[v2] io_uring/openclose: add support for IORING_OP_FIXED_FD_INSTALL

Commit Message

Jens Axboe Dec. 8, 2023, 10:18 p.m. UTC

io_uring can currently open/close regular files or fixed/direct
descriptors. Or you can instantiate a fixed descriptor from a regular
one, and then close the regular descriptor. But you currently can't turn
a purely fixed/direct descriptor into a regular file descriptor.

IORING_OP_FIXED_FD_INSTALL adds support for installing a direct
descriptor into the normal file table, just like receiving a file
descriptor or opening a new file would do. This is all nicely abstracted
into receive_fd(), and hence adding support for this is truly trivial.

Since direct descriptors are only usable within io_uring itself, it can
be useful to turn them into real file descriptors if they ever need to
be accessed via normal syscalls. This can either be a transitory thing,
or just a permanent transition for a given direct descriptor.

By default, new fds are installed with O_CLOEXEC set. The application
can disable O_CLOEXEC by setting IORING_FIXED_FD_NO_CLOEXEC in the
sqe->install_fd_flags member.

Suggested-by: Christian Brauner <[e-mail]>
Reviewed-by: Christian Brauner <[e-mail]>
Signed-off-by: Jens Axboe <[e-mail]>
---

v2:
	- add reviewed/suggested-by
	- default to O_CLOEXEC
	- pull in vfs.misc to get new receive_fd()

Also see expanded test case here:

https://git.kernel.dk/cgit/liburing/tree/test/fd-install.c

 include/uapi/linux/io_uring.h |  9 +++++++
 io_uring/opdef.c              |  9 +++++++
 io_uring/openclose.c          | 46 +++++++++++++++++++++++++++++++++++
 io_uring/openclose.h          |  3 +++
 4 files changed, 67 insertions(+)

Patch

diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
index f1c16f817742..db4b913e6b39 100644
--- a/include/uapi/linux/io_uring.h
+++ b/include/uapi/linux/io_uring.h
@@ -71,6 +71,7 @@  struct io_uring_sqe {
 		__u32		uring_cmd_flags;
 		__u32		waitid_flags;
 		__u32		futex_flags;
+		__u32		install_fd_flags;
 	};
 	__u64	user_data;	/* data to be passed back at completion time */
 	/* pack this to avoid bogus arm OABI complaints */
@@ -253,6 +254,7 @@  enum io_uring_op {
 	IORING_OP_FUTEX_WAIT,
 	IORING_OP_FUTEX_WAKE,
 	IORING_OP_FUTEX_WAITV,
+	IORING_OP_FIXED_FD_INSTALL,
 
 	/* this goes last, obviously */
 	IORING_OP_LAST,
@@ -386,6 +388,13 @@  enum {
 /* Pass through the flags from sqe->file_index to cqe->flags */
 #define IORING_MSG_RING_FLAGS_PASS	(1U << 1)
 
+/*
+ * IORING_OP_FIXED_FD_INSTALL flags (sqe->install_fd_flags)
+ *
+ * IORING_FIXED_FD_NO_CLOEXEC	Don't mark the fd as O_CLOEXEC
+ */
+#define IORING_FIXED_FD_NO_CLOEXEC	(1U << 0)
+
 /*
  * IO completion data structure (Completion Queue Entry)
  */
diff --git a/io_uring/opdef.c b/io_uring/opdef.c
index 799db44283c7..6705634e5f52 100644
--- a/io_uring/opdef.c
+++ b/io_uring/opdef.c
@@ -469,6 +469,12 @@  const struct io_issue_def io_issue_defs[] = {
 		.prep			= io_eopnotsupp_prep,
 #endif
 	},
+	[IORING_OP_FIXED_FD_INSTALL] = {
+		.needs_file		= 1,
+		.audit_skip		= 1,
+		.prep			= io_install_fixed_fd_prep,
+		.issue			= io_install_fixed_fd,
+	},
 };
 
 const struct io_cold_def io_cold_defs[] = {
@@ -704,6 +710,9 @@  const struct io_cold_def io_cold_defs[] = {
 	[IORING_OP_FUTEX_WAITV] = {
 		.name			= "FUTEX_WAITV",
 	},
+	[IORING_OP_FIXED_FD_INSTALL] = {
+		.name			= "FIXED_FD_INSTALL",
+	},
 };
 
 const char *io_uring_get_opcode(u8 opcode)
diff --git a/io_uring/openclose.c b/io_uring/openclose.c
index 74fc22461f48..20efb022563d 100644
--- a/io_uring/openclose.c
+++ b/io_uring/openclose.c
@@ -31,6 +31,11 @@  struct io_close {
 	u32				file_slot;
 };
 
+struct io_fixed_install {
+	struct file			*file;
+	unsigned int			o_flags;
+};
+
 static bool io_openat_force_async(struct io_open *open)
 {
 	/*
@@ -254,3 +259,44 @@  int io_close(struct io_kiocb *req, unsigned int issue_flags)
 	io_req_set_res(req, ret, 0);
 	return IOU_OK;
 }
+
+int io_install_fixed_fd_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_fixed_install *ifi;
+	unsigned int flags;
+
+	if (sqe->off || sqe->addr || sqe->len || sqe->buf_index ||
+	    sqe->splice_fd_in || sqe->addr3)
+		return -EINVAL;
+
+	/* must be a fixed file */
+	if (!(req->flags & REQ_F_FIXED_FILE))
+		return -EBADF;
+
+	ifi = io_kiocb_to_cmd(req, struct io_fixed_install);
+
+	flags = READ_ONCE(sqe->install_fd_flags);
+	if (flags & ~IORING_FIXED_FD_NO_CLOEXEC)
+		return -EINVAL;
+
+	/* default to O_CLOEXEC, disable if IORING_FIXED_FD_NO_CLOEXEC is set */
+	if (flags & IORING_FIXED_FD_NO_CLOEXEC)
+		ifi->o_flags = 0;
+	else
+		ifi->o_flags = O_CLOEXEC;
+
+	return 0;
+}
+
+int io_install_fixed_fd(struct io_kiocb *req, unsigned int issue_flags)
+{
+	struct io_fixed_install *ifi;
+	int ret;
+
+	ifi = io_kiocb_to_cmd(req, struct io_fixed_install);
+	ret = receive_fd(req->file, NULL, ifi->o_flags);
+	if (ret < 0)
+		req_set_fail(req);
+	io_req_set_res(req, ret, 0);
+	return IOU_OK;
+}
diff --git a/io_uring/openclose.h b/io_uring/openclose.h
index 4b1c28d3a66c..8a93c98ad0ad 100644
--- a/io_uring/openclose.h
+++ b/io_uring/openclose.h
@@ -12,3 +12,6 @@  int io_openat2(struct io_kiocb *req, unsigned int issue_flags);
 
 int io_close_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe);
 int io_close(struct io_kiocb *req, unsigned int issue_flags);
+
+int io_install_fixed_fd_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe);
+int io_install_fixed_fd(struct io_kiocb *req, unsigned int issue_flags);