Commit 7380c60b authored by Linus Torvalds's avatar Linus Torvalds
Browse files

Merge tag 'io_uring-6.15-20250509' of git://git.kernel.dk/linux 

Pull io_uring fixes from Jens Axboe:

 - Fix for linked timeouts arming and firing wrt prep and issue of the
   request being managed by the linked timeout

 - Fix for a CQE ordering issue between requests with multishot and
   using the same buffer group. This is a dumbed down version for this
   release and for stable, it'll get improved for v6.16

 - Tweak the SQPOLL submit batch size. A previous commit made SQPOLL
   manage its own task_work and chose a tiny batch size, bump it from 8
   to 32 to fix a performance regression due to that

* tag 'io_uring-6.15-20250509' of git://git.kernel.dk/linux:
  io_uring/sqpoll: Increase task_work submission batch size
  io_uring: ensure deferred completions are flushed for multishot
  io_uring: always arm linked timeouts prior to issue
parents 29fe5d50 92835ceb

io_uring/io_uring.c

+23 −35
Original line number Diff line number Diff line
@@ -448,24 +448,6 @@ static struct io_kiocb *__io_prep_linked_timeout(struct io_kiocb *req) 
	return req->link;

}



static inline struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req)

{

	if (likely(!(req->flags & REQ_F_ARM_LTIMEOUT)))

		return NULL;

	return __io_prep_linked_timeout(req);

}



static noinline void __io_arm_ltimeout(struct io_kiocb *req)

{

	io_queue_linked_timeout(__io_prep_linked_timeout(req));

}



static inline void io_arm_ltimeout(struct io_kiocb *req)

{

	if (unlikely(req->flags & REQ_F_ARM_LTIMEOUT))

		__io_arm_ltimeout(req);

}



static void io_prep_async_work(struct io_kiocb *req)

{

	const struct io_issue_def *def = &io_issue_defs[req->opcode];
@@ -518,7 +500,6 @@ static void io_prep_async_link(struct io_kiocb *req) 


static void io_queue_iowq(struct io_kiocb *req)

{

	struct io_kiocb *link = io_prep_linked_timeout(req);

	struct io_uring_task *tctx = req->tctx;



	BUG_ON(!tctx);
@@ -543,8 +524,6 @@ static void io_queue_iowq(struct io_kiocb *req) 


	trace_io_uring_queue_async_work(req, io_wq_is_hashed(&req->work));

	io_wq_enqueue(tctx->io_wq, &req->work);

	if (link)

		io_queue_linked_timeout(link);

}



static void io_req_queue_iowq_tw(struct io_kiocb *req, io_tw_token_t tw)
@@ -869,6 +848,14 @@ bool io_req_post_cqe(struct io_kiocb *req, s32 res, u32 cflags) 
	struct io_ring_ctx *ctx = req->ctx;

	bool posted;



	/*

	 * If multishot has already posted deferred completions, ensure that

	 * those are flushed first before posting this one. If not, CQEs

	 * could get reordered.

	 */

	if (!wq_list_empty(&ctx->submit_state.compl_reqs))

		__io_submit_flush_completions(ctx);



	lockdep_assert(!io_wq_current_is_worker());

	lockdep_assert_held(&ctx->uring_lock);
@@ -1724,15 +1711,22 @@ static bool io_assign_file(struct io_kiocb *req, const struct io_issue_def *def, 
	return !!req->file;

}



#define REQ_ISSUE_SLOW_FLAGS	(REQ_F_CREDS | REQ_F_ARM_LTIMEOUT)



static inline int __io_issue_sqe(struct io_kiocb *req,

				 unsigned int issue_flags,

				 const struct io_issue_def *def)

{

	const struct cred *creds = NULL;

	struct io_kiocb *link = NULL;

	int ret;



	if (unlikely((req->flags & REQ_F_CREDS) && req->creds != current_cred()))

	if (unlikely(req->flags & REQ_ISSUE_SLOW_FLAGS)) {

		if ((req->flags & REQ_F_CREDS) && req->creds != current_cred())

			creds = override_creds(req->creds);

		if (req->flags & REQ_F_ARM_LTIMEOUT)

			link = __io_prep_linked_timeout(req);

	}



	if (!def->audit_skip)

		audit_uring_entry(req->opcode);
@@ -1742,8 +1736,12 @@ static inline int __io_issue_sqe(struct io_kiocb *req, 
	if (!def->audit_skip)

		audit_uring_exit(!ret, ret);



	if (unlikely(creds || link)) {

		if (creds)

			revert_creds(creds);

		if (link)

			io_queue_linked_timeout(link);

	}



	return ret;

}
@@ -1769,7 +1767,6 @@ static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags) 


	if (ret == IOU_ISSUE_SKIP_COMPLETE) {

		ret = 0;

		io_arm_ltimeout(req);



		/* If the op doesn't have a file, we're not polling for it */

		if ((req->ctx->flags & IORING_SETUP_IOPOLL) && def->iopoll_queue)
@@ -1824,8 +1821,6 @@ void io_wq_submit_work(struct io_wq_work *work) 
	else

		req_ref_get(req);



	io_arm_ltimeout(req);



	/* either cancelled or io-wq is dying, so don't touch tctx->iowq */

	if (atomic_read(&work->flags) & IO_WQ_WORK_CANCEL) {

fail:
@@ -1941,15 +1936,11 @@ struct file *io_file_get_normal(struct io_kiocb *req, int fd) 
static void io_queue_async(struct io_kiocb *req, int ret)

	__must_hold(&req->ctx->uring_lock)

{

	struct io_kiocb *linked_timeout;



	if (ret != -EAGAIN || (req->flags & REQ_F_NOWAIT)) {

		io_req_defer_failed(req, ret);

		return;

	}



	linked_timeout = io_prep_linked_timeout(req);



	switch (io_arm_poll_handler(req, 0)) {

	case IO_APOLL_READY:

		io_kbuf_recycle(req, 0);
@@ -1962,9 +1953,6 @@ static void io_queue_async(struct io_kiocb *req, int ret) 
	case IO_APOLL_OK:

		break;

	}



	if (linked_timeout)

		io_queue_linked_timeout(linked_timeout);

}



static inline void io_queue_sqe(struct io_kiocb *req)

io_uring/sqpoll.c

+1 −1
Original line number Diff line number Diff line
@@ -20,7 +20,7 @@ 
#include "sqpoll.h"



#define IORING_SQPOLL_CAP_ENTRIES_VALUE 8

#define IORING_TW_CAP_ENTRIES_VALUE	8

#define IORING_TW_CAP_ENTRIES_VALUE	32



enum {

	IO_SQ_THREAD_SHOULD_STOP = 0,