io_uring: switch cancel_hash to use per entry spinlock

Add a new io_hash_bucket structure so that each bucket in cancel_hash
has separate spinlock. Use per entry lock for cancel_hash, this removes
some completion lock invocation and remove contension between different
cancel_hash entries.
Signed-off-by: NHao Xu <howeyxu@tencent.com>
Signed-off-by: NPavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/05d1e135b0c8bce9d1441e6346776589e5783e26.1655371007.git.asml.silence@gmail.comSigned-off-by: NJens Axboe <axboe@kernel.dk>

io_uring/cancel.c

@@ -93,14 +93,14 @@ int io_try_cancel(struct io_kiocb *req, struct io_cancel_data *cd)
 	if (!ret)
 		return 0;
 
-	spin_lock(&ctx->completion_lock);
 	ret = io_poll_cancel(ctx, cd);
 	if (ret != -ENOENT)
 		goto out;
+	spin_lock(&ctx->completion_lock);
 	if (!(cd->flags & IORING_ASYNC_CANCEL_FD))
 		ret = io_timeout_cancel(ctx, cd);
-out:
 	spin_unlock(&ctx->completion_lock);
+out:
 	return ret;
 }
 
@@ -192,3 +192,13 @@ int io_async_cancel(struct io_kiocb *req, unsigned int issue_flags)
 	io_req_set_res(req, ret, 0);
 	return IOU_OK;
 }
+
+void init_hash_table(struct io_hash_bucket *hash_table, unsigned size)
+{
+	unsigned int i;
+
+	for (i = 0; i < size; i++) {
+		spin_lock_init(&hash_table[i].lock);
+		INIT_HLIST_HEAD(&hash_table[i].list);
+	}
+}

io_uring/cancel.h

@@ -4,3 +4,9 @@ int io_async_cancel_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe);
 int io_async_cancel(struct io_kiocb *req, unsigned int issue_flags);
 
 int io_try_cancel(struct io_kiocb *req, struct io_cancel_data *cd);
+void init_hash_table(struct io_hash_bucket *hash_table, unsigned size);
+
+struct io_hash_bucket {
+	spinlock_t		lock;
+	struct hlist_head	list;
+} ____cacheline_aligned_in_smp;

io_uring/fdinfo.c

@@ -13,6 +13,7 @@
 #include "io_uring.h"
 #include "sqpoll.h"
 #include "fdinfo.h"
+#include "cancel.h"
 
 #ifdef CONFIG_PROC_FS
 static __cold int io_uring_show_cred(struct seq_file *m, unsigned int id,
@@ -157,17 +158,19 @@ static __cold void __io_uring_show_fdinfo(struct io_ring_ctx *ctx,
 		mutex_unlock(&ctx->uring_lock);
 
 	seq_puts(m, "PollList:\n");
-	spin_lock(&ctx->completion_lock);
 	for (i = 0; i < (1U << ctx->cancel_hash_bits); i++) {
-		struct hlist_head *list = &ctx->cancel_hash[i];
+		struct io_hash_bucket *hb = &ctx->cancel_hash[i];
 		struct io_kiocb *req;
 
-		hlist_for_each_entry(req, list, hash_node)
+		spin_lock(&hb->lock);
+		hlist_for_each_entry(req, &hb->list, hash_node)
 			seq_printf(m, "  op=%d, task_works=%d\n", req->opcode,
 					task_work_pending(req->task));
+		spin_unlock(&hb->lock);
 	}
 
 	seq_puts(m, "CqOverflowList:\n");
+	spin_lock(&ctx->completion_lock);
 	list_for_each_entry(ocqe, &ctx->cq_overflow_list, list) {
 		struct io_uring_cqe *cqe = &ocqe->cqe;
 

io_uring/io_uring.c

@@ -89,6 +89,7 @@
 #include "fdinfo.h"
 #include "kbuf.h"
 #include "rsrc.h"
+#include "cancel.h"
 
 #include "timeout.h"
 #include "poll.h"
@@ -260,11 +261,13 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
 	if (hash_bits <= 0)
 		hash_bits = 1;
 	ctx->cancel_hash_bits = hash_bits;
-	ctx->cancel_hash = kmalloc((1U << hash_bits) * sizeof(struct hlist_head),
-					GFP_KERNEL);
+	ctx->cancel_hash =
+		kmalloc((1U << hash_bits) * sizeof(struct io_hash_bucket),
+			GFP_KERNEL);
 	if (!ctx->cancel_hash)
 		goto err;
-	__hash_init(ctx->cancel_hash, 1U << hash_bits);
+
+	init_hash_table(ctx->cancel_hash, 1U << hash_bits);
 
 	ctx->dummy_ubuf = kzalloc(sizeof(*ctx->dummy_ubuf), GFP_KERNEL);
 	if (!ctx->dummy_ubuf)

io_uring/io_uring_types.h

@@ -224,7 +224,7 @@ struct io_ring_ctx {
 		 * manipulate the list, hence no extra locking is needed there.
 		 */
 		struct io_wq_work_list	iopoll_list;
-		struct hlist_head	*cancel_hash;
+		struct io_hash_bucket	*cancel_hash;
 		unsigned		cancel_hash_bits;
 		bool			poll_multi_queue;
 

io_uring/poll.c

@@ -19,6 +19,7 @@
 #include "opdef.h"
 #include "kbuf.h"
 #include "poll.h"
+#include "cancel.h"
 
 struct io_poll_update {
 	struct file			*file;
@@ -73,10 +74,22 @@ static struct io_poll *io_poll_get_single(struct io_kiocb *req)
 static void io_poll_req_insert(struct io_kiocb *req)
 {
 	struct io_ring_ctx *ctx = req->ctx;
-	struct hlist_head *list;
+	u32 index = hash_long(req->cqe.user_data, ctx->cancel_hash_bits);
+	struct io_hash_bucket *hb = &ctx->cancel_hash[index];
 
-	list = &ctx->cancel_hash[hash_long(req->cqe.user_data, ctx->cancel_hash_bits)];
-	hlist_add_head(&req->hash_node, list);
+	spin_lock(&hb->lock);
+	hlist_add_head(&req->hash_node, &hb->list);
+	spin_unlock(&hb->lock);
+}
+
+static void io_poll_req_delete(struct io_kiocb *req, struct io_ring_ctx *ctx)
+{
+	u32 index = hash_long(req->cqe.user_data, ctx->cancel_hash_bits);
+	spinlock_t *lock = &ctx->cancel_hash[index].lock;
+
+	spin_lock(lock);
+	hash_del(&req->hash_node);
+	spin_unlock(lock);
 }
 
 static void io_init_poll_iocb(struct io_poll *poll, __poll_t events,
@@ -220,8 +233,8 @@ static void io_poll_task_func(struct io_kiocb *req, bool *locked)
 	}
 
 	io_poll_remove_entries(req);
+	io_poll_req_delete(req, ctx);
 	spin_lock(&ctx->completion_lock);
-	hash_del(&req->hash_node);
 	req->cqe.flags = 0;
 	__io_req_complete_post(req);
 	io_commit_cqring(ctx);
@@ -231,7 +244,6 @@ static void io_poll_task_func(struct io_kiocb *req, bool *locked)
 
 static void io_apoll_task_func(struct io_kiocb *req, bool *locked)
 {
-	struct io_ring_ctx *ctx = req->ctx;
 	int ret;
 
 	ret = io_poll_check_events(req, locked);
@@ -239,9 +251,7 @@ static void io_apoll_task_func(struct io_kiocb *req, bool *locked)
 		return;
 
 	io_poll_remove_entries(req);
-	spin_lock(&ctx->completion_lock);
-	hash_del(&req->hash_node);
-	spin_unlock(&ctx->completion_lock);
+	io_poll_req_delete(req, req->ctx);
 
 	if (!ret)
 		io_req_task_submit(req, locked);
@@ -437,9 +447,7 @@ static int __io_arm_poll_handler(struct io_kiocb *req,
 		return 0;
 	}
 
-	spin_lock(&ctx->completion_lock);
 	io_poll_req_insert(req);
-	spin_unlock(&ctx->completion_lock);
 
 	if (mask && (poll->events & EPOLLET)) {
 		/* can't multishot if failed, just queue the event we've got */
@@ -540,32 +548,31 @@ __cold bool io_poll_remove_all(struct io_ring_ctx *ctx, struct task_struct *tsk,
 	bool found = false;
 	int i;
 
-	spin_lock(&ctx->completion_lock);
 	for (i = 0; i < (1U << ctx->cancel_hash_bits); i++) {
-		struct hlist_head *list;
+		struct io_hash_bucket *hb = &ctx->cancel_hash[i];
 
-		list = &ctx->cancel_hash[i];
-		hlist_for_each_entry_safe(req, tmp, list, hash_node) {
+		spin_lock(&hb->lock);
+		hlist_for_each_entry_safe(req, tmp, &hb->list, hash_node) {
 			if (io_match_task_safe(req, tsk, cancel_all)) {
 				hlist_del_init(&req->hash_node);
 				io_poll_cancel_req(req);
 				found = true;
 			}
 		}
+		spin_unlock(&hb->lock);
 	}
-	spin_unlock(&ctx->completion_lock);
 	return found;
 }
 
 static struct io_kiocb *io_poll_find(struct io_ring_ctx *ctx, bool poll_only,
 				     struct io_cancel_data *cd)
-	__must_hold(&ctx->completion_lock)
 {
-	struct hlist_head *list;
 	struct io_kiocb *req;
+	u32 index = hash_long(cd->data, ctx->cancel_hash_bits);
+	struct io_hash_bucket *hb = &ctx->cancel_hash[index];
 
-	list = &ctx->cancel_hash[hash_long(cd->data, ctx->cancel_hash_bits)];
-	hlist_for_each_entry(req, list, hash_node) {
+	spin_lock(&hb->lock);
+	hlist_for_each_entry(req, &hb->list, hash_node) {
 		if (cd->data != req->cqe.user_data)
 			continue;
 		if (poll_only && req->opcode != IORING_OP_POLL_ADD)
@@ -577,21 +584,21 @@ static struct io_kiocb *io_poll_find(struct io_ring_ctx *ctx, bool poll_only,
 		}
 		return req;
 	}
+	spin_unlock(&hb->lock);
 	return NULL;
 }
 
 static struct io_kiocb *io_poll_file_find(struct io_ring_ctx *ctx,
 					  struct io_cancel_data *cd)
-	__must_hold(&ctx->completion_lock)
 {
 	struct io_kiocb *req;
 	int i;
 
 	for (i = 0; i < (1U << ctx->cancel_hash_bits); i++) {
-		struct hlist_head *list;
+		struct io_hash_bucket *hb = &ctx->cancel_hash[i];
 
-		list = &ctx->cancel_hash[i];
-		hlist_for_each_entry(req, list, hash_node) {
+		spin_lock(&hb->lock);
+		hlist_for_each_entry(req, &hb->list, hash_node) {
 			if (!(cd->flags & IORING_ASYNC_CANCEL_ANY) &&
 			    req->file != cd->file)
 				continue;
@@ -600,12 +607,12 @@ static struct io_kiocb *io_poll_file_find(struct io_ring_ctx *ctx,
 			req->work.cancel_seq = cd->seq;
 			return req;
 		}
+		spin_unlock(&hb->lock);
 	}
 	return NULL;
 }
 
 static bool io_poll_disarm(struct io_kiocb *req)
-	__must_hold(&ctx->completion_lock)
 {
 	if (!io_poll_get_ownership(req))
 		return false;
@@ -615,17 +622,23 @@ static bool io_poll_disarm(struct io_kiocb *req)
 }
 
 int io_poll_cancel(struct io_ring_ctx *ctx, struct io_cancel_data *cd)
-	__must_hold(&ctx->completion_lock)
 {
 	struct io_kiocb *req;
+	u32 index;
+	spinlock_t *lock;
 
 	if (cd->flags & (IORING_ASYNC_CANCEL_FD|IORING_ASYNC_CANCEL_ANY))
 		req = io_poll_file_find(ctx, cd);
 	else
 		req = io_poll_find(ctx, false, cd);
-	if (!req)
+	if (!req) {
 		return -ENOENT;
+	} else {
+		index = hash_long(req->cqe.user_data, ctx->cancel_hash_bits);
+		lock = &ctx->cancel_hash[index].lock;
+	}
 	io_poll_cancel_req(req);
+	spin_unlock(lock);
 	return 0;
 }
 
@@ -719,18 +732,23 @@ int io_poll_remove(struct io_kiocb *req, unsigned int issue_flags)
 	struct io_poll_update *poll_update = io_kiocb_to_cmd(req);
 	struct io_cancel_data cd = { .data = poll_update->old_user_data, };
 	struct io_ring_ctx *ctx = req->ctx;
+	u32 index = hash_long(cd.data, ctx->cancel_hash_bits);
+	spinlock_t *lock = &ctx->cancel_hash[index].lock;
 	struct io_kiocb *preq;
 	int ret2, ret = 0;
 	bool locked;
 
-	spin_lock(&ctx->completion_lock);
 	preq = io_poll_find(ctx, true, &cd);
-	if (!preq || !io_poll_disarm(preq)) {
-		spin_unlock(&ctx->completion_lock);
-		ret = preq ? -EALREADY : -ENOENT;
+	if (!preq) {
+		ret = -ENOENT;
+		goto out;
+	}
+	ret2 = io_poll_disarm(preq);
+	spin_unlock(lock);
+	if (!ret2) {
+		ret = -EALREADY;
 		goto out;
 	}
-	spin_unlock(&ctx->completion_lock);
 
 	if (poll_update->update_events || poll_update->update_user_data) {
 		/* only mask one event flags, keep behavior flags */