From: Jens Axboe <axboe@kernel.dk>
Date: Thu, 7 Nov 2019 17:57:36 +0000 (-0700)
Subject: io_uring: properly mark async work as bounded vs unbounded
X-Git-Tag: v5.5-rc1~204^2~22
X-Git-Url: https://asedeno.scripts.mit.edu/gitweb/?a=commitdiff_plain;h=5f8fd2d3e0a7aa7fc9d97226be24286edd289835;p=linux.git

io_uring: properly mark async work as bounded vs unbounded

Now that io-wq supports separating the two request lifetime types, mark
the following IO as having unbounded runtimes:

- Any read/write to a non-regular file
- Any specific networked IO
- Any poll command

Signed-off-by: Jens Axboe <axboe@kernel.dk>
---

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 831bea0fbc75..02a4f5e8a6e4 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -506,6 +506,20 @@ static inline bool io_prep_async_work(struct io_kiocb *req)
 		case IORING_OP_WRITEV:
 		case IORING_OP_WRITE_FIXED:
 			do_hashed = true;
+			/* fall-through */
+		case IORING_OP_READV:
+		case IORING_OP_READ_FIXED:
+		case IORING_OP_SENDMSG:
+		case IORING_OP_RECVMSG:
+		case IORING_OP_ACCEPT:
+		case IORING_OP_POLL_ADD:
+			/*
+			 * We know REQ_F_ISREG is not set on some of these
+			 * opcodes, but this enables us to keep the check in
+			 * just one place.
+			 */
+			if (!(req->flags & REQ_F_ISREG))
+				req->work.flags |= IO_WQ_WORK_UNBOUND;
 			break;
 		}
 		if (io_sqe_needs_user(req->submit.sqe))
@@ -3745,7 +3759,7 @@ static int io_sq_offload_start(struct io_ring_ctx *ctx,
 
 	/* Do QD, or 4 * CPUS, whatever is smallest */
 	concurrency = min(ctx->sq_entries, 4 * num_online_cpus());
-	ctx->io_wq = io_wq_create(concurrency, ctx->sqo_mm, NULL);
+	ctx->io_wq = io_wq_create(concurrency, ctx->sqo_mm, ctx->user);
 	if (IS_ERR(ctx->io_wq)) {
 		ret = PTR_ERR(ctx->io_wq);
 		ctx->io_wq = NULL;