[PATCH for-next 2/6] io_uring: don't inline __io_get_cqe()

[Pavel Begunkov <asml.silence@gmail.com>]

__io_get_cqe() is not as hot as io_get_cqe(), no need to inline it, it
sheds ~500B from the binary.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 io_uring/io_uring.c | 35 +++++++++++++++++++++++++++++++++++
 io_uring/io_uring.h | 36 +-----------------------------------
 2 files changed, 36 insertions(+), 35 deletions(-)

diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 7ffb8422e7d0..a3b1339335c5 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -165,6 +165,11 @@ static inline void io_submit_flush_completions(struct io_ring_ctx *ctx)
 		__io_submit_flush_completions(ctx);
 }
 
+static inline unsigned int __io_cqring_events(struct io_ring_ctx *ctx)
+{
+	return ctx->cached_cq_tail - READ_ONCE(ctx->rings->cq.head);
+}
+
 static bool io_match_linked(struct io_kiocb *head)
 {
 	struct io_kiocb *req;
@@ -673,6 +678,36 @@ bool io_cqring_event_overflow(struct io_ring_ctx *ctx, u64 user_data, s32 res,
 	return true;
 }
 
+/*
+ * writes to the cq entry need to come after reading head; the
+ * control dependency is enough as we're using WRITE_ONCE to
+ * fill the cq entry
+ */
+struct io_uring_cqe *__io_get_cqe(struct io_ring_ctx *ctx)
+{
+	struct io_rings *rings = ctx->rings;
+	unsigned int off = ctx->cached_cq_tail & (ctx->cq_entries - 1);
+	unsigned int shift = 0;
+	unsigned int free, queued, len;
+
+	if (ctx->flags & IORING_SETUP_CQE32)
+		shift = 1;
+
+	/* userspace may cheat modifying the tail, be safe and do min */
+	queued = min(__io_cqring_events(ctx), ctx->cq_entries);
+	free = ctx->cq_entries - queued;
+	/* we need a contiguous range, limit based on the current array offset */
+	len = min(free, ctx->cq_entries - off);
+	if (!len)
+		return NULL;
+
+	ctx->cached_cq_tail++;
+	ctx->cqe_cached = &rings->cqes[off];
+	ctx->cqe_sentinel = ctx->cqe_cached + len;
+	ctx->cqe_cached++;
+	return &rings->cqes[off << shift];
+}
+
 static bool io_fill_cqe_aux(struct io_ring_ctx *ctx,
 			    u64 user_data, s32 res, u32 cflags)
 {
diff --git a/io_uring/io_uring.h b/io_uring/io_uring.h
index ce6538c9aed3..51032a494aec 100644
--- a/io_uring/io_uring.h
+++ b/io_uring/io_uring.h
@@ -16,44 +16,10 @@ enum {
 	IOU_ISSUE_SKIP_COMPLETE	= -EIOCBQUEUED,
 };
 
+struct io_uring_cqe *__io_get_cqe(struct io_ring_ctx *ctx);
 bool io_cqring_event_overflow(struct io_ring_ctx *ctx, u64 user_data, s32 res,
 			      u32 cflags, u64 extra1, u64 extra2);
 
-static inline unsigned int __io_cqring_events(struct io_ring_ctx *ctx)
-{
-	return ctx->cached_cq_tail - READ_ONCE(ctx->rings->cq.head);
-}
-
-/*
- * writes to the cq entry need to come after reading head; the
- * control dependency is enough as we're using WRITE_ONCE to
- * fill the cq entry
- */
-static inline struct io_uring_cqe *__io_get_cqe(struct io_ring_ctx *ctx)
-{
-	struct io_rings *rings = ctx->rings;
-	unsigned int off = ctx->cached_cq_tail & (ctx->cq_entries - 1);
-	unsigned int shift = 0;
-	unsigned int free, queued, len;
-
-	if (ctx->flags & IORING_SETUP_CQE32)
-		shift = 1;
-
-	/* userspace may cheat modifying the tail, be safe and do min */
-	queued = min(__io_cqring_events(ctx), ctx->cq_entries);
-	free = ctx->cq_entries - queued;
-	/* we need a contiguous range, limit based on the current array offset */
-	len = min(free, ctx->cq_entries - off);
-	if (!len)
-		return NULL;
-
-	ctx->cached_cq_tail++;
-	ctx->cqe_cached = &rings->cqes[off];
-	ctx->cqe_sentinel = ctx->cqe_cached + len;
-	ctx->cqe_cached++;
-	return &rings->cqes[off << shift];
-}
-
 static inline struct io_uring_cqe *io_get_cqe(struct io_ring_ctx *ctx)
 {
 	if (likely(ctx->cqe_cached < ctx->cqe_sentinel)) {
-- 
2.36.1