From a7c41b4687f5902af70cd559806990930c8a307b Mon Sep 17 00:00:00 2001
From: Xiaoguang Wang <xiaoguang.wang@linux.alibaba.com>
Date: Mon, 30 May 2022 21:15:20 +0800
Subject: [PATCH] io_uring: let IORING_OP_FILES_UPDATE support choosing fixed
file slots
Git-commit: a7c41b4687f5902af70cd559806990930c8a307b
Patch-mainline: v5.19-rc1
References: bsc#1205205
One big issue with the file registration feature is that it needs user
space apps to maintain free slot info about io_uring's fixed file table,
which really is a burden for development. io_uring now supports choosing
free file slot for user space apps by using IORING_FILE_INDEX_ALLOC flag
in accept, open, and socket operations, but they need the app to use
direct accept or direct open, which not all apps are prepared to use yet.
To support apps that still need real fds, make use of the registration
feature easier. Let IORING_OP_FILES_UPDATE support choosing fixed file
slots, which will store picked fixed files slots in fd array and let cqe
return the number of slots allocated.
Suggested-by: Hao Xu <howeyxu@tencent.com>
Signed-off-by: Xiaoguang Wang <xiaoguang.wang@linux.alibaba.com>
[axboe: move flag to uapi io_uring header, change goto to break, init]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Gabriel Krisman Bertazi <krisman@suse.de>
---
fs/io_uring.c | 72 ++++++++++++++++++++++++++++++++++++------
include/uapi/linux/io_uring.h | 6 +++
2 files changed, 68 insertions(+), 10 deletions(-)
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -571,6 +571,7 @@ struct io_close {
struct file *file;
int fd;
u32 file_slot;
+ u32 flags;
};
struct io_timeout_data {
@@ -1304,7 +1305,9 @@ static int io_req_prep_async(struct io_k
static int io_install_fixed_file(struct io_kiocb *req, struct file *file,
unsigned int issue_flags, u32 slot_index);
-static int io_close_fixed(struct io_kiocb *req, unsigned int issue_flags);
+static int __io_close_fixed(struct io_kiocb *req, unsigned int issue_flags,
+ unsigned int offset);
+static inline int io_close_fixed(struct io_kiocb *req, unsigned int issue_flags);
static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer);
static void io_eventfd_signal(struct io_ring_ctx *ctx);
@@ -5641,14 +5644,18 @@ static int io_statx(struct io_kiocb *req
static int io_close_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
{
- if (sqe->off || sqe->addr || sqe->len || sqe->rw_flags || sqe->buf_index)
+ if (sqe->off || sqe->addr || sqe->len || sqe->buf_index)
return -EINVAL;
if (req->flags & REQ_F_FIXED_FILE)
return -EBADF;
req->close.fd = READ_ONCE(sqe->fd);
req->close.file_slot = READ_ONCE(sqe->file_index);
- if (req->close.file_slot && req->close.fd)
+ req->close.flags = READ_ONCE(sqe->close_flags);
+ if (req->close.flags & ~IORING_CLOSE_FD_AND_FILE_SLOT)
+ return -EINVAL;
+ if (!(req->close.flags & IORING_CLOSE_FD_AND_FILE_SLOT) &&
+ req->close.file_slot && req->close.fd)
return -EINVAL;
return 0;
@@ -5664,7 +5671,8 @@ static int io_close(struct io_kiocb *req
if (req->close.file_slot) {
ret = io_close_fixed(req, issue_flags);
- goto err;
+ if (ret || !(req->close.flags & IORING_CLOSE_FD_AND_FILE_SLOT))
+ goto err;
}
spin_lock(&files->file_lock);
@@ -7633,6 +7641,41 @@ static int io_files_update_prep(struct i
return 0;
}
+static int io_files_update_with_index_alloc(struct io_kiocb *req,
+ unsigned int issue_flags)
+{
+ __s32 __user *fds = u64_to_user_ptr(req->rsrc_update.arg);
+ unsigned int done;
+ struct file *file;
+ int ret, fd;
+
+ for (done = 0; done < req->rsrc_update.nr_args; done++) {
+ if (copy_from_user(&fd, &fds[done], sizeof(fd))) {
+ ret = -EFAULT;
+ break;
+ }
+
+ file = fget(fd);
+ if (!file) {
+ ret = -EBADF;
+ break;
+ }
+ ret = io_fixed_fd_install(req, issue_flags, file,
+ IORING_FILE_INDEX_ALLOC);
+ if (ret < 0)
+ break;
+ if (copy_to_user(&fds[done], &ret, sizeof(ret))) {
+ ret = -EFAULT;
+ __io_close_fixed(req, issue_flags, ret);
+ break;
+ }
+ }
+
+ if (done)
+ return done;
+ return ret;
+}
+
static int io_files_update(struct io_kiocb *req, unsigned int issue_flags)
{
struct io_ring_ctx *ctx = req->ctx;
@@ -7646,10 +7689,14 @@ static int io_files_update(struct io_kio
up.resv = 0;
up.resv2 = 0;
- io_ring_submit_lock(ctx, issue_flags);
- ret = __io_register_rsrc_update(ctx, IORING_RSRC_FILE,
- &up, req->rsrc_update.nr_args);
- io_ring_submit_unlock(ctx, issue_flags);
+ if (req->rsrc_update.offset == IORING_FILE_INDEX_ALLOC) {
+ ret = io_files_update_with_index_alloc(req, issue_flags);
+ } else {
+ io_ring_submit_lock(ctx, issue_flags);
+ ret = __io_register_rsrc_update(ctx, IORING_RSRC_FILE,
+ &up, req->rsrc_update.nr_args);
+ io_ring_submit_unlock(ctx, issue_flags);
+ }
if (ret < 0)
req_set_fail(req);
@@ -9765,9 +9812,9 @@ err:
return ret;
}
-static int io_close_fixed(struct io_kiocb *req, unsigned int issue_flags)
+static int __io_close_fixed(struct io_kiocb *req, unsigned int issue_flags,
+ unsigned int offset)
{
- unsigned int offset = req->close.file_slot - 1;
struct io_ring_ctx *ctx = req->ctx;
struct io_fixed_file *file_slot;
struct file *file;
@@ -9804,6 +9851,11 @@ out:
return ret;
}
+static inline int io_close_fixed(struct io_kiocb *req, unsigned int issue_flags)
+{
+ return __io_close_fixed(req, issue_flags, req->close.file_slot - 1);
+}
+
static int __io_sqe_files_update(struct io_ring_ctx *ctx,
struct io_uring_rsrc_update2 *up,
unsigned nr_args)
--- a/include/uapi/linux/io_uring.h
+++ b/include/uapi/linux/io_uring.h
@@ -46,6 +46,7 @@ struct io_uring_sqe {
__u32 rename_flags;
__u32 unlink_flags;
__u32 hardlink_flags;
+ __u32 close_flags;
};
__u64 user_data; /* data to be passed back at completion time */
/* pack this to avoid bogus arm OABI complaints */
@@ -240,6 +241,11 @@ enum io_uring_op {
#define IORING_ACCEPT_MULTISHOT (1U << 0)
/*
+ * close flags, store in sqe->close_flags
+ */
+#define IORING_CLOSE_FD_AND_FILE_SLOT (1U << 0)
+
+/*
* IO completion data structure (Completion Queue Entry)
*/
struct io_uring_cqe {