Merge tag 'io_uring-6.14-20250131' of git://git.kernel.dk/linux

Pull more io_uring updates from Jens Axboe:

 - Series cleaning up the alloc cache changes from this merge window,
   and then another series on top making it better yet.

   This also solves an issue with KASAN_EXTRA_INFO, by making io_uring
   resilient to KASAN using parts of the freed struct for storage

 - Cleanups and simplications to buffer cloning and io resource node
   management

 - Fix an issue introduced in this merge window where READ/WRITE_ONCE
   was used on an atomic_t, which made some archs complain

 - Fix for an errant connect retry when the socket has been shut down

 - Fix for multishot and provided buffers

* tag 'io_uring-6.14-20250131' of git://git.kernel.dk/linux:
  io_uring/net: don't retry connect operation on EPOLLERR
  io_uring/rw: simplify io_rw_recycle()
  io_uring: remove !KASAN guards from cache free
  io_uring/net: extract io_send_select_buffer()
  io_uring/net: clean io_msg_copy_hdr()
  io_uring/net: make io_net_vec_assign() return void
  io_uring: add alloc_cache.c
  io_uring: dont ifdef io_alloc_cache_kasan()
  io_uring: include all deps for alloc_cache.h
  io_uring: fix multishots with selected buffers
  io_uring/register: use atomic_read/write for sq_flags migration
  io_uring/alloc_cache: get rid of _nocache() helper
  io_uring: get rid of alloc cache init_once handling
  io_uring/uring_cmd: cleanup struct io_uring_cmd_data layout
  io_uring/uring_cmd: use cached cmd_op in io_uring_cmd_sock()
  io_uring/msg_ring: don't leave potentially dangling ->tctx pointer
  io_uring/rsrc: Move lockdep assert from io_free_rsrc_node() to caller
  io_uring/rsrc: remove unused parameter ctx for io_rsrc_node_alloc()
  io_uring: clean up io_uring_register_get_file()
  io_uring/rsrc: Simplify buffer cloning by locking both rings

include/linux/io_uring/cmd.h

@@ -19,8 +19,8 @@ struct io_uring_cmd {
 };
 
 struct io_uring_cmd_data {
-	struct io_uring_sqe	sqes[2];
 	void			*op_data;
+	struct io_uring_sqe	sqes[2];
 };
 
 static inline const void *io_uring_sqe_cmd(const struct io_uring_sqe *sqe)

include/linux/io_uring_types.h

@@ -222,7 +222,8 @@ struct io_alloc_cache {
 	void			**entries;
 	unsigned int		nr_cached;
 	unsigned int		max_cached;
-	size_t			elem_size;
+	unsigned int		elem_size;
+	unsigned int		init_clear;
 };
 
 struct io_ring_ctx {

io_uring/Makefile

@@ -13,7 +13,7 @@ obj-$(CONFIG_IO_URING)		+= io_uring.o opdef.o kbuf.o rsrc.o notif.o \
 					sync.o msg_ring.o advise.o openclose.o \
 					epoll.o statx.o timeout.o fdinfo.o \
 					cancel.o waitid.o register.o \
-					truncate.o memmap.o
+					truncate.o memmap.o alloc_cache.o
 obj-$(CONFIG_IO_WQ)		+= io-wq.o
 obj-$(CONFIG_FUTEX)		+= futex.o
 obj-$(CONFIG_NET_RX_BUSY_POLL) += napi.o

io_uring/alloc_cache.c

@@ -0,0 +1,44 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include "alloc_cache.h"
+
+void io_alloc_cache_free(struct io_alloc_cache *cache,
+			 void (*free)(const void *))
+{
+	void *entry;
+
+	if (!cache->entries)
+		return;
+
+	while ((entry = io_alloc_cache_get(cache)) != NULL)
+		free(entry);
+
+	kvfree(cache->entries);
+	cache->entries = NULL;
+}
+
+/* returns false if the cache was initialized properly */
+bool io_alloc_cache_init(struct io_alloc_cache *cache,
+			 unsigned max_nr, unsigned int size,
+			 unsigned int init_bytes)
+{
+	cache->entries = kvmalloc_array(max_nr, sizeof(void *), GFP_KERNEL);
+	if (!cache->entries)
+		return true;
+
+	cache->nr_cached = 0;
+	cache->max_cached = max_nr;
+	cache->elem_size = size;
+	cache->init_clear = init_bytes;
+	return false;
+}
+
+void *io_cache_alloc_new(struct io_alloc_cache *cache, gfp_t gfp)
+{
+	void *obj;
+
+	obj = kmalloc(cache->elem_size, gfp);
+	if (obj && cache->init_clear)
+		memset(obj, 0, cache->init_clear);
+	return obj;
+}

io_uring/alloc_cache.h

@@ -1,11 +1,30 @@
 #ifndef IOU_ALLOC_CACHE_H
 #define IOU_ALLOC_CACHE_H
 
+#include <linux/io_uring_types.h>
+
 /*
  * Don't allow the cache to grow beyond this size.
  */
 #define IO_ALLOC_CACHE_MAX	128
 
+void io_alloc_cache_free(struct io_alloc_cache *cache,
+			 void (*free)(const void *));
+bool io_alloc_cache_init(struct io_alloc_cache *cache,
+			 unsigned max_nr, unsigned int size,
+			 unsigned int init_bytes);
+
+void *io_cache_alloc_new(struct io_alloc_cache *cache, gfp_t gfp);
+
+static inline void io_alloc_cache_kasan(struct iovec **iov, int *nr)
+{
+	if (IS_ENABLED(CONFIG_KASAN)) {
+		kfree(*iov);
+		*iov = NULL;
+		*nr = 0;
+	}
+}
+
 static inline bool io_alloc_cache_put(struct io_alloc_cache *cache,
 				      void *entry)
 {
@@ -23,52 +42,30 @@ static inline void *io_alloc_cache_get(struct io_alloc_cache *cache)
 	if (cache->nr_cached) {
 		void *entry = cache->entries[--cache->nr_cached];
 
+		/*
+		 * If KASAN is enabled, always clear the initial bytes that
+		 * must be zeroed post alloc, in case any of them overlap
+		 * with KASAN storage.
+		 */
+#if defined(CONFIG_KASAN)
 		kasan_mempool_unpoison_object(entry, cache->elem_size);
+		if (cache->init_clear)
+			memset(entry, 0, cache->init_clear);
+#endif
 		return entry;
 	}
 
 	return NULL;
 }
 
-static inline void *io_cache_alloc(struct io_alloc_cache *cache, gfp_t gfp,
-				   void (*init_once)(void *obj))
+static inline void *io_cache_alloc(struct io_alloc_cache *cache, gfp_t gfp)
 {
-	if (unlikely(!cache->nr_cached)) {
-		void *obj = kmalloc(cache->elem_size, gfp);
+	void *obj;
 
-		if (obj && init_once)
-			init_once(obj);
+	obj = io_alloc_cache_get(cache);
+	if (obj)
 		return obj;
-	}
-	return io_alloc_cache_get(cache);
+	return io_cache_alloc_new(cache, gfp);
 }
 
-/* returns false if the cache was initialized properly */
-static inline bool io_alloc_cache_init(struct io_alloc_cache *cache,
-				       unsigned max_nr, size_t size)
-{
-	cache->entries = kvmalloc_array(max_nr, sizeof(void *), GFP_KERNEL);
-	if (cache->entries) {
-		cache->nr_cached = 0;
-		cache->max_cached = max_nr;
-		cache->elem_size = size;
-		return false;
-	}
-	return true;
-}
-
-static inline void io_alloc_cache_free(struct io_alloc_cache *cache,
-				       void (*free)(const void *))
-{
-	void *entry;
-
-	if (!cache->entries)
-		return;
-
-	while ((entry = io_alloc_cache_get(cache)) != NULL)
-		free(entry);
-
-	kvfree(cache->entries);
-	cache->entries = NULL;
-}
 #endif

io_uring/filetable.c

@@ -68,7 +68,7 @@ static int io_install_fixed_file(struct io_ring_ctx *ctx, struct file *file,
 	if (slot_index >= ctx->file_table.data.nr)
 		return -EINVAL;
 
-	node = io_rsrc_node_alloc(ctx, IORING_RSRC_FILE);
+	node = io_rsrc_node_alloc(IORING_RSRC_FILE);
 	if (!node)
 		return -ENOMEM;
 

io_uring/futex.c

@@ -36,7 +36,7 @@ struct io_futex_data {
 bool io_futex_cache_init(struct io_ring_ctx *ctx)
 {
 	return io_alloc_cache_init(&ctx->futex_cache, IO_FUTEX_ALLOC_CACHE_MAX,
-				sizeof(struct io_futex_data));
+				sizeof(struct io_futex_data), 0);
 }
 
 void io_futex_cache_free(struct io_ring_ctx *ctx)
@@ -320,7 +320,7 @@ int io_futex_wait(struct io_kiocb *req, unsigned int issue_flags)
 	}
 
 	io_ring_submit_lock(ctx, issue_flags);
-	ifd = io_cache_alloc(&ctx->futex_cache, GFP_NOWAIT, NULL);
+	ifd = io_cache_alloc(&ctx->futex_cache, GFP_NOWAIT);
 	if (!ifd) {
 		ret = -ENOMEM;
 		goto done_unlock;

io_uring/io_uring.c

@@ -315,16 +315,18 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
 	INIT_LIST_HEAD(&ctx->cq_overflow_list);
 	INIT_LIST_HEAD(&ctx->io_buffers_cache);
 	ret = io_alloc_cache_init(&ctx->apoll_cache, IO_POLL_ALLOC_CACHE_MAX,
-			    sizeof(struct async_poll));
+			    sizeof(struct async_poll), 0);
 	ret |= io_alloc_cache_init(&ctx->netmsg_cache, IO_ALLOC_CACHE_MAX,
-			    sizeof(struct io_async_msghdr));
+			    sizeof(struct io_async_msghdr),
+			    offsetof(struct io_async_msghdr, clear));
 	ret |= io_alloc_cache_init(&ctx->rw_cache, IO_ALLOC_CACHE_MAX,
-			    sizeof(struct io_async_rw));
+			    sizeof(struct io_async_rw),
+			    offsetof(struct io_async_rw, clear));
 	ret |= io_alloc_cache_init(&ctx->uring_cache, IO_ALLOC_CACHE_MAX,
-			    sizeof(struct io_uring_cmd_data));
+			    sizeof(struct io_uring_cmd_data), 0);
 	spin_lock_init(&ctx->msg_lock);
 	ret |= io_alloc_cache_init(&ctx->msg_cache, IO_ALLOC_CACHE_MAX,
-			    sizeof(struct io_kiocb));
+			    sizeof(struct io_kiocb), 0);
 	ret |= io_futex_cache_init(ctx);
 	if (ret)
 		goto free_ref;

io_uring/io_uring.h

@@ -226,21 +226,16 @@ static inline void io_req_set_res(struct io_kiocb *req, s32 res, u32 cflags)
 }
 
 static inline void *io_uring_alloc_async_data(struct io_alloc_cache *cache,
-					      struct io_kiocb *req,
-					      void (*init_once)(void *obj))
+					      struct io_kiocb *req)
 {
-	req->async_data = io_cache_alloc(cache, GFP_KERNEL, init_once);
-	if (req->async_data)
-		req->flags |= REQ_F_ASYNC_DATA;
-	return req->async_data;
-}
+	if (cache) {
+		req->async_data = io_cache_alloc(cache, GFP_KERNEL);
+	} else {
+		const struct io_issue_def *def = &io_issue_defs[req->opcode];
 
-static inline void *io_uring_alloc_async_data_nocache(struct io_kiocb *req)
-{
-	const struct io_issue_def *def = &io_issue_defs[req->opcode];
-
-	WARN_ON_ONCE(!def->async_size);
-	req->async_data = kmalloc(def->async_size, GFP_KERNEL);
+		WARN_ON_ONCE(!def->async_size);
+		req->async_data = kmalloc(def->async_size, GFP_KERNEL);
+	}
 	if (req->async_data)
 		req->flags |= REQ_F_ASYNC_DATA;
 	return req->async_data;

io_uring/msg_ring.c

@@ -89,15 +89,15 @@ static void io_msg_tw_complete(struct io_kiocb *req, struct io_tw_state *ts)
 static int io_msg_remote_post(struct io_ring_ctx *ctx, struct io_kiocb *req,
 			      int res, u32 cflags, u64 user_data)
 {
-	req->tctx = READ_ONCE(ctx->submitter_task->io_uring);
-	if (!req->tctx) {
+	if (!READ_ONCE(ctx->submitter_task)) {
 		kmem_cache_free(req_cachep, req);
 		return -EOWNERDEAD;
 	}
 	req->cqe.user_data = user_data;
 	io_req_set_res(req, res, cflags);
 	percpu_ref_get(&ctx->refs);
 	req->ctx = ctx;
+	req->tctx = NULL;
 	req->io_task_work.func = io_msg_tw_complete;
 	io_req_task_work_add_remote(req, ctx, IOU_F_TWQ_LAZY_WAKE);
 	return 0;