[PATCHSET 0/3] Misc cleanups

[PATCHSET 0/3] Misc cleanups

[Jens Axboe]

Hi,

Nothing particularly interesting in here, just a bit of code cleanup
I came across while doing other changes.

-- 
Jens Axboe

[PATCH 1/3] io_uring: rename trace_io_uring_submit_sqe() tracepoint

[Jens Axboe]

It has nothing to do with the SQE at this point, it's a request
submission. While in there, get rid of the 'force_nonblock' argument
which is also dead, as we only pass in true.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 include/trace/events/io_uring.h | 15 ++++++---------
 io_uring/io_uring.c             |  3 +--
 2 files changed, 7 insertions(+), 11 deletions(-)

diff --git a/include/trace/events/io_uring.h b/include/trace/events/io_uring.h
index 936fd41bf147..69454f1f98b0 100644
--- a/include/trace/events/io_uring.h
+++ b/include/trace/events/io_uring.h
@@ -360,19 +360,18 @@ TRACE_EVENT(io_uring_complete,
 );
 
 /**
- * io_uring_submit_sqe - called before submitting one SQE
+ * io_uring_submit_req - called before submitting a request
  *
  * @req:		pointer to a submitted request
- * @force_nonblock:	whether a context blocking or not
  *
  * Allows to track SQE submitting, to understand what was the source of it, SQ
  * thread or io_uring_enter call.
  */
-TRACE_EVENT(io_uring_submit_sqe,
+TRACE_EVENT(io_uring_submit_req,
 
-	TP_PROTO(struct io_kiocb *req, bool force_nonblock),
+	TP_PROTO(struct io_kiocb *req),
 
-	TP_ARGS(req, force_nonblock),
+	TP_ARGS(req),
 
 	TP_STRUCT__entry (
 		__field(  void *,		ctx		)
@@ -380,7 +379,6 @@ TRACE_EVENT(io_uring_submit_sqe,
 		__field(  unsigned long long,	user_data	)
 		__field(  u8,			opcode		)
 		__field(  u32,			flags		)
-		__field(  bool,			force_nonblock	)
 		__field(  bool,			sq_thread	)
 
 		__string( op_str, io_uring_get_opcode(req->opcode) )
@@ -392,16 +390,15 @@ TRACE_EVENT(io_uring_submit_sqe,
 		__entry->user_data	= req->cqe.user_data;
 		__entry->opcode		= req->opcode;
 		__entry->flags		= req->flags;
-		__entry->force_nonblock	= force_nonblock;
 		__entry->sq_thread	= req->ctx->flags & IORING_SETUP_SQPOLL;
 
 		__assign_str(op_str, io_uring_get_opcode(req->opcode));
 	),
 
 	TP_printk("ring %p, req %p, user_data 0x%llx, opcode %s, flags 0x%x, "
-		  "non block %d, sq_thread %d", __entry->ctx, __entry->req,
+		  "sq_thread %d", __entry->ctx, __entry->req,
 		  __entry->user_data, __get_str(op_str),
-		  __entry->flags, __entry->force_nonblock, __entry->sq_thread)
+		  __entry->flags, __entry->sq_thread)
 );
 
 /*
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 536940675c67..775b53730c2f 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -2305,8 +2305,7 @@ static inline int io_submit_sqe(struct io_ring_ctx *ctx, struct io_kiocb *req,
 	if (unlikely(ret))
 		return io_submit_fail_init(sqe, req, ret);
 
-	/* don't need @sqe from now on */
-	trace_io_uring_submit_sqe(req, true);
+	trace_io_uring_submit_req(req);
 
 	/*
 	 * If we already have a head request, queue this one for async
-- 
2.39.2

[PATCH 2/3] io_uring: cap io_sqring_entries() at SQ ring size

[Jens Axboe]

We already do this manually for the !SQPOLL case, do it in general and
we can also dump the ugly min3() in io_submit_sqes().

Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 io_uring/io_uring.c | 2 +-
 io_uring/io_uring.h | 4 +++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 775b53730c2f..a0b64831c455 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -2434,7 +2434,7 @@ int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr)
 	if (unlikely(!entries))
 		return 0;
 	/* make sure SQ entry isn't read before tail */
-	ret = left = min3(nr, ctx->sq_entries, entries);
+	ret = left = min(nr, entries);
 	io_get_task_refs(left);
 	io_submit_state_start(&ctx->submit_state, left);
 
diff --git a/io_uring/io_uring.h b/io_uring/io_uring.h
index c33f719731ac..193b2db39fe8 100644
--- a/io_uring/io_uring.h
+++ b/io_uring/io_uring.h
@@ -262,9 +262,11 @@ static inline bool io_sqring_full(struct io_ring_ctx *ctx)
 static inline unsigned int io_sqring_entries(struct io_ring_ctx *ctx)
 {
 	struct io_rings *rings = ctx->rings;
+	unsigned int entries;
 
 	/* make sure SQ entry isn't read before tail */
-	return smp_load_acquire(&rings->sq.tail) - ctx->cached_sq_head;
+	entries = smp_load_acquire(&rings->sq.tail) - ctx->cached_sq_head;
+	return min(entries, ctx->sq_entries);
 }
 
 static inline int io_run_task_work(void)
-- 
2.39.2

[PATCH 3/3] io_uring: get rid of io_preinit_req()

[Jens Axboe]

Just assign where we setup the ring anyway, splitting the init into
two doesn't really buy us anything and it's a bit more fragile. With
this, io_init_req() handles the whole thing while the cacheline is
pulled in anyway.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 io_uring/io_uring.c | 19 ++++---------------
 1 file changed, 4 insertions(+), 15 deletions(-)

diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index a0b64831c455..7f5d0b833955 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -1034,19 +1034,6 @@ void io_req_defer_failed(struct io_kiocb *req, s32 res)
 	io_req_complete_defer(req);
 }
 
-/*
- * Don't initialise the fields below on every allocation, but do that in
- * advance and keep them valid across allocations.
- */
-static void io_preinit_req(struct io_kiocb *req, struct io_ring_ctx *ctx)
-{
-	req->ctx = ctx;
-	req->link = NULL;
-	req->async_data = NULL;
-	/* not necessary, but safer to zero */
-	req->cqe.res = 0;
-}
-
 static void io_flush_cached_locked_reqs(struct io_ring_ctx *ctx,
 					struct io_submit_state *state)
 {
@@ -1097,7 +1084,6 @@ __cold bool __io_alloc_req_refill(struct io_ring_ctx *ctx)
 	for (i = 0; i < ret; i++) {
 		struct io_kiocb *req = reqs[i];
 
-		io_preinit_req(req, ctx);
 		io_req_add_to_cache(req, ctx);
 	}
 	return true;
@@ -2172,14 +2158,17 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
 	int personality;
 	u8 opcode;
 
-	/* req is partially pre-initialised, see io_preinit_req() */
 	req->opcode = opcode = READ_ONCE(sqe->opcode);
 	/* same numerical values with corresponding REQ_F_*, safe to copy */
 	req->flags = sqe_flags = READ_ONCE(sqe->flags);
 	req->cqe.user_data = READ_ONCE(sqe->user_data);
+	req->cqe.res = 0;
+	req->ctx = ctx;
 	req->file = NULL;
 	req->rsrc_node = NULL;
 	req->task = current;
+	req->async_data = NULL;
+	req->link = NULL;
 
 	if (unlikely(opcode >= IORING_OP_LAST)) {
 		req->opcode = 0;
-- 
2.39.2