return -EINVAL;
+	}
 }
 
 /*
@@ -266,7 +343,7 @@ int io_unregister_napi(struct io_ring_ctx *ctx, void __user *arg)
 
 	WRITE_ONCE(ctx->napi_busy_poll_dt, 0);
 	WRITE_ONCE(ctx->napi_prefer_busy_poll, false);
-	WRITE_ONCE(ctx->napi_enabled, false);
+	WRITE_ONCE(ctx->napi_track_mode, IO_URING_NAPI_TRACKING_INACTIVE);
 	return 0;
 }
 
diff --git a/io_uring/napi.h b/io_uring/napi.h
index 4ae622f37b30..fa742f42e09b 100644
--- a/io_uring/napi.h
+++ b/io_uring/napi.h
@@ -44,7 +44,7 @@ static inline void io_napi_add(struct io_kiocb *req)
 	struct io_ring_ctx *ctx = req->ctx;
 	struct socket *sock;
 
-	if (!READ_ONCE(ctx->napi_enabled))
+	if (READ_ONCE(ctx->napi_track_mode) != IO_URING_NAPI_TRACKING_DYNAMIC)
 		return;
 
 	sock = sock_from_file(req->file);